Products of Secondary Metabolism:wiley Encyclopedia

Biotechnology
Second Edition Volume 7
Products of Secondary Metabolism
VCH
A Wiley company
4b
Biotechnology
Second Edition
Fundamentals
Volume 1 Biological Fundamentals Volume 2 Genetic Fundamentals and Genetic Engineering Volume 3 Bioprocessing Volume 4 Measuring, Modelling, and Control
Special Topics
Volume 9 Enzymes, Biomass, Food and Feed Volume 10 Special Processes Volumes l l a and b Environmental Processes Volume 12 Legal, Economic and Ethical Dimensions
Products
Volume 5 Recombinant Proteins, Monoclonal Antibodies, and Therapeutic Genes Volume 6 Products of Primary Metabolism Volume 7 Products of Secondary Metabolism Volume 8 Biotransformations
A Multi-Volume Comprehensive Treatise
Second, Completely Revised Edition Edited by H.-J. Rehm and G. Reed in cooperation with A. Piihler and P. Stadler
Volume 7
Biotechnology
Products of Secondary Metabolism

Edited by H. Kleinkauf and H. von Dohren
VCH
A Wiley company
4b
Series Editors: Prof. Dr. H.-J. R eh m Institut f u r Mikrobiologie Universitat Munster CorrensstraBe 3 D-48149 Munster
Dr. G. R e e d 1914 N. Prospect Ave. #61 Milwaukee, WI 53202-1401

USA Prof. Dr. P. J. W. Stadler Bayer AG Verfahrensentwicklung Biochemie Leitung Friedrich-Ebert-StraBe 217 D-42096 Wuppertal
FRG
Prof. Dr. A. Piihler Biologie VI (Genetik) Universitat Bielefeld
Volume Editors: Prof. Dr. H. Kleinkauf Dr. H. von D o h r en Institut f u r Biochemie Technische Universitat Franklin-StraBe 29 A-10587 Berlin Germany
P.O. Box 100131 D-33501 Bielefeld FRG
FRG
This book was carefully produced. Nevertheless, authors, editors and publisher do not warrant the information contained therein to be free of errors. Readers are advised to keep in mind that statements, data, illustrations, procedural details or other items may inadvertently be inaccurate.
Executive Editor: Dr. Hans-Joachim Kraus Editorial Director: Karin Dembowsky Production Manager: Hans-Jochen Schmitt Library of Congress Card No.: applied for British Library Cataloguing-in-Publication Data: A catalogue record for this book is available from the British Library
Die Deutsche Bibliothek - CIP-Einheitsaufnahme Biotechnology : a multi volume comprehensive treatise I ed. by H.-J. Rehm and G. Reed. In cooperation with A. Piihler and P. Stadler. 2., completely rev. ed. -VCH. ISBN 3-527-28310-2 (Weinheim ...) NE: Rehm, Hans J. [Hrsg.] Vol. 7. Products of secondary metabolism I ed. by H. Kleinkauf and H. von Dohren - 1997 ISBN 3-S27-28317-X OVCH Verlagsgesellschaft mbH, D-69451 Weinheim (Federal Republic of Germany), 1997 Printed on acid-free and chlorine-free paper.
All rights reserved (including those of translation into other languages). No part of this book may be reproduced in any form -by photoprinting, microfilm, or any other means-nor transmitted or translated into a machine language without written permission from the publishers. Registered names, trademarks, etc. used in this book, even when not specifically marked as such, are not to be considered unprotected by law. Composition and Printing: Zechnersche Buchdruckerei, D-67330 Speyer. Bookbinding: J. SchSiffer, D-67269 Griinstadt. Printed in the Federal Republic of Germany
Preface
In recognition of the enormous advances in biotechnology in recent years, we are pleased to present this Second Edition of Biotechnology relatively soon after the introduction of the First Edition of this multi-volume comprehensive treatise. Since this series was extremely well accepted by the scientific community, we have maintained the overall goal of creating a number of volumes, each devoted to a certain topic, which provide scientists in academia, industry, and public institutions with a well-balanced and comprehensive overview of this growing field. We have fully revised the Second Edition and expanded it from ten to twelve volumes in order to take all recent developments into account. These twelve volumes are organized into three sections. The first four volumes consider the fundamentals of biotechnology from biological, biochemical, molecular biological, and chemical engineering perspectives. The f next four volumes are devoted to products o industrial relevance. Special attention is given here to products derived from genetically engineered microorganisms and mammalian cells. The last four volumes are dedicated to the description of special topics. The new Biotechnology is a reference work, a comprehensive description of the state-of-the-art, and a guide to the original literature. It is specifically directed to microbiologists, biochemists, molecular biologists, bioengineers, chemical engineers, and food and pharmaceutical chemists working in industry, at universities or at public institutions. A carefully selected and distinguished Scientific Advisory Board stands behind the
series. Its members come from key institutions representing scientific input from about twenty countries. The volume editors and the authors of the individual chapters have been chosen for their recognized expertise and their contributions to the various fields of biotechnology. Their willingness to impart this knowledge to their colleagues forms the basis of Biotechnology and is gratefully acknowledged. Moreover, this work could not have been brought to fruition without the foresight and the constant and diligent support of the publisher. We are grateful to VCH for publishing Biotechnology with their customary excellence. Special thanks are due to Dr. HansJoachim Kraus and Karin Dembowsky, without whose constant efforts the series could not be published. Finally, the editors wish to thank the members of the Scientific Advisory Board for their encouragement, their helpful suggestions, and their constructive criticism. H.-J. Rehm G. Reed A. Puhler P. Stadler
Scientific Advisory Board
Pro$ Dr. M. J. Beker

August Kirchenstein Institute of Microbiology Latvian Academy of Sciences Riga, Latvia
Pro$ Dr. T. K. Ghose

Biochemical Engineering Research Centre Indian Institute of Technology New Delhi, India
Pro$ Dr. J. D. BuLock

Weizmann Microbial Chemistry Laboratory Department of Chemistry University of Manchester Manchester, UK
Pro$ Dr. I. Goldberg

Department of Applied Microbiology The Hebrew University Jerusalem, Israel
Pro$ Dr. C. L. Cooney

Department of Chemical Engineering Massachusetts Institute of Technology Cambridge, MA, USA
Pro$ Dr. G. Goma

Departement de GCnie Biochimique et Alimentaire Institut National des Sciences AppliquCes Toulouse, France
Pro$ Dr. H. W. Doelle

Department of Microbiology University of Queensland St. Lucia, Australia
Sir D. A. Hopwood
Department of Genetics John Innes Institute Norwich, UK
Prof Dr. J. Drews

F. Hoffmann-La Roche AG Basel, Switzerland
Pro$ Dr. E. H. Houwink

Organon International bv Scientific Development Group Oss. The Netherlands
Pro$ Dr. A. Fiechter

Institut fur Biotechnologie Eidgenossische Technische Hochschule Zurich, Switzerland
Pro$ Dr. A. E. Humphrey

Center for Molecular Bioscience and Biotechnology Lehigh University Bethlehem, PA, USA
VIII
Scientific Advisory Board
Prof. Dr. I. Karube

Research Center for Advanced Science and Technology University of Tokyo Tokyo, Japan
Prof. Dr. K. Schiigerl

Institut fur Technische Chemie Universitat Hannover Hannover, Germany
Prof. Dr. M . A. Lachance

Department of Plant Sciences University of Western Ontario London, Ontario, Canada
Prof. Dr. P. Sensi

Chair of Fermentation Chemistry and Industrial Microbiology Lepetit Research Center Gerenzano, Italy
Prof. Dr. Y. Liu

China National Center for Biotechnology Development Beijing, China
Prof. Dr. Y. H. Tan

Institute of Molecular and Cell Biology National University of Singapore Singapore
Prof. Dr. J. F. Martin

Department of Microbiology University of Leon Leon, Spain
Prof. Dr. D. Thomas

Laboratoire de Technologie Enzymatique UniversitC de Compibgne Compibgne, France
Prof. Dr. B. Mattiasson

Department of Biotechnology Chemical Center University of Lund Lund, Sweden
Prof. Dr. W . Verstraete

Laboratory of Microbial Ecology Rijksuniversiteit Gent Gent, Belgium
Prof. Dr. M . Roehr

Institut fur Biochemische Technologie und Mikrobiologie Technische Universitat Wien Wien, Austria
Prof. Dr. E.-L. Winnacker

Institut fur Biochemie Universitat Munchen Munchen, Germany
Prof. Dr. H. Sahm

Institut fur Biotechnologie Forschungszentrum Julich Julich, Germany
Contents
H. von Dohren, H. Kleinkauf

General Aspects of Secondary Metabolism 1 H. von Dohren, U. Grafe Regulation of Bacterial Antibiotic Production 57 K. Chater, M. Bibb Screening of Novel Receptor-Active Compounds of Microbial Origin 107 H. Tanaka, S. Omura Microbial Lipids 133 C. Ratledge Microbial Siderophores 199 G. Winkelmann, H. Drechsel Advances in the Molecular Genetics of PLactam Antibiotic Biosynthesis 247 P. 15.Skacrud, T. Schwecke, H. v. Liempt, M. B. Tobin Peptide Antibiotics 277 H. Kleinkauj H. von Dohren Lantibiotics 323 R. Jack, F. Gotz, G. Jung
Introduction
9 Glycopeptide Antibiotics (Dalbaheptides) 369 G. Lancini, B. Cavalleri 10 Aminoglycosides and Sugar Components in Other Secondary Metabolites 397 W. Piepersberg, J. Distler 11 Products from Basidiomycetes 489 G. Erkel, T. Anke 12 Cyclosporins: Recent Developments in Biosynthesis, Pharmacology and Biology, and Clinical Applications 535 J. Kallen, V. Mikol, V. F. J. Quesniaux, M. D. Walkinshaw,E. Schneider-Scherzer, K. Schorgendorfer, G. Weber, H. Fliri 13 Secondary Products from Plant Cell Cultures 593 J. Berlin 14 Biotechnical Drugs as Antitumor Agents 641 U. Grafe, K . Dornberger, H.-P. Saluz
Index 707
Contributors
Prof. Dr. Timm Anke

Lehrbereich Biotechnologie Universitat Kaiserslautern Postfach 3049 D-67618 Kaiserslautern Germany Chapter I1
Prof. Keith Chater

John Innes Centre Norwich Research Park Colney Lane Colney, Norwich NR4 7UH UK Chapter 2
Dr. Jochen Berlin

Gesellschaft fur Biotechnologische Forschung Mascheroder Weg 1 D-38124 Braunschweig Germany Chapter 13
Jurgen Distler
Bergische Universitat G H Mikrobiologie - FB 19 Gauss-StraSe 20 D-42097 Wuppertal Germany Chapter 10
Dr. Mervin Bibb

John Innes Centre Norwich Research Park Colney Lane Colney, Norwich NR4 7UH UK Chapter 2
Dr. Hans von Dohren

Institut fur Biochemie Technische Universitat Franklin-Str. 29 D-10587 Berlin Germany Chapters I, 7
Dr. Bruno Cavalleri

MMDRI - Lepetit Research Center Via R. Lepetit, 34 1-21040 Gerenzano (Varese) Italy Chapter 9
Dr. Klausjiirgen Dornberger

Hans-Knoll-Institut fur Naturstoff-Forschung Bereich Naturstoffchemie BeutenbergstraBe 11 D-07745 Jena Germany Chapter 14
XI1
Contributors
Dr. Hartmut Drechsel

Mikrobiologie und Biotechnologie Universitat Tubingen Auf der Morgenstelle 1 D-72076 Tubingen Germany Chapter 5
Prof. Dr. Gunter Jung

Universitat Tubingen Institut fur Organische Chemie Auf der Morgenstelle 18 D-72076 Tubingen Germany Chapter 8
Dr. Gerhard Erkel

Lehrbereich Biotechnologie Universitat Kaiserslautern Postfach 3049 D-67618 Kaiserslautern Germany Chapter I1
Dr. Jorg Kallen

Sandoz Pharma Ltd. Preclinical Research CH-4002 Basel Switzerland Chapter I2
Dr. Hans Fliri

RhBne Poulonc Rorer S.A. Centre de Recherche de Vitry-Alfortville 13, quai Jules Guesde F-94403 Vitry-sur-Seine Cedex France Chapter 12
Prof. Dr. Horst Kleinkauf

Institut fur Biochemie Technische Universitat Franklin-Str. 29 D-10587 Berlin Germany Chapter 7
Dr. Friedrich Gotz

Lehrstuhl fur Mikrobielle Genetik Universitat Tubingen Auf der Morgenstelle 18 D-72076 Tubingen Germany Chapter 8
Dr. Giancarlo Lancini

MMDRI - Lepetit Research Center Via R. Lepetit, 34 1-21040 Gerenzano (Varese) Italy Chapter 9
Prof. Dr. Udo Grafe

Hans-Knoll-Institut fur Naturstoff-Forschung Bereich Naturstoffchemie BeutenbergstraSe 11 D-07745 Jena Germany Chapters I, 14
Dr. Henk van Liempt

DRL, BT-FDG SudstraSe 125 D-53175 Bonn Germany Chapter 6
Dr. Ralph Jack

Universitat Tubingen Institut fur Organische Chemie Auf der Morgenstelle 18 D-72076 Tubingen Germany Chapter 8
Dr. Vincent Mikol

Sandoz Pharma Ltd. Preclinical Research CH-4002 Basel Switzerland Chapter 12
Contributors
XI11
Prof. Dr. Satoshi Omura

School of Pharmaceutical Sciences Kitasato University The Kitasato Institute 9-1, Shirokane 5-chome Minato-ku, Tokyo 108 Japan Chapter 3
Dr. Kurt Schorgendorfer

Biochemie GmbH A-6330 Kufstein-Schaftenau Austria Chapter 12
Prof. Dr. Wolfgang Piepersberg

Bergische Universitat G H Mikrobiologie - FB 19 Gauss-Strafie 20 D-42097 Wuppertal Germany Chapter 10
Dr. Torsten Schwecke

Institute of Biochemistry University of Cambridge Tennis Court Road Cambridge, CB2 1QW UK Chapter 6
Dr. Valkrie F.J. Quesniaux

Dr. Paul L. Skatrud

Infectious Diseases Research Eli Lilly and Company Lilly Corporate Center Indianapolis, IN 46285 USA Chapter 6
Prof. Colin Ratledge

The University of Hull Department of Applied Biology Hull HU6 7RX UK Chapter 4
Dr. Haruo Tanaka

School of Pharmaceutical Sciences Kitasato University The Kitasato Institute Minato-ku, Tokyo 108 Japan Chapter 3
Prof. Dr. habil. Hans-Peter Saluz

Hans-Knoll-Institut fur Naturstoff-Forschung Bereich Naturstoffchemie BeutenbergstraSe 11 D-07745 Jena Germany Chapter 14
Dr. Matthew B. Tobin

Infectious Diseases Research Eli Lilly and Company Lilly Corporate Center Indianapolis, IN 46285 USA Chapter 6
Dr. Elisabeth Schneider-Scherzer

Biochemie GmbH A-6330 Kufstein-Schaftenau Austria Chapter 12
Dr. Malcolm D. Walkinshaw

XIV
Contributors
Dr. Gerhard Weber

Biochemie GmbH A4330 Kufstein-Schaftenau Austria Chapter 12
Prof. Dr. Giinther Winkelmann

Mikrobiologie und Biotechnologie Universitat Tubingen Auf der Morgenstelle 1 D-72076 Tubingen Germany Chapter 5
Biotechnology
Edited by H.-J. Rehm and G. Reed OVCH Verlagsgesellschaft mbH, 1997
This volumes provides an overview of secondary metabolites illustrating most aspects of their discovery, formation, exploitation, and production. Compared to the first edition the focus when has clearly shifted towards the molecular genetic background of the producing organisms. These efforts serve not only our understanding of the production processes to permit improvements by genetic manipulations, but also promote our appreciation of the environmental significance of secondary metabolites. The term secondary metabolite has been discussed widely, and a shift in perception took place in the last years. From a playground of nature leading to mostly disparable products ideas focus now on special purpose products promoting evolutionary advantages. This shift is connected to the impressive elucidation of the genetics of multistep synthetic processes of secondary metabolite formation. Genes encoding biosynthetic reaction sequences have been found clustered together with resistance or export genes and are under the control of specific signals. Biosynthetic functions or unit operations reside on modules, and these modules in their functional protein state interact to assure the fidelity of the multistep processes. The genetic burden for many of these processes seems remarkable, and genes assembled from modules often display sizes of 10 to more than 45 kilobases. Since some of the now established microbial genomes are devoid of such multistep pathways, their unique placement in other genomes indicates important functions for their producers.
Still largely unconnected to the background of their producers secondary metabolites generally are high-value compounds established mainly in pharmacology, veterinary medicine, agriculture, and biochemical and medical research. The introductory chapter points to product fields and to the genetic investigation of biosynthetic unit operations. Regulatory mechanisms are then considered in the most advanced fields of the prokaryotes. As the central field of present drug discovery approaches target-based screenings are discussed. Compound groups considered are lipids siderophores, aminoglycosides, and peptides (p-lactams, dalbaheptides, cyclosporins, lantibiotics). Producer groups presented are basidiomycetes and plant cells. As a target group antitumor drugs are evaluated. An updated chapter on macrolides as secondary metabolites including reprogramming strategies will be included in Volume 10 of the Second Edition of Biofechnofogy(see also Volume 4 of the First Edition). Further chapters to be consulted are especially on biopolymers and surfactants (Volume 6), on the overproduction of metabolites and the treatment of producer organisms like bacilli, streptomycetes and filamentous fungi (Volume 1) as well as on reactor modeling (Volume 3). We thank our colleagues for their valuable contributions, the publisher for their patience and cooperativity, and the series editors for many helpful suggestions. Berlin, March 1997 Hans von Dohren Horst Kleinkauf
Biotechnology
1 General Aspects of Secondary Metabolism
HANS VON DOHREN

Berlin, Germany
UDOGRAFE
Jena, Germany
1 Introduction: The Importance of Secondary Metabolites as Drugs 2 2 Secondary Metabolism, an Expression of Cellular and Organismic Individuality 11 2.1 Roles of Secondary Metabolites in Producing Organisms 11 2.2 Regulation of Microbial Secondary Metabolism 17 2.2.1 Genetic Organization of Product Formation 17 2.2.2 Regulatory Mechanisms 23 2.2.3 Genetic Instability 26 2.2.4 Developmental Processes 27 2.2.5 The A-Factor and the Signal Cascade of Cytodifferentiation in Streptomyces 27 2.2.6 Overproduction of Microbial Secondary Metabolites and Precursor Pools 29 2.2.7 Biotechnical Production of Secondary Metabolites 31 3 The Biosynthetic Pathways 31 3.1 Precursors and the Main Biosynthetic Pathways 31 3.2 Secondary Metabolites Formed through Biosynthetic Modifications of a Single Precursor 31 3.3 Polyketides 32 3.4 Terpenes 35 3.5 Sugar-Derived Oligomeric Structures 35 3.6 Oligo- and Polypeptides 36 3.7 Biosynthetic Modifications of Structures and Precursor-Directed Biosyntheses 37 4 Variability of Structures of Secondary Metabolites 38 4.1 Secondary Metabolites as Products of Biological Unit Operations 38 4.2 Structural Classifications of Secondary Metabolites 38 5 Future Perspectives: New Products of the Secondary Metabolism 40 6 References 41
crease in resistant nosocomial and opportunistic pathogens (particularly dangerous to immunosuppressed AIDS and tumor patients) requires both improvement of known drugs and search for new drugs (GRAFE, 1992; LANCINI LORENZETTI, and 1993; HUTCHINSON, 1994). Microbial products such as doxorubicin, Today, bioactive secondary metabolites of microorganisms and of plants, and their syn- bleomycin, and mitomycin C are indispensathetic derivatives as well, are among the most ble as cancerostatics (Fox, 1991). The same is frequently used therapeutics in human and true for plant metabolites such as the vinca alveterinary medicine (Scrip, 1993). The inven- kaloids, taxol, and their chemical derivatives tion of antibiotic therapy contributed greatly which exert excellent antitumor activity by into the successful control of most of the epi- teraction with the cellular mitotic system and demic infectious diseases and even promoted (NOBLE, 1990 Fox, 1991; HEINSTEIN 1994 POTIERet al., 1994). their disappearance. Moreover, it contributed CHANG, However, even the non-therapeutic fields to the general increase in the lifespan of man, not only in industrialized countries. New ap- of application, such as in animal husbandry plications for bioactive biotechnical products and plant protection, contributed to a high in medical care like their use as immunosup- degree to the continuing interest in secondary pressants or antiatherosclerotics, and as ani- metabolite production. Last but not least, natmal growth promoters and pesticides in agri- ural products of biotechnical and agricultural culture rendered research on new secondary origin play an important role as biochemical metabolites an apparently endless story (SAN- tools in molecular biology and in the investiGLIER and LARPENT, 1989; ComitC Editorial, gation of cellular functions. 1992; LANCINI LORENZE-ITI, and 1993; VINMore than loo00 antibiotics and similar bioactive secondary metabolites have been ING and STUTTARD, 1995). isolated so far from microbes, and a comparaIn the past, natural products supplied 5 4 % of the annual increase in the worlds total bly higher number of drugs was derived from pharmaceutical market. The list of the 25 plants and even from animals (see, e.g., maworldwide best-selling drugs for application rine tunicates, molluscs, toxic insects, snakes, in humans in 1992 includes a series of drugs and toads) (BERDYet al., 1980; LAATSCH, of microbial origin which are used either in 1994). Approximately 500 new representatheir native structures or as chemical derivatives of low-molecular weight compounds are tives (see, e.g., Mevacor, Cefaclor or other published every year. In addition to this huge and still growing cephalosporins, Augmentin, Sandimmun) number of bioactive molecules, more than (Scrip, 1993). Many plant products, from digitalis glyco- 1OOOOO derivatives as representatives of some sides and neuroactive alkaloids to the pyre- few basic structures (e. g., p-lactams, macrothrines, serve as therapeutics for human dis- lides, aminoglycosides, tetracyclines, anthraeases and as agricultural agents (Comitk Edi- cyclines) were obtained by means of synthetic 1994). Irrespective torial, 1992). Sometimes, the experiences derivatizations (LAATSCH, made in folk medicine initiated the discovery of this plethora of drug molecules a little of new plant-derived antitumor drugs, anti- more than a hundred basic structures gained neuralgic, antihypertonic, antidepressant, in- practical importance. We owe much progress in the detection of secticidal, nematicidal, and other bioactive new drug structures to modern physicocompounds. Antiinfective chemotherapy once was the chemical approaches such as mass-spectromeclassical domain of biotechnical drug produc- try, high-field nuclear magnetic resonance tion due to the discovery of p-lactam antibiot- spectroscopy and X-ray diffractometry. Comics, such as penicillins, cephalosporins, clavu- pilations of the numerous structural data lanates, and carbapenems. Even today, the in- (BERDY et al., 1980; BYCROFT, 1988;
1 Introduction: The Importance of Secondary Metabolites as Drugs
I Introduction: The Importance of Secondary Metabolites as Drugs
LAATSCH, 1994) provide indispensable assistance in the identification of new drug molecules. Thus, the enormous number of already known metabolites from microbes and plants increased the detection and isolation of already known structures dramatically. A compilation of about 200 recently described products illustrates the current trends in screening efforts (Tab. 1). These have been published during the last two years. It is evident from these data that highly selective screens prevail and yet the majority of compounds originate from the classical Actinomycete pool. Rare bacteria and fungi, marine microorganisms and plants now have a significant share. It is obvious that well-known organisms again contribute with newly isolated substances to new, e. g., receptor targeted screens. Strategies of such screens are discussed in this volume in Chapter 3 by TANAKA and OMURA. The development of new drugs from natural sources is common practice of the pharmaceutical industry. 6000 to loo00 chemicals have to be tested in a given assay system to obtain one single compound suitable as a therapeutical agent (OMURA,1992; KROHN et al., 1993). No wonder that research and development for a new approved drug may cost up to one billion US$. In most cases, a new natural leading structure is intensively modified by chemical means to improve its activity and to reduce side effects. Chemistry is also extremely helpful if rather rare natural products occurring in low amounts or in organisms from sensitive ecological areas have been proposed as drugs. For example, 40000 yew trees, i. e., the whole population of Northern America, would be required to produce 25 kg of taxol, a new promising cancerostatic drug, and even this amount would not be sufficient to treat every cancer patient. Fortunately, taxol derivatives of similar activity (taxotere) can be obtained by chemical derivatization of taxoid metabolites which are obtainable in large quantities from the dried and leaves of European yews (HEINSTEIN CHANG,1994). Alternatively, cell cultures (ELLISet al., 1996) or endophytic fungi such as Pestalotiopsis microspora (STIERLEet al., 1994, 1995; STROBEL al., 1996) of Taxus et species could be exploited for production.
From the recently completed chemical synthesis of taxol it is evident that, as in bicyclic plactams, classical approaches cannot compete with natural producers. Instead, increasing attention is given to the recruitment of biocatalysts for certain key reactions in metabolite production. In addition, directed biosynthesis in microbial cultures (THIERICKE and ROHR, 1993), production of plant products in cell cultures (BERLIN, Chapter 14, this volume), and cell free in vitro systems of enzymatic synthesis and peptide and protein producing translation systems are considered as complementary methods in structure-function studies (ALAKHOV and VON DOHREN, unpublished data). Only 30% of the total developmental efforts have been spent to the search for a new drug. However, for the estimation of its efficacy and evaluation of safety often more than 50% are needed. Taking into account a quota of approximately 1:15000 for a hit structure, the challenges of modern pharmaceutical development become visible. In general, natural products seem to offer greater chances than synthetically derived agents. Hence, a great research potential is still dedicated to the discovery of new natural drugs and their biotechnical production. Classical strategies of drug development are being more and more supplemented by new biomedical approaches and ideas and by the use of genetically engineered microbes and cells as screening organisms (TOMODA OMURA, and 1990; ELDERet al., 1993). These tools initiated a renaissance in the search for new leading structures. New sources of bioactive material, such as marine organisms, and new microbes from ecological niches promoted the recent advances in the discovery of drugs (WILLIAMS and VICKERS 1986; RINEHART and SHIELD 1988; MONAGHAN TKACZ,1990; JACOB and and ZASLOFF,1994; JENSENand FENICAL, 1994) (Tab. 1). Present research activities were also stimulated by the discovery of block busters (Scrip, 1993) such as cyclosporin A (KAHAN, 1987), avermectins (CAMPBELL,1989), acarbose (MULLER, 1989), and monacolin (ENDO, 1979) in microbial cultures. A series of very promising new screening drugs (zaragozic acid, squalestatins) (HASUMI, 1993), erbstatin
I General Aspects of Secondary Metabolism
Tab. 1 Selected Natural Products Detected by Screening Efforts Published in 1995196 .
Compound Reference
Producing Organisms
Structural Type'
Selected Properties
Research Group Involved
Antimicrobial Drugs: Actinomycete Griseusin (unidentified) derivatives Streptomyces BE-24566B violaceus-niger Amicenomycin Streptomyces sp.
PK PK PK-GLYC PK, mod. PEP acyl AA PEP PK PEP PK PEP PK, mod. PK PK PEP + PK PK PK, mod. PK PK PK PEP ALK PEP
antibacterial antibacterial antibacterial antibacterial, MDR strains antibacterial antibacterial antibacterial antibacterial antibacterial antifungal antibacterial antibacterial antilegionella antibacterial antibacterial antibacterial antibacterial antibacterial antibacterial antibacterial antibacterial antibacterial antibacterial antibacterial antibacterial, cytotoxic antibacterial antibacterial antibacterial antibacterial
Institute of Microbial Chemistry Banyu Pharm. Co. Institute of Microbial Chemistry Yamanouchi Pharm. Co. and PT Kalbe Pharma Lepetit Institute of Microbial Chemistry Lepetit Hoechst Hans Knoll Institute and Univ. Tiibingen Bristol Myers Squibb Lepetit Cheil Foods & Chem. Inc and NIH Korea GBF GBF Sankyo Univ. Tiibingen and Hans Knoll Institute Lederle Institute of Microbial Chemistry RIKEN Institute of Microbial Chemistry Institute of Microbial Chemistry Institute of Microbial Chemistry Univ. Alcala Hans Knoll Institute Univ. Tlibingen, Univ. Gattingen, Hans Knoll Institute Univ. Lund and Univ. Kaiserslautern Panlabs Bristol Myers Squibb
Kalimantacins A21459 Epoxyquinomycins GE 37468 Phencomycin Chrysoapermin Bacillaene GE2270 AL072 Ripostatin Sorangiolid Thiomarinol Echinoserine 07F275 Pyralomycins RS-22 Ochracenomycins Azicemycins Amythiamycin APHE
31~
Alcaligenes sp. Actinoplanes sp. Amycolatopsis Streptomyces sp. Streptomyces sp. Apiocrea chrysosperma Bacillus subtilis Planobispora rosea Streptomyces sp. Sorangium cellulosum Sorangium cellulosum Alteromonas rava (marine) Streptomyces tendae unidentified fungus Actinomadura spiralis Streptomyces violaceusniger Amycolatopsis sp. Amycolatopsis sulphurea Amycolaotopsis sp. Streptoverticillium griseocarnum Streptomyces aurantiacus Streptomyces griseoviridus Lachnum payraceum Penicillium chrysogenum Penicillium chrysogenum
Aurantimycin Cineromycins Papyracon Cephem derivatives Sorrentanone
TERP PEP, mod. PK

Tab. 1 (Continued) .
Compound Reference Benzastatin
Producing Organisms Streptomyces nitrosporeus Sorangium cellulosum Fusarium sp. unidentified mushroom Sphaerellopsis filu Bacillus polymyxa Mycogone rosea Actinomycete Streptomyces sp. Sorangium cellulosum Sorangium cellulosum Apiocrea chrysosperma Pterula sp. Fusarium sambucinum Favolaschia A ureobasidium pullulans Streptomyces sp. Streptomyces aurantiogriseus Bacillus sp. Sporomiella australis Streptomyces sp. Streptomyces sp. Actinomadura spinosa Actinomadura spinosa Ascotricha amphitricha Mucor hiemalis Fusarium sp. Fusarium sp. Tricothecium sp. Streptomyces sp. Stachybotrys sp.
Structural Type ' ALK
Selected Properties antifungal, antiviral free radical scavenger antifungal antifungal antifungal antibacterial, antifungal antifungal, antibacterial antifungal, antibacterial antifungal antifungal antifungal cytotoxic antifungal antifungal antifungal antifungal antifungal antifungal antifungal antifungal antifungal antifungal antibacterial antifungal antifungal antifungal antifungal antifungal antifungal, cytotoxic antifungal antifungal antifungal antiviral, HIV HIV protease inhib., endothelin antag.
Research Group Involved KRIBB
Jerangolides BE29602 Dibefurin Darlucins Fusaricidin Helioferin Azalomycin Liposidolide Chirosazol Ratjadon Chrysospermin Hydroxystrobilurin Fusacandin Favolon Aureobasidins Phosmidosine NP-1OlA YM-47522 Australifungin AKD-2 UK-2AIBICID Prodimicin Pradimicin Ascosteroside Epothilone Fusarielin Saricandin Furanocandin Siamycin L-671,776, derivatives
PK PK-GLYC PK PK-mod. PEP-PK PEP-PK PK PK PK PK PEP PK PK-GLYC TERP PEP NUC A-mod. PK PK PK PEP PK PK TERP-GLYC PK-AA PK PK PK-GLYC PEP PK-AA
GBF Banyu Pharm. Co. Abbott Univ. Kaiserslautern and Univ. Munich Wakunaga Pharm. Co. and PT Kalbe Pharma Hans Knoll Institute Hoechst, AgrEvo RIKEN GBF GBF Hans Knoll Institute Univ. Kaiserslauern and Univ. Munich Abbott Univ. Kaiserslautern and Univ. Munich Takara Shuzo Co. RIKEN and SynPhar Lab. Inc. Hokkaido Univ. Yamanouchi Pharm. Co Merck Sharp & Dohme Univ. Osaka City Osaka Univ. and Suntory Ltd. Meijo U, Toyama Pref. Univ. Toyama Pref. Univ. and Bristol Myers Squibb Bristol Myers Squibb GBF Univ. Tokyo Abbott Meiji Seika Kaishi Ltd and Mitsubishi Chem. Corp. Bristol Myers Squibb Ciba-Geigy
Tab. 1 (Continued) . Compound Reference Benzastatins Triterpenesulfates Quinoxapeptides Karalicin AH-758 Eulicin Sattabacin Sattazolin GE20372 Isochromophilones Producing Organisms Streptomyces nitrosporeus Fusarium compactum Betula papyrifera Pseudomonas fluorescens Streptomyces sp. Streptomyces sp. Bacillus sp. Bacillus sp. Streptomyces sp. Penicillium sp. Structural Type' ALK TERP PEP-PK PK PK PEP-mod. A-mod. AA-mod. PEP PK Selected Properties free radical scavenger antifungal, antiviral rhinovirus protease inhib. antiviral: HIV1,2,RT antiviral: HSV antiviral: HSV antiviral: HIVl antiviral: HSV antiviral: HSV antiviral: HIV antiviral: HIV DGAT, ACTAT inhib. Research Group Involved KRIBB Abbott Merck Sharp & Dohme Univ. Cagliari and Univ. Cattolin (Rome) Kumamoto Univ. Jikei Univ., Institute of Microbial Chemistry Univ. Cagliari and Univ. Rome Univ. Cagliari and Univ. Rome Lepetit Kitasato
Antitumor Drugs Sch5290011 Gliocladiurn sp. Micromonospora sp. Rakicidins Actinomadura Esperamicin verrucosospora Ossamycin Streptomyces hygroscopicus Acetophthalidin Penicillium sp. (marine)
PEP PEP-PK PK-GLYC PK PK PEP-TERP NUC-mod. PK PEP PEP PK PK PK PK-GLYC PK PEP-PK PK, mod. PK PK
antitumor cytotoxic antitumor cytotoxic cell cycle inhibitor cell cycle inhib. proliferation mod. antitumor antitumor cytotoxic cytotoxic cytocidal antitumor antitumor cancerostatic aromatase inhib. anticancer oncogen function inhib. cytotoxic
Schering-Plough Bristol Myers Squibb Bristol Myers Squibb Lilly RIKEN RIKEN Toyama Pref. Univ. Kirin Brewery Co. Bristol Myers Squibb GBF Univ. Tokyo Kitasato Kyowa Hakko Kogyo Snow Brand Milk Co. and Kamagawa Univ. Institute of Microbial Chemistry and Showa College Fujisawa Nippon Kayaku Keio Univ. and Institute of Microbial Chemistry Abbott
Tryprostatins Sparoxomycin Cochleamycins Himastatin Chondramide Anguinomycin Clovalicin Clecarmycin Piericidin derivatives Hydroxymycotrienine FR901537 Medelamine Naphthablin Macquarimicin
Aspergillus fumigatus Streptomyces sparsogenus Streptomyces sp. Streptomyces hygroscopicus Chondromyces crocatu'S Streptomyces sp. Sporothrix sp. Streptomyces sp. Streptomyces sp. Bacillus sp. Bacillus sp. Streptomyces sp. Streptomyces sp. Micromonospora sp.
I Introduction: The Importance of Secondary Metabolites as Drugs Tab. 1. (Continued) Compound Reference Thiazinotrienomycin Cremeduycin Tryprostatin Sch50673,6 Terpentecin FD-211 Cytogenin Producing Organisms Streptomyces sp. Streptomyces cremeus Aspergillus fumigatus Nattrassia mangiferae Streptomyces sp. Myceliophthora lutea Streptoverticillium eurocidium Streptomyces sp. Penicillium patulum Streptomyces sp. Streptosporangium amethystogenes unidentified fungus Structural Type' PK-PEP Selected Properties cytostatic (cancer) Research Group Involved Institute of Microbial Chemistry and Showa College Univ. Illinois RIKEN Schering-Plough Kyowa Hakko Kogyo
A, mod. PEP, mod. PK PK

PK PK
Enaminedonin Dihydroepiepoformin EI-1507-1/2 TAN-15 11 CJ-12,371,2
PEP-PK PK-mod. PK PEP-PK PK
cytotoxic cell cycle inhib. antitumor antitumor: topoisomerase inhib. cytotoxic: MDR Taisho Pharm. Co. antitumor Institute of Chemotherapy (Shizuoka) and Institue of Microbial Chemistry detransforming RIKEN tumor cells IL-1 receptor Upjohn antag. IL-1-converting Kyowa Hakko Kogyo enz. inhib. induces Takeda cytokines DNA gyrase Pfizer inh.
Pharmacological Activities FR901,483 Cladybotryum sp. Streptomyces prunicolor PA-48,153 27-0-demethyl- Steptomyces hygroscopicus Rapamycin Streptomyces sp. NFAT 68,133 Penicillium sp. Stevastatin Streptomyces sp. Trichstatin Cytosporin Leustroducsin Plactins TAN1323CID Monamidocin A-72363 Trachyspic acid Carbazoquinocins Cytospora sp. Streptomyces platensis Agonomycetales Streptomyces purpurescens Streptomyces sp. Streptomyces nobilis Talaromyces trachyspermus Streptomyces violaceus
ALK-P-ester PK PK-AA PK PEP-PK PK PK PK-P-ester PEP PK PEP GLYC PK AA-PK
immunosuppr. immunosuppr. immunosuppr. immnuosuppr. immunosuppr. immunosuppr., histidine decarboxylase inhib. angiotensin bdg. inhib. t hrombocytosis inhib. stimulates fibrinolytic activity angiogenesis inhib. fibrinogen rec. antag. heparanase inhib. hep a r a na se inhib. antioxidant
Fujisawa Pharm. Co. Shionogi Smith Kline Beecham Abbott Nippon Kayaku Kyowa Hakko Kogyo Merck Sharp & Dohme Sankyo Co. Tokyo Noko Univ. Takeda Nippon Roche Sankyo Co. Sankyo and Univ. Tokyo Univ. Tokyo
Compound Reference Phenopyrazin

,
Producing Organisms Penicillium sp. Streptomyces sp.
Structural Type' PK-AA PK ALK PK PK, mod. PK PK PK-AA PK PEP PK-AA PK PK TERP-PK PEP ALK TERP PK PK-S PEP PEP-PK PK PK TERP-mod. TERP-PK TERP-PK PK
Selected Properties radical scavenger protein kinase inhib. protein kinase C inhib. protein tyrosine kinase inhib. myoinositol Pase inhib. myosin light chain kinase inhib. endothelin converting enzyme inhib. endothelin rec. antag. endothelin rec. antag. endothelin rec. antag. endothelin rec. antag. vasodilatory endothelin rec. antag. entothelin rec. bdg. platelet aggr. inhib. platelet aggr. inhib. squalene synth. inhib. cell adhesion inhib. Willebrand factor rec. antag. ACAT inhib. ACAT inhib. ACAT inhib. ACAT inhib. DGAT inhib. ACAT inhib. ACAT inhib. cholesterol esterase inhib. cholesteryl ester transfer protein inhib.
Research Group Involved Kitasato Ciba-Geigy Ciba Geigy Ciba Geigy and Panlabs Lepetit Kyowa Hakko Kogyo Fujisawa Asahi Kyowa Hakko Kogyo Kyowa Hakko Kogyo Ciba-Geigy Mercian Corp. Xenova and Parke Davis Xenova Taisho Pharm. Kitasato Sankyo Kitasato Nippon Roche Tokyo Noko Univ. Tokyo Noko Univ. KRIBB Kitasato and Pfizer Kitasato Kitasato Sankyo Tokyo Tanabe Co. and Univ. Tokyo Cornell Univ. and Schering-Plough
Balmoralmycin
Staurosporine Streptomyces longisporoflavus analogs Paeciloquinones Paecilomyces carneus Factor AIC MS-444 WS79089B Stachybocin RES-1149 RES-701 L-671,776 derivatives Mer-A2026 ET Drirnane-sesquiterpenes Bassiatin Herquline Schizostatin Macrosphelide Sulfobacins Lateritin Isohalobacillin GERI-BP002-A Pyripyropenes Amidepsine Terpendole Epi-cochlioquinone F1839 CETPI unidentified fungus Micromonospora sp. Streptosporangium roseum Stachybotrys sp. Aspergillus sp. Streptomyces sp. Stachybotrys sp. Streptomyces pactum Penicillium sclerotium Aspergillus ustus Beauveria bassina Penicillium herquei Schizophyllum commune Microsphaeropsis sp. Chryseobacter sp. Gibberella lateritium Bacillus sp. Aspergillus fumigatus Aspergillus furnigatus Humicola sp. Albophoma yamanashiensis Stachybotrys bisbyi Stachybotrys Cytospora (insect associated)

~~
Compound Reference Fluvirucin Thermorubin Salfredins Panosialins Xenovulene Arisugacin Nerfilin I Michigazones Aestivophoerin Lavanduquinocin Epolactaene MQ-387 YL-01869P YM 4714112 Poststatin Cathstatins BE-40644 RPR113228 Andrastin Saquayamycins
Producing Organisms Streptomcyces sp. Thermoactinomyces sp. Crucibulum sp. Streptomyces sp. Acremonium strictum Penicillium sp. Streptomyces halstedii Streptomyces halstedii
Structural Type' PK-GLYC PK PK-mod. PK-mod. TERP TERP PEP-PK PEP
Selected Properties phospholipase inhib. aldose reductase inhib. aldose reductase inhib. glycosidase inhib. GABA-benzodiazepine receptor binding AChE-inhib. neurite outgrowth ind. neuronal cell protecting neuronal cell protecting neuronal cell protecting neuritogenic aPase N inhib. matrix metalloproteinase inhib. elastase inhib. Pro-endopeptidase inhib. proteinase inhib.
Research Group Involved Univ. Keio UNITIKA Co. and Univ. Osaka Shionogi Kitasato Xenova Kitasato Somtech and Univ. Tokyo Univ. Tokyo Univ. Tokyo Univ. Tokyo RIKEN and Kaken Pharm. Co. KRIBB Sankyo
PK-mod. Streptomy ces purpeofuscus Streptomyces virdochromogenes Penicillium sp. (marine) PK Streptomyces nayagawaensis Actinomadura ultramentaria Flexibacter sp. Streptomyces virdochromogenes Microascus longirostris Actinoplanes sp. PEP PEP-mod. PEP PEP PEP-mod. PK
Chrysosporium lobatum TERP Penicillium sp. Actinomycetes TERP-PK PK
Yamanouchi Pharm. Co. Institute of Microbial Chemistry SynPhar Lab Inc. and Institute of Marine Bioscience (Halifax) t hioredoxin Tsukuba Res. and inhib. Banyo Pharm. Co. farnesyl protein R h h e Poulenc Rorer transferase inhib. farnesyl protein Kitasato and Keio Univ. transferase inhib. farnesyl protein Keio Univ. and Institute of Microbial Chemistry transferase inhib. plant growth regulator plant growth regulator herbicidal Univ. Tokyo and Ajinimoto Nippon Kayaku Univ. Paul Sabatier (Toulouse)
Agricultural Uses Streptomy ces Rotihibin graminofaciens Streptomyces sp. Pironetin
PEP-PK PK PK
Phthoxazolin
Streptomyces griseoaurantiacus
10
Compound Reference Methylstreptimidon-derivatives Fudecalone Arohynapene Xanthoquinodin Hydrantomycin Iturins Trichorzins Azalom ycin Phthoxazolines Phenamide Patulodin Gualamycin NK-374200 Melanoxadin Albocycline CI-4 Oligosperons Isocoumarins Milbemycins Sulfinemycin Musacins Lachnumlactone
Producing Organisms Streptomyces sp. Penicillium sp. Penicillium sp. Humicola sp. Streptomyces sp. Bacillus subtilis Trichoderma harzianum Actinomycete Streptomyces hygroscopicus Streptomyces sp. Streptomyces albospinus Penicillium urticae Streptomyces sp. Taralomyces sp. Trichoderma sp. Streptomyces sp. Pseudomonas sp. (marine) Arthrobotyrys oligospora Lachnum sp. (Ascomycete) Streptomyces sp. Streptomyces albus Streptomyces griseoviridis Lachnum papyraceum
Structural Type' PK-mod. PK PK PK PK PEP-PK PEP PK PK-mod. AA-mod. PK GLYC NUC-PEP PK PEP TERP PK PK PK-mod.
Selected Properties herbicidal anticoccidial anticoccidial anticoccidial herbicidal antibiotic phytopathogens antifungal antifungal antifungal antifungal antifungal acaricidal insecticidal melanine bios. inhib. melanogenesis inhib. chitinase inhib. nematocidal nematocidal antihelminthic antihelminthic antihelminthic nematocidal, cytotoxic
Research Group Involved Hoechst India Kitasato Kitasato Kitasato Kitasato USDA, Univ. Texas and Univ. Purdue CNRS (Paris) Hoechst and AgrEvo Merck Sharp and Dohme Kitasato Monsanto Osaka Univ. Nippon Kayaku Co. Nippon Kayaku Co. Teikyo Univ. and Tokyo Univ. Kitasato Shimizu Labs. Australian National Univ. Univ. Kaiserslautern (FRG) and Univ. Lund (Sweden) Smith Kline Beecham Lederle Univ. Gottingen, Univ. Tubingen, Hans Knoll Institute Univ. Lund and Univ. Kaiserslautern
PK
'
*
Structural type: PEP - peptide, PK - polyketide, TERP - terpenoid, GLYC - glycoside, A A - amino acid, NUC - nucleoside, mod. - modified. Property: antag. - antagonist; bios. - biosynthesis; ind. - inducer; inhib. - inhibitor; rec. - receptor. Group identification: Univ. - University of.
(AZUMA, 1987), bestatin (OCHIAI, 1987), topostins (SUZUKI al., 1990), etc., are to be et introduced into future therapy. The large-scale biotechnical production of bioactive compounds has been developed in a highly effective manner. Fermentations of high-producing microorganisms are carried
out up to a volume of more than 300 m3. The yield is sometimes more than 40 g L-' (VANDAMME, 1984), and up to 1OOgL-' in penicillin fermentations. This demonstrates the efficiency of strain selection which supported knowledge of biosynthesis and strain genetics. Optimum bioprocess control and suitable fer-
2 Secondary Metabolism, an Expression of Cellular and Organismic Individuality
11
mentation equipment were developed as further prerequisites of a highly efficienct production of biotechnical drugs. As an introduction to this volume, this chapter summarizes some of the general aspects of secondary metabolism in microorganisms such as:
small, but systematically defined groups of organisms (e.g., special species and genera of microbes, plants, animals) and point to the enormous variability of chemical structures (ComitC Editorial, 1992). In microbes, the capacity to generate secondary metabolites is frequently lost by genomic mutations, but this feature misses any concomitant effect on the - the biological role of bioactive compounds vegetative development of the pertinent 1989; OLESKIN, 1994). An strains (SHAPIRO, in the producer strains, inverse correlation is usually observed be- the biosynthetic pathways and their organitween specific growth rate and the formation zation, - natural and induced variations of second- of secondary metabolites such as antibiotics. ary metabolite structures and problems of Particular features of morphological differentiation in surface or submerged cultures, such their structural classification. as the formation of spores and conidia, seem Finally, future perspectives of drug screen- to be related to the production capacity of secondary metabolism. Moreover, a maxiing from microbial sources are discussed. mum production rate of antibiotics and other secondary metabolites (pigments, alkaloids, mycotoxins, enzyme inhibitors, etc.) has frequently been observed when growth-promoting substrates were depleted from the me1992). This phenomenon was dium (DEMAIN, called catabolite regulation (DEMAIN, 1974). This may be one of the reasons for the phase-dependency of biosynthesis of many microbial drugs. Thus, during the microbial growth phase (trophophase) secondary metabolism is often suppressed, but increased later during the 1986). Sometimes this idiophase (VINING, 2.1 Roles of Secondary feature is not present and depends on the parMetabolites in Producing ticular strains and growth conditions. For inOrganisms stance, the formation of phytotoxins by some phytopathogenic microbes such as Alternaria The majority of bioactive products of mi- and Fusarium strains is not a subject of catabcroorganisms and plants is generated by sec- olite regulation and even occurs in a growthondary metabolism. This part of the meta- associated manner (REUTER,1989). On the bolic machinery of microbes, plants, and ani- other hand, the production of antifungal efmals may play no essential role in the vegeta- fectors including peptaibol trichorzianine may tive development of the producing organisms, be induced, as shown in Trichoderma harbut seems to convey advantages to the perti- zianum by cell walls of the plant pathogen et nent species concerning its long-term survival Botrytis cinerea (SCHIRMBOCK al., 1994). in the biological community and environment Likewise, certain plant metabolites may in(LUCKNER al., 1977; KLEINKAUF VON duce the synthesis of peptide antibiotics in et and DOHREN, 1986; WILLIAMS al., 1989; the respective pathogenic Pseudomonas et and WHITE,1994; MO et LUCKNER, 1989 VINING, 1992; WILLIAMS strains (MAZZOLA et al., 1992; CAVALIER-SMITH, OLESKIN, al., 1995). In general, the phase-dependency 1992; 1994; VINING STUTTARD, and 1995) (Tab. 2). or specific inducibility indicates that the secFurther interpretations imply the formation ondary metabolism is strictly governed by inof certain secondary metabolites by relatively herent regulatory systems (see Sect. 2.2).
12
Tab. 2.
Presumed Roles of Secondary Metabolites in Their Producer Organism Endogenous role in the Exogenous role in the producing organism environment endogenous regulatory protection against competing signals triggering morphoorganisms - genesis endogenous signals regulating regulation of commensalism mating processes such as and cohabitation pheromones protection against endogenous detoxification physicochemical noxes (UV light) of metabolites acquisition of trace elements supply of special building material of the cell wall detoxification of trace elements endogenous reserve material not accessible to other microorganisms
Most of the secondary metabolites are biosynthesized in microbes and plants via complex multistep pathways involving many enzymatic and even non-enzymatic events. These appear to be integrated in a coordinated manner into the global microbial processes of cytodifferentiation such as formation of spores, conidia, and aerial mycelia (LUCKNER, 1989), or in the processes of invasion or defense. The same is true for plants in which secondary metabolite formation occurs in different tissues, e. g., roots, leaves, flowers, and seeds. Hence, it seems obvious that secondary metabolism does not reflect an occasional feature but is the result of a very long evolutionary development. As was shown for the tetracycline antibiotics from Sfrepfomyces spp. more than 200 genes may affect the biosynthetic pathway (VANEKand HOSTALEK, 1985). No wonder that speculation about the endogenous function and roles of secondary metabolites in the producing organisms themselves never came to an end (VANEK et al., 1981; VINING, 1992; OLESKIN, 1994; VINING STUTTARD, 1995). and To maintain such a great number of genes, generally linked into clusters, during evolution should be of advantage to the pertinent organism. Obviously, in plants many secondary metabolites are involved in the protection against microorganims and animals (CUND-
1992; JOHNSON and ADAMS, 1992). Others act as chemoattractants or as repellents towards insects fructifying flowers or damaging plant tissues. A series of plant hormones (cytokinins, gibberelic acid, jasmonic acid, etc.) are similar in structure but per definitionem are not secondary metabolites. Another function of secondary metabolites in plants is the detoxification of poisonous metabolites via an endogenous compartmen1989). The role of tized storage (LUCKNER, secondary metabolism in microbes is even more difficult to understand. Cellular efforts needed for secondary pathways are rather low in the wild-type strains (only a small amount of the overall substrate intake is converted to bioactive secondary metabolites). This part of metabolism would possibly have been eliminated during phylogenesis without any selective advantage of secondary metabolite production. It appears to be a generally accepted view that microbial secondary metabolites play an important but not generalizable rote, at least in special situations, e. g., in warranting the survival in particular environmental systems, during limitation of nutrient supply or even in the course of morphological development (LUCKNER al., et 1977; KLEINKAUFand VON DOHREN, 1986; VINING, 1992; KELLet al., 1995; VINING and STUTTARD, 1995). From this point of view,
LIFFE,
13
lated by secondary metabolites in heterothe formation of large amounts of antibiotics logous populations. by high-producing strains (substrate conver>0.1) would be considsion rates Yglucose,drug The self-protecting mechanism in antibiotered as a pathophysiological problem (VAic-producing microbes should be mentioned NEK et al., 1981). In order to better understand the general roles of secondary metabol- as a further evidence of an ecological function ites in microbes one could refer to the color of antibiotics, as a weapon against competiet et of hairs and feathers in animals, their odorous tors (ZAHNER al., 1983; BRUCKNER al., 1989,1992; WILLIAMS and pheromones, and other metabolic products 1990; CUNDLIFFE, 1992). By this means the miwhich do not contribute per se to the vegeta- MAPLESTONE, tive life of the pertinent species. But they crobe prevents suicide due to its own secondcould have outstanding importance during ary metabolite either by enzymatic modificathe adaptation to changing media, in the pro- tions of the drug, by alteration of its biologitection against competing organisms, and in cal target, or by an active transport-directed the regulation of sexual and asexual processes export (see, e. g., the tetracycline efflux) of genetic exchange. General discussions of (JOHNSON and ADAMS, 1992; NIKAIDO, secondary metabolite formation in microbes 1994). Usually, resistance mechanisms of the consider four major fields of importance antibiotic-producing microorganisms are the (LUCKNER al., 1977; KLEINKAUF et and VON same as in antibiotic-resistant bacteria. The DOHREN,1986; LUCKNER, 1989; WILLIAMS analysis of the gene sequences encoding reet al., 1989, 1992; VINING, 1992; CAVALIER- sistance determinants support the idea that SMITH,1992; OLESKIN, 1994; VININGand the transfer of resistance occurs from the antibiotic producers to the non-producing miSTU~TARD, 1995) (Tab. 2): crobes (JOHNSONand ADAMS, 1992; SA1995; DAV(1) The formation of secondary metabolites LYERS et al., 1995; HIRAMATSU, facilitates the adaptation to metabolic im- IES, 1994). In addition, the emergence of new balances as a kind of a metabolic valve, types of resistance factors by the formation of which is needed to remove an excess of mosaic genes has been analyzed in P-lactamtoxic, endogenous metabolites that other- resistant pneumococci (SPRATT,1994; COFwise are accumulated during a partial lim- FEY et al., 1995). The great variation of both active and inacitation of substrates. (2) Secondary metabolism could be a source tive secondary metabolites, that were obof individual building blocks of cells or of served in microorganims and plants supplied metabolic reserves which warrant the in- the main arguments against their determined dividuality and particular functionality of function. Obviously, the formation of a bioactive secondary metabolite, such as an antithe given strain. (3) Secondary metabolites could be regarded biotic, rather appears as an exception than as as endogenous signals triggering particu- a rule. Frequently, many inactive shunt-melar stages of morphogenesis and the ex- tabolites and congenors are produced in adchange of genetic material (see Fig. 1). dition to the few active metabolites. It is not This hypothesis was particularly sup- reasonable to believe that all these metaboported by the observation that the major- lites are needed in a single organism. It might ity of the good producers (e.g., actino- be that a function of a secondary metabomycetes, fungi, bacteria) display a life cy- lite could become apparent only in a particucle involving several stages of morpholog- lar, exceptional situation or in special stages of development. Hence, the selection presical differentiation. (4) Secondary metabolite formation is partic- sure on structures and secondary pathways is ularly important in biosystems as a signal necessarily low (ZAHNERet al., 1983). As a of interspecific communication be- consequence, spontaneously evolving variants tween microbes and other microbes, and mutants could survive with the same plants, and animals. Symbiosis, commen- probability as their parents, and modifications salism, and antagonism could be regu- of secondary pathways and structures would
14
A-factor
VB-factorS
autoinducer from Vibrio Rscheri
Butalactin
M0
, , J ,
factorfrom Str. vlrldochromugenes
differolide
Q
0
$ ?
Gennicidin
OH
on
Basidifferquinone
ChOH
no
antheridiol (for references, see text).
oogonlol
trlsporlc acid C
sirenin
Fig. 1. Structures of some representatives of signaling molecules from bacteria (streptomycetes) and fungi
be preserved (secondary metabolism as a tochromes, chlorophylls, sexual pheromones playground of evolution) (ZAHNER et al., of fungi and bacteria, etc. might have been 1983). This might explain the existence of the evolved similarly. Some of them may be atnumerous similar structures. According to tested to defined functions of microbial secthis hypothesis, the limited substrate specifici- ondary metabolites (Tab. 2, Fig. 1). A role of secondary metabolism in the adty of some enzymes of secondary metabolism has to be mentioned (LUCKNER, 1989). How- aptation to changing nutrient conditions is a ever, it should be noted that in many multi- realistic position since an excessive supply of step processes this limited specificity is re- metabolic intermediates (precursors) usually stricted to certain steps and thus less re- induces or stimulates drug production (DEstricted structural regions of the compounds MAIN 1974,1984,1992). Growth may become (KLEINKAUF and VON DOHREN, press). A imbalanced and precursors are accumulated in few secondary metabolites, out the pool of during the limitation of a given substrate in the many non-functional metabolites, have the medium, while others are still available in apparently acquired an essential role in excess. Excessive precursors could be regrowth and differentiation. The siderophores, leased into the medium or converted to harde.g., are microbial vehicles of iron transport ly metabolizable products which would not formed in variable structures as constitutive support the growth of competitors. Moreover, parts of the iron uptake system (VON DER colored secondary metabolites, such as pigHELMand NEILANDS, 1987; WINKELMANN, ments, could protect cells and spores from 1991; WINKELMANN DRECHSEL, and Chapter damage by ultraviolet radiation or also could 5, this volume). Per definitionern, they should promote the acquisition of rare elements via not be regarded as secondary metabolites. complex formation as, e. g., siderophores. Highly specialized biomolecules such as cy- Complex formation could also protect the
15
cells from high concentrations of toxic heavy metals. The incorporation of secondary metabolites into cellular structures has been suggested to contribute to their individual characteristics. Thus, streptomycin and its building moiety, streptidine, were established as a constitutent of the cell wall of the producing Sfrepfomyces griseus (DEMAIN, 1984; DISTLER et al., 1992). Otherwise, the production of secondary metabolites (so-called idiolites) (DEMAIN, 1992), could serve as a kind of a metabolic reserve which cannot be metabolized by other microbes. Some antibiotics (anthracyclines, tetracyclines, cyclosporins, etc.), e. g., are stored within the mycelium and their complete degradation requires a series of specialized enzymatic steps. Otherwise, bioconversions of antibiotics are a constitutive part of the self-protecting mechanisms of the producer strain. Moreover, concentrations of several antibiotics were shown to decrease in the course of prolonged cultivation, thus indicating the onset of degradative processes. Some fungi are well-known to degradate their own polyketides such as, e.g., citrinin (BARBER al., et 1988) and zearalenon and even to use them for additional syntheses. Active antibiotics were usually not detected in soil samples, although recently sensitive procedures have permitted the detection of phenazines (COOK et al., 1995). Their complete degradation under natural conditions seems very likely. Most likely, a series of signaling molecules is supplied by the secondary metabolism that possess interspecific (ecological) or speciesdependent functions, e. g., as signals triggering morphogenesis and the exchange of genetic material (Fig. l). By growth inhibition of competing microbes a producer strain could attain an advantage (c. f. the production of herbicidal antibiotics by phytopathogenic bacteria which damage plant tissues and facilitate nutrient acquisition from the host) (KOHMOTOand YODER, 1994; MAZZOLA and WHITE,1994; M o et al., 1995). Vice versa, secondary metabolism could confer a particular advantage in symbiotic systems, such as Pseudomonaslplant roots, to both the producing strain and the symbiont. An example is the control of phytopathogenic Fusarium or
Rhizocfonia fungi on plant roots by products of cohabiting streptomycetes and bacteria. Interspecific effects have also been postulated for volatile compounds which are formed, e. g., by streptomycetes and cyanobacteria. Geosmin, isoborneol, and mucidon are the constituents of the typical earthy odor. It has been shown that sclerin and scleroid from the fungus Sclerofinia liberfiana stimulate the biosynthesis of aminoglycosides by streptomycetes, but also the growth of some plants (KUBOTA al., 1966; OXFORDet al., 1986). et The formation of phytotoxins by phytopathogenic microbes is mentioned as another interspecific communication system (KOHMOTO and YODER, 1994). Constituents of the microbial cell wall (elicitors such as p1,3-1,6glucans from Phytophfora megasperma) are recognized by specific plant cell membrane receptors. Subsequently, a series of protective mechanisms is induced in the plant (e.g., hypersensitivity reactions, de novo synthesis of tissues, secretion of enzymes lysing microorganisms, and formation of antimicrobial phytoalexins). On the other hand, some of the phytoalexins are inactivated by enzymes of phytopathogenic microbes. In the natural habitat genetic information can be transferred from one microbe to another interspecifically. Both biosynthetic procedures and resistance mechanisms thus can be spread among various heterologous species and genera. Apparently this is also true for genetic exchanges between plants and microbes. A recent intriguing example is the discovery of a taxol producing fungus living in taxol producing yew trees (STIERLEet al., 1994). Typical plant hormones such as gibberellins and jasmonic acid are also produced by some microorganisms. Aflatoxins formed via complicated biosynthetic pathways in fungi, such as Aspergillus, have been established in actinomycetes. Sequence analyses of the genes encoding penicillin and cephalosporin biosynthetic clusters (ACV synthase, isopenicillin N-synthase, acyltransferase, deacetoxycephalosporin C-synthase, and deacetoxycephalosporin C-hydroxylase) in Penicillium chrysogenum, Acremonium chrysogenum, and Streptomyces spp. strongly suggested that fungi received the pertinent genes from the prokaryotic actinomycetes during evolution
16
(LANDAN al. 1990 MILLERand INGOLIA, et 1993; BUADESand MOYA, 1996). The production of cephabacins, chitinovorins, clavulanates, olivanic acids, carbapenems, and thiopeptides by unicellular bacteria and streptomycetes may indicate that an original biosynthetic pathway was spread horizontally among different microbes, thus giving rise to evolutionary variations of structures and pathways. The evolution of secondary metabolism even appears to create hybrid structures by the combination of genetic material originating from heterologous hosts. Recently, thiomarinol (SHIOZAWA al., 1993) was isolated et from the marine bacterium Alteromonas rava as a composite compound formed by the esterification of pseudomonic acid (found in Pseudomonas fluorescens) and holomycin (a pyrrothine antibiotic, found in Streptomyces Spa). The involvement of secondary metabolism in the regulation of microbial cytodifferentiation seems to be important, at least in some cases. The morphogenesis of antibiotic-producing microorganisms (streptomycetes, fungi, Mycobacteria, etc.) is obviously mediated by a plethora of biochemical steps, which display a high specificity for the given organism. The pathways are regulated by individual signals in a highly coordinated manner (Fig. 1) (LUCKNER, 1989). During morphogenesis, silent genes are activated that have not been expressed during the growth phase. Accordingly, several endogenous non-antibiotic regulators of the cell cycle were discovered in Streptomyces cultures, and their structure was elucidated (see below) (KHOKHLOV, 1982; GRAFE, 1989 HORINOUCHIand BEPPU, l990,1992a,b, 1995; BEPPU,1992,1995). Correlations between the biogenesis of some peptidic antibiotics and morphogenesis were also described for synchronously growing Bacillus cultures (MARAHIEL al., 1979). Tyrocidin, et gramicidin, and bacitracin are produced during the onset of sporulation, suggesting that their function concerns the control of transcription, spore permeability, dormancy of spores, and their temperature stability (MARAHIEL et al., 1979,1993). The y-butyrolactones represent a particularly important group of endogenous regula-
tors of Streptomyces differentiation (Fig. 2) (KHOKHLOV, 1982; GRAFE 1989; HORINOUCHI and BEPPU,l990,1992a, b, 1995; BEPPU, 1995). They are required as microbial hormone-like substances in few species such as streptomycin, virginiamycin or anthracycline producing strains. These effectors permit the formation of antibiotics and aerial mycelium by some blocked, asporogenous, antibioticnegative mutants even in very low concentrations. Several other autoregulators of morphogenesis have been investigated (see, e. g., factor C) (SZESZAK al., 1991). Otherwise, et germicidin B (PETERSEN al., 1993) from et Streptomyces violaceusniger inhibits germination of its own spores by interference with endogenous ATPase. Antibiotics such as hormaomycin (ROSSNER al., 1990) and pamaet mycin (KONDOet al., 1986) were shown to have autoregulatory functions. Moreover, streptomycetes can produce interspecific inducers such as anthranilic acid and basidifferquinone (Fig. 1) which affect basidiomycetes and the formation of fruiting bodies (AZUMA et al., 1980; MURAOet al., 1984). Moreover, regulatory molecules inducing cytodifferentiation were isolated from fungi and molds confirming that morphogenesis can be mediated by the aid of an agency of specialized endogeneous factors (HAYASHI et al., 1985). They can be regarded as secondary metabolites since they do not possess any function in vegetative development. In addition, sexual factors from fungi and yeasts can be considered as functionalized secondary metabolites. They trigger zygospore formation by haploid cells belonging to different mating types (GOODAY, 1974). During the evolution of signal systems, from the simple pro- and eukaryotes up to the hormonal control in mammalians, some structures and activities have been conserved. The alpha-factor of the yeast Saccharornyces cerevisiae as one of its sexual pheromones, e.g., appears to be partially homologous to the human gonadotropin releasing hormone (LouMAYE et al., 1982). Moreover, inducers of differentiation of Friend leukemia cells were isolated from soil organism such as Chaetomium sp. These chlorine containing substituted diphenols (Fig. 1) also induce morphogenesis (stalk cell differentiation) of Dictyostelium
17
Fig. 2. Regulatory events suggested to be involved in morphogenesis and secondary metabolism of Strepand tomyces griseus (P: promotor) (HORINOUCHI BEPPU,1992a).
discoideum, suggesting the similarity of mammalian and fungal control of the cell cycle (KUBOHARA al., 1993). Recently, the ocet currence of sexual pheromones was even established for the prokaryote Streptococcus faecalis. Its pheromones stimulate or inhibit the transfer of conjugative plasmids from donor to recipient strains (WIRTH al., 1990). et Peptides triggering competence in Bacillus subtilis have been characterized and were termed pheromones (DSOUZAet al., 1994; SOLOMONet al., 1995; HAMOEN et al., 1995).
2.2 Regulation of Microbial Secondary Metabolism 2.2.1 Genetic Organization of Product Formation
A large number of biosynthetic genes were isolated and characterized and, in general, they have been found assembled in clusters (Tab. 3). Such clusters may contain information for the biosynthesis of the basic structure of the metabolite, its modification, resistance determinants, e. g., promoting modification of products, targets, altered targets, or export systems, as well as regulatory elements; individual gene products which might as well exert regulatory functions.
18
Tab. 3. Biosynthetic Clusters Identified
Compound A54145 Aflatoxins Actinomycin Anguibactin Astaxanthin Avermectin Avilamycin Bacitracin Bialaphos Carbomycin Carotinoids Carotinoids Carotinoids Clavulanic acid Cephalosporin Cephamycin Coronatin Cyclosporin Daptomycin Daunomycin, Daunorubicin, Doxorubicin Destruxin Elloramycin Fatty acids Fatty acids Fengymycin Ferrichrome Frenolicin Geldanamycin Gramicidin S Granaticin Griseusin HC-toxin HET? Immunomycin Jadomycin B
Type acylpeptidolactone polyketide chromopeptidolactone modified peptide carotinoid polyketide polyketide branched cyclopeptide peptide polyketide terpenoids terpenoids terpenoids modified peptide modified peptide modified peptide modified polyketide cyclopeptide acylpeptidolactone polyketide
0rganism
Streptomyces fradiae Aspergillus parasiticus, Aspergillus fzavus Streptomyces chrysomallus Vibrio anguillarum Agrobacterium aurantiacum Streptomyces avermitilis Streptomyces viridochromogenes Bacillus licheniformis Streptomyces viridochromogenes Streptomyces thermotolerans Rhodobacter capsulatus Myxococcus xanthus Synecococcus PCC7942 Streptomyces clavuligerus Acremonium chrysogenum Nocardia lactamdurans Pseudomonas syringae Tolypocladium niveum Streptomyces roseosporus Streptomyces C51'peucetius
Selected References' BALTZet al., 1996' BROWN al. 1996 et MAHANTI al., 1996 et KELLER al., 19962 et CHENet al., 1996 MISAWA al., 1995 et MACNEIL, 1995 BECHTHOLD al., 1996' et HERZOG-VELIKONJA et al., 1994 SCHWARTZ al., 1996 et ARISAWA al., 1995 et ARMSTRONG, 1994 ARMSTRONG, 1994 ARMSTRONG, 1994 HODGSON al., 1995 et MART~N GUTIERREZ, and 1995 COQUE al., 1993, et 1995a,b; PETRICH al., et 1994 BENDER al., 1996 et WEBER al., 1994 et BALTZet al., 1996' YE et al., 1994; GRIMM et al., 1994; FILIPPINI al., et 1995; MADDURI HUTand CHINSON, 1995a, b; DICKENS al., 1996 et BAILEY al., 1996 et DECKER al., 1995 et SUMMERS al., 1995 et ROCKand CRONAN, 1996 Liu et al., 1996' LEONGet al., 1996' BIBBet al., 1994 ALLENand RITCHIE, 1994 TURGAY MARAHIEL, and 1995 SHERMAN al. 1989; et BECHTHOLD al., 1995 et Yu et al., 1994 PITKIN al., 1996 et BLACK and WOLK,1994 MOTAMEDI al., 19962 et YANGet al., 1995b, 1996b
peptidolactone polyketide polyketide polyketide peptide cyclopeptide polyketide polyketide cyclopeptide polyketide polyketide cyclopeptide polyketide? modified polyketide polyketide
Metarhizium anisopliae Streptomyces olivaceus Streptomyces glaucescens Escherichia coli Bacillus subtilis Ustilago maydis Streptomyces roseofulvus Streptomyces hygroscopicus Bacillus brevis ATCC9999 Streptomyces violaceoruber Streptomyces griseus Helminthosporium carbonum Anabaena sp. Streptomyces sp. Streptomyces venezuelae

Tab. 3. (Continued)
19
Compound Me1anin Landomycin 6-Methylsalicylic acid Microcystin Mithramycin unknown Nikkornycin Nodusmicin Nogalamycin Oleandomycin Oxytetracyclin Penicillin Phenazin Pristinamycin A Pristinamycin M Purornycin Pyoverdin Rapamycin Saframycin Soraphen A Sterigmatocystin Streptomycin
Type
Organism
Selected References' TAKANO al.. 1995 et BECHTHOLD al., 1996* et BECKet al., 1990 MEISSNER al., 1996 et LOMBd et al., 1996 ARROWSMITH al., 1992 et BORMANN al., 1996 et LE GOUILL al., 1993 et YLIHONKO al., 1996 et Q U I R ~ S SALAS, and 1995 KIM et al., 1994 SMITH al., 1990, et MACCABE al., 1990; et DfEz et al., 1990 PIERSON al., 1995 et DE CRECY-LAGARD, personal Communication BECKet al.. 1990 TERCERO al., 19% et STINTZI al., 1996 et SCHWECKE al., 1995 et POSPIECH al., 1996 et SCHUPP al., 1995 et BROWN al., 1996 et BEYER al., 1996 et FERNANDEZ-MORENO et al., 1996 COSMINA al., 1993 et SHENand HUTCHINSON, 1994 MERSON-DAVIES and CUNDLIFFE, 1994 DECKER al., 1995 et DAVIS and CHATER, 1990 ARMSTRONG, 1994, HUNDLE al., 1994 et
poly ket ide, glycosylated polyketide cyclopeptide polyketide polyketide modified peptide polyketide polyketide polyketide polyketide modified peptide heterocycle acylpeptidolactone polyketide/peptide modified aminoglucoside branched cycloacylpeptide modified polyketide modified peptide modified polyketide polyketide aminoglycoside
Aspergillus nidulans Colletotrichum lagenarium Streptomyces sp. Penicillium patulum Microcystis aeruginosa Streptomyces argillaceus Streptomyces cinnamonensis Streptomyces tendae Saccharopolyspora hirsuta Streptomyces nogalater Streptomyces antibioticus Streptomyces rimosus Aspergillus nidulans, Penicillium chrysogenum Pseudomonas aureofaciens Streptomyces pristinaespiralis Streptomyces sp. Streptomyces alboniger Pseudomonas fluorescens Streptomyces hygroscopicus Myxococcus xanthus Sorangium cellulosum Aspergillus nidulans Streptomyces glaucescens Streptomyces griseus Streptomyces rochei Bacillus subtilis Streptomyces glaucescens Streptomyces fradiae Streptomyces fradiae Streptomyces coelicolor Erwinia herbicola, Erwinia uredovora
Streptothricin modified aminoglucoside Surfactin Tetracenomycin Tylosin Urdamycin Whi, spore pigment Zeaxanthin peptidolactone polyketide polyketide polyketide polyket ide terpenoid (carotinoid)
Presented at the conference Genetics and Molecular Biology of Industrial Microorganisms. Bloomington 1996. Presented at the symposium Enzymology of Biosynthesis of Natural Products. Berlin 1996. Abstracts available from the authors on request.
'
20
The techniques employed include reverse genetics if sequence data of relevant enzymes is available, the use of homologous gene probes or probes constructed from key sequences, the generation by PCR of specific probes flanked by conserved key motifs, complementation of idiotrophic mutants, expression of pathways or single step enzymes in heterologous hosts, cloning of resistance determinants followed by isolation of flanking sequences, identification and cloning of regulatory genes or sequences (promoters, regulatory protein binding sites, pleiotropic genes, master genes, etc.). To improve product levels, the addition of extra copies of positive regulators (CHATER, 1992; HOPWOOD al., 1995; CHATERand et BIBB,Chapter 2, this volume), extra copies of biosynthetic genes possibly representing bottlenecks (SKATRUD al., Chapter 6, this volet ume), or the alteration of promoters of key enzymes are under investigation. The analysis of clusters has revealed a wealth of information including biosynthetic unit operations and their surprisingly complex organization. The majority of large proteins now known are multifunctional enzymes involved in peptide and polyketide formation, with sizes ranging from 165 kDa to 1.7 MDa. Other systems also forming polyketides, peptides, aminoglycosides, etc., are comprised of non-integrated enzyme activities, still performing the synthesis of highly complex structures. The details of various biosynthetic clusters are described in the respective chapters on regulatory mechanisms (CHATER and BIBB,Chapter 2, this volume), peptides (VON DOHREN and KLEINKAUF, Chapter 7, this volume), plactams (SKATRUD al., Chapter et 6, this volume), lantibiotics (JACK et al., Chapter 8, this volume), and aminoglycosides (PIEPERSBERG DISTLER, and Chapter 10, this volume). Recent highlights of the elucidation of such data have been the rapamycin and immunomycin clusters in Streptomyces, the erythromycin cluster in Succharopolysporu, the surfactin and gramicidin S clusters in Bacillus, various plactam clusters, and the sterigmatocystin cluster in Aspergillus nidulans. An overview of examples is presented in Tab. 3. The amplification of biosynthetic clusters in highly selected strains has been a fascinat-
ing key result, as shown for the industrial penicillin producer (FIERRO et al., 1995; MARTfN and GUTIERREZ,1995). The main findings with regard to sequencing of complete genomic fragments are as follows:
- The identification of biosynthetic genes follows by the detection of core sequences. Such sequences permit the recognition of types of biosynthetic unit operations like polyketide condensation reactions, the specificities of the respective transferase sites (HAYDOCKet al., 1995), the number of elongation steps, amino acid activation sites; in the case of repetitive cycles where certain sites are reused, as in type I1 polyketide forming systems or, e. g., cyclodepsipeptide synthetases, where the number of steps remains uncertain. - Additional genes for modification reactions like oxygenases and transferases are readily identifiable by standard structural alignments as well as possible regulatory proteins. At present, however, the unambiguous correlation of product and biosynthetic machinery is not possible without the support of various genetic techniques or, if not available due to the lack of transformation systems, structural details from protein chemistry of isolated enzymes or multienzymes. To illustrate a few concepts, we will point to some recent examples of cluster analysis: PLactam antibiotics as classical examples of modified peptides are still leading antibacterial drugs. Some efforts have been directed to understand at the molecular level the performance of industrial overproducers selected for decades (SKATRUD al., Chapter 6, this et volume). Following the reverse genetics approach in isolation of the isopenicillin N synthase gene (SAMSON al., 1985), which cataet lyzes the formation of the penem bicycle from the tripeptide precursor ACV, the clustering of biosynthetic genes was demonstrated in both pro- and eukaryotic producers (BARTON et al., 1990). The two key enzymes, ACV synthetase and isopenicillin N synthase showed extensive similarities in both bacteria and fungi, and a horizontal intergenic transfer has been suggested (LANDAN al., 1990; MILLet
21
and INGOLIA, 1993; BUADESand MOYA, 1996). The linkage of these adjacent genes illustrates well basic principles of cluster organet ization (Fig. 3) (AHARONOWITZal., 1992). In bacteria both genes are transcribed unidirectionally within an operon linked to sets of other genes the products of which are required for the modifying reactions of the cephem nucleus to cephamycin, and the formation of the plactamase inhibitor clavulanic acid (WARDand HODGSON, 1993). Such extensive linkages have been termed superclusters. In fungi the encoding genes for ACVS and isopenicillin N synthase are bidirectionally transcribed, separated by intergenic regions of about 1 kbp. A variety of environmental conditions are known to affect fungal plactam production at the transcriptional level (ESPESOand PENALVA,1996; SUAREZand PENALVA,1996; BRAKHAGE and TURNER, 1995). The bidirectionally oriented promoters between acvA (pcbAB) and inpA (pcbC) may permit the asymmetrical expression of both genes, and indeed different levels of expression have been obtained in constructs emER
ploying different reporter genes which allowed to measure the expression of both genes simultaneously (BRAKHAGE al., et 1992; BRAKHAGEand TURNER, 1995; BRAGKHAGE VAN DEN BRULLE, and 1995; THENBERGet al., 1996). Such results suggest possible additional functions for the penicillin tripeptide precursor, besides its role in the formation and the still unclear excretion of penicillins. The 872 bp intergenic region between the A . nidulans acvA (pcbAB) and ipnA (pcbC) permits the complex and sensitive regulation involving several protein factors (for P. chrysogenum, see FENG et al., 1995; CHU et al., 1995). The current knowledge of regulatory factors and putative factors implied by the identification and characterization of trans-acting mutations specifically involved in the regulation of the A. nidulans biosynthetic genes is summarized in Fig. 3b. One of these factors, designated PACC, was shown to activate at least the ipnA gene transcription in response to shifts to alkaline pH values (SHAHet al., 1991; ESPESOet al., et 1993;TILLBURN al., 1995; ARST,1996). For
a
E F
&lactarns
AB
C
'
10kbp
'
aflatoxin
raparnycin
c
A
N
P
'
acvA
aat
b -
Fig. 3%Organization of the biosynthetic clusters of plactams, rapamycin, and sterigmatocystin, b regulatory sites identified in the penicillin biosynthetic cluster in Aspergillus nidulans.
22
PACC seven binding sites with different affinities have been mapped in this intergenic region (SUAREZand PENALVA, 1996). Another binding site containing a CCAAT motif was detected, bound by a protein complex designated PENRl (THENBERG et al., 1996). PENRl also binds to a CCAAT-containing DNA region in the promoter of the aat gene encoding acyl-CoA:isopenicillin N acyltransferase which is located 3 of the ipnA gene (LITZKA al., 1996). Deletion analysis and et mutagenesis experiments indicated that the binding of PENRl represses the expression of acvA and increases that of both ipnA and aat (THEN BERG et al., 1996; LITZKA et al., 1996). PENRl thus represents the first example of a regulatory protein controlling the regulation of the whole plactam biosynthesis gene cluster in fungi. However, many promoters of eukaryotic genes are known to contain CCAAT motifs which are bound by distinct gene regulatory proteins (JOHNSON and MCKNIGHT, 1989). At the time being, it is unknown what kind of CCAAT binding protein PENRl represents and whether it is a global acting factor specific for the regulation of /3lactam biosynthesis genes. Using a genetic approach which is feasible for the ascomycete A. nidulans, three recessive trans-acting mutations were identified designated prgAllprgB1 for penicillin regulation (BRAKHAGE and VAN DEN BRULLE, et 1995) and npeEl (P~REZ-ESTEBAN al., 1995). These mutations formally correspond to positively acting regulatory genes. Mutants carrying one of the mutations mentioned produced reduced amounts of penicillin. For prgAl and prgBl it was shown that the expression of both genes acvA and ipnA was affected (BRAKHAGE VAN DEN BRULLE, and 1995), whereas npeEl controls at least ipnA expression (P~REz-ESTEBAN al., 1995). et The major nitrogen regulatory protein NRE of Penicillium chrysogenum has also been found to specifically attach to three GATA/ GATT pairs within this intergenic region (HAAS and MARZLUFF, 1995). The pairwise attachment sites indicate a possible dimeric state of this GATA family transcription factor and as well connect this regulatory site with nitrogen assimilation. This example illustrates that similar biosynthetic genes are un-
der the regime of organizationally specific mechanisms of regulation. The respective regulatory mechanisms will be evaluated comparatively in a variety of pro- and eukaryotic hosts. Regulation of the formation of secondary metabolites in eukaryotes, however, does not need to be this complex, as will be discussed below in the case of sterigmacystin/aflatoxin biosynthesis. As a second example for the organization of biosynthetic information the POlyketide immunosuppressant rapamycin has been selected (SCHWECKE al., 1995). This et polyketide with an iminoacyl residue is of interest as an immunosuppressor in autoimmune disease and transplantation. Its biosynthesis proceeds by 16 successive condensation and 21 modification reactions of 7 acetyl and propionyl residues, respectively, followed by pipecolate onto the cyclohexane carboxylic acid starter unit. The respective cluster has been identified in Streptomyces hygroscopicus by LEADLAY al. (SCHWECKE al., 1995) et et using polyketide synthase gene probes of erythromycin synthase from Saccharopolyspora erythrea). The sequence of 107.3 kbp has been determined as well as the boundary sequences, to assure the completeness of the effort. The key part of the cluster is represented by four genes encoding multifunctional enzymes with sizes of 900 (A), 1070 (B), 660 (C), and 154.1 kDa (P) responsible for the formation of the macrolactam ring. These four genes of 25.7, 30.7, 18.8, and 4.6 kb unambiguously correlate with the structural features of the product, however, module 3 and 6 contain catalytic sites for the reduction of the polyketide intermediates, which actually are not found in rapamycin. The solution of this problem remains to be found and plausible explanations are either non-functionality due to, e.g., point mutations, or a possible transient reduction of the intermediates to facilitate folding, which is reversed later. These key genes are flanked by additional 24 open reading frames, most of which have been assigned tentative functions including modification of the macrolactam, export, and regulation. Standard identification procedures are hampered by the non-availability of genetic operations for this strain.
23
The essential data in this case are the presence of large polyfunctional genes in prokaryotic clusters and the surprising lack of strict correlation of expected biosynthetic unit operations within the predicted modules with the actual gene structures found. A similar observation has also been made in the case of the avermectin biosynthetic cluster (MCNEIL et al., 1995). As a recent eukaryotic example the sterigmatocystin biosynthetic cluster in A. niduluns is considered (BROWN al., 1996). Sterigmaet tocystin is the penultimate intermediate in the biosynthesis of aflatoxins. Both polyketides are highly mutagenic and thus carcinogenic. They spoil food upon fungal colonization, especially by A. flavus and A. parasiticus. These losses may be reduced by a detailed understanding of the regulation of the biosynthetic events. So, e. g., the induction of aflatoxin formation has been shown to be strongly suppressed by jasmonate, a phytohormone (GOODRICHTANRIKULU al., 1995). Deet tailed genetic studies have confirmed the linkage and coregulation of sterigmatocystin and aflatoxin biosynthesis (TRAILet al., 1995a, b; KELLER and ADAMS, 1995; BROWNet al., 1996). The recent sequencing of the sterigmatocystin biosynthetic cluster in A. niduluns revealed within a 60 kb region 25 transcripts, the expression of which is coordinated under conditions of toxin production. The cluster is flanked by genes also expressed under nonproduction conditions. The regulatory gene aflR and its A. flavus homolog both specifically induce gene expression within the cluster. Among the identified genes are a fatty acid synthase, five monoxoygenases, four dehydrogenases, an esterase, an O-methyltransferase, a reductase, and an oxidase, all functionally implied in the proposed reaction sequence. Comparative evaluation of the respective cluster in A. parasiticus shows conservation of clustering, but no strict conservation of the gene order (TRAIL al., 1995a, b; et Yu and LEONARD,1995). Conservation of clustering has been suggested to serve both purposes of global regulation and horizontal movement of biosynthetic activities among species. The striking features of the tremendous efforts so far show the integration of a specific fatty acid synthase into a secondary
product cluster. These types of genes have been commonly referred to as primary pathway enzymes. The respective hexanoyl structure serves as a starter and is elongated by a type I1 system forming an aromatic polyketide. So far, such systems have been found only in prokaryotes. Gene characteristics, however, do not suggest a horizontal transfer as in the plactam case (BROWN al., 1996). et Finally, a specific transcription factor is a key element in the expression of the enzyme system, and no evidence has yet been obtained for complex timing and differential gene expression as in the penicillin pathway in A. niduluns. Inspection of other clusters included in Tab. 3 suggests extensive similarities of certain groups which, at first sight, look like structurally unrelated compounds. Certain types of regulatory genes are implied in the formation of various metabolites. There seems to be a non-species-related separation of type I and type I1 systems, e.g., in polyketide formation, but the various degrees of integration of biosynthetic modules catalyzing unit operations may be dictated by the chemistry of their products. Finally, the clustering of pathways also suggests their genetic transfer between various hosts. Within the evolutionary frame, adaptation of pathways to various targets has been proposed, e. g., for Aspergilli adapting to insect colonization and perhaps moving to other target organisms (WICKLOW al., 1994). The structures of et metabolites with key roles in invasive processes would then adapt to new targets by evolutionary processes.
2.2.2 Regulatory Mechanisms

Mechanisms involved in the regulation of secondary metabolite expression have been reviewed recently, focussing on global control in bacterial systems (DOULL and VINING, 1995), bacterial mechanisms in detail (CHATER and BIBB, Chapter 2, this volume), antibiotic formation in Streptomyces coelicolor (HOPWOOD al., 1995), and autoregulators et (HORINOUCHIand BEPPU, 1995; BEPPU, 1995). Eukaryotic systems except for P-lac-
24
tams have not been in focus regarding special metabolites. Recent reviews cover plactams (BRAKHAGE TURNER, and 1995; SKATRUD et al., Chapter 6, this volume; JENSENand DEMAIN, 1995). A variety of stress conditions have been documented to lead to secondary metabolite production (DEMAIN, 1984; DOULLand VINING,1995; VINING and STUTTARD,1995). Besides physical parameters (temperature shock, radiation) chemical signals will trigger the formation of various small response molecules, which are the subject of this volume. Such signals include both high and low concentrations of oxygen (oxidative stress, lack of oxygen, or shift to anaerobic growth), acidity (pH shift), but generally the response to nutrient alterations. Phase-dependency of secondary metabolite formation in microbial cultures and its correlation to morphological changes suggest that secondary metabolism is subject to general regulatory mechanisms governing cellular development (BARABAS et al., 1994). Only some of the regulatory features have been elucidated in the past and many are still to be unraveled. Nutrient shift regulation of growth is closely coupled to differention through a series of common metabolic signals and regulations such as mediated by sigma factors and transcriptional enhancers. In this context, two major questions are addressed: (1) Why are microbial secondary metabolism and morphogenesis suppressed during growth on media which are rich in carbon, nitrogen, or phosphorus and what is the cause of catabolite regulation? (2) What is the nature of the general signals governing a plethora of metabolic events and how do they cooperate within the cellular frame of developmental programs? There are indeed drastic variations in the extent of responses upon nutritional stress. Obvious morphological changes like sporulation or formation of aerial mycelia are caused by an undetermined number of respective genes, reading to sets of proteins and mediators promoting alterations in the cellular composition. Such changes include altered cell
wall composition and changes in the metabolic spectrum. The changes may not be obvious and some work has been conducted on model systems such as Escherichia coli, Bacillus subtilis, and Aspergillus nidulans. Besides nutrient depletion as envisioned and studied in chemostate-like environments employed in fermentation, a generally neglected field is the response to environmental factors indicating the presence of alike or competing organisms. According to our understanding of the basic role of many of the metabolites employed in the control of invasive processes. Such approaches seem obvious. It has been shown that cell density critically affects antibiotic production (WILLIAMSet al., 1992; FUCQUA al., 1994; SANCHEZ et and BRANA, 1996). The induction of nisin formation by nisin itself, as mediated by its cluster-inherited signal system, is another intriguing example (RA et al., 1996; DERUYTERet al., 1996). Likewise the presence of phytopathogenic fungi induces responses, e. g., in rhizosphere colonizing bacteria including the production of antifungals (KAJIMURA al., 1995; PIERet SON and PIERSON, 1996). While the presence of resistant microorganisms has been applied in selection processes for antimicrobial agents the identification of response signals is still an open field.
Stress Conditions Related to Nutrient Limitations In connection with nutrient depletion carbon, nitrogen, and phosphate starvation are considered in general. The differential induction of metabolite forming processes has been excellently demonstrated by BUSHELLand FRYDAY (1983). Extensive studies of this aspect have also been conducted in the antibiotic fermentation of gramicidin S in Bacillus brevis. Formation of this cyclopeptide has been found in a variety of stress conditions, including sporulation and non-sporulation conditions and, surprisingly, two phosphate concentration ranges (KLEINKAUF VON and DBHREN, 1986) differing from other phosphate-effected systems (LIRASet al., 1990). Thus, in many cases less specific induction
25
and maintenance by interacting regulatory bon sources causing slow growth promote devices are implied and manipulation may be production. This has been demonstrated nicely in the case of bacitracin formation in Bucilexerted by growth rate control. Nutritional downshift in the media caused lus licheniformis (HANLONand HODGES, by limitation of particular metabolites (amino 1981). Glucose-6-phosphate suppresses the acids, ATP, sugars, etc.) promotes excessive synthetase enzymes in penicillin biosynthesis formation of some metabolites due to an im- (JENSENand DEMAIN,1995). However, as balanced metabolism (supra) (MART~N al., was shown for ACV-synthase, IPN-synthase, et 1986; LIRAS et al., 1990). Accumulation of and expandase in penicillin and cephalospothese precursors is known to induce sec- rin producing fungal and Streptomyces strains ondary pathways (see, e.g., the induction of inhibition or repression by glucosed-phosergotamin alkaloid formation by tryptophane phate, ammonium, and phosphate ions deet in Cluviceps strains) (HOTTER,1986). On the pend on the given strain (AHARONOWITZ other hand, limitation of some endogenous al., 1992). Carbon uptake systems have been studied metabolites could be important which inhibit global regulatory mechanisms governing aer- in several organisms including enteric bacteial mycelium and spore formation. In this re- ria in which the phosphoenolpyruvate<arbopressing or inhibitory effects on the second- hydrate phosphotransferase system controls ary pathways and on morphogenesis could be uptake and transport (POSTMAet al., 1993). diminished. Both features, accumulation of This phosphorylation-controlled multistep precursors and limitation of repressing process involves adenylate cyclase and metabolites seem to be involved (DEMAIN, CAMP-mediated gene regulation. Other 1974, 1992; MART~N al., 1986; HORINOU- mechanisms operate in gram-positive bacteria et CHI et al., 1990, LIRASet al., 1990). The perti(STEWART, 1993) and streptomycetes (CHAnent regulatory mechanism may be similar to TER and BIBB,Chapter 2, this volume). Nitrogen depletion again is a determining those shown for other global microbial regulfactor in many antibiotic fermentations (SHAations. Metabolite formation has been studied in PIRO,1989). These effects are attributed to nidetail in model cases of surfactin (B. subtilis), trogen catabolite repression. A two-compostreptomycin (Streptomyces griseus), or peni- nent system sensing the glutamine and a-kecillin (A. niduluns and P. chrysogenum). It is toglutarate levels activates transcription of controlled by superimposed regulatory cas- catabolic enzymes releasing ammonia or othcades or networks. Such networks include in- er nitrogen sources by autophosphorylation and VINING, tracellular and extracellular components and of a His protein kinase (DOULL might include regulators, transducers, signal- 1995). The activation of glutamine synthetase ing systems, interacting repressors, and acti- is included in this process, the activity of vators, as well as modification and expression which is as well controlled by several factors systems. In the case of streptomycin the term including the glutamine level. Actinomycetes decision phase has been coined as a model contain two types of glutamine synthetases. In for a variety of production processes (PIE- process analysis ammonia has been found to and PERSBERG,1995; PIEPERSBERG DIST- repress secondary metabolite formation. Roles of various nitrogen sources have not LER, Chapter 10, this volume). Despite this complexity, manipulations of single genes been evaluated in detail, but are discussed in et may have substantial effects on production the case of plactams (SKATRUD al., Chapter 6, this volume). Ammonium ions are also levels. Most information on the respective deple- catabolite repressors of plactam biosynthesis tion events have come from model organisms, (cephalosporin C, cephamycin C) in Acrebut they proved to be useful in a variety of monium and some Streptomyces spp. (JENSEN 1989). Deamicases. Carbon sources are known but poorly and DEMAIN,1995; DEMAIN, understood tools in natural product pro- nation of L-valine in the biosynthesis of tylocesses. Readily assimilated compounds, e. g., sin is subject to catabolite regulation by am1986). glucose repress production while other car- monium ions (TANAKA,
26
In bacteria, a stringent response is caused by nitrogen limitation (CASHEL,1975) and the appearance of non-acylated tRNAs. A concomitant increase of guanosine-3 ',5 '-tetraphosphate (ppGpp) concentration switches off unfavorable biosynthetic processes. Ribosomal protein synthesis is reduced, but the degradation of amino acids continues. This fact is due to the binding of ppGpp to RNA polymerase and the alteration of its promoter recognition. Thus, transcription of many genes might be stimulated while the expression of others declines in a coordinated manner. The molecule of guanosine-3 ' , 5 '-tetraphosphate might be involved in the regulation of the secondary metabolism and also in sporulation of streptomycetes (OCHI, 1990). The heterogeneity of promotor structures and the complementation of bacterial RNA polymerases by sigma factors could provide another rational basis for the understanding of the developmental regulation of gene expression (CHATER and BIBB,Chapter 2, this volume). RNA polymerase consists of a core enzyme composed of each of two a- and two Psubunits. Bacterial promotor recognition is u43, etc.) atregulated by sigma factors tached to the core enzyme. Depending on the type of the individual sigma factor, either general (e.g., the factors needed for vegetative growth) or specialized genes (e. g., those responsible for secondary metabolism and cytodifferentiation) can be transcribed. In Streptomyces griseus MARCOSet al. (1995) identified three sigma factors differentially expressed under specific nutritional conditions. The sigma factors whiG and sigF, each controlling certain events in the development of spore chains in Streptomyces coelicolor, are controlled by transcriptional and posttranscriptional events involving additional proteins (KELEMEN al., 1996). et Recent approaches of molecular genetics showed that DNA-binding protein factors are crucial for the transcription of both eukaryotic and prokaryotic genes (HORINOUCHI and BEPPU,1992b; CHATER, 1992; THEN BERGet al., 1996). They often occur as dimers and stimulate activity by binding to particular promotor regions. An example is the regulatory system of the y-butyrolactones (A-factor) involving proteinaceous transcriptional activa-
tors (AfsR protein) (HORINOUCHI BEPand PU, 1992b; BEPPU,1995). The response to exogenous phosphate has been studied in E. coli and the involvement of more than 30 genes in the PHO regulon has been established (WANNER,1993). Respective efforts in antibiotic production have been reviewed (LIRASet al., 1990). So p-aminobenzoate synthetase by S. griseus as a key enzyme of candicidin synthesis is negatively regulated by inorganic phosphate (MART~N, 1989). An upstream promotor region of 113 bp length and rich in AT was identified as a binding site of a general phosphate-dependent repressor protein. If phosphate-insensitive genes such as the Pgalactosidase gene were coupled to this fragment and transferred in other Streptomyces hosts (such as S. lividam) they became subject to phosphate control.
2.2.3 Genetic Instability

The formation of secondary metabolites often is genetically instable and many explanations for this phenomenon have been given (DYSON and SCHREMPF, 1987; ALTENBUCHNER, 1994). The occurrence of extracellular plasmids containing transposon structures and IS elements was discussed initially. These could be integrated into the genome and induce genomic rearrangements and gene disruptions (HORNEMANN al., 1993). et Streptomycetes contain only one single linear chromosome (8 Mb) (ALTENBUCHNER, 1994; CHENet al., 1994; REDENBACH al., et 1996). Gene mapping experiments, complementation of blocked mutants, and heterologous expression of genes in different Streptomyces hosts have shown that the genes of secondary metabolite production are localized on chromosomal gene clusters (HOPWOOD et al., 1983; LIU et al., 1992; STUTTARD and VINING, 1995). Clusters which are responsible for the polyketide and aminoglycoside syntheses contain the genes of self-resistance protecting against the toxicity of the own secondary metabolite (SENOand BALTZ,1989). Moreover, regulatory gene products are involved which integrate secondary metabolism
(a7,
27
into cellular developmental regulations (DISTLER al., 1988; HORINOUCHI BEPet and PU, 1990 BEPPU,1992). An organizational principle is the formation of large amplified genomic structures (DYSON and SCHREMPF,1987). Such sequences could be used as amplifying tools for biosynthetic pathways in the future. A few examples demonstrate that the enlarged gene dosage contributes to improved drug production in high-yielding strains (TURNER, 1992; MARTfN and GUTIERREZ, 1995). COmmercial penicillin producing strains derived in decades of random selections may contain more than 20 copies of the biosynthetic cluster linked by conserved hexanucleotide spacer elements (FIERROet al., 1995). The frequent deletion of the cluster has been related to these hot spots of recombination, since non-producer mutants have lost the entire region within these boundaries (FIERROet al., 1996). In the event of transpositions, frameshift mutations may lead to disruptions of both structural and regulatory genes (gene deletions) (HORNEMANN al., 1993). Activation et of silent gene sequences may occur in mutant strains because of the same reason. Thus, regeneration of protoplasts or curing of plasmids may yield mutants of streptomycetes experiencing a completely altered pattern of secondary metabolism (see, e. g., the formation of curromycin, indolizomycin, and iremycin) (OGARA et al., 1985; OKAMIet al., 1988).
thesis. Accordingly, leucyl-tRNA could signify a marker governing gene transcription of differentiation-dependent pathways, at least in streptomycetes (DAVIESand CHATER, 1992; CHATER,1992). As a summary, initial factors suppressing or stimulating cytodifferentiation and secondary metabolism of the microbes are excessive nutrients converted to regulatory metabolites, nutrient downshifts, the accompanying changes of general regulatory metabolites (such as ppGpp), and accumulations of precursors due to metabolic imbalances. Low and high molecular-weight mediators are needed to trigger the coordination of numerous pathways and cellular events in cytodifferentiation. The A-factor and similar y-butyrolactones are mentioned here as a particularly intriguing example of how the complex developmental programs are organized in streptomycetes (HORINOUCHI and BEPPU, 1990, 1992a, b; BEPPU,1995).
2.2.5 The A-Factor and the Signal Cascade of Cytodifferentiation in Streptomy ces
The A-factor and its dihydro derivatives (Fig. 1) are formed by numerous streptomycetes. In some Streptomyces strains such as S. griseus and S. virginiae y-butyrolactone autoregulators are required as a kind of a microbial hormone for antibiotic production and even for sporulation (PLINER al., 1976; Hoet 2.2.4 Developmental Processes RINOUCHI and BEPPU,1990; ISHIZUKA al., et 1992). In 1975, KHOKHLOV coworkers (PLINand Another feature of regulation of gene expression in the development of streptomy- ER et al., 1976) found an idiotrophic mutant cetes implies regulatory genes, such as whi, strain of S. griseus which neither produced bld, afs, and abs in S. lividans and S. coelico- streptomycin (an aminoglycoside antibiotic) lor (HOPWOOD, 1988 DAVIESand CHATER, nor formed aerial mycelia and spores. It re1992). Thus, deletion of the whiB gene causes gained normal cytodifferentiation in the presthe concomitant loss of aerial mycelium for- ence of the culture liquid of the parental mation. BldA was shown to specify the leu- strain. Later, the structure of the autoregulatRNA (UUA codon). Further evidence sug- tory A-factor and a series of 2-dihydro dergested that ITA codons in the DNA are ab- ivatives was elucidated (PLINERet al., 1976; et sent from all genes involved in vegetative ISHIZUKA al., 1992). The A-factor (2R-hygrowth, but are present in the regulatory or droxymethyl - 3 - oxocaproyl) - y - butyrolactone, resistance gene clusters of antibiotic biosyn- its homologs and analogs are capable not only
28
General Aspects of Secondary Metabolism 1995). Gene disruption of afsK in S. coelicolor caused the reduction of actinorhodin formation without effecting growth. Residual biosynthetic activities may be regulated by other kinases and the two-component system ufsQllafsQ2 controlling actinorhodin production in S. coelicolor. So the phosphorylated afsR protein seems to bind to regulatory DNA sequences near the ufsA gene (in S. griseus), act genes (in S. coelicolor), and red genes (in S. lividans) and to enhance their transcription. The afsA gene encodes for the biosynthesis of A-factor-like molecules, which is accomplished by the fusion of phosphorylated glycerol and P-ketofatty acids (SAKUDAet al., 1992). Intracellular recognition of the A-factor occurs via an A-factor binding protein acting as the represand sor of the X-gene (HORINOUCHI BEPPU, 1992a). Inactivation by A-factor thus permits formation of the X-protein acting as a transcriptional enhancer of the strR and aphD genes. While AphD is responsible for the selfresistance of the producer strain to streptomycin, strR appears as a transcriptional antiterminator of streptomycin biosynthesis. Although the scheme is still incomplete it suggests that numerous events of sporulation and secondary metabolism could be governed by afsR, X- and strR gene products in a concerted manner. In analogy to S. griseus, a virginiae butanolide (VB-C) binding protein (Mr 36000 Da) was isolated from S. virginiae which is suggested to be involved in the mechanism of pleiotropic signal transduction. The binding activity of this protein towards VB-C decreased by 40% in presence of DNA in a similar manner as shown for other regulatory proteins and transcriptional factors (KIM et al., 1990; SAKUDA al., 1992). The pertinent et gene was sequenced and displayed considerable homology (6244%) to the amino acid sequences of ribosomal protein L ll of diverse origins (rpIK) and to the essential protein of E. coli. This suggested it to be a part of an essential gene cluster encoding general components of the transcriptional and translational systems. Exogenously added factors have been used here to improve metabolite production in the case of virginiamycin (YANGet al., 1995a, 1996a).
to induce streptomycin biosynthesis and aerial mycelium formation but also daunorubicin, virginiamycin, and carbapenem production in other Streptomyces strains, bioluminescence in Vibrio fischeri, nodulation of plant associated bacteria, and toxin production in Pseudomonas aeruginosa (BEPPU, 1995). For the dihydro derivatives of the A-factor isolated from Streptomyces viridochromogenes and S. bikiniensis, the 2R,3R- and the 2S,3R-configuration was initially proposed (SAKUDAand YAMADA,1991) but later the absolute stereochemistry was established as 2S,3R,2' R and 2S,3R,2 S, respectively (YAMADA et al., 1987; SAKUDA al., 1992; LI et et al., 1992). The latter structure, but not the Afactor-type 2 '-0x0-butyrolactones, induce the production of the peptide antibiotic virginiamycin by S. virginae (virginiae butanolides) (YAMADAet al., 1987). Recently, 2'-deoxy derivatives (NFX factors) were even shown to stimulate virginiamycin production in the same manner, and NFX-2 ((2R,3R,4S)-2-hexyl-3-hydroxy-4-pentanolide) proved to be identical with blastomycinol lactole (a compoet nent of antimycin A l ) (YAMADA al., 1987; KIMet al., 1990 OKAMOTO al., 1992). et Thus y-butyrolactones play an outstanding role as regulatory signals inducing cytodifferentiation and formation of quite different secondary metabolite structures such as aminoglycosides, polyketides, and peptides (HORINOUCHI and BEPPU, l990,1992a, b SAKUDA et al., 1992; BEPPU,1995). Genetical and biochemical experiments contributed much to the present knowledge of the regulatory cascade of cytodifferentiation of S. griseus which involves the A-factor and its congeners as signal transmitters (Fig. 2). AfsR (a 100 kDa protein encoded by the afsR gene containing ATP and DNAbinding domains) represents an early event in the cytodifferentiation of S. griseus. It is active in its phosphorylated form AfsR-P as a transcriptional activator of several other genes and it can be phosphorylated by afsK, a respective kinase. The N-terminal region of this kinase shows significant similarity to other Ser/Thr kinases including the P-adrenergic receptor kinase, the Rous sarcoma oncogene product, and a Myxococcus enzyme (BEPPU,
29
Possibly, the afsR gene of S. griseus is also controlled by other genes which have not been identified so far. The whiB gene of S. coelicolor, e. g., is responsible for early sporulation events due to the formation of a small transcription factor-like protein which is dispensable for growth, but essential for sporulation (DAVIESand CHATER,1992; CHATER, 1992). Moreover, S. griseus mutants which were recently investigated, produce the A-factor but nevertheless miss the normal sporulation behavior (MCCUEet al., 1992). An open coding gene sequence ( O W 1590) was identified which is possibly responsible for the synthesis of two polypeptidic transcription factors (P 56 and P49.5). Dimerization of P5 6 was suggested to induce the onset of sporulation, but P49.5 prevents this event. In the above mutant imbalanced regulation of the syntheses of P 56 and P 49.5 have been proposed to cause lack of sporulation. As another type of event, ADP-ribosylation of proteins catalyzed by NAD-glycohydrolase and ADP-ribosyltransferase seems to participate in cytodifferentiation of S. griseus. Failure to ADP-ribosylate certain cellular proteins in mutant strains was thought to cause impaired differentiation (SZESZAKet al., 1991; OCHIet al., 1992). A y-butyrolactone derivative, phydroxybutyryl-homoserine lactone, is the autoinducer of light emission by Vibrio hurveyi (MEIGHEN, 1991; WILLIAMS al., 1992; Fuet QUA et al., 1994; GEIGER,1994). Similar to other photobacteria, luminescence is strongly influenced by the density of the cell culture. V. hurveyi synthesizes the above small extracellular molecule, which accumulates in the growth medium and induces luminescence by luciferase and FMNH2-coupled oxidation of a long-chain fatty aldehyde. Vibrio fischeri forms a similar autoinducer, P-ketocaproyl homoserine lactone (FUQUAet al., 1994). Previously, similar molecules have been reported to regulate carbapenem biosynthesis by Erwiniu curotovora (BAINTON et al., 1992). The signaling pathway of light emission which is induced in presence of the above mentioned butyrolactones seems to involve transmembrane signaling proteins as recep-
tors. They possess enzymic domains at the inner site of the membrane (MEIGHEN,1991; FUQUA al., 1994; GEIGER, et 1994). Early evidence for autoregulatory functions of special metabolites in the differentiation and diploidization was presented for a series of fungi (GOODAY, 1974; ZAKELJMAVRIC et al., 1995). In some molds there are sex hormones like antheridiol, sirenin, oogoniol, and trisporic acids (Fig. l), which trigger zygospore formation and the subsequent exchange of genetic material. In the aquatic fungus Achlya the signaling chain of the fungal sterol antheridiol displays similarity to mammalian cells. Here the response to steroidal sex hormones is also mediated by membrane receptors (ZAKELJMAVRIC al., 1995). et
2.2.6 Overproduction of Microbial Secondary Metabolites and Precursor Pools

Much experience has been obtained in the past with empirical selections of high-yielding strains of antibiotic producing microorganisms. Comparison of the high-producing mutants of streptomycetes and fungi with the low-yielding wild-type strains suggested that a series of heritable metabolic changes had been introduced (OCHIet al., 1988; VANEK and HOSTALEK, 1988), for instance: the elimination of bottle-necks in the production of biosynthetic precursors, - the suppression of negative catabolite regulations concomitant with increased production of synthetases, - improved resistance of the producer strain against its own toxic product, and - the absence of negative feedback regulation of the formed secondary metabolite on its biogenesis.
-
To realize these prerequisites of high productivity, the natural regulatory mechanism of the wild-type strains, permitting only little product formation had to be altered in a stepby-step selection procedure. The alterations concern both the genetic and the physiologi-
30
cal system of the pertinent strain, the secondary pathways, and the cellular morphology (VANEK and MIKULIK, 1978). Many of the high-producing strains overproduce the pertinent precursors. An excessive precursor supply thus appears to determine high secondary metabolite production. Moreover, when several alternative precursors can be used by the same biosynthetic pathway the availability of the individual precursors governs the quality of formed products. Wild-type strains often produce a series of homologous structures due to the usage of several intracellularly supplied precursors (SANGLIERand LARPENT, 1989). During strain improvement by mutagenesis and selection empirical pathway engineering was done. Sometimes, the selection promoted excessive formation of a single precursor and, consequently, a single product was formed instead of a series of homologous structures (CLARIDGE, 1983; THIERICKE ROHR,1993). and Precursor-directed biosynthesis, mutational and hybrid biosyntheses signify microbiological techniques (CLARIDGE,1983; THIERICKE ROHR, 1993), which have and successfully been used in the past to alter product formation by excessive feeding of precursors or biosynthetic intermediates to parental strains and their mutants. Even f when structural analogs o the special precursor were fed to the medium they could be used as a substitute of the natural structure. In this manner, the formation of many new and unusual secondary metabolites was demonstrated (SHIERet al., 1969). During the rapid (balanced) growth of microbial cultures no excess of intermediary metabolites is available, but when some substrates become rate-limiting while others are still available a metabolic imbalance arises which promotes the accumulation of precursors (imbalanced growth) (DEMAIN,1974, 1992). Apparently, the size of precursor pools is of regulatory importance in secondary metabolite formation and determines the production rate. Investigations of the plactam biosynthesis illustrate well that penicillin formation by P. chrysogenum is subject to negative feedback control by L-lysine, and to a lesser extent by L-valine (MARTINet al., 1986). The former
inhibits and suppresses homocitrate synthetase in the low-producing strains as a negative feedback regulator. The high-producing strains display greatly reduced sensitivity to lysine (MARTIN and DEMAIN, 1980). This branched-pathway model of regulation was also reported for the biogenesis of candicidin by S. griseus. It is reduced by excessive tryptophan in the medium due to the feedback inhibition of the p-aminobenzoic acid synthetase (MARTIN, 1978). L-cysteine needed for p-lactam production can be produced either from sulfide and 0acetylserine or by reverse transsulfuration of 0-acetylhomoserine using L-methionine as a donor of sulfur. In P. chrysogenurn (forming penicillin G ) cysteine is produced mainly by the sulfate reduction pathway, in Acremonium chrysogenum (producing cephalosporin C) via transsulfuration (MARTIN,1978). In the latter strain, feeding of L-methionine highly stimulates cephalosporin biosynthesis concomitant with the formation of arthrospores in submerged fermentations (MARTIN et al., 1986). High-producing strains were shown to synthesize precursors by particular metabolic sequences. Carboxylation of acetyl coenzyme A by oxaloacetate to yield malonyl coenzyme A and the activation of D-glucose by polyphosphate glucokinase are characteristics of some streptomycetes (QUEENER al., 1986; VAet NEK et al., 1978). These peculiar pathways enhance precursor supply in the biosyntheses of tetracyclines, erythromycin, and macrolide polyenes. Compartmentation of the precursor- and energy-generating metabolism plays an important but yet incompletely understood role in eukaryotic microorganisms. The biosynthesis of benzodiazepines by Penicillium cyclopium depends on precursor pools stored within vacuoles. Their membranes become permeable during the production phase due to the appearance of a particular permeabiliz1986). ing factor (Roos and LUCKNER,
3 The Biosynthetic Pathways
31
2.2.7 Biotechnical Production of Secondary Metabolites

For more than 50 years semi-empirical rules determined the scaling-up of microbial procedures for the production of secondary metabolites. Maximum production rate of a given secondary metabolite usually is attained below the maximum growth rate. Consequently, fermentations are carried out under partial substrate limitation. Mostly, complex nutrient sources are employed or slow feeding of substrates such as glucose which cause a vigorous development of biomass concomitant with catabolite repression of secondary pathways. But, an optimized fermentation process is characterized by moderate development of biomass. Secondary metabolism thus occurs parallel to submaximal but continuous growth. A major goal of the bioengineer is to grow high concentrations of producing biomass in the fermenter. Finally, the available oxygen concentration in the fermenter is the critical value for high productivity (CALAM, 1987; FIECHTER, 1988). Oxygen intake is dependent on fermenter geometry and impeller performance. Promotion of impeller speed increases shear stress of the mycelia and causes fragmentations and reduction of the mycelial production rate. More than other microbial processes fermentations of secondary metabolites require producer strains displaying an optimal morphological behavior under the given technical conditions. Changes of the mycelial morphology not only cause alterations in the rheological behavior of the fermentation broth, but also affect the intake of oxygen into the culture. Moreover, nutrient penetration into the cells is affected by the formation of pellets and mycelial aggregations (STEELE 1991). and STOWERS, Another serious problem in large-scale biotechnical production of secondary metabolites is heat formation. Maintenance metabolism of high biomass concentrations burns a great part of substrate without product formation. Hence, strains selected for low heat production appear particularly promising.

3.1 Precursors and the Main Biosynthetic Pathways
Secondary metabolites are formed from few starter molecules acting as precursors. They will either be modified to yield new chemical derivatives of the initial molecule or they will be coupled to oligomeric material which is subsequently modified. An outstanding variability of structures arises from the latter biosynthetic principle which combines homologous and even heterologous building moieties in a polycondensation process. Moreover, oligomeric structures once formed, such as the aglycones of macrolides, angucyclines, and anthracyclines, can be linked to other moieties like biosynthetically modified sugars. Only a few precursor structures are used in secondary metabolite formation: coenzyme A derivatives of lower fatty acids (acetyl-, propionyl-, n- and isobutyryl-CoA, etc.), mevalonate (also derived from acetyl-CoA), amino acids and shikimate, sugars (preferably glucose), and nucleosides (purines and pyrimidines). These precursors are also needed in primary metabolism to form cellular materials such as proteins, nucleic acids, cell wall constituents, and membrane lipids (ZAHNER and ZEECK,1987). Numerous biotransformations of single molecules are known, but oligomerizations of the above mentioned basic structures only occur by three pathways: glycosylation of activated sugars and polycondensations involving either activated fatty acids or amino acids.
3.2 Secondary Metabolites Formed through Biosynthetic Modifications of a Single Precursor

Structurally modified monosaccharides, such as valienamin in acarbose (MULLER, 1989), are derived from glucose via a series of
34
detectable intermediate of erythromycin bio- system is formed from the same intermediate synthesis. Altered structures of polyketides polyketide (ROHR et al., 1993). Obviously, may be engineered by point mutations within daunomycin, tetracyclines, tetracenomycines, functional domains (KATZ and DONADIO,and some angucyclines arise from nonaketide 1995), by positional alterations of domains, precursors which are cyclicized in a quite dife.g., the terminating thioesterase domain ferent manner in the course of polyketide (WIESMANN al., 1995), or by domain ex- processing (Fig. 6). Various successful atet changes (BEDFORD al., 1996; OLIYNYK et et tempts have been made to deduce the functions of proteins detected in type I1 polykeal., 1996). The genes of the aromatic type I1 polyketide biosynthetic clusters (KIM et al., 1995). tide synthases from different streptomycetes This has led to the concept of a minimal polydisplay extensive sequence homology suggest- ketide forming system containing the coning only minor differences in the substrate densing enzyme, the acyl carrier protein, and et specificity and in the sequence of reactions a malonyl-CoA transferase (MCDANIEL (O'HAGAN, 1991; DONADIO al., 1991; al., 1994). Additional proteins may then funcet HOPWOOD KHOSLA, and 1992). But the indi- tion as chain length factors determining the vidual manner of folding of the intermediate number of elongation steps and as cyclases dienzyme-bound polyketides determines in a recting the mode of cyclization (HUTCHINSON large measure what kind of cyclic aromatic and FUJII,1995). A number of new polyke-
one sugar
1 to 3 S U M R
Ho
O H 0
OM.
additiorurl lactone structures In the macrooligdides
n alkyls
c= 0 c=c
up to seven conjugated doublebonds
macrolides and polyenes
Fig. 6. Variations of polyketide structures (tetracyclines, anthracyclines, macrolides, polyethers) occurring in microorganisms. The substituents may vary in dependence of the given compound. The polyether structure shown above is highly variable with regard to the arrangement of the structural elements (rings, hydroxyketo structure, substituents).
E r '
tetracyclines
*
rug.r
anthracyclines
glycosylation by 0 to 2 sugars
Ito 8 tetrahydropyranyi and tetra-
0 -to 3 integratedrings and hemlketd structures
10 to 80 memberedring
-W
Ito 3 suga
1 "
insteadof0
polyethers
35
tides have been formed by new strains with various combinations of minimal systems and factors leading to first combinatorial biosynthetic approaches (TSOIand KHOSLA,1995; KAO et al., 1995). Without detailed structural knowledge of the proteins involved the results remain highly unpredictable (MEURER and HUTCHINSON, 1995), but exciting procedures for the generation of new compounds have been opened up (HUTCHINSON, 1994).
as 1-0-dTDP and 1-0-dUDP derivatives and mutual coupling to other activated sugars generate more than 200 oligosaccharide structures in actinomycetes (BERDYet al., 1980 BYCROFT,1988; LAATSCH,1994). Aminocyclitols and other secondary metabolites thus originate from a few sugar moieties (HOITA et al., 1995). The biosynthetic pathways leading to some therapeutically important representatives of sugar-derived structures such as streptomycin, kanamycin, and lincomycin have been investigated in detail (WRIGHT, 3.4 Terpenes 1983; PIEPERSBERG, 1994, 1995; PIEPERSBERG and DISTLER, Chapter 10, this volume). A plethora of mono-, sesqui-, di-, and tri- L-Glucosamine, streptidine, and L-streptose terpenoid structures of secondary metabolites as constitutive parts of the streptomycin molare formed from acetyl coenzyme A via me- ecule are formed via three independent, mulvalonate and isopentenyl pyrophosphate. The tistep pathways. Thus, dTDP-L-dihydrostrepinitial steps of their biosynthesis (e. g., forma- tose formation is started from 1-0-dTDP-glution of Phydroxy methylglutaryl coenzyme cose followed by dehydratation, 35-epime1-iA, isopentenyl pyrophosphate, geranyl pyro- zation, and reduction in an initial series of phosphate, farnesyl pyrophosphate) are the reactions (WRIGHT, 1983; PIEPERSBERG, same as in the formation of triterpenoid ste- 1994, 1995). Streptidine is synthesized by S. roids and hopanoids as essential cellular con- griseus from glucose via a series of at least stituents of fungi and bacteria (CANE et al. twelve enzymic steps. By linkage of the three 1992; CANE,1992, 1995). Terpenoid second- subunits hydrostreptomycin-6-0-phosphate is ary metabolites frequently occur as secondary formed intracellularly which is inactive as an 1994). During its metabolites in plants and fungi, but they are antibiotic (PIEPERSBERG, rather unusual in bacteria (see, e. g., pentale- transport through the cytoplasmic membrane nolacton, arenaemycin) (BERDYet al., 1980). outside the cells oxidation occurs and phosFinal steps of fungal terpenoid biosynthesis phate is split off to yield the active streptomy1978). Strepto(e. g., trichothecens, germacrine, aristolo- cin (WALKERand WALKER, chene, etc.) are carried out by specialized mycin biosynthesis was studied in more detail cyclases (CANE,1992). Many cyclizations inby the investigation of the pertinent genes volve the protonation or alkylation of a dou- and the corresponding enzymes (PIEPERSand ble bond or an epoxide and the ionization of BERG, 1994, 1995; PIEPERSBERG DISTan allylic diphosphate ester. Thereafter, car- LER, Chapter 10, this volume). It provides a bocationic intermediates are formed by the nice example of the formation of sugar-deelectrophilic attack of the resulting species to rived secondary metabolites. Moreover, the an olefinic bond followed by proton elimina- same biosynthetic mechanism, steps of sugar tion and a reaction with water as a nucleo- activation and transformation, are suggested phile. A series of terpenoid cyclases have to be involved in the formation of mixed-type been investigated recently by labeling and structures such as, e.g., macrolides (KATZ gene cloning experiments (CANEet al., 1992; and DONADIO,1995), anthracycline antibiotics (HUTCHINSON, 1995), and glycopeptides CANE,1992, 1995). (ZMIJEWSKI FAYERMAN, and 1995; LANCINI and CAVALLERI, Chapter 9, this volume).
3.5 Sugar-Derived Oligomeric Structures
Biotransformations of simple monosaccharides, their activation as 1-0-nucleosides such
36
tion of the peptidic bonds occurs through translocation of the growing nascent peptide Three ways of peptide bond formation are chain involving a phosphopantothenoyl carknown in secondary metabolism (KLEINKAUF rier moiety (Fig. 7). and VON DOHREN, 1990, 1996): The terminating reactions are carried out by specified enzymic subunits of the same - coupling of amino acids by single enzymes multienzyme complex. Cyclizations can occur to form small peptides with up to five ami- to form cyclo- and depsipeptides as well as reno acids (e. g., glutathione, peptidoglycan), ductions, oxidations, and methylations which - non-ribosomal biosynthesis of larger pep- introduce, e.g., a disulfide bond (see, e.g., tides (containing up to about 50 amino triostins) (VON DOHREN, 1990 BERDYet al., acids) by multienzyme complexes, and 1980) or reduce a carboxylic acid to the perti- ribosomal mechanisms. nent aldehyde (see, e. g., pepstanone) (BERDY et al., 1980). Oligopeptide biosynthesis on multienzyme A major difference of template-directed complexes (as the most important mecha- mechanisms as compared to the ribosomal nism) has been described for many bacterial formation of peptidic bonds is the acceptance products such as gramicidins, bacitracin, tyro- of non-proteinogenic amino acids and even of cidin, and fungal secondary metabolites such hydroxy acids and fatty acids either as buildas enniatins and cyclosporins (KLEINKAUF ing blocks of the oligomer formation or as and VON DOHREN, 1987,1990,1996). carbon and nitrogen terminal substituents The individual amino acids are first acti- (KLEINKAUF and VON DOHREN, 1987; VON vated via adenylate formation and thereafter DOHREN, 1990). This peculiarity of the are bound as thioesters to the non-ribosomal non-ribosomal mechanism contributes in a synthase multienzyme complex. Subsequent- particular manner to the structural diversity ly, they are coupled in a step-by-step proce- of low-molecular weight peptides produced as dure to form large polypeptides which are secondary metabolites by so many microorsometimes composed of several subunits. The ganisms (BERDYet al., 1980 BYCROFT, 1988; sequence of the amino acids in the peptide is LAATSCH, 1994). exactly the same as that of the amino acids Genetic analysis of peptide forming enactivated on the multienzyme complex (thio- zyme systems has revealed a modular structemplate-directed non-ribosomal peptide syn- ture of the enzymes involved. As in the case thesis on a protein matrix). Stepwise forma- of polyketides various degrees of integration
3.6 Oligo- and Polypeptides
Condensation domain (optional)
Z L,A-site { Carrier domaindomain 2 ~ T ~ 2

- 4 &
Carrier domain 1 ondensation domain 1 sat site (P) tr
t * -
P-site -site
domain 3
limeization domain
nidsterase domain
Exit site
Fig. 7. Schematic view of the multiple carrier protein model of enzymatic peptide formation (thiotemplate model).
37
are found, with eukaryotic systems generally being fully integrated (KLEINKAUF VON and DOHREN,1996, and Chapter 7, this volume). Genetic exchange of modules specifying amino acids or related substrates in the protein code may lead to new peptides of altered composition (STACHELHAUS al., 1995a, et b). Polypeptide-type secondary metabolites such as, e.g., microcins, tendamistat, subtilin, and lantibiotics (epidermin, gallidermin, nisin) are biosynthesized in microorganisms on the ribosomes as larger prepeptides. During their export into the medium, proteolytic processing occurs to yield the bioactive structures. A series of posttranslational alterations, such as the linkage to chromogenic and other groups, the formation of lanthionine, methyl lanthione, and disulphide units increases the number of possible homologs and creates the bioactive structures (SAHLet al., 1995; GASSON,1995; JACKet al., Chapter 8, this volume; MORENO al., 1995). et
3.7 Biosynthetic Modifications of Structures and Precursor-Directed Biosyntheses

The secondary metabolism is carried out by specified enzymes acting within the frame of long biosynthetic chains. Modified structures can frequently be obtained due to the comparably low substrate specificity of some enzymes (LUCKNER, 1989). In many cases, feeding of a tentative precursor molecule or interruption of its biosynthesis, e.g., by the addition of metabolic inhibitors, has been used successfully to direct the secondary metabolism toward the formation of one single component of a mixture of naturally occurring metabolites (precursor-directed biosynthesis) (SADAKANE al., 1983; DUTTON et al., et 1991). Some of the producer strains even accept structural homologs of the natural precursor to form unusual derivatives of the original molecule(s) (SADAKANE al., 1983; et BALDWIN al., 1991; MARTINEZ-BLANCO et et al., 1991; LUENGO, 1995). The term mutational biosynthesis signifies the use of blocked idiotrophic mutants
which are unable to carry out the complete biosynthetic pathway (SHIER et al., 1969; CLARIDGE,1983; THIERICKE and ROHR, 1993). Biosynthesis is initiated, again, when the missing intermediate is fed to the medium. Feeding of chemically derived analogs of the pertinent intermediate can yield new structural variants of the initial products. This technique was invented already in 1969 by SHIER (SHIERet al., 1969), and in a few cases (avermectins, cyclosporins) (see, e. g., DUTTON et al., 1991) more powerful compounds were obtained. Similar to the mutational biosynthesis the hybrid biosynthesis employs idiotrophic mutants which are blocked in a particular step of the secondary biosynthetic pathway (SADAKANE al., 1983). Some of et them accumulate intermediates of the interrupted biosynthetic chain due to the lack of a transforming enzymic step. Such kinds of intermediates (e.g., the protylonolide from Streptomyces fradiae) were fed to blocked mutants of another strain missing the formation of a similar intermediate (e. g,. spiramycino lactone in idiotrophs of Streptomyces ambofaciens forming spiramycin). Sometimes the fed heterologous metabolite (e. g., protylonolide) can be used in the same manner as the native metabolite (spiramycino lactone). In this way, chimeramycins were formed as hybrids of secondary metabolite structures from two different Streptomyces strains (SADAKANE et al., 1983). As was demonstrated with biosynthetic enzymes, e. g., acyltransferase and isopenicillinN-synthase of Penicillium chrysogenum, cyclosporin synthase of Beauveria niveum, enniatin synthase of Fusarium oxysporum, and gramicidin S synthases of Bacillus brevis, directed biosyntheses can also be carried out very efficiently by cell-free enzymes (BALDWIN et al., 1991; MARTINEZ-BLANCO, 1991; LAWEN and TRABER,1993; KLEINKAUF and VON DOHREN,1996). The above biocatalysts convert a series of synthetic acyl coenzyme-A derivatives and homologs of the ACV-tripeptide to form novel penicillins which have not occurred as microbial products so far. Reference should also be given here to the use of enzymes in biotransformations of secondary metabolites (see, e. g., the enzymatic hydrolyses of the side chains of penicillins and ce-
38
phalosporin C). The total synthesis of various terpenes by some plants). This is the reason cyclopeptides and depsipeptides has been car- why the search for new structures turns to unusual sources such as plants, animals, and ried out up to the milligram scale. Moreover, growing evidence attests to the microorganismsfrom special ecosystems (e.g., outstanding possibilities of molecular genetics marine animals and bacteria, special fungi, in the modification of already known struc- lichens, algae). Plants referred to in folk medtures, and in the generation of new structures icine and marine tunicates, toxic snakes, and (HOPWOOD, 1989; HOPWOOD SHERMAN, toads offer an advantageous field of research and 1990; HUTCHINSON, 1994; HUTCHINSON and on new leading structures. Moreover, the FUJII,1995). Genes of Sfrepfomyces type I1 biosynthetically available modifications of bapolyketide synthases have recently been sic structures such as macrolides, peptides, transferred to other Sfreptomyces hosts, and polyethers, etc. follow distinct rules: some the biosynthesis of new and modified aromat- derivatives occur frequently, but others are ic structures (hybrid antibiotics) is being ex- very rare. In general, the anthracyclines, e. g., ploited (HOPWOOD, 1989; HUTCHINSON and occur as glycosylated derivatives, whereas the tetracyclines are usually non-glycosylated. FUJII,1995). But previously, the dactylocyclins were detected in cultures of Ducfylosporungiumsp. as the first glycosylated representatives of the tetracycline family (TYMIAK al., 1993). et Otherwise, small structural changes of a given basic structure will often cause major changes in biological activities. The macrolide antibiotics from streptomycetes are an example which are similar in structure but possess antibacterial, antifungal, insecticidal, nemato4.1 Secondary Metabolites as cidal, immunosuppressant, and cytotoxic Products of Biological Unit properties. Traditional rescreening of comOperations pounds in newly established biological screens leads to the detection of unsuspected Starting from a few molecular structures as biological activities. In addition, chemical precursors, the secondary pathways of the mi- derivatization of side chains is an established crobial kingdom produce much more than and especially effective procedure to arrive at 10000, and the secondary pathways of plants functionally improved structures. produce more than 100OOO different chemical individuals (VERALL, 1985; GROOMBRIDGE, 1992). At first glance, this huge number ap- 4.2 Structural Classifications of pears to be incredibly high, but the observer soon recognizes that the majority of struc- Secondary Metabolites tures are representatives of some few structural classes. Many homologs of a basic strucThe large number of known secondary meture have been disclosed, not only in a given tabolites needs classification. This could be strain but also in different species and genera achieved by considering their biosynthesis, (BERDY et al., 1980; BYCROFT, 1988; the producing organisms (bacteria, fungi, LAATSCH, 1994). The detection of a novel plants, animals, etc.), their biological activistructural class of natural drugs structurally ties, and also their chemical structures. Few unrelated to the already known compounds examples can be mentioned here to show how appears to be rather rare. Some organisms the structural variability of secondary metabare characterized by the preferred production olites is channeled by classifications according of a particular secondary class of metabolites to biosynthetic origin and chemical nature (c. f., the frequent formation of polyene mac- (BERDYet al., 1980; LANCINI and LORENrolides by streptomycetesor of sesqui- and di- ZEITI, 1993).
4 Variability of Structures of Secondary Metabolites
4 Variability of Structures of Secondary Metabolites
39
Moreover, one to three sugars are attached to the non-polyene macrolide aglycones Peptidic drugs are produced by numerous which are excessively substituted by hydroxy, bacteria, fungi, plants and even animals (c.f. methoxy, methyl, and epoxy functions. Even open-chain polyenic fatty acids (e. g., the magainins and other skin antibiotics of toxic toads) (JACOB and ZASLOFF, 1994). enacyloxin) are produced by strains which Peptide antibiotics from microbial sources oc- cannot carry out the final step of lactonization et cur as linear homopeptides so far composed (WATANABE al., 1992). of a maximum of 45 amino acids (KLEINKAUF Structures of the antitumor anthracyclines and VON DOHREN,1996). Substitutions by also demonstrate the diversity which has been fatty acids are a characteristic feature of the introduced by a few modifications into a basic lipopeptides. Many cyclic peptides are known structure of a tetrahydro naphthacenequi(c. f., the cyclosporins as undecapeptides) and none backbone. Up to ten sugars are linked the amino acids are often replaced by a-and to several molecule positions. In addition, the Phydroxy acids in an irregular or even a reg- number of hydroxy, carboxymethyl, and keto ular manner (peptidolactones and depsipep- groups varies in the individual representatives to form approximately 300 different structides). Even non-proteinaceous amino acids can tures. Many aromatic polycyclic compounds are be constitutive parts of peptidic drugs (KEINKAUF and VON DOHREN,1986, 1987, 1990). also derived from the polyketide pathway. Small peptide chains can be linked to other During their biosynthesis ring closures involvunique structures such as fatty acids and chro- ing nitrogen and oxygen substituents are fremophoric groups whereas combinations with quent features. In this way, even heterocyclic sugars, macrolides, and anthracyclines seem structures such as carbazols, phenoxazins, and phenazines are formed (BERDYet al., 1980 to be unusual. 1994). Variable structures have also been unrav- BYCROFT,1988; LAATSCH In fungi mycotoxins such as the aflatoxins eled in the high-molecular weight peptide antibiotics from streptomycetes such as and ochratoxins are likewise polyketide and ALDRIDGE, 1983; lantibiotics (SAHLet al., 1995; GASSON, 1995; metabolites (TURNER et JACK al., Chapter 8, this volume) and en- BROWN al., 1996). et zyme inhibitors (subtilin, streptinoplasmin, etc.) (BERDYet al., 1980). The peptide chains are formed by ribosomal mechanisms and Terpenoid Structures posttranslational modifications create the inRich sources are plants and fungi, while dividual bioactive structures. terpenoid structures rarely occur in bacteria. Characteristic fungal terpenoids are mycotoxins such as trichothecens (BERDYet al., 1980 TURNER and ALRIDGE, 1983; LUCKNER, Polyketide Drugs 1989). Important bioactive terpenoid strucActinomycetes are rich sources of polyke- tures from plants are the vinca alkaloids (Notide metabolites like macrolides, polycyclic BLE, 1990 Fox, 1991) and taxol (HEINSTEIN 1994). Moreover, the triterpenaromatic and semi-aromatic compounds like and CHANG, tetracyclines, anthracyclines and angucy- oid steran backbone is widely distributed in natural structures such as digitalis glycosides, clines, polyethers, and ansamycins. The variability of macrolide structures in- saponins, etc. Fungal antibiotics such as fuvolves ring sizes ranging from 10 to 60 (as re- sidic acid, cephalosporin P, azasterols, and cently found in quinolidomycin) (HAYAKA- toad toxins (see, e. g., bufadienolides) are also representatives of the same structural class. WA et al., 1993). Up to seven conjugated and additionally isolated double bonds can be Marine tunicates offer a rich source of unique present in the macrocycle (BERDY et al., steroid-type molecules (CAVALIER-SMITH, 1992). 1980; B Y C R O ~ , LAATSCH, 1988; 1994). Peptides
40
Oligoglycosides Up to several hundred sugars (see, e. g., shizophyllan, lentinan, avilamycins) can be linked in a linear or even a cyclic manner. Therapeutically useful oligomeric representatives of the oligosaccharides are the aminocyclitols which contain amino inositols such as streptidine (in streptomycin) and 2-deoxystreptamin (in neomycins, gentamicins, kanamycins, istamycins, etc.) (UMEZAWAand HOOPER,1982; DIMITRIU, 1996). Nucleoside Antibiotics Approximately 150 nucleoside analogs are known from natural sources like actinomycetes and fungi. They are characterized by the presence of false nucleobases as aglycones and/or by false sugars (ISONO,1990, 1995). The biosynthetic strategies imply their formation from a normal nucleoside (see, e.g., guanosine in the biogenesis of tomaymycin) and/ or sugars such as ribose. These few examples mentioned above demonstrate that the structural characteristics of secondary metabolites can serve as a tool for their classification despite their outstanding diversity. Many compounds combine structural elements of several basic classes as can be seen in Tab. 1. Thus, biosynthetic pathway enzymes of different types interact in a highly specific manner.
5 Future Perspectives: New Products of the Secondary Metabolism

As DAVIDPERLMAN, of the pioneers one of modern industrial microbiology once stated microbial capacity is rather unlimited, and if mankind asks the microbes the proper questions they will truly answer. This idea applies to all of the living organisms when they are considered as a source of new bioactive structures. Screening for new structures is still
a growing business stimulating other fields of biotechnology and biomedicine (VERALL, 1985; MONAGHAN and TKACZ, 1990) (Tab. 1). As far as the sources of new drugs are concerned, only a small percentage of the presumed microbial world population has been explored so far, and only a minor part of the existing microbial strains has been deposited in strain collections (CHICARELLI-ROBINSON et al., 1994). The traditional sources of bioactive microbial metabolites, actinomycetes, bacilli, and sporulating fungi, still appear promising. Special genera such as the Myxobacteria (REICHENBACH HOFLE,1993) have been and demonstrated as particularly rich producers of unique structures. Future interest is focused on ecological niches and microhabitats which might harbor peculiar organisms. They look promising because they could miss special metabolic control due to their particular adaptation to the natural environment. In this context, microorganisms from extremly poor grounds, marine systems, plant rhizoand endospheres became subject to detailed investigations (OMURA, 1992). Plants still provide an apparently inexhaustible reservoir of new bioactive structures. More than in the past, increasing knowledge on the molecular, cellular, and organismic causes of diseases promotes the search for new drugs possessing more specific activity (TOMODA OMURA, and and 1990; OMURA, 1992; TANAKA OMURA, Chapter 3, this volume). Today, the screening assay determines what kind of novel natural product will be detected. In the last decade, an increasing number of publications on enzyme inhibitors and receptor antagonists attests that classical screening for simply antibiotic molecules has been extended and rationalized on the basis of modern and biochemical pharmacy and molecular biology (see Tab. 1). Even viral targets such as, e.g., viral proteinase and adhesive proteins (GP 120) became amenable to the search for new inhibitors. The following screening assays are now commonplace: Mammalian cell cultures to detect receptor agonists and antagonists, inducers and inhibitors of cytodifferentiation, and effectors of cell-to-cell and cell-to-virus interactions. Mammalian cell lines trans-
6 References
41
formed by the expression of oncogenes are used in the search for new antitumor agents promoting redifferentiation of cancer cells. A particular advantage of these assays is that they reduce animal trials which otherwise would be necessary in the development of new pharmacological agents. In the same manner the use of plant and insect cell cultures permits more rational and time-saving drug discovery in screening for new phytoeffectors and insecticides. Cloning of genes encoding for enzymes, receptors, protein factors, etc. and their expression in heterologous organisms supplies another promising approach to drug discovery. The insertion of regulatory and reporter gene sequences into microorganisms and cells may be useful for the detection of compounds which interfere with DNA-binding proteins and transcriptional regulators. In this way, new specific inhibitors of oncogenesis and viral replication may be uncovered. As a conclusion, secondary metabolism in microbes, plants, and animals still promises new leading structures for future drug development. This promise is due to the apparently inexhaustible pool of organisms and structures and the rapid development in biomedical and biochemical disease research. Moreover, secondary metabolism supplies biochemical tools which allow deeper insights into cellular processes (see, e.g., the previous discovery of inhibitors of protein kinases and protein phosphatases as effectors of the mammalian cell cycle). It seems reasonable to believe that the discovery of new leading structures of antibiotics, anticancer and antiviral drugs, pharmacological agents, crop protecting and insecticidal compounds, etc. will continue to promote the future development of biotechnology and medicine. Every new structure provides a challenge to the biochemist exploring its mode of action, to the chemist wanting to disclose the structure-activity relationships, to the pharmacologist studying the activity in macroorganisms, to the molecular biologist investigating the genes of the biosynthetic pathway, to the fermentation engineer developing a new biotechnical procedure, and last but not least, to those who could benefit for their health from new and better drugs.
Acknowledgements We are gratefully indebted to Dr. HANSPETER SALUZfor helpful comments and critical manuscript revision and to Dr. AXEL BRAKHAGE helpful comments and part of for Fig. 3.
6 References
AHARONOWITZ, COHEN,G., MART~N, F. Y., J. (1992), Penicillin and cephalosporin biosynthetic genes: structure, organization, regulation, and evolution, Annu. Rev. Microbiol. 46, 461495. I. D. ALLEN, W., RITCHIE, A. (1994), Cloning and analysis of DNA sequences from Streptomyces hygroscopicus encoding geldanamycin biosynthesis, MoC. Gen. Genet. 243, 593-599. ALTENBUCHNER, J. (1994), Hohe genetische Instabilitat von Streptomyceten durch chromosomale Deletionen und DNA-Amplifikationen, Bio/Engineering 3, 33-46. K., ARISAWA, TSUNEKAWA, OKAMURA, A., H., OKAMOTO,R. (1995), Nucleotide sequence analysis of the carbomycin biosynthetic genes including the 3-0-acyltransferase gene from Streptomyces thermotolerans, Biosci. Biotech. Biochem. 59,582-588. G ARMSTRONG,. A. (1994), Eubacteria show their true colors: genetics of carotinoid pigment biosynthesis from microbes to plants, J. Bacteriol. 176,47954802. F., ARROWSMITH, J., MALPARTIDA, SHERMAN, T. A., D. D. H., BIRCH, HOPWOOD, A., ROBINSON, J. A. (1992), Characterization of actl-homologous DNA encoding polyketide synthase genes from the monensin producer Streptomyces cinnamonensis, Mol. Gen. Genet. 234,256264. ARST,H. N., JR. (1996), Regulation of gene expression by pH, in: The Mycota III (ESSER, K., LEMKE, A,, Eds.), Biochemistry and MolecuP. Car Biology (BRAMBLE, MARZLUFF, A., R., G. Eds.), pp. 235-240. Berlin, Heidelberg: SpringerVerlag. AZUMA, (1987), Developments of immunostimuI. lants in Japan, in: Immunostimulants Now and Tomorrow (AZUMA, JOLLES,G., Eds.), pp. I., 41-56. Tokyo, Berlin: Japan Scientific Press/ Springer. AZUMA,M., HORI, K., HASHI,Y.-O., YOSHIDA, M., HORINOUCHI, BEPPU,T. (1980), BasidifS., ferquinone, a new inducer of fruiting-body for-
42
I General Aspects of Secondary Metabolisn
T. mation of a basidiomycete Farvolus arcularius BEPPU, (1992), Secondary metabolites as chemical signals for cellular differentiation, Gene 115, from a Streptomyces strain. 11. Structure of ba159-165. sidifferquinone, Agric. Biol. Chem. 54, 14471452. BEPPU,T. (1995), Signal transduction and secondary metabolism: prospects for controlling proBAILEY, M., KERSHAW, J., HUNT,B. A., PAA. M. ductivity, Trends Biotechnol. 13, 264-269. I. A. S. TERSON, C., CHARNLEY, K., REYNOLDS, J., A., M., E., CLARKSON, M. (1996), Cloning and se- BERDY, ASZALOS, BOSTIAN, MCNITT, J. K. L. (1980), CRC Handbook of Antibiotic quence analysis of an intron-containing domain Compounds, 1980-1984. Boca Raton, FL: CRC from a peptide synthetase of the entomopathoPress. genic fungus Metarhizium anisopliae, Gene 173, BEYER,S., DISTLER, PIEPERSBERG, (1996), 195-197. J., W. The Str cluster for the biosynthesis of 5'-hydroxyBAINTON, J., STEAD,P., CHABRA, R., BYN. S. streptomycin in Streptomyces glaucescens G. CROFT,B. W., SALMOND, P. C., STEWARD, Gla.O(ETH 22794) - new operons and evidence G. S. A. B., WILLIAMS, (1992), N-(3-OxohexP. for pathway specific regulation by StrR, Mol. anoyl)-L-homoserine lactone regulates carbaGen. Genet. 250,775-784. penem antibiotic production in Envinia carotoBIBB,M. J., SHERMAN, H., OMURA, HOPD. S., vora, Biochem. J. 288, 977-1004. WOOD, A. (1994), Cloning, sequencing and D. BALDWIN, E., BRADLEY, ABBOTT,S. D., J. M., deduced functions of a cluster of Streptomyces ADLINGTON, M. (1991), New penicillins from R. genes probably encoding biosynthesis of the poisopenicillin N-synthase, Tetrahedron 47, 5309lyketide antibiotic frenolicin, Gene 142, 31-39. 5328. BIRCH, W., ROBINSON, A. (1995), Polyethers, A. J. BARABAS, PENYIGE, HIRANO, (1994), G., A., T. in: Genetics and Biochemistry of Antibiotic ProHormone-like factors influencing differentiation L. duction (VINING, C., STUITARD, C., Eds.), of Streptomyces cultures, FEMS Microb. Rev. 14, pp. 443476. Boston, M A Butterworth-Heine75-82. mann. BARBER, CHAPMAN, C., HOWARD, D., J., A. T. BIRCH, W., LEISER, ROBINSON, A. (1993), A. A., J. TEBB, G. (1988), Recycling of polyketides by Cloning, sequencing, and expression of the gene fungi: the degradation of citrinin by Penicillium encoding methylmalonyl coenzyme A mutase citrinum, Appl. Microb. Biotechnol. 29, 387from Streptomyces cinnamonensis, J. Bacteriol. 391. 175,3511-3519. BEDFORD, JACOBSEN, Luo, G., CANE, D., J. R., D. BLACK, A., WOLK,C. P. (1994), Analysis of a T. E., KHOSLA, (1996), A functional chimeric C. Het(-) mutation in Anabaena Sp strain PCC modular polyketide synthase generated via do7120 implicates a secondary metabolite in the main replacement, Chemistry & Biology 3, 827regulation of heterocyst spacing, J. Bacteriol. 831. 176,2282-2292. BECHTHOLD, SOHNG, K., SMITH, M., CHU, BORMANN, A., J. T. C., MOHRLE, V., BRUNTNER,C. X., FLOSS, G. (1995), Identification of StrepH. (1996), Cloning and heterologous expression of tomyces violaceoruber Tii22 genes involved in the entire set of structural genes for nikkomycin the biosynthesis of granaticin, Mol. Gen. Genet. synthesis from Streptomyces tendae Tii901 in 248,610-620. Streptomyces lividans, J. Bacteriol. 178, 1216BECK,J., RIPKA,S., SIEGNER, SCHILTZ, A., E., 1218. E. SCHWEIZER, (1990), The multifunctional 6- BRAKHAGE, A., TURNER, (1995), BiotechniA. G. methylsalicylic acid synthase gene of Penicillium cal genetics of antibiotic biosynthesis, in: The patulum. Its gene structure relative to that of P. Mycota IZ (ESSER,K., LEMKE, A., Eds.), Geother polyketide synthases, Eur. J. Biochem. netics and Biotechnology (KUCK, U., Ed.), pp. 192,487498. 263-385. Berlin, Heidelberg: Springer-Verlag. BBHAL, HUNTER, S. (1995) Tetracyclines, in: BRAKHAGE, A., VAN DEN BRULLE, (1995), V., I. A. J. Genetics and Biochemistry of Antibiotic ProducUse of reporter genes to identify recessive transL. C., tion (VINING, C., STUTTARD, Eds.), pp. acting mutations specifically involved in the reg359-384. Boston: Butterworth-Heinemann. ulation of Aspergillus nidulans penicillin biosynBENDER,C., PALMER,D., PENALOZAVAZQUEZ, thetic genes, J. Bacteriol. 177, 2781-2788. A. A., RANGASWAMY, ULLRICH, (1996), BRAKHAGE, A., BROWNE,P., TURNER,G. V., M. Biosynthesis of coronatine, a thermoregulated (1992), Regulation of Aspergillus nidulans peniphytotoxin produced by the phytopathogen cillin biosynthesis and penicillin biosynthesis genes acvA and ipnA by glucose, J. Bacteriol. Pseudomonas syringae, Arch. Microbiol. 166, 174,3789-3799. 71-75.
6 References
43
BROWN, W., Yu, J.-H., KELKAR, S., FER- CHEN,C. W., LIN, Y.-S., YANG,Y.-L., Tsou, M.D. H. F., CHANG, H.-M., KIESER, M., HOPWOOD, H. NANDES, M., NESBITT, C., KELLER,N. P., T. D. A. (1994), The linear chromosomes of StrepADAMS, H., LEONARD, J. (1996), TwentyT. T. tomyces: structure and dynamics, Actinomycetofive coregulated transcripts define a sterigmatologica 8, 103-112. cystin gene cluster in Aspergillus nidulans, Proc. A. M. CHEN,Q., WERTHEIMER, M., TOLMASKY, Natl. Acad. Sci. USA 93, 1418-1422. J. BRUCKNER, BLECHSCHMIDT, SEMBDNER, E., CROSA, H. (1996), The AngR protein and B., D., the siderophore anguibactin positively regulate G. SCHNEIDER, (1990), Fungal gibberellin G. the expression of iron-transport genes in Vibrio production, in: Biotechnology of Vitamins, Piganguillarum, Mol. Microbiol. 22, 127-134. ments, and Growth Factors (VANDAMME, E., J.-S., XIA, T.-H., MEI, B.-G., JIN, Z. K., Ed.), pp. 383-429. London, New York: Else- CHIAO, Gu, W.-L. (1995), Ansamycins, in: Genetics and vier. Biochemistry of Antibiotic Production (VINING, BUADES, MOYA, (1996), Phylogenetic analyC., A. L. C., STUTTARD, Eds.), pp. 477-498. BosC., sis of isopenicillin N synthetase horizontal gene ton, MA: Butterworth-Heinemann. transfer, J. Mol. Evol. 42, 537-542. M. BUSHELL, E., FRYDAY, (1983), The applica- CHICARELLI-ROBINSON,I., Fox, F., NISBETH, M. A. C. (1994), The value of microbial screening in tion of materials balancing to the characterizadrug discovery (abstract), in: New Strategies in tion of sequential secondary metabolite formaSearching Drug Leads, pp. 1-3. Airport Sympotion in Streptomyces cattleya NRRL 8057, J. Gen. sium, Frankfurt, 28 February. Microbiol. 129, 1733-1741. G. BYCROFT, J. (1988), Dictionary of Antibiotics. CHU, Y. W., RENNO,D., SAUNDERS, (1995), B. Detection of a protein which binds specificallyto London, New York: Chapman and Hall. the upstream region of the pcbAB gene in PeniCALAM, T. (1987), Process Design and Control C. cillium chrysogenum, Curr. Genet. 28, 184-189. in Antibiotic Fermentations. Cambridge: CamCLARIDGE, A. (1983), Mutasynthesis and diC. bridge University Press. rected biosynthesis for the production of new CAMPBELL, C. (Ed.) (1989). Ivermectin and AvW. antibiotics, in: Basic Biology and New Developermectin. New York, Heidelberg: Springer-VerA., ments in Biotechnology (HOLAENDER, LASlag. P., KIN,A. I., ROGERS, Eds.), pp. 231-269. New CANE, E. (1992), Terpenoid cyclases: design and D. York: Plenum Press. function of electrophilic catalysts, in: Secondary C. Metabolites: Their Function and Evolution, Ciba COFFEY,T. J., DOWSON, G., DANIELS,M., SPRATT, G. (1995), Genetics and molecular B. D. Foundation Symposium 171 (CHADWICK, J., biology of beta-lactam-resistant Pneumococci, WHELAN,J., Eds.), pp. 163-168. Chichester, Microb. Drug Res. - Mechanisms, Epidemiology New York: J. Wiley & Sons. and Disease 1,29-34. CANE,D. E. (1995), Isoprenoid antibiotics, in: GeEDITORIAL (1992), Medicaments antibionetics and Biochemistry of Antibiotic Production COMITB tiques. Traite de chimie therapeutique, Vol. 2. (VINING, C., STUTTARD, Eds.), pp. 633L. C., Paris: Tec & Doc Lavosier. 656. Boston, MA: Butterworth-Heinemann. D. L. CANE, E., YANG, COATES, M., PYNN, D. G., R. H., COOK,R. J., THOASHOW, S., WELLER, M., FUJIMOTO, MAZZOLA, D., M., BANGERA, B., HOLM,T. M. (1992), Trichodiene synthase. SynKIM, D. (1995) Molecular mechanisms of deergistic inhibition by inorganic phosphate and fense by rhizobacteria against root disease, Proc. aza-analogs of the bisabolyl cation, J. Org. Natl. Acad. Sci. USA 92, 4197-4201. Chem. 51, 3454-3462. J. CASHEL, (1975), Regulation of bacterial ppGpp COQUE,J. J. R., LIRAS,P., MART~N, F. (1993), M. Genes for a p-lactamase, a pencillin binding proand pppGpp, Ann. Rev. Microbiol. 29, 301-335. tein and a transmembrane protein are clustered CAVALIER-SMITH,(1992), Origins of secondary T. with the cephamycin biosynthetic genes in Nometabolism, in: Secondary Metabolites: Their cardia lactamdurans, EMBO J. 12, 631-639. Function and Evolution, Ciba Foundation SymJ. F. J. D. J., COQUE, J. R., ENGUITA, J., MARTIN, F., LIposium 171 (CHADWICK, J., WHELAN, RAS,P. (1995a), A two protein component 7-ceEds.), pp. 64-87. Chichester, New York: J. Wiley phem-methoxylase encoded by two genes of the & Sons. cephamycin C cluster converts cephalosporin to CHATER, F. (1992). Genetic regulation of secK. 7-methoxycephalosporin C, J. Bacteriol. 177, ondary metabolic pathways in streptomyces, in: 2230-2235. Secondary Metabolites: Their Function and EvoJ. F. lution, Ciba Foundation Symposium I71 (CHAD- COQUE, J. R., PEREZ-LLARENA, J., ENGUITA, F. J., FUENTE, L., MARTfN, J. F., LIRAS, J. P. WICK, D. J., WHELAN, J., Eds.), pp. 144-162. (1995b), Characterization of the cmcH genes of Chichester, New York J. Wiley & Sons.
44
1 General Aspects of Secondary Metabolism1
Nocardia lactamdurans encoding a functional 3hydroxymethylcephem 0-carbamoyltransferase for cephamycin biosynthesis, Gene 162,21-27. COSIMA,P., RODRIGUEZ, DE FERRA, F., F., G., GRANDI, PEREGO, G., M., VENEMA, VAN SINDEREN, (1993), Sequence and analysis of D. the genetic locus responsible for surfactin synthesis in Bacillus subtilis, Mol. Microbiol. 8, 821831. CUNDLIFFE, (1989), How do antibiotic-producE. ing organisms avoid suicide, Annu. Rev. Microbiol. 43, 207-233. CUNDLIFFE, (1992), Self-protection mechanisms E. in antibiotic producers, in: Secondary Metabolites: Their Function and Evolution, Ciba FoundaD. tion Symposium 171 (CHADWICK, J., WHELAN,J., Eds.), pp. 199-214. Chichester, New York: J. Wiley & Sons. DAVIES, (1994), Inactivation of antibiotics and J. the dissemination of resistance genes, Science 264,375-382. DAVIS, K., CHATER, F. (1990), Spore colour N. K. in Streptomyces coelicolor A3(2) involves the developmentally regulated synthesis of a compound biosynthetically related to polyketide antibiotics, Mol. Microbiol. 4, 1679-1691. DAVIS, K., CHATER, F. (1992), The StreptoN. K. myces coelicolor whiB gene encodes a small transcription factor-like protein dispensable for growth but essential for sporulation, Mol. Gen. Genet. 232,351-358. DECKER, ROHR,J., MOTAMEDI, ZAHNER, H., H., H., HUTCHINSON, R. (1995), Identification of C. Streptomyces olivaceus Tii2353 genes involved in the production of the polyketide elloramycin, Gene 166,121-126. DEMAIN, L. (1974), Nutrition and the function A. of secondary metabolites, Adv. Appl. Microbiol. 16, 177-202. DEMAIN, L. (1984), Biology of antibiotic formaA. tion, in: Biotechnology of Industrial Antibiotics (VANDAMME, J., Ed.), pp. 33-39. New York E. Marcel Dekker. DEMAIN, L. (1989), Carbon source regulation of A. idiolite biosynthesis in actinomycetes, in: Regulation of Secondary Metabolism in Actinomycetes (SHAPIRO, Ed.), pp. 127-134. Boca Raton, S., F L CRC Press. DEMAIN, L. (1992), Microbial secondary metabA. olism: a new theoretical frontier for academia, a new opportunity for industry, in: Secondary Metabolites: Their Function and Evolution, Ciba D. Foundation Symposium 171 (CHADWICK, J., J., WHELAN, Eds.). Chichester, New York: J. Wiley & Sons. 0. DE RUYTER, G. G. A., KUIPERS, P., BEERP. I., THUYZEN, M., VAN ALENBOERRIGTER, M.
DEVOS,W. M. (1996), Functional analysis of promoters in the nisin gene cluster of Lactobacillus lactis, J. Bacteriol. 178, 3434-3439. DICKENS, L., YE, J. S., STROHL, R. (1996), M. W. Cloning, sequencing, and analysis of aklaviketone reductase from Streptomyces sp. strain C5, J. Bacteriol. 178, 3384-3388. D~Ez,B., GUTIBRREZ, BARREDO, L., VAN s., J. SOLINGEN, VAN DER VOORT, L. H. M., P., MARTfN, J. F. (1990), The cluster of penicillin biosynthetic genes. Identification and characterization of the pcbAB gene encoding a-aminodipyl-cysteinyl-valine synthetase and linkage to the pcbC and penDE genes, J. Biol. Chem. 265, 16358-16365. DIMITRIU, (Ed.) (1996), Polysaccharides in MedS. ical Applications. New York, Basel, Hongkong: Marcel Dekker. DISTLER, EBERT,A., HEINZEL, MANSOUW., P., G., K. RI,K., MAYER, PISSOWOTZKI, (1988), Expression of genes for streptomycin biosynthesis, Y., in: Biology of Actinomycetes 88 (OKAMI, H., BEPPU,T., OGAWARA, Eds.), pp. 86-91. Toyko: Japan Scientific Society Press. DISTLER, MANSOURI, STOCKMANN, PIEJ., K., M., PERSBERG, (1992), Streptomycin biosyntheW. sis and its regulation in Streptomycetes, Gene 115,105-111. DONADIO, STARER,M. J., MCALPINE, B., S., J. S. SWANSON, J., KATZ,L. (1991), Modular organization of genes required for complex polyketide biosynthesis, Science 252,675-679. DOULL, L., VINING, C. (1995), Global physioJ. L. logical controls, in: Genetics and Biochemistry of L. Antibiotic Production (VINING, C., STUTTARD,C., Eds.), pp. 9-64. Boston, MA: Butterworth-Heinemann. DSOUZA, NAKANO, M., ZUBER, (1994), C., M. P. Identification of comS, a gene of the srfA operon that regulates the establishment of genetic competence in Bacillus subtilis, Proc. Natl. Acad. Sci. USA 91, 9397-9401. DUITON, C. J., GIBSON,S. P., GOODIE,A. C., HOLDOM, S., PACEY,M. S., RUDDOCK, K. J., BULOCK, D., RICHARDS, K. (1991), Novel J. M. avermectins produced by mutational biosynthesis, J. Antibiot. 44, 357-365. DYSON, SCHREMPF, (1987), Genetic instabiliP., H. ty and DNA-amplification in Streptomyces lividuns 66,J. Bacteriol. 169, 47694803. ELDER, J. H., SCHNOLZER, M., HASSELKUSD. LIGHT, C. S., HENSON,M., LERNER, A., T. P. PHILLIPS, R., WAGAMAN, C., KENT,S. B. H. (1993), Identification of proteolytic processing sites within the gag and pol polyproteins of feline immunodeficiency virus, J. Virol. 67,18691876.
6 References
ELLIS, D. D., ZELDIN,E. L., BRODHAGEN, M., W. RUSSIN, A., M c C o w , B. H. (1996), Taxol production in nodule cultures of Taxus, J. Nat. Prod. 59, 246-250. ENDO,A. (1979), Monacolin K, a new hypercholesterolemic agent produced by Monascus species, J. Antibiot. 32, 852-854. ESPESO,E. A., PENALVA,M. A. (1996), Three binding sites for the Aspergillus nidulans PacC zinc-finger transcription factor are necessary and sufficient for regulation by ambient pH of the isopenicillin N synthase gene promoter, J. Biol. Chem. 271,28825-28830. ESPESO, A., TILBURN, ARST, H. N., JR., E. J., PENALVA, A. (1993), pH regulation is a maM. jor determinant in expression of a fungal biosynthetic gene, EMBO J. 12,3947-3956. FENG, FRIEDLIN, MARZLUFF, A. (1995), B., E., G. Nuclear DNA-binding proteins which recognize the intergenic control region of penicillin biosynthetic genes, Curr. Genet. 27, 351-358. FERNANDEZ-MORENO, A., VALL~N, MALM. C., F. PARTIDA, (1996), Isolation and characterization of streptothricin biosynthetic genes from a newly Streptomyces isolated strain, Proc. Con$ Biol. Streptomycetes, Ohrbeck, Abstract V51. FIECHTER, (1988), Bioprocess development, in: A. Overproduction of Macrolide Metabolites (VANEK, Z., HOSTALEK, Eds.), pp. 231-259. BosZ., ton, London: Buttenvorth. FIERRO,F., BARREDO, L., DIEZ, B., GUTIERJ. REZ, S., FERNANDEZ, J., MARTIN, J. F. F. (1995), The penicillin gene cluster is amplified in tandem repeats linked by conserved hexanucleotide sequences, Proc. Natl. Acad. Sci. USA 92, 62004204. FIERRO,F., MONTENEGRO, GUTIERREZ, E., S., MART~N, F. (1996), Mutants blocked in peniJ. cillin biosynthesis show a deletion of the entire penicillin gene cluster at a specific site within a conserved hexanucleotide sequence, Appl. Microbiol. Biotechnol. 44, 597-604. FILIPPINI,S., SOLINAS, M. M., BREME, U., SCHLUTER, B., GABELLINI, BIAMONTI, M. D., L. G., COLOMBO, L., GAROFANO, (1995), A. Streptomyces peucetius daunorubicin biosynthesis gene, dnrR sequence and heterologous expression, Microbiology 141, 1007-1016. Fox, B. W. (1991), Natural products in cancer treatment from bench to the clinic, Trans. R. SOC. Trop. Med. Hyg. 85,22-85. FUQUA, C., WINANS, C., GREENBERG, P. W. S. E. (1994), Quorum sensing in bacteria: the LuxRLux I family of cell-density-responsive transcriptional regulators, J. Bacteriol. 176, 269-275. GASSON, J. (1995), Lantibiotics, in: Genetics and M. Biochemistry of Antibiotic Production (VINING,
45
L. C., STUTTARD, Eds.), pp. 283-306. BosC., ton, MA: Butterworth-Heinemann. GEIGER,0. (1994), N-Acyl-homoserinlacton-Autoinduktor-Signalmolekule bei gramnegativen Bakterien, BioEngineering 5, 40-46. GOODAY,G . W. (1974), Fungal sex hormones, Ann. Rev. Biochem. 1974,3549. GOODRICHTANRIKULU, MAHONEY, M., M. E., RODRIGUEZ, B. (1995), The plant growth regS. ulator methyl jasmonate inhibits aflatoxin production by Aspergillus flavus, Microbiology 141, 2831-2837. GRAFE,U. (1989), Autoregulatory secondary metabolites from actinomycetes, in: Regulation of Secondary Metabolism in Actinomycetes (SHAPIRO, S., Ed.), pp. 75-126. Boca Raton, F L CRC Press. GRAFE,U. (1992), Biochemie der Antibiotika. Heidelberg: Spektrum Verlag. K., GRIMM, MADDURI, ALI, A., HUTCHINSON, A., C. R. (1994), Characterization of the Streptomyces peucetius ATCC 29050 genes encoding doxorubicin polyketide synthase, Gene 151, 110. GROOMBRIDGE, (Ed.) (1992), Global BiodiverB. sity. Status of the Earths Living Resources. London: Chapman and Hall. HAAS,H., MARZLUFF, A. (1995), NRE, the maG. jor nitrogen regulatory protein of Penicillium chrysogenum, binds specifically to elements in the intergenic promoter regions of nitrate assimilation and penicillin biosynthetic gene clusters, Curr. Genet. 28, 177-183. HAMOEN, W., ESHUIS, JONGBLOED, VENL. H., J., D. EMA, G., VAN SINDEREN, (1995), A small gene, designated comS, located within the coding region of the fourth amino acid-activation domain of srfA, is required for competence development in Bacillus subtilis, Mol. Microbiol. 15, 55-63. HANLON, W., HODGES, A. (1981), Bacitracin G. N. and protease production in relation to sporulation during exponential growth of Bacillus lichenijormis on poorly utilized carbon and nitrogen source, J. Bacteriol. 147, 427-431. HASUMI,K. (1993), Competitive inhibition of squalene synthetase by squalestatin-I, J. Antibiot. 46,689-691. HAYAKAWA, MATSUOKA, Y., M., SHIN-YA,K., SETO,H. (1993), Quinolidomycins, A,, AZ, and B1, novel 60-membered macrolide antibiotics. I. Taxonomy, fermentation, isolation, physicochemical properties and biological activity, J. Antibiot. 46, 1557-1562. HAYASHI, TARUI, MURAO, (1985), IsolaH., N., S. tion and identification of cyclooctasulfur, a fruiting-body-inducing substance, produced by Strep-
46
tomyces albulus T0447, Agric. Biol. Chem. 49, HORINOUCHI, BEPPU,T. (1990), AutoregulatoS., 101. ry factors of secondary metabolism and morphogenesis in actinomycetes, CRC Crit. Rev. BioHAYDOCK, F., APARICIO, F., MOLNAR,I., S. J. technol. 10,191-204. SCHWECKE, KHAW, L. E., K ~ N I GA., T., , S., T. I. MARSDEN, F. A., GALLOWAY, S., STAUN- HORINOUCHI, BEPPU, (1992a), Regulation of A. secondary metabolism and cell differentiation in P. TON,J., LEADLAY, F. (1995), Divergent sestreptomycetes:A-factor as a microbial hormone quence motifs correlated with the substrate speand the AfsR protein as a component of a twocificity of (methy1)malonyl-CoA-acyl carrier component regulatory system, Gene 115, 167protein transacylase domains in modular polyke172. tide syntheses, FEBS Lett. 374, 246-248. S., HEINSTEIN, F., CHANG, J. (1994), Taxol, Ann. HORINOUCHI, BEPPU,T. (1992b), AutoregulaP. C. tory factors and communication in actinomyRev. Plant. Physiol. Plant Mol. Biol. 45, 663cetes, Annu. Rev. Microbiol. 46, 377-393. 674. S., Z., HERZOG-VELIKONJA, PODLESEK, GRAB- HORINOUCHI, BEPPU,T. (1995), AutoregulaB., tors, in: Genetics and Biochemistry of Antibiotic NAR, M. (1994), Isolation and characterization C., Production (VINING,L. C., STUTTARD, of Tn917-generated bacitracin deficient mutants Eds.), pp. 103-120. Boston, MA: Butterworthof Bacillus licheniformis, FEMS Microbiol. Lett. Heinemann. 121,147-152. S., M., HIRAMATSU, (1995), Molecular evolution of HORINOUCHI, KITO, M., NISHIYAMA, FuK. RUYA, K., HONG,S. K., MIYAKE, BEPPU,T. K., MRSA, Microbiol. Immunol. 39, 531-543. (1990), Primary structure of AfsR, a global reguHODGSON, E., FOSBERRY, P., RAWLINSON, J. A. latory protein for secondary metabolite formaN. S., Ross, H. N. M., NEAL,R. J., ARNELL, J. tion in Streptomyces coelicolor A3(2), Gene 95, C., EARL,A. J., LAWLOR, J. (1995), ClavuE. 49-56. lanic acid biosynthesis in Streptomyces clavuligeU., rus: gene cloning and characterization, Gene 166, HORNEMANN, ZHANG,X. Y., OTTO, J. C. (1993), Transferable Streptomyces DNA amplifi49-55. cation and coamplification of foreign DNA seHOPWOOD, A. (1988), Towards an understandD. quences, J. Bacteriol. 175, 1126-1133. ing of gene switching in Streptomyces, the basis HOTTA, K., DAVIES,J., YAGISAWA, (1995), M. of sporulation and antibiotic production, Proc. Aminoglycosides and aminocyclitols, in: GeR. SOC. London B: Biological Sciences 235, 121netics and Biochemistry of Antibiotic Production 138. (VINING, C., STU-ITARD, Eds.), pp. 571L. C., HOPWOOD, A. (1989), Antibiotics: opportuniD. 596. Boston, MA: Butterworth-Heinemann. ties for genetic manipulation, Philos. Trans. R. HUNDLE,B., ALBERTI,M., NIEVELSTEIN, V., SOC. (London) 324,549-562. BEYER,P., KLEINIG, ARMSTRONG, A., H., G. HOPWOOD, A., KHOSLA, (1992), Genes for D. C. J. BURKE, H., HEARST, E. (1994), Functional D. polyketide secondary metabolic pathways in miassignment of Erwinia herbicola EholO caroticroorganisms and plants, in: Secondary Metabolnoid genes expressed in Escherichia coli, Mol. ites: Their Function and Evolution, Ciba FoundaGen. Genet. 245, 406-416. D. tion Symposium 171 (CHADWICK, J., WHE- HUTTER,R. A. (1986), Overproduction of microLAN, J., Eds.), pp. 88-112. Chichester, New bial metabolites, in: Biotechnology, 1st Edn., York J. Wiley & Sons. Vol. 4 (REHM,H. J., REED,G., Eds.), pp. 3-14. HOPWOOD, A., SHERMAN, A. (1990), MolecD. D. Weinheim: VCH. ular genetics of polyketides and its comparison HUTCHINSON, R. (1994), Drug synthesis by geC. to fatty acid biosynthesis, Annu. Rev. Genet. 24, netically engineered microorganisms, BiolTech37-66. nology 12,375-380. HOPWOOD, A., KIESER,T., WREIGHT,E. M., HUTCHINSON, R. (1995), Anthracyclines, in: GeD. C. BIBB,M. J. (1983), Plasmids, recombination and netics and Biochemistry of Antibiotic Production chromosome mapping in Streptomyces lividans (VINING, C., S T U ~ A R D , Eds.), pp. 331L. C., 66, J. Gen. Microb. 129, 2257-2269. 358. Boston, MA: Butterworth-Heinemann. HOPWOOD, A., CHATER,K. F., BIBB, M. J. HUTCHINSON, R., FUJII,I. (1995), Polyketide D. C. (1995), Genetics and antibiotic production in synthase gene manipulation: a structure-funcStreptomyces coelicolor A3(2), a model Streptotion approach in engineering novel antibiotics, mycete, in: Genetics and Biochemistry of AntiAnnu. Rev. Microbiol. 49,201-238. L. c., ISHIZUKA, HORINOUCHI, KIESER,H. M., biotic Production (VINING, c., STUTTARD, H., S., Eds.), pp. 65-102. Boston, MA: ButterworthD., HOPWOOD, BEPPU,T. (1992), A putative Heinemann. two component regulatory system involved in
6 References
47
secondary metabolism in Streptomyces species, J. KELLER, P., ADAMS, H. (1995), Analysis of a N. T. Bacteriol. 174, 7585-7594. mycotoxin gene cluster in Aspergillus nidulans, SAAS Bull. Biochem. Biotechnol. 8, 14-21. ISONO, (199O), Nucleoside antibiotics: Structure, K. A. biological activity, and biosynthesis, J. Antibiot. KHOKHLOV, S. (1982), Low molecular weight 41, 1711-1752. microbial bioregulators of secondary metabolism, in: Overproduction of Microbial Products ISONO,K. (1995), Polyoxins and related nucleotides, in: Genetics and Biochemistry of Antibiotic (KRUMPHANZ, SIKYTA,B., VANEK, Z., V., Production (VINING, L. C., STUITARD, C., Eds.), pp. 97-109. London: Academic Press. Eds.), pp. 597418. Boston, M A Butterworth- KIM, H. S., TADA,H., NIHIRA,T., YAMADA, Y. Heinemann. (1990), Purification and characterization of virginiae butanolide C-binding protein, a possible JACOB,L., ZASLOFF, (1994), Potential theraM. pleotropic signal-transducer in Streptomyces virpeutic applications of magainins and other anginiae, J. Antibiot. 43, 692-706. timicrobial agents of animal origin, in: Antimicrobial Peptides, Ciba Foundation Symposium KIM,E. S., BIBB,M. J., BUTLER, J., HOPWOOD, M. 186 (BOMAN, G., MARSH, GOODE, A., H. J., J. D. A., SHERMAN, H. (1994), Sequences of D. Eds.), Chichester: J. Wiley & Sons. the oxytetracycline polyketide synthase-encoding otc genes from Streptomyces rimosus, Gene JENSEN, E., DEMAIN, L. (1995), Beta-lactams, S. A. 141, 141-142. in: Genetics and Biochemistry of Antibiotic ProL. duction (VINING, C., STUTTARD,C., Eds.), KIM,E.-S., CRAMER, D., SHREVE, L., SHERK. A. pp. 239-268. Boston, MA: Butterworth-HeineMAN, D. H. (1995), Heterologous expression of mann. an engineered biosynthetic pathway: functional dissection of type I1 polyketide synthase compoW. JENSEN, R., FENICAL, (1994), Strategies for P. the discovery of secondary metabolites from nents in Streptomyces species, J. Bacteriol. 177, 1202-1207. bacteria: ecological perspectives, Annu. Rev. Microbiol. #, 559-584. KLEINKAUF, VON DOHREN, (1986), Peptide H., H. Antibiotics, in: Regulation of Secondary MetabR., J. JOHNSON, ADAMS, (1992), The ecology and olite Formation (KLEINKAUF, VON DOHREN, H., evolution of tetracycline resistance, TREE 7 , H., DORNAUER, NESEMANN, Eds.), pp. H., G., 295-299. 173-207. Weinheim: VCH. JOHNSON, F., MCKNIGHT, L. (1989), EukaP. S. H. H., ryotic transcriptional regulatory proteins, Annu. KLEINKAUF, VON DOHREN, (1987), Biosynthesis of peptide antibiotics, Annu Rev. MicroRev. Biochem. 58,799-839. biol. 41, 259-289. KAHAN, D. (Ed.) (1987), Cyclosporins. London: B. KLEINKAUF, VON DOHREN, (Eds.) (199O), H., H. Grune & Stratton. Biochemistry of Peptide Antibiotics. Recent AdKAJIMURA, Y., SUGUYAMA, KANDEA,M. M., vances in the Biochemistry of o-Lactams and Mi(1999, Bacillopeptins, new cyclic lipopeptide crobial Bioactive Peptides. Berlin: DeGruyter. antibiotics from Bacillus subtilis FR-2, J. AntiKLEINKAUF, VON DOHREN, (1996), A nonH., H. biot. #, 1095-1103. ribosomal system of peptide biosynthesis, Eur. J. KAO, C. M., Luo, G., KATZ, L., CANE,D. E., Biochem. 236, 335-51. KHOSLA, (1999, Manipulation of macrolide C. H. H., ring size by directed mutagenesis of a modular KLEINKAUF, VON DOHREN, (in press), Biopolyketide synthase, J. Am. Chem. SOC. 117, synthesis of cyclosporins and related peptides, T., 9105-91 06. in: Fungal Biotechnology (ANKE, Ed.). London: Chapman and Hall. KATZ,L., DONADIO, (1995), Macrolides, in: GeS. K., netics and Biochemistry of Antibiotic Production KOHMOTO, YODER,0. C. (Eds.) (1994), HostSpecific Toxin. Biosynthesis, Receptor and Mo(VINING, C., STUITARD,C., Eds.), pp. 385L. lecular Biology. Tottori University Press. 420. Boston, M A Butterworth-Heinemann. S., K., M., KELEMEN, H., BROWN, L., KORMANEC, KONDO, YASUO, MATSUMA, KATAYAG. G. J., MA,M., MARUMO, (1986), Pamamycin-6@7, a S. POTUCKOVA, CHATER, F., BUTTNER, L., K. M. new macrodiolide antibiotic from Streptomyces J. (19%), The positions of the sigma factor alboniger, J. Antibiot. 41, 1196-1204. genes, whiG and sigF, in the hierarchy controlling the development of spore chains in the aerial KROHN, KIRST, A., MAAG, (Eds.) (1993), K., H. H. hyphae of Streptomyces coelicolor A3(2), Mol. Antibiotics and Antiviral Compounds. WeinMicrobiol. 21, 593-603. heim: VCH. K., Y., KELL,D. B., KAPRELYANTS, S., GRAFEN,A. KUBOHARA, OKAMOTO, TANAKA, AsA. Y., (1995). Pheromones, social behaviour and the N. ALIN, SAKURAI, TAKAHASHI, (1993), K., A., functions of secondarv metabolism in bacteria. Differanisol A. an inducer of the differentiation Trends Ecol. Evol. lO.126-129. of Friend leukemic cells induces stalk cell differ-
48
1 General Aspects of Secondary Metabolisnrl

New York, Heidelberg, Berlin, Tokyo: SpringerVerlag. LUENGO, M. (1995), Enzymatic synthesis of hyJ. drophobic penicillins, J. Antibiot. 48, 11951212. MACCABE, P., RIACH, B. R., UNKLES, E., A. M. S. KINGHORN, R. (1990), The Aspergillus niduJ. lam npeA locus consists of three contiguous genes required for penicillin biosynthesis, EMBO J. 9,279-287. MADDURI, HUTCHINSON, R. (1995a), FuncK., C. tional characterization and transcriptional analysis of the dnrR, locus, which controls daunorubicin biosynthesis in Streptomyces peucetius, J. Bacteriol. 177, 1208-1215. MADDURI, HUTCHINSON, R. (1995b), FuncK., C. tional characterization and transcriptional analysis of a gene cluster governing early and late steps in daunorubicin biosynthesis in Streptomyces peucetius, J. Bacteriol. 177, 3879-3884. MAHANTI, N., BHATNAGAR, CARY,J. W., D., JOUBRAN, LINZ,J. E. (19%), Structure and J., function of FAS-lA, a gene encoding a putative fatty acid synthetase directly involved in aflatoxin biosynthesis in Aspergillus parasiticus, Appl. Environ. Microbiol. 62, 191-195. MARAHIEL, A., DANDERS, KRAUSE,M., M. W., H. KLEINKAUF, (1979), Biological role of gramicidin S in spore function, Eur. J. Biochem. 99, 49-58, MARAHIEL, A,, NAKANO, M., ZUBER,P. M. M. (1993), Regulation of peptide antibiotic production in Bacillus, Mol. Microbiol. I , 631-636. MARCOS, T., GUTIERREZ, DIEZ, B., FERA. S., J. NANDEZ, J., OGUIZA, A., MARTIN, F. F. J. (1995), Three genes hrdB, hrdD and hrdT of Streptomyces griseus IMRU 3570, encoding sigma-factor-like proteins, are differentially expressed under specific nutritional conditions, Gene 153,41-48. MART~N,F. (1978), Manipulation of gene expresJ. sion in the development of antibiotic production, in: Antibiotics and Other Secondary Metabolites (HOTTER, LEISINGER, NOESCH, J., R., J., WEHRLI,W., Eds.), pp. 19-38. London, New York: Academic Press. MARTIN, F. (1989), Molecular mechanisms for J. the control by phosphate of the biosynthesis of antibiotics and other secondary metabolites, in: Regulation of Secondary Metabolism in ActinoS., mycetes (SHAPIRO, Ed.), pp. 213-233. BocaRaton, FL: CRC Press. MART~N, F., DEMAIN, L. (1980), Control of J. A. antibiotic biosynthesis, Microbiol. Rev. 44, 230251. MART~N, F., GUTIERREZ, (1995), Genes for J. S. beta-lactam antibiotic biosynthesis, Antonie van
entiation in Dictyostelium discoideum, FEBS Lett. 322, 73-75. KUBOTA, TOKOROYAMA, T., KAMIKAWA, SAT., TAMURA, (1966), The structures of sclerin Y. and scleroide, metabolites of Sclerotinia libertianu, Tetrahedron Lett. 1966, 5205. LAATSCH, (1994), Antibase, Database. HeidelH. berg: Chemical Concepts. LANCINI, LORENZETTI, (1993), BiotechnoloG., R. gy of Antibiotics and Other Bioactive Microbial Metabolites. New York, London: Plenum Press. LANDAN, G., COHEN,G., AHARONOWITZ, Y., SHUALI, GRAUR, SHIFFMAN, (1990), Y., D., D. Evolution of isopenicillin N synthase may have involved horizontal gene transfer, Mol. Biol. Evol. 7,399406. LAWEN, TRABER, (1993), Substrate specifiA., R. cities of cyclosporin synthetase and peptolide SDZ 214-103 synthetase, J. Biol. Chem. 268, 20452-20465. LE GOUILL,C., DESMARAIS, DERY, C. V. D., (1993), Saccharopolyspora hirsuta 367 encodes clustered genes similar to ketoacyl synthase, ketoacyl reductase, acyl carrier protein, and biotin carboxyl carrier protein, Mol. Gen. Genet. 240, 146-150. LI, W., NIKIRA, SAKUDA, HISHIDA, YAT., S., T. MADA, (1992), New inducing factors for virY. giniamycin production from Streptomyces antibioticus, J. Ferment. Biotechnol. 14, 214-217. LIRAS,P., ASTURIAS, A., MARTIN, J. F. (1990), J. Phosphate sequences involved in transcriptional regulation of antibiotic biosynthesis, Trends Biotechnol. 8, 184-189. LITZKA, THENBERG, K., BRAKHAGE, A. O., A. (1996), The Aspergillus nidulans penicillin biosynthesis gene aat (penDE) is controlled by a CCAAT containing DNA element, Eur. J. Biochem. 238,675-682. LIU, Y. S., KIESER,H. M., HOPWOOD, A., D. CHEN,C. W. (1992), The chromosomal DNA of Streptomyces lividans 66 is linear, Mol. Microbiol. 10, 923-933. LOMBb, F., BLANCO, FERNANDEZ, MENG., E., DEZ,C., SALAS, A. (1996), Characterization of J. Streptomyces argillaceus genes encoding a polyketide synthase involved in the biosynthesis of the antitumor mithramycin, Gene 172, 87-91. LOUMAYE, THORNER, GAIT, K. J. (1982), E. J., Yeast mating pheromone activates mammalian gonadotrophs: evolutionary conservation of a reproductive hormone, Science 218,1323-1325. LUCKNER, (1989), Secondary Metabolism in MiM. crobes, Plants and Animals. Jena: G. Fischer. LUCKNER, NOVER,L., BOHM,H. (1977), SecM., ondary Metabolism and Cell Differentiation.
6 References
49
Leeuwenhoek Int. J. Gen. Mol. Microbiol. 67, 181-200. MARTIN, J. F., LOPEZ-NIETO, J., CASTRO,J. M. M., CORTES,J., RAMOS,F. R., CANTORAL, J. M. N., ALVAREZ, RAMIREZ, G., BARREDO, E., J. C., LIRAS,P. (1986), Enzymes involved in plactam biosynthesis controlled by carbon and nitrogen regulation, in: Regulation of Secondary Metabolite Formation (KLEINKAUF, H., VON DOHREN,H., DORNAUER NESEMANN, H., G., Eds.), pp. 41-47. Weinheim: VCH. MARTINEZ-BLANCO, REGLERS, LUMENGO, H., A., J. M. (1991), I n vifro-syntheses of different naturally-occurring semisynthetic and synthetic penicillins using a new effective enzymatic coupled system, J. Antibiot. 44, 1252-1254. MAZZOLA, WHITE,F. F. (1994), A mutation in M., the indole-3-acetic acid biosynthesis pathway of Pseudomonas syringae PV syringae affects growth in Phaseolus vulgaris and syringomycin production, J. Bacteriol. 176, 1374-1382. MCCUE,L. A., KWAK,J., BABCOCK, J, KENM. DRICK,K. E. (1992), Molecular analysis of sporulation in Streptomyces griseus, Gene 115, 173179. MCDANIEL,R., EBERT-KOHOSLA, Fu, H., S., HOPWOOD, A., KHOSLA,C. (1994), EngiD. neered biosynthesis of novel polyketides: influence of a downstream enzyme on the catalytic specificity of a minimal aromatic polyketide synthase, Proc. Natl. Acad. Sci. USA 91, 1154211346. MCNEIL,D. J. (1995), Avermectins, in: Genetics and Biochemistry of Antibiotic Production (VINING, L. c., STUTTARD, Eds.), pp. 421442. C., Boston, MA: Butterworth-Heinemann. MEIGHEN, A. (1991), Molecular biology of bacE. terial luminescence, Microbiol. Rev. 55, 123142. MEISSNER, DITTMANN, BORNER, (1996), K., E., T. Toxic and nontoxic strains of the cyanobacterium Microcystis aeruginosa contain sequences homologous to peptide synthetase genes, FEMS Microbiol. Lett. 135, 295-303. MERSON-DAVIES, A., CUNDLIFE, (1994), L. E. Analysis of five tylosin biosynthetic genes from the tyIIBA region of the Streptomyces fradiae genome, Mol. Microbiol. 13, 349-355. MEURER, HUTCHINSON, R. (1995), DaunoG., C. rubicin type I1 polyketide synthase enzymes DpsA and DpsB determine neither the choice of starter unit nor the cyclization pattern of aromatic polyketides, J. Am. Chem. SOC.117, 58995900.
MILLER, R., INGOLIA, D. (1993), Cloning and J. T. characterization of &lactam biosynthetic genes, Mol. Gen. Genet. 3,689495.
MISAWA, SATOMI, KONDO,K., YOKOYAN., Y., MA, A., KAJIWARA, SAITO,T., OHTANI, s., T., MIKI,W. (1995), Structure and functional analysis of a amrine bacterial carotinoid biosynthesis gene cluster and astaxanthin biosynthetic pathway proposed at the gene level, J. Bacteriol. 177, 65754584. Mo, Y. Y., GEIBEL, BONSALL, F., GROSS, M., R. D. C. (1995), Analysis of sweet cherry (Prunus avium L.) leaves for plant signal molecules that activate the syrB gene required for synthesis of the phytotoxin, syringomycin, by Pseudomonas syringae pv. syringae, Plant Physiol. 107, 603612. MONAGHAN, L., TKACZ, S. (1990), Bioactive R. J. microbial products: focus upon mechanism of action, Annu. Rev. Microbiol. 44,271-301. MORENO, MILLAN, L. S., HERNANDEZ-CHIF., J. c o , C., KOLTER,R. (1995) Microcins, in: Genetics and Biochemistry of Antibiotic Production (VINING, C., STUITARD,C., Eds.), pp. 307L. 322. Boston, MA: Butterworth-Heinemann. MOLLER,L. (1989), Chemistry, biochemistry and therapeutic potential of microbial a-glucosidase inhibitors, in: Novel Microbial Products for MedA. icine and Agriculture (DEMAIN, L., SOMKUTI, G . A., HUNTER-CEVERA, C., ROSSMORE, J. H. W., Eds.), pp. 109-116. Amsterdam: Elsevier. MURAO,S., HAYASHI, TARUI,H. (1984), AnH., thranilic acid, as a fruiting body inducing substance in Favolus arcularius, from a strain TA7 of actinomycetes, Agric. Biol. Chem. 48, 16691677. NIKAIDO, (1994), Prevention of drug access to H. bacterial targets: permeability barriers and active efflux, Science 264, 382-388. NOBLE, L. (1990), The discovery of the vinca alR. kaloids - chemotherapeutic agents against cancer, Biochem. Cell. Biol. 68, 1344-1351. OCHI,K. (1990), Streptomyces relC mutants with an altered ribosomal protein ST-Ul and genetic analysis of a Streptomyces griseus relC mutant, J. Bacteriol. 172, 4008-4016. OCHI,K., TSURUMI, SHIGEMATSU, IWANI, Y., N., M., UMEHARA, OKUHARA, (1988), PhysK., M. iological analysis of bicozamycin high-producing Streptomyces griseoflavus used at industrial level, J. Antibiot. 41, 1106-1115. OCHI,K., PENYIGE, BARABAS, (1992), The A., G. f possible role o ADP-ribosylation in sporulation and streptomycin production by Streptomyces, J. Gen. Microbiol. 138, 1745-1750. OCHIAI, (1987), Critical view on the use of imT. munostimulants in cancer treatment, in: ImmuI., nostimulants: Now and Tomorrow (AZUMA, JOLLES,G., Eds.), pp. 167-172. TokyolBerlin: Japan Scientific Societies PresslSpringer.
50
OGARA, NAKAYAMA, FURIHATA, SHIM., H., K., MADZU, SETO,H., OTAKE, (1985), StrucA., N. ture of a new antibiotic curromycin A, produced by a genetically modified strain of Streptomyces hygroscopicus, a polyether antibiotic producing organism, J. Antibiot. 38, 669-673. OHAGAN, (Ed.) (1991), The Polyketide MetabD. olites. Chichester: Ellis Horwood. S., T., H., OKAMOTO, NIHIRA, KATAOKA, SUZUKI, A., YAMADA, (1992), Purification and molecY. ular cloning of a butyrolactone autoregulator receptor from Streptomyces virginiae, J. Biol. Chem. 267, 1093-1098. OLESKIN, V. (1994), Social behaviour of microA. bial populations, J. Basic Microbiol. 34, 425439. OLIYNYK, BROWN,M. J. B., CORTES, J., M., STAUNTON, LEADLAY, F. (1996), A hybrid J., P. modular polyketide synthase obtained by domain swapping, Chemistry & Biology 3, 833839. S. OMURA, (Ed.) (1992), The Search of Bioactive Metabolites. New York: Springer-Verlag. OXFORD, S., FIELDS, J., REEVES, S. (Eds.) J. H. D. (1986), Drug Resistance in Viruses, Other Microbes and Eukaryotes. London, New York: Academic Press. PJ~REZ-ESTEBAN, B., G6MEZ-PARD0, E., PENALVA, M. A. (1995), A lacZ reporter fusion method for the genetic analysis of regulatory mutations in pathways of fungal secondary metabolism and its application to the Aspergillus nidulans penicillin pathway, J. Bacteriol. 177, 60696076. PETERSEN, ZAHNER, H., METZGER,J. W., F., S., R. FREUND, HUMMEL, P. (1993). Germicidin, an autoregulative germination inhibitor of Streptomyces viridochromogenes NRRL B-1551, J. Antibiot. 46, 1126-1138. PETRICH, K., LESKIW, K., PARADKAR, S., A. B. A. JENSEN, E. (1994), Transcriptional mapping S. of the genes encoding the early enzymes of the cephamycin biosynthestic pathway of Streptomyces clavuligerus, Gene 142,4143, PIEPERSBERG, (1994), Biosynthesis of secondW. ary carbohydrate components in actinomycetes and other bacteria. Perspectives of a pathway engineering (in German), BioEngineering 6, 2734. PIEPERSBERG, (1995), Streptomycin and related W. aminoglycosides, in: Genetics and Biochemistry of Antibiotic Production (VINING, C., STUTL. TARD, Eds.), pp. 531-570. Boston, MA: Butc., terworth-Heinemann. E. PIERSON, S., PIERSON, A. (1996), Phenazine L. antibiotic production in Pseudomonas aureofaciens - role in rhizosphere ecology and pathogen
suppression, FEMS Microbiol. Lett. 136, 101108. T., F. PIERSON, S., GAFFNEY, LAM,S., GONG, C. L. (1995), Molecular analysis of genes encoding phenazine biosynthesis in the biological control bacterium Pseudomonas aureofaciens 30-84, FEMS Microbiol. Lett. 134, 299-307. PITKIN, W., PANACCIONE, G., WALTON, D. J. D. J. (1996), A putative cyclic peptide efflux pump encoded by the toxA gene of the plant pathogenic fungus Cochliobolus carbonum, Microbiology 142, 1557-1565. PLINER, A., KLEINER,E. M., KORNITSKAYA, S. E. Y., TOVAROVA, I., ROZYNOV, V., SMIRI. B. NOVA, M., KHOKKLOV, S . (1976), IsolaG. A. tion and primary characteristics of the A-factor, Sov. J. Bioorg. Chem. (English translation) 2, 825-832. POSPIECH,A., BIETENHADER, SCHUPP, T. J., (1996), Two multifunctional peptide synthetases and an 0-methyltransferase are involved in the biosynthesis of the DNA-binding antibiotic and x antitumor agent saframycin M l from Myxococcus xanthus, Microbiology 142, 741-746. G. POSTMA, W., LENGELER, W., JACOBSON, P. J. R. (1993), Phosphoeno1pyruvate:carbohydrate phosphotransferase systems of bacteria, Microbiol. Rev. 57, 543-594. POTIER, GUERI~TE-VOEGELEIN, P., F., GUENARD, D. (1994), Taxoids, a new class of antitumor agents of plant origin: recent results, Nouv. Rev. Fr. Hematol. 36,521-523. QUEENER, S . W., WILKERSON, TUNIN, R., S., D. J. J. MCDERMO~T,P., CHAPMAN, K., NASH,C., WESTPHELING, (1986), Cephalosporin C ferJ. mentation biochemical and regulatory aspects of sulfur metabolism, in: Biotechnology of IndusE. trial Antibiotics (VANDAMME, J., Ed.), pp. 141-170. New YorkBasel: Marcel Dekker. QUIROS, L. M., SALAS, A. (1995), Biosynthesis J. of the macrolide oleandomycin by Streptomyces antibioticus, J. Biol. Chem. 270, 18234-18239. T. P. RA, S. R., QIAO, M. Q., IMMONEN, I., SARIS, E. J. (1996), Genes responsible for nisin synthesis, regulation and immunity from a regulon of two operons and are induced by nisin in Lactococcus lactis N8,Microbiology 142, 1281-1288. REDENBACH, KIESER,H. M., DENAPAITE, M., D., H., EICHNER, CULLUM, KINASHI, HOPA., J., D. WOOD, A. (1996), A set of ordered cosmids and a detailed genetic and physical map for the 8 Mb Streptomyces coelicolor A3(2) chromosome, Mol. Microbiol. 21, 77-96. H., REICHENBACH, HOFLE,G. (1993), Biologically active secondary metabolites from Myxobacteria, Biotechnol. Adv. 11, 219-277.
6 References
51
REUTER, (1989), Enzymatic regulation of micro- SANGLIER, J., LARPENT, P. (1989), BiotechnoG. J. J. bial phytoeffector biosynthesis, Progr. Ind. Milogie des Antibiotiques. Paris, Milano, Barcelona, Mexico: Masson. crobiol. 27, 271-282. M., RINEHART, L., SHIELD, S. (1988), Biological SCHIRMBOCK, LORITO,M., WANG, Y. L., K. L. I., HAYES,C. K., ARISAN-ATAC, SCALA,F., active marine natural products, in: Horizons on G. C. Antibiotic Research (DAVIES, D., ICHIKAWA, HARMAN, E., KUBICEK, P. (1994), Parallel B. formation and synergism of hydrolytic enzymes T., MAEDA, MITCHER, A., Eds.), pp. 194K., L. and peptaibol antibiotics, molecular mechanisms 227. Tokyo: Japan Antibiotics Association. involved in the antagonistic action of TrichoderROCK,C. 0.. CRONAN, E. (1996), Escherichia J. ma harzianum against phytopathogenic fungi, coli as a model for the regulation of dissociable Appl. Environ. Microbiol. 60, 43644370. (type 11) fatty acid biosynthesis, Biochim. BioSCHUPP, TUOPET,C., CLUZEL,B., NEFF, S., T., phys. Acta 1302, 1-16. HILL,S., BECK, J., LIGON, M. (1995), A SoJ. J. ROHR, J., SCHONEWOLF, UDVARNOKI, M., G., rangium cellulosum (Myxobacterium) gene clusECKARDT, SCHUMANN, WAGNER,C., K. G., ter for the biosynthesis of the macrolie antibiotic BEALE, M., SOREY,S. D. (1993), InvestigaJ. sorphen A-cloning, characterization, and homotions on the biosynthesis of the angucycline logy to polyketide synthase genes from Actinogroup antibiotics, J. Org. Chem. 58, 2547-2551. mycetes, J. Bacteriol. 177, 3637-3679. Roos, W., LUCKNER, (1986). The spatial organM. D., B., ization of secondary metabolism in microbial SCHWARTZ, ALIJAH,R., NUSSBAUMER, S., W. PELZER, WOHLLEBEN, (1996), The pepand plant cells, in: Cell Growth and Management tide synthetase gene phsA from Streptomyces (SUBRAMANIAN,V. (Ed.), pp. 45-73. Boca A. viridochromogenes is not juxtaposed with other Raton, F L CRC Press. genes involved in nonribosomal biosynthesis of E., ROSSNER, ZEECK,A., KONIG,W. A. (1990), peptides, Appl. Environ. Microbiol. 62, 570Structure elucidation of hormaomycin (in Ger577. man), Angew. Chem. 102, 84-85. I., SCHWECKE, APARICIO, F., MOLNAR, KOT., J. SADAKANE, TANAKA, OMURA, (1983), N., Y., S. NIG, A., KHAW, E., HAYDOCK, F., OLIYL. S. Hybrid biosynthesis of a new macrolide antibiotNYK, M., CAFFREY, CORTES, LESTER, P., J., J. ic by a daunomycin-producing microorganism, J. B., BOHM,G. A., STAUNTON, LEADLAY, J., P. Antibiot. 36, 921-922. F. (1995), The biosynthetic gene cluster for the SAHL,H.-G., JACK, W., BIERBAUM, (1995), R. G. polyketide immunosuppressant rapamycin, Proc. Biosynthesis and biological activities of Natl. Acad. Sci. USA 91, 7839-7843. lantibiotics with unique post-translational modif- Scrip, World Pharmacentical News (London) ications, Eur. J. Biochem. WO,827-853. (1993), 1825,23. SAKUDA, YAMADA, (1991), Stereochemistry SENO, T., BALTZ, H. (1989), Structural organS., Y. E. R. of butyrolactone autoregulators from streptomyization and regulation of antibiotic biosynthesis cetes, Tetrahedron Lett. 32, 1817-1820. and resistance genes in actinomycetes, in: RegulSAKUDA, HIGASHI, TANAKA, NIHIRA, S., A., S., ation of Secondary Metabolism in Actinomycetes T., YAMADA, (1992), Biosynthesis of virginY. (SHAPIRO, Ed.), pp. 243. Boca Raton, FL: S., iae butanolide A, a butyrolactone autoregulator CRC Press. from Streptornyces, J. Am. Chem. SOC.114,663- SHAH, J., TILBURN, ADLARD, W., ARST, A. J., M. 668. H. N., JR. (1991), pH regulation of penicillin SALYERS, A., SHOEMAKER, A., STEVENS, A, N. A. production in Aspergillus nidulans, FEMS MiM., LI, L.-Y. (1995), Conjugative transposons: crobiol. Lett. 77, 209-212. an unusual and diverse set of integrated gene SHAPIRO, (Ed.) (1989), Regulation of Secondary S. transfer elements, Microbiol. Rev. 59, 579-590. Metabolism in Actinomycetes. Boca Raton, F L CRC Press. SAMSON, M., BELAGAJE, BLANKENSHIP, S. R., D. D., P. SHEN,B., HUTCHINSON, R. (1994), Triple hyT., CHAMPMAN,L., PERRY, SKATRUD, J. C. L., VAN FRANK,R. M., ABRAHAM, P., E. droxylation of tetracenomycin A2 to tetracenoS. T. BALDWIN, E., QUEENER, W., INGOLIA, J. mycin C in Streptomyces glaucescens, J. Biol. D. (1985), Isolation, sequence determination Chem. 269,30726-30733. and expression in Escherichia coli of the isopeni- SHERMAN, H., MALPARTIDA, BIBB,M. J., D. F., cillin N synthetase gene from Cephalosporium KIESER, M., HOPWOOD, A. (1989), StrucH. D. acremonium, Nature 318, 191-194. ture and deduced function of the granaticin producing polyketide synthase gene cluster of StrepSANCHEZ, BRANA, F. (1996), Cell density inL., A. tomyces violaceoruber Tii22, EMBO J. 8, 2717fluences antibiotic biosynthesis in Streptomyces 2725. clavuligerus, Microbiology 142, 1209-1220.
52
G., SHIER, T., RINEHART, L., JR., GO~TLIEB, W. K. D. STROBEL, YANG,X. S., SEARS,J., KRAMER, R., SIDHU,R. S., HESS, W. M. (1996), Taxol (1969), Preparation of four new antibiotics from from Pestalotiopsis microspora, an endophytic a mutant of Streptomyces fradiae, Proc. Natl. fungus of Taxus wallachiana, Microbiology 142, Acad. Sci USA 63, 198-204. 435-440. T., T., SHIOZAWA, KAGASAKI, KINOSHITA, H., HARUYAMA, DOMON,H., UTSUI,Y., Ko- SUAREZ, PENALVA, A. (1996), CharacterizaH., T., M. DAMA, K., TAKAHASHI, (1993), Thiomarinol, s. tion of a Penicillium chrysogenum gene encoda new hybride antimicrobial antibiotic produced ing a p a c e transcription factor and its binding by a marine bacterium, J. Antibiot. 46, 1834sites in the divergent pcbAB-pcbC promoter of the penicillin biosynthetic cluster, Mol. Micro1842. biol. 20, 529-540. M. SMITH, J., BURNHAM, K. R., BULL,J. H., D. P., SUMMERS, G., ALI, A., SHEN,B., WESSEL, HODGSON, E., WARD, J. M., BROWNE, J. R. W. J., B., BROWN, BARTON, EARL,A. J., TURNER, A., HUTCHINSON, R. (1995), Malonyl-coenC. G. (1990), PLactam antibiotic biosynthetic zyme A-acyl carrier protein acyltransferase of genes have been conserved in clusters in prokaStreptomyces glaucescens - a possible link beryotes and eukaryotes, EMBO J. 9,741-747. tween fatty acid and polyketide biosynthesis, SOLOMON, M., MAGNUSON, SRIVASTAVA, Biochemistry 34, 9389-9402. J. R., A., GROSSMAN, D. (1995), Convergent sens- SUZUKI, YAMAGUCHI, MIYAZAKI, NAA. K., H., S., ing pathways mediate response to two extracelGORI, K. (1990), Topostin, a novel inhibitor of lular competence factors in Bacillus subtilis, mammalian DNA topoisomerase I from FlexiGenes Dev. 9,547-558. bacter topostinus sp. nov. I. Taxonomy, and fermentation of producing strain, J. Antibiot. 43, SPRA-IT, B. G. (1994), Resistance to antibiotics me154-157. diated by target alterations, Science 264, 388393. SZESZAK, VITALIS, BEKESI,I., SZABO,G. F., S., STACHELHAUS, SCHNEIDER, MARAHIEL, (1991), Presence of factor C in streptomycetes T., R., and other bacteria, in: Genetics and Product ForM. A. (1995a), Rational design of peptide antiS., mation in Streptomycetes (BAUMBACH, KRUbiotics by targeted replacement of bacterial and GEL, H., NOACK,D., Eds.), pp. 11-18. New fungal domains, Science 269, 69-72. York: Plenum Press. STACHELHAUS, SCHNEIDER, MARAHIEL, T., R., M. A. (1995b), Engineered biosynthesis of pep- TAKANO, KUBO,Y., SHIMIZU, MISE, K., Y., K., tide antibiotics, Biochem. Pharmacol. 52, 177I. OKUNO, FURUSAWA, (1995), Structural T., analysis of PKSl, a polyketide synthase gene in186. volved in melanin biosynthesis in Colletotrichum STEELE,D. B., STOWERS, D. (1991), TechM. lagenarium, Mol. Gen. Genet. 249, 162-167. niques for selection of industrially important microorganisms, Annu. Rev. Microbiol. 45, 89- TANAKA, (1986), Biosynthesis of tylosin and its Y. 106. regulation by ammonium and phosphate, in: Regulation of Secondary Metabolite Formation STEWART, C. (1993), Catabolite repression in G. (KLEINKAUF, H., VON DOHREN, H., DORthe gram-positive bacteria: generation of negaNAUER, H., NESEMANN, Eds.), pp. 305-337. G., tive regulators of transcription, J. Cell Biochem. Weinheim: VCH. 51,25-28. J. J. R. STIERLE, STIERLE, STROBEL, BIGNANI, TERCERO, A., ESPINOSA, C., LACALLE, A., H., D., G., JIMBNEZ, (1996), The biosynthetic pathway A. G., GROTHANS, (1994), Endophytic fungi of P. of the aminonucleoside antibiotic puromycin, as pacific yew (Taxus brevifolia) as a source of taxdeduced from the molecular analysis of the pur 01, taxanes and other pharmacophores, ACS cluster of Streptomyces alboniger, J. Biol. Chem. Symp. Ser. 557 (Bioregulators for crop protec271,1579-1590. tion and pest control), 64-77. A. STIERLE, STROBEL, STIERLE,D., GROT- THENBERG, K., LITZKA,O., BRAKHAGE, A. A., G., (1996), Identification of a major cis-acting eleP., G. HAUS, BIGNAMI, (1995), The search for a ment controlling bidirectionally transcribed pentaxol-producing microorganism among the enicillin biosynthesis genes acvA (pcbAB) and dophytic fungi of the pacific yew, Taxus brevifolipnA (pcbC) of Aspergillus nidulans, J. Bacteriol. ia, J. Nat. Prod. 58, 1315-1324. 178,3908-3916. STINTZI, A., CORNELIS,P., HOHNADEL,D., R., MEYER, J.-M., DEAN,C., POOLE,K., KOURAM- THIERICKE, ROHR,J. (1993), Biological variation of microbial metabolites by precursor-diBAS, S., KRISHNAPILLAI, (19%), Novel P O V. YV erdine biosynthesis gene(s) of Pseudornonas aerected biosynthesis, Nat. Prod. Rep. 12, 265289. ruginosa PAO, Microbiology 142, 1181-1190.
6 References
53
TILBURN, SARKAR, WIDDICK, A., ESPEJ., S., D. so, E. A., OREJAS,M., MUNGROO, PENALJ., VA,M. A., ARST,H. N., JR. (1995), The Aspergillus PacC zinc finger transcription factor mediates regulation of both acidic- and alkaline-expressed genes by ambient pH, EMBO J. 14,779790. TOMODA, OMURA, (1990), New strategies H., S. for discovery of enzyme inhibitors: screening with intact mammalian cells or intact microorganisms having special functions, J. Antibiot. 38, 1207-1222. TRAIL, MAHANTI, LINZ,J. E. (1995a), MoF., N., lecular biology of aflytoxin biosynthesis, Microbiology 141, 755-765. TRAIL,F., MAHANTI, RARICK, MEHIGH, N., M., R.,LIANG, H., ZHOU,R., LINZ,J. E. (1995b), S. Physical and transcriptional map of an aflatoxin gene cluster in Aspergillus parasiticus and functional disruption of a gene involved early in the aflatoxin pathway, Appl. Environ. Microbiol. 61, 2665-2673. Tsor, C. J., KHOSLA, C. (1995), Combinatorial biosynthesis of "unnatural" natural products: the polyketide example, Chem. Biol. 2, 355-362. TURGAY, MARAHIEL, A. (1995), The gtcRS K., M. operon coding for two-component system regulatory proteins is located adjacent to the grs operon of Bacillus brevis, DNA Sequence 5, 283290. TURNER, (1992), Genes for the biosynthesis of G. plactam compounds in microorganisms, in: Secondary Metabolites: Their Function and Evolution, Ciba Foundation Symposium 171 (CHADJ., WICK,D. J., WHELAN, Eds.), pp. 113-128. Chichester, New York: J. Wiley & Sons. TURNER, B., ALDRIDGE, C. (1983), Fungal W. D. Metabolites II. London: Academic Press. TYMIAK, A., Ax, H. A., BOLGAR,M. S., A. KAHLE, D., PORUBCAN, A., ANDERSEN, A. M. N. H. (1993), Dactylocyclines, novel tetracycline derivatives produced by Dactylosprangium sp. 11. Structure elucidation, J. Antibiot. 45, 18991%. UMEZAWA, HOOPER, R. (Eds.) (1982), AmiH., I. noglycoside Antibiotics. New York, Heidelberg, Berlin, Tokyo: Springer-Verlag. VANDAMME, J. (Ed.) (1984), Biotechnology of E. Industrial Antibiotics. New York, Basel: Marcel Dekker. VANEK,Z., HOSTALEK, (1985), Tetracycline Z. biosynthesis, in: The Tetracyclines (HLAVKA, J. J., BROTHE,J. H., Eds.), pp. 137-138. New York, Heidelberg, Berlin, Tokyo: Springer-Verlag. VANEK, HOSTALEK, (Eds.) (1988), OverproZ., Z. duction of Microbial Metabolites: Strain Im-
provement and Process Control Strategies. Bostonllondon: Butterworth. VANEK, MAJOR,J. (1967), Macrolide antibiotZ., ics, in: Antibiotics, Vol. 2 (GOTTLIEB, SHAW, D., P. D., Eds.), pp. 154-188. New York, Heidelberg, Berlin, Tokyo: Springer-Verlag. Z., K. VANEK, MIKULIK, (1978), Microbial growth and production of antibiotics, Folia Microbiol. 23,309-328. VANEK, BEHAL,V., JECHOVA, CURDOVA, Z., V., E., BLUMAUEROVA, HOSTALEK, (1978), M., Z. Formation of tetracycline antibiotics, in: Antibiotics and Other Secondary Metabolites (HUTI., TER, R., LEISINGER, NOESCH,J., WEHRLI, W., Eds.), pp. 101-112. London, New York, San Francisco: Academic Press. VANEK, CUDLIN, BLUMAUEROVA, HosZ., J., H., TALEK, (1981), Physiology and PathophysioZ. logy of the Production of Excessive Metabolites. Institute of Microbiology, CS Academy of Sciences, Prague. VERALL, S. (Ed.) (1985), Discovery, and IsolaM. tion of Microbial Products. Chichester: Ellis Horwood. L. VINING, C. (1986), Secondary metabolism, in: Biotechnology, 1st Edn., Vol. 4 (REHM,H.-J., REED,G., Eds.), pp. 19-38. Weinheim: VCH. VINING, C. (1992), Role of secondary metabolL. ites from microbes, in: Secondary Metabolites: Their Function and Evolution, Ciba Foundation D. J., Symposium 171 (CHADWICK, J., WHELAN, Eds.), pp. 184-198. Chichester, New York: J. Wiley & Sons. VINING, L. C., STUTTARD, (1995), ChlorC. amphenicol, in: Genetics and Biochemistry of Antibiotic Production (VINING,L. C., STUTTARD,C., Eds.), pp. 505-530. Boston, MA: Butterworth-Heinemann. S., Z., VINING, C., SHAPIRO, AHMED, VATS, S., L. DOULL,J. STUTTARD, (1986), Genetic and C. physiological control of chloramphenicol production, in: Regulation of Secondary Metabolite H., H., Formation (KLEINKAUF, VON DOHREN, DORNAUER, NESEMANN, Eds.), pp. 209H., G., 224. Weinheim: VCH. VOGEL, R. (1984), Natiirliche Enzyminhibitoren. Stuttgart: Thieme. VON DER HELM, NEILANDS, B. (Eds.) (1987), P., J. Iron Transport in Microorganisms, Plants and Animals. Weinheim: VCH. VON DOHREN, H. (1990), Compilation of peptide structures: a biogenetic approach, in: BiochemisH., try of Peptide Antibiotics (KLEINKAUF, VON D ~ H R E N , Eds.), pp. 411-507. Berlin, New H., York: de Gruyter. VON D ~ H R E N , (1995), Peptides, in: Genetics and H. Biochemistry of Antibiotic Production (VINING,
54
L., STUTTARD,C., Eds.), pp. 129-172. Boston, MA: Butterworth-Heinemann. J. M. WALKER, B., WALKER, S. (1978), Enzymatic synthesis of streptomycin as a model system for study of the regulation and evolution of antibiotic biosynthesis, in: Overproduction of Microbial V., Products (KRUMPHANZL, SIKYTA,B., VAZ., NEK, Eds.), pp. 423-438. London, New York: Academic Press. WANNER, L. (1993), Gene regulation by phosB. phate in enteric bacteria, J. Cell. Biochem. 51, 47-54. WARD,J. M., HODGSON. E. (1993), The biosynJ. thetic genes for clavulanic acid and cephamycin production occur as super-cluster in three Streptomyces, FEMS Microbiol. 110, 239-242. WATANABE, SHIRMA, IZAKI,K., SUGIYAT., K., MA, T. (1992), New polyenic antibiotic active against gram-positive and gram-negative bacteria. VII. lsolation and structure of enacyloxin IVa, possible biosynthetic intermediate of enacyloxin IIa, J. Antibiot. 45, 575-576. WEBER,G., SCHORGENDORFER, SCHNEIDERK., SCHERZER, LEITNER, (1994), The peptide E., E. synthetase catalyzing cyclosporine production in Tolypocladium niveum is encoded by a giant 45.8 kilobase open reading frame, Curr. Genet. 26, 120-125. WICKLOW, T., DOWD, P. F., GLOER,J. B. D. (1994), Antiinsectian effects of Aspergillus metabolites, in: The Genus Aspergillus from Taxonomy and Genetics to Industrial Applications (POWELL, RENWICK, PEBERDY, Eds.), K., A., J., pp. 93-114. New York: Plenum Press. WIESMANN, E. H., CORTBS,J., BROWN, J. K. M. B., CUTTER, L., STAUNTON, LEADLEY, A. J., P. F. (1995), Polyketide synthesis in vitro on a modular polyketide synthase, Chemistry & Biology 2,583-589. WILLIAMS, H., MAPLESTONE, A. (1992), D. R. Why are secondary metabolites synthesized? Sophistication in the inhibition of cell wall biosynthesis by vancomycin group antibiotics, in: Secondary Metabolites: Their Function and Evolution, Ciba Foundation Symposium I71 (CHADWICK, D. J., WHELAN, Eds.), pp. 45-63. ChiJ., Chester, New York: J Wiley & Sons. WILLIAMS, T., VICKERS, L. (1986), The ecoloS. J. 43gy of antibiotic production, Microb. Ecol. U, 52. WILLIAMS, H., STONE,M. J., HAUCK,P. R., D. RAHMAN, R. (1989), Why are secondary meS. tabolites (natural products) biosynthesized? J. Nut. Prod. 52, 1189-1208. WILLIAMS, BAINTON,N. J., SWIFT,S., SCHAP., G., BRA,C. R., WILSON,M. K., STEWARD, MIG. CHAEL, K., SALMAND, P. c.,BYROFT,B. w.
(1992), Small-molecule dependent density-control of gene expression in prokaryotes. Bioluminescence and the biosynthesis of carbapenem antibiotics, FEMS Microbiol. Lett. 100, 161-167. WINKELMANN, (Ed.) (1991), Handbook of MiG. crobial Iron Chelates. Boca Raton, FL: CRC Press. WIRTH,R., WANNER, GALLI,D. (1990), The G., sex pheromone system of Enterococcus faecalis: a unique mechanism of plasmid uptake (in German), Forum Mikrobiol. 6, 321-332. WRIGHT, C. (1983), The leucomycin-celesticidD. in-anthramycin group, in: Biochemistry and Genetic Regulation of Commercially Important AnC. tibiotics (VINING, C., Ed.), pp. 311-328. Reading: Addison-Wesley. YAMADA, SUGAMURA, KONDO,K., YAY., K., NAGIMOTO, OKADA, (1987), The strucM., H. f tures o inducing factors or virginiamycin production in Streptomyces virginiae, J. Antibiot. #, 496-504. YANG,Y. K., SHIMIZU, SHIOYA, SUGA,K.H., S., I., NIHIRA, YAMADA, (1995a), Optimum T., Y. autoregulator addition strategy for maximum virginiamycin production in batch cultivation of Streptomyces virginiae, Biotechnol. Bioeng. 46, 437-442. YANG,K. Q., HAN,L., VINING, C. (1995b), RegL. ulation of jadomycin B production in Streptomyces venezuelae ISP5230 - involvement of a repressor gene, jadR(2), J. Bacteriol. 177, 61116117. YANG, Y. K., MORIKAWA, M., SHIMIZU, H., T., SHIOYA, SUGA, K.-I., NIHIRA, YAMADA, S., Y. (1996a), Maximum virginiamycin production by optimization of cultivation conditions in batch culture with autoregulator addition, Biotechnol. Bioeng. 49, 437444. YANG,K. Q., HAN,L., AYER,S. W., VINING, C. L. (1996b), Accumulation of the angucycline antibiotic rabelomycin after disruption of an oxygenase gene in the jadomycin B biosynthetic gene cluster of Streptomyces venezuelae, Microbiology 142,123-132. YE, J., DICKENS, L., PLATER,R., LI, Y., LAWM. RENCE, J., STROHL, w. R. (1994), Isolation and sequence analysis of polyketide synthase genes from the daunomycin-producing Streptomyces sp. strain C5, J. Bacteriol. 176, 6270-6280. YLIHONKO, TUIKKANEN, JUSSILA, CONG, K., J., S., P. L. N., MANTSALA, (1996), A gene cluster involved in nogalamycin biosynthesis from Streptomyces nogalater - sequence analysis and complementation of early block mutations in the anthracycline pathway, Mol. Gen. Genet. 251, 113120.
6 References
55
tion and secondary pathways, in: Secondary MeYu, J. H., LEONARD, J. (1995), Sterigmatocystin T. tabolism and Cell Differentiation in Fungi (BENbiosynthesis in Aspergillus nidulans requires a J. , A., N E ~ W., CIEGLER, Eds.), pp. 51-70. Lonnovel type I polyketide synthase, J. Bacteriol. don, New York: Academic Press. 177,4792-4800. T., M., W. Yu, T . W., BIBB,M. J., REVILL, P., HOPWOOD, ZAKELJMAVRIC, KASTELICSUHADOLC, PLEMENITAS, A., RIZNER,T. L., BELIC, I. D. A. (1994), Cloning, sequencing and analysis (19959, Steroid hormone signalling system and of the griseusin polyketide synthase gene cluster fungi, Comp. Biochem. Physio1.-B. Comp. Biofrom Streptomyces griseus, J. Bacteriol. 176, chem. 112,6374142. 2627-2634, J. M. ZAHNER,H., ZEECK,A. (1987), Mikrobieller Se- ZMIJEWSKI, J., JR., FAYERMAN, T. (1995), Glycopeptides, in: Genetics and Biochemistry of kundarstoffwechsel, in: Jahrbuch der BiotechnoAntibiotic Production (VINING, C., STUTL. logie (PRAvE, P., Ed.), pp. 93-113. Miinchen, TARD, C., Eds.), pp. 269-282. Boston, MA: ButWien: Oldenbourg. terworth-Heinemann. ZAHNER, ANKE,H., ANKE,T. (1983), EvoluH.,
Biotechnology
2 Regulation of Bacterial Antibiotic Production
KEITH F. CHATER MERVYN BIBB J.

Norwich, UK
1 Introduction 59 1.1 The Scope of this Chapter 59 1.2 Cellular Efficiency Involves Extensive Regulation of Metabolism in Response to Growth Conditions 59 1.3 Antibiotic Production Does Not Usually Occur in Rapidly Growing Cultures 59 2 Themes in the Regulation of Antibiotic Production Illustrated by Examples from Unicellular Bacteria 61 2.1 Intracellular Signals Associated with Starvation and Low Growth Rate Activate Microcin C7 Synthesis in E. coli 61 2.2 A Critical Cell Population Density Signaled by an Autogenous Extracellular Signal Molecule Triggers Carbapenem Synthesis by Erwinia carotovora 65 2.3 The Non-Ribosomal Production of Peptide Antibiotics in Various Bacillus spp. Is One of a Number of Alternative Stationary-Phase Fates Determined by a Network of Transition State Regulators Involving Protein Phosphorylation 67 2.4 What Has Been Learned from Studies of Antibiotic Production in Unicellular Bacteria? 70 3 Regulation of Antibiotic Production in Streptomycetes and their Relatives 70 3.1 Introduction to the Organisms 70 3.2 General Physiological Aspects of the Regulation of Antibiotic Production in Streptomyces 71 3.2.1 Metabolite Interference with Antibiotic Production in Streptomycetes 72 3.2.2 Antibiotic Production and Imbalances in Metabolism 74 3.2.3 The Possible Role of Growth Rate and ppGpp in Antibiotic Production 75 3.2.4 Antibiotic Production and the Accumulation of Small Diffusible Signaling Compounds 78 3.2.5 Summary 80
58
3.3 Genetics of Antibiotic Production 80 3.3.1 Organization of Antibiotic Biosynthetic Genes and Clusters 80 3.3.2 Genes that Pleiotropically Affect Antibiotic Production in Streptomyces coelicolor: Introduction and Overview 81 3.3.2.1 afsB, afsR, and afsK - A Role for Protein Phosphorylation in Triggering the Onset of Antibiotic Production? 81 3.3.2.2 afsQZ and afsQ2 - A Two-Component Regulatory System that Can Influence Antibiotic Production 83 3.3.2.3 abaA Influences the Production of Three of the Four Antibiotics Made by Streptomyces coelicolor 83 3.3.2.4 absA and absB - Mutants Isolated on the Basis of a Pleiotropic Defect in Antibiotic Production 83 3.3.2.5 mia - Multicopy Inhibition of Antibiotic Production 84 3.3.2.6 Genes that Affect Both Antibiotic Production and Morphological Differentiation 84 3.3.2.7 An Outline Scheme for the Interactions of Pleiotropic Antibiotic Regulatory Genes in Streptomyces coelicolor 85 3.3.3 Streptomyces griseus - The A-Factor Cascade 87 3.3.4 Pathway-Specific Regulatory Genes 88 3.3.4.1 The actZZ-ORF4, redD, and dnrl Family of Pathway-Specific Activator Genes 89 3.3.4.2 srmR - A Regulatory Gene for Spiramycin Production in Streptomyces ambofaciens 90 3.3.4.3 strR Encodes a DNA-Binding Protein that Regulates at Least One of the Streptomycin Biosynthetic Genes in Streptomyces griseus 90 3.3.4.4 brpA and dnrN - Regulatory Genes for Bialaphos Production in Streptomyces hygroscopicus and for Daunorubicin Production in Streptomyces peucetius which Show Different Degrees of Similarity to Response Regulator Genes of Two-Component Systems 91 3.3.4.5 Negative Regulation of Methylenomycin Production in Streptomyces coelicolor 91 3.3.5 Induction of Antibiotic Resistance in Antibiotic Producing Streptomycetes Antibiotics as Inducers of Gene Expression 92 3.3.5.1 actZZ-ORF1/2 of Streptomyces coelicolor and tcmWA of Streptomyces glaucescens GLA.0 - Regulatory Cassettes for Antibiotic Export 92 3.3.5.2 srmB of Streptomyces ambofaciens - A Probable ATP-Dependent Efflux System Induced by its Antibiotic Substrate 92 3.3.5.3 Induction of tlrA in Streptomyces fradiae - A Role for Transcriptional Attenuation? 93 3.3.5.4 Induction of Resistance to Novobiocin in Streptomyces sphaeroides and Streptomyces niveus - Roles for DNA Supercoiling and a Diffusible Signaling Molecule 93 3.3.5.5 Regulation of Isoforms of the Target for Pentalenolactone Inhibition in the Producing Organism 93 4 Concluding Remarks 94 5 References 96
I introduction
59
steps in the pathways and also cause repression of synthesis of many of the biosynthetic enzymes, mostly by mechanisms involving re1.1 The Scope of this Chapter pressor proteins or transcriptional or translational attenuation. (We are not aware of any Study of the molecular basis for the regula- such feedback regulatory loops in antibiotic tion of antibiotic biosynthesis has gradually synthesis.) Likewise, the assimilation of carexpanded since production genes were first bon usually involves specific induction of cloned from Streptomyces spp. more than 10 genes for the relevant enzymes, usually by inet years ago (see HOPWOOD al., 1983 and teraction of a specific transcriptional represMART~N GIL,1984 for reviews of the ear- sor or activator protein with the substrate or a and ly history of this subject). This has led to a simple derivative of it generated in the cell by and a constitutive low level of the pathway enplethora of recent reviews (e.g., MART~N LIRAS,1989; SENOand BALTZ,1989; CHAT- zyme(s). More global regulation is also necesER, 1990, 1992; CHAMPNESS and CHATER, sary. For example, most free-living microbes 1994; HUTCHINSON al., 1994) which have possess an integrated system of carbon cataet dealt almost exclusively with Streptomyces bolite repression which ensures that the most and related actinomycetes. Recently, knowl- readily utilized and ergogenic substrate (often edge of the regulation of antibiotic biosynthe- glucose) is used in preference to less favorasis in non-actinomycete bacteria has emerged ble substrates. In E. coli, such repression opalmost as a by-product of studies of the switch erates in two ways: via interference with upfrom rapid growth to stationary phase in take of the less favorable compound, preventmodel organisms such as Escherichia coli and ing it from participating in induction of the Bacillus subtilis and of mechanisms of plant enzymes needed for its assimilation (inducer pathogenicity in Erwinia carotovora. This has exclusion) and through more direct ingiven a new opportunity, in this review, to ex- fluences on transcription of the genes for amine both general themes common to the these enzymes (SAIER,1989; POSTMAet al., regulation of production of diverse antibiotics 1993; Fig. 1). In the latter case, the pattern of by diverse bacteria, and the idiosyncrasies of available carbon sources determines the indifferent bacterial genera or different path- tracellular pool of CAMP, which in turn directly interacts with a CAMP-binding protein, ways within particular bacteria. and modulates its ability to interact with the transcriptional initiation complex at promoters of various gene sets for utilization of car1.2 Cellular Efficiency Involves bon sources (Fig. 1). Carbon catabolite reExtensive Regulation of pression may, however, be exerted by different mechanisms in different microbes (SAIER, Metabolism in Response to 1991).
1 Introduction
Growth Conditions
When nutrients are abundant and readily available, microorganisms grow fast. In mixed communities, rapid conversion of nutrients to biomass is the overriding theme of metabolism and its efficiency is maximized by the well-known regulatory systems that govern such assimilation (many of which are reviewed by NEIDHARDT al., 1987). Some of et these are pathway-specific. Thus, in the feedback loops of amino acid biosynthetic pathways of E. coli, the end products typically inhibit the activity of enzymes for the earliest
1.3 Antibiotic Production Does Not Usually Occur in Rapidly Growing Cultures
In bacteria, antibiotic production nearly always takes place only after rapid growth has ceased. To a large extent, this chapter reviews the molecular and physiological mechanisms that underpin this general observation, but in this section we briefly reflect on the possible adaptive significance of this regulatory pat-
60

Mtl jout)
Glucose (out)
Fig. 1 Global regulation of carbon catabolite-repressible operons in E. coli involves multiple influences of . membrane-bound and cytoplasmic components. During uptake of glucose by EIIBCG1' (components B and C of FTSG'', the glucose phosphotransferase system), the sugar is phosphorylated by the phosphorylated form of the cytoplasmic EIIAG" which is thereby itself dephosphorylated. This has three important regulatory consequences. (1) The activity of adenylate cyclase is reduced, because it depends on interaction with EIIG1'-P. This causes a drop in the level of CAMPavailable to bind to and activate the catabolite repression protein (CRP). The CAMP-CRP complex is necessary for efficient transcription of many glucose-repressible promoters. (2) Unphosphorylated EIIAG'' directly inhibits non-PTS sugar permeases, excluding the relevant sugars from the cytoplasm and, therefore, preventing them from inducing the relevant genes for sugar catabolism. (3) The unphosphorylated EIIAG" competes with other PTS systems such as EIIABCM", the uptake system for Mfl, for the phosphate-donating protein HPr P, thereby indirectly inhibiting uptake of other PTS sugars EI, part of the enzyme cascade responsible for the transfer of phosphate from PEP to PTS sugars.
tern. We assume that antibiotic activities observed experimentally have actual evolutionary significance: i.e., that antibiotics help the producing organisms by inhibiting their competitors in natural environments. Why, then, do organisms not produce antibiotics throughout growth to maximize this competitive advantage? Perhaps the answer lies in the comparatively high diversion of resources away from biomass accumulation that might be required if the few cells present during early growth are to produce an inhibitory level of antibiotic. This might conflict with the need to grow as rapidly as possible in the competition for the nutritional resources of a new environment. On the other hand, the effective production of chemical weapons at relatively high population density (i.e., no earlier than the last few cell divisions before nutrient exhaustion) can be achieved with a much smaller proportion of each cell's metab-
olism being devoted to secondary metabolism. Even at this late stage in exploiting an environment there is still potential advantage to be gained from inhibiting competitors. This could take several forms: inhibiting the development of more persistent resting stages of competitors; greater competitiveness in the hidden population dynamics of stationary phase (during which minor subpopulations of cells grow at the expense of the majority of the population) (ZAMBRANO al., 1993); or, et in the case of developmentally complex organisms such as streptomycetes, to prevent invasion of colonies by competitors after the lysis of some of the cells within colonies, which may provide nutrition for spore development (CHATER and MERRICK, 1979; MBNDEZet al., 1985; but see also O'CONNOR and ZusMAN, 1988). Viewed in this way, one important aspect of the regulation of antibiotic production should be the mechanisms by which
2 Themes in the Regulation of Antibiotic Production Illustrated
61
information about population density or nutrient availability is perceived by cells. This information must then be interpreted by the cell and ultimately used to activate the specific relevant pathways of antibiotic biosynthesis.
2 Themes in the Regulation of Antibiotic Production Illustrated by Examples from Unicellular Bacteria
(2) responsible for expression of many stationary-phase genes (MULVEY LOEWEN, and et 1989; TANAKA al., 1993; NGUYENet al., 1993; see Fig. 2 for a summary of u factor structure and function). (The variation in mcc transcription among E. coli strains is consistent with the finding that cultures left in stationary phase are often taken over by mutants in which the C-terminus of 6s is deleted ZAMBRANO 1993.) Thus microcin C7 is et al., produced only during stationary phase because, directly or indirectly, its production genes are activated via #. The nature of mcc promoters and the conserved features of usdependent promoters in general remain to be fully elucidated. There is, however, some overlap between the promoter class recognized by 6s and that recognized by the princiet pal sigma factor, u70 (TANAKA al., 1993; NGUYEN al., 1993). This is consistent with et 2.1 Intracellular Signals the close similarities between the regions of usand u70expected to make sequence-specifAssociated with Starvation and ic DNA contacts (regions 2.4 and 4.2 in Fig. Low Growth Rate Activate 2). Some promoters that are u7O-dependent Microcin C7 Synthesis in E. coli and require activators during rapid growth may perhaps be utilized during stationary Bacterial antibiotic production is generally phase in an activator-independent manner by found in organisms such as Streptomyces spp. RNA polymerase containing 2 (KOLTERet that undergo complex differentiation (CHAT- al., 1993). How is 2 activity increased on entry into ER and MERRICK, 1979), but many simple unicellular bacteria are also producers. Thus, stationary phase? GENTRY et al. (1993) some E. coli strains produce microcins, a het- showed that uslevels respond to intracellular erogeneous collection of inhibitory com- changes in the concentration of the important pounds many (but not all) synthesized non- signaling molecule ppGpp, best known for its ribosomally. The regulation of production of role in mediating the stringent response one such compound, microcin C7, provides a (Fig.3). Levels of ppGpp increase in E. coli nice example of the activation of antibiotic when cultures are limited for amino acids, synthesis in response to a shift-down in cellu- inorganic nitrogen, or carbon (IRR,1972; CASHEL and RUDD,1987), and probably also for lar metabolism. et Microcin C7 is a ca. loo0 Da oligopeptide phosphate (GENTRY al., 1993), so ppGpp is antibiotic whose production, in post-exponen- probably a regulator of entry into stationary tial phase, is specified by genes (mcc) located phase. Consistent with this view, increasing on the plasmid pMccC7 (NOVOAet al., 1986). the steady-state level of ppGpp either by the Only certain laboratory strains of E. coli K-12 use of mutants deficient in ppGpp degradaf support transcription of mcc-lac2 fusions tion or by manipulating the expression o a (DfAZ-GUERRA et al., 1989), other Strains truncated ppGpp synthetase gene leads to a bearing mutations in a locus that turned out reduction in growth rate under conditions of et to coincide with appR, initially studied as a nutritional sufficiency (SARUBBI al., 1988 et regulatory gene for acid phosphatase synthe- SCHREIBER al., 1991). Interestingly an E. sis. The appR gene, formerly also referred to coli strain unable to make ppGpp has a pheas nur and katF, is now known as rpoS and notype somewhat like that of an rpoS mutant et encodes an RNA polymerase sigma factor (GENTRY al., 1993).
62
a

promoter
b
Primaryandclosely
/
Heat shock
Bacillus
sporulation flagellar SigB Idarillus)
5 elongation
complex
/6factorrelease
Fig. 2. (a) The role of u factors in initiation of transcription. Nearly all u factors are related to each other and share certain features. At (1) we emphasize two regions that interact with promoters; region 2.4 of any particular uinteracts about 10 bp upstream of the transcription start point ( + 1)at sequences characteristic of promoters dependent on that u,and region 4.2 interacts with DNA about two helical turns further upstream (-35). (2) The u-DNA interaction is largely dependent on association of u with RNA polymerase core enzyme to give RNA polymerase holoenzyme which is thereby directed to appropriate promoters. (3) Initially a closed promoter complex is formed in which the DNA remains double-stranded. RNA polymerase-promoter interactions may be significantly affected at this and the next stage by contacts with regulatory proteins, especially transcriptional activators, bound a short distance upstream. (4) The u factor plays an important role in melting the DNA around +1 to give an open complex. (5) RNA polymerase begins to transcribe, and the u factor is ejected and may become associated with another core enzyme particle to reinitiate the cycle. (6) Eventually the completed mRNA and the RNA polymerase core enzyme are released. The core enzyme may now potentially associate with a different u factor to initiate transcription from a promoter of a different class. (b) Phylogeny of u factors. The familial relationships of most known u factors from diverse bacteria are shown here. Arrows indicate u factors referred to in this chapter. The diagram is basically that of LONETTOet al. (1994).
It is not clear how the increased ppGpp levels might cause increases in the level of 2 (indeed, the mechanism of the stringent response is still elusive). HUISMAN and KOLTER (1994b) suggested a speculative model for the effect of ppGpp, contingent on the well-known association of increased ppGpp levels with increased expression of genes for amino acid biosynthesis (Fig. 4). It predicts that the resultant increase in the threonine pool would cause feedback inhibition of threonine biosynthesis, and so an increase in the pool of homoserine and homoserine phosphate (which normally feed into threo-
nine biosynthesis). These intermediates could then be cyclized to homoserine lactone via a known interaction with tRNA synthetases. Homoserine lactone is proposed to be the critical intracellular regulator for induction of rpoS. The key pieces of evidence to support this model are: (1) the discovery of an E. coli gene, rspA, encoding a product resembling a known lactonizing enzyme, which switches off rpoS transcription when present at high copy number (RspA may degrade the homoserine lactone); and (2) the elimination of rpoS expression by mutations blocking homoserine synthesis. The model was proposed following

Amino acid stwation Nutrient limitation
\\
Uncharged tRNA
A
Active ribosomes
Occupation of ribosome A site
\1 .
Carbon starvation
Activity of RelA
( ribosome-bound
\f
Re A inactive
u
~
63
Stringent response Increased synthesis o f amino acids
THREONINE
THREONINE
ppGpp synthetase)
Binding t o tRNA synthetases
LACTONE
Fig. 3. Activating ppGpp synthesis in E. coli. Most is known about how amino acid starvation causes ppGpp synthesis. The effects of carbon and especially phosphate starvation are comparatively little studied. SpoT can apparently cause ppGpp synthesis in a relA null mutant (HERNANDEZ BREand MER, 1991; XIAOet al., 1991), but its major physiological role is its ppGpp-degrading activity which is inhibited under carbon-starved conditions (GENTRY et al., 1993).
Increased expression o f rpoSls) bs
Fig. 4. Speculative model connecting increased ppGpp levels with increased production of d in E. coli. The diagram represents the model formulated by HUISMAN KOLTER and (1994b).
(Fig. 5). It has even turned out that a segment of the rpoS mRNA responds to osmotic stress, and a model has been proposed in the discovery that homoserine lactones are which (by analogy with the situation for the of widespread as regulatory molecules (see mRNA for the heat shock sigma factor d2 E. colz) an internal segment of the mRNA Sect. 2.2). The regulation of us is proving to be very binds to the translational initiation region, intricate (see HUISMAN KOLTER,1994a, acting as an antisense inhibitor (LANCEand and 1994). The stability of the for a review). Not only is its transcription ap- HENGGE-ARONIS, parently susceptible to subtle regulation by sense-antisense interaction is postulated to various metabolic influences, but there is also depend on an osmotically sensitive protein. regulation at the levels of translation and pro- Also, us half-life increases from about 2 min tein stability (HECKER and SCHROETER, during rapid growth to 10-20 min during staand 1985; LOEWENet al., 1993; MCCANNet al., tionary phase (LANCE HENGGE-ARONIS, et 1993; TAKAYANAGI al., 1994; LANGEand 1994; TAKAYANAGI al., 1994). Increased et HENGGE-ARONIS, 1994): indeed, posttrans- translation of rpoS mRNA at late exponential criptional regulation is the overriding in- phase also implies that cell density informafluence on us activity in minimal medium tion may be a component of 2 regulation,
64

Stationary phase,
/ -
.Degradatioy
Highcell density, Osmotic shock
Fig. 5. Complex control of the 2 subunit of E. coli RNA polymerase. The diagram is based on the data (1994) and TAKAYANAGI (1994). Transcription of rpoS et al. and models of LANGEand HENGGE-ARONIS is initiated from as many as four promoters. The CAMP-CRP complex (Fig. 1) inhibits use of one (or more) of these promoters, and ppGpp stimulates use of one or more of them, possibly by causing increased intracellular concentrations of homoserine lactone (HSL) (Fig. 4). Translation of the rpoS mRNA is thought to be limited by a (protein-stabilized?) secondary structure that sequesters the ribosome-binding site (RBS). This secondary structure is destabilized under some conditions, such as during osmotic shock, releasing the RBS for translation. The resulting usprotein is rapidly degraded during the growth phase, but is more stable in stationary phase, allowing it to direct RNA polymerase to transcribe stationary phaseassociated genes such as those determining microcin C biosynthesis.
perhaps involving extracellular substances (LANCEand HENGGE-ARONIS, 1994). Examples of such situations in other bacteria are discussed in the following sections. Not all stationary-phase activities in E. coli are regulated by us (LANCEand HENGGEARONIS,1991a; MCCANNet al., 1991). For example, induction of many proteins by glucose starvation is $-independent and requires the cAMP/CRP system (Fig. 1). A case in point is provided by some of the E. coli genes (notably the &CAY operon) for stationary phase-associated synthesis of glycogen under conditions of nitrogen limitation (ROMEOand PREISS,1989). Transcription of glgCA Y is unaffected by rpoS mutations (HENGGE-ARONIS FISCHER,1992), and and it is not clear what form of RNA polymerase is involved in glgCA Y promoter recognition (ROMEO and PREISS,1989). Both ppGpp and cAMP/CRP have significant regulatory impact on glgCA Y , particularly when the carbon source is not glucose. The &CAY operon is also subject to repression during growth by a 6.8 kDa protein encoded by the csrA gene which may also be an important regulator of
stationary phase genes (ROMEOet al., 1993). Interestingly, one of the genes needed for glycogen synthesis, glgS, maps away from the csrAlcAMP-CRPlppGpp-regulatedglg genes and shows a clear dependence on us (HENGGE-ARONIS FISCHER,1992). The and intricacy of stationary phase regulation is further illustrated by the finding that CAMP/ CRP may also influence the expression of rpoS (LANCEand HENGGE-ARONIS, 1994). Most of these stationary-phase regulatory devices are trans-acting, but at least one cisacting mechanism is known: the gearbox promoter. Such promoters may be defined by their property of enhanced relative expression at low growth rates, coupled with resemblance to a particular unusual - 10 consensus sequence (VICENTEet al., 1991). Gearbox promoters have been discussed mostly in the context of some cell cycle-related genes, the expression of which may have a specially important relationship to growth rate, as well as in relation to stationary phase. The promoter of the mcbA-G operon, which specifies the ribosomally synthesized antibiotic microcin B17, has gearbox kinetics and a gearbox-like
65
-10 region (VICENTE al., 1991). There is unable to make OHHL (BAINTONet al., et little information about what gives these pro- 1992a, b). Group 2 mutants are also pleiomoters their property of gearbox kinetics, but tropically defective in the production of varit is not the result of recognition by a particu- ious exoenzymes associated with the degradalar u factor (LANGEand HENGGE-ARONIS, tion of plant tissues during the disease process, and this entire phenotype can be re1991b). versed by OHHL. OHHL had been identified earlier as an extracellular signaling molecule in a different 2.2 A Critical Cell Population context: it is required to trigger light emission Density Signaled by an by the marine organism Vibrio fischeri Autogenous Extracellular Signal (MEIGHEN, 1991), as a very effective signal of increasing cell density (WILLIAMS, 1994; Molecule Triggers Carbapenem Fig. 7). OHHL synthesis requires the action Synthesis by Erwinia carotovora of the luxl gene product in a single-step biosynthesis from intermediary metabolism (posPLactam antibiotics are produced by div- sibly from S-adenosyl methionine and 3-0x0et erse bacteria including species of Strepto- hexanoyl coenzyme A) (EBERHARD al., myces, Nocardia, Flavobacterium, and Erwi- 1981). During growth at low cell densities, nia, as well as by fungi. There is surprisingly low-level expression of luxl results in a slow little information about the genetic regulation accumulation of OHHL in the environment. of plactam biosynthesis. The extreme amen- As cultures become denser, so the concentraability of many purple gram-negative bacte- tion of OHHL builds up. Since OHHL is preria, including the carbapenem producer Er- dicted to be freely diffusible through memwinia carotovora, to rapid genetic manipula- branes (because of its lipophilic side chain), tion has provided the most penetrating infor- the intracellular levels also increase until they mation so far. However, the different life- are high enough (lo- M) for effective bindstyles and ecologies of the different producers ing to a specific cytoplasmic receptor protein may mean that their regulatory systems have (LuxR). The LuxR-OHHL complex can diverged much more than the structural stimulate transcription of luxl, thereby causing increased OHHL synthesis and reinforcgenes. In apparent contrast to the situation de- ing the signal that activates luminescence. scribed above for microcin C7 production by Since luxl is part of an operon that also enE. coli, production of carbapenem by E. caro- codes luciferase, light emission is strongly actovora is regulated by a specialized extracel- tivated in response to a threshold level of lular signal. Carbapenem non-producing OHHL. Regulatory systems that recognize (Car-) mutants fall into two classes: group 1 critical levels of population density have been mutants produce N-(3-oxohexanoyl)-~-homo-called quorum sensors (FUQUA et al., serine lactone (OHHL) (Fig. 6; EBERHARD1994). In the model described earlier impliet al., 1981) which restores the Car+ pheno- cating homoserine lactone as a hypothetical type to group 2 mutants which are themselves intra- (rather than inter-) cellular regulatory factor in E. coli it is supposed that the lactone has no lipophilic side chain, so it is not released from the cell (HUISMAN KOLTER and 1994b). OHHL and functional homologs of it have turned out to be widespread. This has been demonstrated by introducing a V. fischeri lux 0 gene set deleted for luxl into E. coli, and exposing the transformed strain to culture suFig. 6 Structure of the luminescence autoinducer . N-(3-oxohexanoyl)-~-homoserine lactone (OHHL) pernatants of different bacteria. Luminescence was induced by samples from 18 strains of Vibrio fischeri.
b&
66
Fig. 7. Cell population-dependent expression of light emission mediated by the autoinducer OHHL of Vibrio fischeri. As the cell density increases, the concentration of freely permeating OHHL becomes high enough to bind to LuxR which then becomes an activator of the 1uxICDABE operon encoding a protein (LuxI) that causes OHHL synthesis (hence giving an autoinducing regulatory loop) and enzymes required for luminescence. The structure of OHHL is shown in Fig. 6.
of gram-negative bacteria of 13 genera and 6 strains of gram-positive bacteria of 5 genera (WILLIAMS, 1994). In several of the gram-negatives OHHL (or a related molecule) acts as an indicator of cell density, and each organism contains a gene that can substitute for luxl (SWIFT al., 1993). The luxl homologs et form a rather widely diverged family of genes, but many are accompanied by luxR-like genes encoding specific lactone-binding regulatory proteins (the LuxR family; FUQUA al., et 1994), each of which is presumably specific for a set of genes that determine a property responsive to cell population density. The emerging picture of these regulatory proteins is of an N-terminal domain for lactone binding and multimerization and a C-terminal domain with DNA-binding and transcription-activating regions. The C-terminal domain resembles those of some other transcriptional activators unconnected with quorum responsiveness, including the UhpA subfamily of response regulators (Sect. 2.3), and even of most (T factors (the LuxR superfamily; FuQUA et al., 1994). LuxR probably binds to a 20 bp inverted repeat centered at about -40 from the transcription start point, and similar sequences occupy equivalent positions in at
least two other OHHL-regulated promoters, for traA in Agrobacterium tumefaciens and lasB of Pseudomonus aeruginosa (FUQUAet al., 1994). Carbapenem production is the only example of antibiotic production known to be quorum-regulated by OHHL-like regulators in non-differentiating bacteria (though quorum regulation is a feature of production of some antibiotics by some differentiating bacteria including Bacillus subtilis and perhaps Streptomyces griseus; see Sects. 2.3,3.2.4, and 3.3.3). However, existing screening systems may well miss some OHHL-related compounds because there is considerable specificity for binding. This has been revealed both by analyzing the efficacy of synthetic analogs of OHHL (WILLIAMS, 1994) and by the finding that the lux genes of another luminescent species, Vibrio hurveyi, respond to a closely related molecule, N-(3-hydroxybutyryl) homoserine lactone (CAO and MEIGHEN, 1989), instead of OHHL.
67
2.3 The Non-Ribosomal Production of Peptide Antibiotics in Various Bacillus spp. Is One of a Number of Alternative Stationary-Phase Fates Determined by a Network of Transition State Regulators Involving Protein Phosphorylation
The dissection of stationary-phase functions has been most extensively analyzed in the gram-positive Bacillus subtilis 168. This reflects widespread interest in two of these functions, the formation of resistant endospores and the occurrence of natural competence for genetic transformation. Bacillus spp. often produce small peptide antibiotics during stationary phase, either through the action of large peptide synthetase enzymes or by the posttranslational modification of ribosomally synthesized propeptides. The non-ribosomally synthesized peptides include antibiotics such as bacitracin (B. licheniformis), gramicidin and gramicidin S (B. brevis), polymyxins (B. polymyxa), tyrocidine (B. brevis ATCC 8185), and surfactin (B. subfilis ATCC 21 332). Understanding of the regulation of production of these antibiotics was reviewed by MARAHIEL al. (1993) and has recently et been extended in the case of surfactin which is the main subject of this section. Surfactin production requires three large peptide synthetases encoded by three of the four consecutive genes in the 27 kb srfA operon (COSMINA al., 1993). The classical geet netic strain 168 of B. subtilis does not make surfactin, even though it contains an intact srfA operon. This deficiency is due to a frameshift mutation (the sfp allele) of the adjacent sfp gene (NAKANOet al., 1992) which may be involved in the export of small peptides (GROSSMAN al., 1993). et Transcription of the srfA operon is the major point of regulation of surfactin production, and it has turned out to involve a complex array of controlling elements including autoregulation, phosphorylation cascades, extracellular signaliig, and interplay with the regulation of different developmental path-
ways. All these influences appear to be transmitted through a single transcriptional activator, ComA, which also plays a key role in the onset of competence for transformation (DUBNAU, 1993; Fig. 8). ComA belongs to the response regulator class of transcriptional activators (WEINRAUCH al., 1989; Fig. 9), et activity of which is generally controlled by phosphorylation of a conserved aspartate residue. This residue, and others involved in forming a phosphorylation pocket, are present in ComA, so it is not surprising that its ability to bind to the srfA promoter in vifro is greatly enhanced by phosphorylation (ROGGIANO and DUBNAU, 1993). In its active configuration, ComA binds to each of two similar dyad elements upstream of the srfA promoter, and the bound proteins are presumed to interact with each other bringing about DNA bending (ROGGIANO and DUBNAU,1993; NAKANOand ZUBER, 1993) (Fig. 8). This multiple interaction stimulates srfA transcription by RNA polymerase bound to the promoter (probably via the major (+ factor, &; NAKANO al., 1991). Phosphorylation of reet sponse regulators such as ComA typically occurs when appropriate environmental conditions are sensed by a membrane-located histidine protein kinase, and ComA is no exception (WEINRAUCH al., 1990). Its partner et kinase, ComP, appears to behave as a quorum sensor, being activated by binding of an extracellular competence pheromone produced by the B. subtilis cells themselves. This pheromone, a 9-10 amino acid oligopeptide with a lipophilic modification, is formed from the C-terminus of the 55 amino acid product of the comX locus, in a process dependent on the product of the immediately adjacent upstream gene, comQ (come and comX are themselves immediately upstream of the genes encoding the ComP-ComA proteins involved in sensing and responding to the ComX pheromone; Fig. 8). Remarkably, a segment of the srfA operon plays an additional role, as part of a regulatory cascade leading to competence (DSOUZA et al., 1993) (Fig. 8). Surfactin itself is not involved in this cascade, since srfA mutations eliminating production do not all eliminate competence; and indeed, B. subtilis 168 which usually makes no surfactin is used as a labora-
68
Fig. 8. Signal cascades and networks leading to surfactin synthesis and other stationary-phase processes in Bacillus subtilis. The onset of surfactin synthesis is transcriptionally dependent on the phosphorylated form of the ComA response regulator. ComA phosphorylation is carried out partly by a cognate histidine protein kinase, ComP, which autophosphorylates a conserved histidine residue (H) when it interacts with an extracellular oligopeptide pheromone, ComX, which is a processed and modified form of the comX gene product. This pheromone and a second one, CSF, accumulate to effective concentrations only under conditions of high cell density. CSF also stimulates ComA phosphorylation by an unknown route involving the SpoOK transporter complex. Production of CSF is autoregulated via the sporulation phosphorelay which is activated partially by S p d K and partially by the histidine protein kinase KinB and another protein kinase, KinA, in response to unknown signals. The phosphorelay passes a phosphoryl group via SpoOF and S p d B to SpoOA, a response regulator protein that is both an activator of sporulation genes and a repressor of abrB, itself encoding a repressor of some sporulation genes and of CSF synthesis. Thus, activation of the phosphorelay leads to relief of CSF synthesis from repression (the figure also shows that the activation of surfactin synthesis is accompanied by the activation of competence; see text for further details).
tory organism because of its competence. Studies of constructed srfA partial deletions had shown that regulation of competence requires only the DNA region encoding the valine-activating domain of one of the peptide SINDEREN al., 1993), and et synthetases (VAN that aminoacylation activity itself is not needed (DSOUZA al., 1993). This unusual et situation has been clarified by the recent discovery that a distinct small open reading
frame, comS, within srfA encodes a critical regulatory element for competence (DSouZA et al., 1994) (Fig. 8). comS is translated in a different reading frame from the sequence encoding the valine-activating domain of srfA,giving a protein which appears to activate competence by interfering with a protein-mediated inhibition of ComK, a critical activator of the late competence genes comC, comG, and comDE (MSADEK al., 1994; et
2 Themes in the Regulation of Antibiotic Production Illustrated SIGNAL
69
Transcriptional activation o f target promoters
Fig. 9. Signal transduction in a bacterial two-component regulatory system. The conserved core elements of such systems are a protein kinase module capable of autophosphorylation at a conserved histidine residue (H), and a target response regulator module to which the activated phosphoryl group is transferred via an acidic pocket consisting of conserved aspartate residues, one of which (Asp5) becomes phosphorylated. These modules are usually on separate proteins though they can also be found within single proteins (ALEXand SIMON, 1994). In the example shown, the histidine kinase module forms the C-terminal, cytoplasmic domain of a protein whose N-terminal domain is a surface receptor for an unspecified extracellular signal. Binding of the ligand activates the kinase which then phosphorylates the N-terminal domain of a cytoplasmic response regulator. This activates a C-terminal DNA-binding domain of the regulator. In this example, binding to a specific DNA target sequence leads to transcriptional activation by contact with RNA polymerase at an adjacent promoter. The DNA-bindingkranscriptional activation modules of such response regulators themselves form subfamilies, some of which are found in regulatory proteins that are not part of two-component systems (e.g., LuxR; Sect. 2.2).
KONG and DUBNAU, 1994). It is not known why surfactin production has evolved this connection with competence. Activation of ComA can be detected, albeit at a reduced level, in comP mutants that lack the cognate protein kinase. This is accounted for by an as yet incompletely defined pathway from a second membrane-bound signal receptor. This receptor, encoded by the complex SPOOK locus, is an aggregate of several proteins and resembles various oligopeptide transport systems (hence the use of the acronym Opp to describe the SpoOK proteins). It is a member of the ATP-binding cassette (ABC) family of transporters (HIGGINS, 1992). The Opp complex responds to a second oligopeptide pheromone, CSF (competence stimulating factor), quite distinct from the ComX pheromone. The precise structure
and biosynthetic origin of CSF are not known, but its production is regulated by a route different from that of ComX pheromone: it is repressed by the AbrB protein, a repressor of several stationary-phase pathways including sporulation. Relief of these pathways from AbrB-mediated repression at the onset of stationary phase is usually brought about by phosphorylation of SpoOA, the central regulator of transition stage genes, via a phosphorelay system that is best known for its role in initiating sporulation (HOCH, 1993). The Opp complex is also one of several routes by which largely undefined physiological signals lead to transmission of phosphoryl groups to SpoOA via the SpoOF and S p d B proteins (RUDNER al., 1991; HOCH, 1993). et SpoOA has an NH2-terminal portion homologous to the phosphorylated domain of re-
70
sponse regulators (HOCH, 1993; Fig. 8). The SpoOA P protein represses ubrB, thereby derepressing CSF production in an autoregulatory loop that biases further development in the direction of sporulation, surfactin production, and competence development. SpoOA P also directly activates some genes, notably some involved in sporulation. This constellation of SpoOA P activities explains the pleiotropic sporulation, antibiotic production, and competence deficiencies of spoOA mutants, and of SPOOK, spoOB, and SPOOFmutants deficient in the phosphorelay. The full expression of spoOA, and hence the expression of srfA, also depends on a minor RNA polymerase 5 factor, &', which directs transcription of spoOA from an alternative, stationary phase-specific promoter. The regulation of &' is itself highly complex, with increased &' activity during entry into stationary phase involving both transcriptional and posttranscriptional control (HEALY et al., 1991). The stimuli for activation of the SpoO phosphorelay are not yet well understood. The multiple steps possibly provide the means to integrate sensory input from diverse sources (HOCH, 1993). For example, SpoOB is specified by the first gene in an operon that also encodes an essential GTP-binding protein, Obg, and Obg might cause phosphorylation of SpoOB in response to intracellular information about the cell cycle or the levels of GTP, which are critical in signaling the onset of sporulation (HOCH,1993). In a further twist of this increasingly complex system, srfA expression can be influenced by another regulatory protein, DegU, which belongs to the same subfamily of response regulators as ComA (HENNER et al., 1988; WEINRAUCH al., 1989): hyperet phosphorylated mutant forms of DegU repress srfA by an unknown mechanism (HAHN and DUBNAU, 1991).
duction with stationary phase is brought about by an almost bewildering variety of mechanisms. The initial switches typically result from a change in a constitutively synthesized regulatory protein, brought about by either intracellular or extracellular chemical signals (e.g., OHHL or oligopeptide pheromones). Extracellular signals may be membrane-diffusible and recognized by cytoplasmic binding proteins, or membrane-non-diffusible and recognized by membrane-bound proteins able to initiate an intracellular phosphorylation cascade. In all cases described so far, the end result of transmission of the initial signal is activation of transcription of structural genes for antibiotic biosynthesis. In the best characterized cases, this activation may arise either because of increased levels of a minor form of RNA polymerase containing a sigma factor such as 2 of E. coli that can recognize the promoters of the structural genes, or by the posttranscriptional activation of transcription factors needed for the major form of RNA polymerase to initiate transcription at the relevant promoters (e.g., phosphorylation of ComA in B. subtilis or OHHL-mediated conformational change of a LuxR homolog in carbapenem-producing E. curotovoru). It is abundantly clear from studies in E. coli and B. subtilis that antibiotic production requires the integration of diverse information through complex regulatory networks.
3 Regulation of Antibiotic Production in Streptomycetes and their Relatives

3.1 Introduction to the Organisms
Streptomyces spp. are the most versatile and commercially important producers of antibiotics and have traditionally been central to screening programs for new chemotherapeutic compounds. They are morphologically and
2.4 What Has Been Learned from Studies of Antibiotic Production in Unicellular Bacteria?
The examples discussed so far have revealed that the association of antibiotic pro-
71
phylogenetically distinct from the bacteria dealt with in the preceding sections. Their branching, mycelial growth habit is probably an adaptation that allows them to grow efficiently on the surface of insoluble organic debris in soil. Dispersal is by means of spores, typically borne on aerial hyphae, and in the laboratory mature Strepfomyces colonies have a furry appearance because of this aerial myto celium. The Acfinomycefules which streptomycetes belong form a division of the grampositive bacteria characterized by a high proportion of G + C in their DNA (on average 74mol% G + C ) . B. subtilis, on the other hand, belongs to the division with low G + C content. These divisions resulted from a very ancient evolutionary separation. Striking progress has been made in the last decade in the molecular analysis of antibiotic production by streptomycetes, notably in two model species, Streptomyces griseus, the producer of streptomycin, and Strepfomyces coelicolor A3(2), genetically the most-studied strain (reviewed by CHATERand HOPWOOD, 1993, and HOPWOOD et al., 1995). Studies in S. coelicolor have been helped by the extensive availability of natural and artificial genetic systems, the development of a combined physical and ge-
netic linkage map of the chromosome (KIESER et al., 1992), and the fact that the strain produces at least four antibiotics, production of one of which (methylenomycin A) is plasmid-specified (Sect. 3.3.4.5); two others are conveniently pigmented (actinorhodin is blue at high pH and red at low pH while undecylprodigiosin is red). Moreover, diverse aspects of the physiology and developmental biology of S. coelicolor have been studied providing important information to relate to studies of secondary metabolism. Much of the information reviewed below is drawn from S. griseus and S. coelicolor.
3.2 General Physiological Aspects of the Regulation of Antibiotic Production in Streptomyces

Like the organisms already described, streptomycetes grown in liquid media generally produce antibiotics during stationary phase or at low growth rates. (In the latter case, this may reflect production by cells inside mycelial pellets that may be nutritionally limited and that have, therefore, entered sta-
1.2
10.
60
0.10
0.08
?J
40
0.06
&
0.04 0
20
a z
2.
0.02
0.
Fig. 10. Growth phase-dependent production of candicidin by S. griseus in liquid culture. A Candicidin; A dry weight; 0 glucose; 0 DNA. Redrawn from MART~N MCDANIEL and (1975).
72
ot
Spores 8
24
40
56 Time [h]
72
88
104
duction of oleandomycin by surfacegrown cultures of s. antibioticus. 0 dry weight; 0 dry weight minus glycogen; A oleandomycin. Redrawn from MENDEZ et al. (1985).
Fig. 11. Growth phase-dependent pro-
tionary phase.) For example, candicidin was In Sects. 3.2.1-3.2.5 we review current unproduced by S. griseus in liquid culture only derstanding of the physiological factors that after net DNA synthesis had ceased (Fig. 10; might play a general role in triggering the onMARTIN and MCDANIEL, 1975). Biomass set of antibiotic synthesis, before moving to continued to increase slowly, presumably consider the genetics of antibiotic production through the accumulation of storage comin Sect. 3.3. pounds such as glycogen (BRANAet al., 1986) or triacyl glycerol (OLUKOSHI PACKTER, and 1994). Stationary phase antibiotic production 3.2.1 Metabolite Interference with could be viewed as a physiological abnormal- Antibiotic Production in ity that results from growing soil organisms in submerged culture, particularly since most Streptomycetes streptomycetes do not differentiate normally in liquid culture. However, antibiotic producGrowth is most rapid when readily utilized tion on solid media is also growth phase-de- carbon, nitrogen, and phosphate sources are pendent; thus, production of oleandomycin abundant, in part because of repression or inby Streptomyces antibioticus on agar began hibition by these metabolites of most inessenonly after growth had ceased (Fig. 11; M ~ N - tial processes. In principle, antibiotic producDEZ et al., 1985). Interestingly, oleandomycin tion could be subject to the same regulatory production appeared to be confined largely to controls, and its occurrence during stationary the substrate mycelium since synthesis of the phase might reflect relief from metabolite inantibiotic was apparently completed before terference after nutrient depletion. While aerial hyphae appeared. In other cases, anti- there are many examples of metabolite interbiotic production coincides approximately ference with antibiotic biosynthesis, especialwith the onset of morphological differentia- ly by glucose, ammonium, and phosphate tion, and the isolation from both S. coelicolor (Tab. l),the underlying mechanisms are genand S. griseus of bld mutants defective in both erally not known. In a few cases, repression of processes suggests at least some common ele- transcription appears to be involved, as in ments of genetic control (Sect. 3.3.2.6). glucose repression of phenoxazinone synth-

Tab. 1 Metabolites that Interfere with Antibiotic Production in Streptomycetes .
73
Interfering metabolite Carbon sources: Citrate Glucose Glycerol Nitrogen sources: Ammonium ions L-Glu, L-Ala, L-Phe, D-Val L-Tyr, L-Phe, ~ - T r p PABA , Inorganic phosphate:
Antibiotic Novobiocin Actinomycin, chloramphenicol, chlortetracycline, kanamycin, mitomycin, neomycin, oleandomycin, puromycin, siomycin, streptomycin, tetracycline, tylosin Actinomycin, cephamycin Actinorhodin, chloramphenicol, leucomycin, streptomycin, streptothricin, tetracycline, tylosin, undecylprodigiosin Actinomycin Candicidin Actinorhodin, candicidin, cephamycin, nanaomycin, nourseothricin, streptomycin, tetracycline, tylosin, undecylprodigiosin, vancomycin
Information from DEMAIN al. (1983). DEMAIN et (1992), and HOBBSet al. (1992).
Fig. 12. Effect of glucose (Gluc) on activity (left) and mRNA levels (right) of phenoxazinone synthase (PHS) in S. antibioticus (Gal, galactose). Redrawn from JONES (1985).
ase, the final enzyme in actinomycin biosyn1985; Fig. 12). thesis in S. antibioticus (JONES, Phosphate also appears to repress transcription of genes required for candicidin biosynthesis by S. griseus (Fig. 13; ASTURIAS al., et 1990) and of those for actinorhodin production in S. coelicolor (HOBBS al., 1992). et Unfortunately, the levels of nutrients and the identification of the growth-limiting com-
ponent(s) during culture have rarely been reported, so the extent of the contribution of metabolite interference to growth phase-dependent antibiotic production is difficult to assess. Furthermore, although metabolite interference might account for some examples of growth phase dependence, there appears to be no general pattern. Even within one species, the production of different secondary
74
2 Regulation of Bacterial Antibiotic Produc:tion

100
3.2.2 Antibiotic Production and Imbalances in Metabolism

An alternative or additional possibility is that antibiotic production is triggered by an imbalance in metabolism. Undecylprodigiosin, the major component of the red antibiotic of S. coelicolor (TSAOet al., 1985), is derived et partly from proline (WASSERMAN al., 1974; GERBER al., 1978). To determine whether et the amino acid incorporated into the antibiotic was synthesized internally or taken up from outside, HOODet al. (1992) isolated and characterized put mutants deficient in proline transport, which turned out to be defective also in proline catabolism. Since proline biosynthesis appears to be constitutive in S. coelicolor, such mutants might be expected to accumulate proline intracellularly. While this has not been determined experimentally, the mutants markedly overproduce the red antibiotic suggesting that undecylprodigiosin serves as a sink for excess proline. The need to remove surplus proline might reflect the role that it plays as an osmoregulant in other bacteria (KILLHAM FIRESTONE, and 1984a, b). It will be interesting to see whether the put mutants produce undecylprodigiosin earlier than the parental strain, as predicted by this hypothesis, and how this is mediated at the level of gene regulation. An imbalance in carbon metabolism may be responsible for triggering the production of methylenomycin A, another S. coelicolor antibiotic. Methylenomycin biosynthesis began at the same time as a rapid drop in the pH of a S. coelicolor fermentation (Fig. 14) caused by the efflux of a-ketoglutarate and pyruvate from the mycelium (HOBBSet al., 1992). Indeed, acid shock alone caused transient methylenomycin biosynthesis, suggesting that this might be a stress response to the change in pH that presumably reflects the imbalance in carbon metabolism (G. HOBBS and S. G. OLIVER, personal communication).
75
f %
3
50
25
0
Time [h]
Fig. 13. Phosphate repression of candicidin production in S. griseus. pabS encodes PABA synthase, and plays an early role in candicidin production. SPG, soya peptone-glucose medium. 0 SPG; W SPG+7.5 mM phosphate. Redrawn from ASTURIAS et al. (1990).
metabolites appears to be triggered by the depletion of different nutrients. In Sfrepfomyces cuttleyu, the production of melanin, cephamycin C, and thienamycin in batch culture occurred on depletion of glucose, ammonia, and phosphate, respectively. In a chemostat, cephamycin C production occurred at low growth rates that could be brought about by limiting for carbon, nitrogen, or phosphate, whereas the production of thienamycin required a low growth rate specifically associated with phosphate deficiency, consistent with the observations made in batch culture (LILLEYet al., 1981).

300 4
75
200 -
i
a z
g* 8
5 1 0
100-
n
0 -
Fig. 14a-c. Methylenomycin production during growth of S. coelicolor A3(2) on minimal medium with alanine as nitrogen source. a Growth and change in extracellular pH. I3 DNA; - Ln% COz (logarithm of the percentage carbon dioxide concentration in the fermenter exhaust gases); - - pH. b Glucose assimilation and methylenomycin production. A glucose; 0 methylenomycin. c Production of pyruvate and a-ketoglutarate. 0 a-ketoglutarate; A pyruvate. Redrawn from HOBBSet al. (1992).
0 250
'
8 0 .E
200
150
100
50
Time [h]
3.2.3 The Possible Role of Growth Rate and ppGpp in Antibiotic Production
Most of the published data are consistent with a role for growth rate, or the cessation of growth, in determining the onset of antibiotic production in streptomycetes. As discussed
earlier (Sect. 2.1), ppGpp is believed by many to play a central roie &-the growth rate con-et trol of gene expression in E. coli (SARUBBI al.. 1988: HERNANDEZ BREMER. and 1990. 1993; SCHREIBER al., 1991). A possible role et for ppGpp in triggering the onset of antibiotic production was first addressed in S. griseus by AN and VINING(1978). They found that streptomycin production occurred only after
76
a
gs
10
20
30
Time [h]
Fig. 15a-d. ppGpp and antibiotic production in S. coelicolor A3(2). a Growth (O),antibiotic, and ppGpp (A) production in S. coelicolor A3(2). The shaded boxes labeled ACT or RED denote the presence of actinorhodin or undecylprodigiosin, respectively. tD, doubling time. b Transcription of actll-ORF4 was monitored by S1 nuclease protection studies using RNA isolated at the times indicated from the culture shown in (a). Probe, the position of the full-length probe. EXP and STAT, exponential and stationary phases, respectively, and the shaded area between them indicates the transition phase. SM, size marker. Similar results were observed for redD transcription. c Growth curve of S. coelicolor A3(2) with (A) and without (0) nutritional shiftdown. The shaded boxes labeled (0RED) or (0ACT) denote the presence of undecylprodigiosin or actinorhodin, respectively in the control culture. SD indicates when the cultures were subjected to shiftdown, the shaded box labeled (A ACT) denotes the presence of actinorhodin in the culture subjected to shiftdown, and (0)the ppGpp level in the shifted culture. d Transcription of actllORF4 after nutritional shiftdown. S1 nuclease protection analysis of actll-ORF4 transcripts in RNA isolated at the times indicated from the culture shown in (c).

d
Hours after shiftdown
77
and HOLT et al. (1992) noted a peak of ppGpp accumulation that coincided with transcription of the pathway-specific activator gene, brpA, before production of bialaphos by Streptomyces hygroscopicus. However, BASCARAN al. (1991) concluded that there et was no obligate relationship between ppGpp and antibiotic production in Streptomyces clavuligerus. Production of cephalosporins (mainly cephamycin C) occurred during a phase of slow exponential growth (doubling Fig. 1Sd times of ca. 7 h and 18 h in rich and minimal media, respectively) and increased in stationary phase, while ppGpp levels remained cona drop in ppGpp levels, so it seemed that stant from the beginning of growth. relC-like ppGpp did not stimulate the initiation of anti- mutants accumulated lower levels of ppGpp biotic synthesis. Subsequently, ppGpp was than the wild-type strain after nutritional detected in several Streptomyces spp. (HA- shiftdown (varying between 8% and 85% of MAGISHI et al., 1980; SIMUTH al., 1979; HA- the wild-type levels), but there was no simple et MAGISHI et al., 1981; NISHINO and MURAO, correlation between the levels of ppGpp and 1981; STASTNA MIKULIK, and 1981), and po- cephalosporin production (some produced sitive correlations were observed between more, and others less, cephalosporin than the ppGpp and antibiotic biosynthesis in Strepto- wild-type strain). myces aureofaciens (SIMUTH al., 1979) and et Recently, the relationship between ppGpp Streptomyces galifaeus (HAMAGISHI al., synthesis and the production of undecylprodiet 1981). OCHI(1986) found that an accumula- giosin and actinorhodin has been examined tion of ppGpp following nutritional shiftdown in S. coelicolor (Fig. 15). Transcription of was accompanied by an 8-fold increase in for- pathway-specific activator genes for each antimycin production by Streptomyces lavendulae biotic (redD for undecylprodigiosin and MA406-A-1; moreover, a mutant deficient in actll-ORF4 for actinorhodin: see Sect. 3.3.4.1) ppGpp accumulation after amino acid deple- conspicuously increased as the culture was tion (the relaxed phenotype) was impaired in transition between growth and stationary in formycin production. Similar relC mutants phase, coinciding with the onset of ppGpp were isolated from S. antibioticus 3720 (OCHI, production; this was followed by transcription 1987), Streptomyces griseojlavus 1805 (OCHI, of representative red and act biosynthetic 1988), S. griseus 13189 (OCHI, 1990a), and S. structural genes and production of the antiet et coeficolor (OCHI, 1990b). The mutants accu- biotics (STRAUCH al., 1991; TAKANO al., mulated low levels of ppGpp after nutritional 1992). In contrast, transcription of a typical shiftdown (on average ca. 15% of wild-type rRNA gene set (rrnD) decreased markedly levels) and were deficient in antibiotic pro- (E. TAKANO M. J. BIBB,unpublished reand duction, leading to the idea that ppGpp plays sults). Thus a positive correlation was oba central role in triggering antibiotic produc- served between ppGpp accumulation and tion (OCHI, 1990b). A streptomycin-produc- transcription of actll-ORF4 and redD at the ing pseudorevertant of a relC mutant of S. gri- end of exponential growth. ppGpp synthesis seus remained defective in (p)ppGpp accumu- could also be induced by a nutritional shiftlation. This suppressor mutation was pro- down during rapid growth, and again resulted posed to identify a gene activated by ppGpp in a rapid and marked decrease in rrnD tranet and involved in triggering streptomycin proscription (STRAUCH al., 1991). Transcripduction. Consistent with OCHIS proposal, tion of actZZ-ORF4 was detected within KELLYet al. (1991) found reduced levels of 30min of shiftdown (Fig. 15) and transcripactinomycin biosynthetic enzymes and tion of the act biosynthetic genes followed and 1994). HowmRNA in a relC mutant of S. antibioticus, 30 min later (TAKANO BIBB,
78
ever, there was no immediate stimulation of redD transcription (TAKANO and BIBB, 1994), suggesting that if ppGpp does play a role in triggering the onset of antibiotic production, it is not always sufficient: the activation of at least some biosynthetic pathways may depend on additional factors. Such variations in regulatory responses of different sets of pathway genes in Streptomyces spp. are perhaps not surprising in view of comparable variations observed in E. coli and Bacillus spp. (see Sects. 2.1 and 2.3). Correlations between ppGpp synthesis and the onset of antibiotic production, where they occur, do not establish a causal relationship, and although the relC mutants isolated by OCHI (1986, 1987, 1988, 1990a, b) showed a marked reduction in antibiotic production, they also grow at about half the maximal rate of the parental strain, presumably reflecting impaired protein synthesis. Thus it is difficult to assess whether the effect on antibiotic production is a direct consequence of reduced levels of ppGpp or an indirect effect of the refC mutation on protein synthesis. It is interesting to note that mutants of E. coli unable to make ppGpp (ppGpp" mutants) do not show the marked reduction in growth rate characteristic of relC mutants (XIAO et al., 1991). Attempts to establish a role for ppGpp in triggering the onset of antibiotic production in Streptomyces spp. will require the isolation of the corresponding ppGpp" mutants and possibly the ability to regulate ppGpp levels in the absence of the physiological trauma associated with nutritional shiftdown, as was done by GENTRY al. (1993) in their et analysis of the effects of ppGpp on rpoS expression in E. coli (see Sect. 2.1).
3.2.4 Antibiotic Production and the Accumulation of Small Diffusible Signaling Compounds
The synthesis of threshold levels of small diffusible signaling molecules appears to play a central role in triggering production of at least some antibiotics in some streptomycetes (HORINOUCHI BEPPU,1992, 1995). Such and molecules might act simply as indicators of
cell population density or they might be synthesized in response to physiological conditions under which antibiotic production would be favorable to the organism. yButyrolactone compounds whose acylated lactone structures somewhat resemble the homoserine lactones found in gram-negative bacteria (see Fig. 6) have been detected in many streptomycetes, and have been implicated in antibiotic production and morphological differentiation in several species (Fig. 16). The most intensively studied example is A-factor (2-isocapryloyl-3R-hydroxymethylybutyrolactone) which is required for streptomycin production and morphological differentiation in s. griseus (Sect. 3.3.3). Five related compounds, virginiae butanolides A-E (Fig. 16a; VB-A-E), are inducers of virginiamycin production by Streptomyces virginiae (YAMADA al., 1987), and compounds with et the same biological activity are found in other streptomycetes (OHASHIet al., 1989). Such butyrolactones probably diffuse readily through membranes, so extracellular and intracellular levels are likely to be the same: hence, just as with LuxR and OHHL, monitoring of extracellular concentrations is done by cytoplasmically located binding proteins with very high affinity and specificity for their ligands (KD ca. 10-9M). To illustrate this specificity, virginiae butanolide C shows no biological activity in vivo with A-factor-deficient mutants of S. griseus and the A-factor et receptor does not bind it in vitro (MIYAKE al., 1989). However, ligand specificity can vary between strains, and A-factor analogs with acyl chains of different lengths, or with a hydroxyl group rather than a carbonyl group at position 6, had some activity in an anthracycline-producing S. griseus strain that requires A-factor for sporulation and antibiotic production (GRAFE et al., 1982, 1983). The receptor proteins have not been characterized - an earlier report that a cytoplasmic binding protein for the S. virginiae factor VB-C showed significant sequence homology to NusG (OKAMOTO al., 1992), an E. coli proet tein believed to play a role in transcriptional antitermination, appears to have been illfounded (PUTTIKHUNT al., 1993). How Aet factor binding protein influences antibiotic production is addressed in Sect. 3.3.3.
79
S. coelicolor does not make A-factor. However, it does make a series of structurally very similar compounds (ANISOVA al., 1984; et EFREMENKOVA 1985; Fig. 16b), and at et al., least some of these can apparently stimulate streptomycin production by S. griseus mutants deficient in A-factor biosynthesis (HARAet al., 1983). afsA mutants of S. coelicolor that had lost the ability to cross-stimulate the S. griseus A-factor-deficient mutants were unaffected in antibiotic production and morphological differentiation. However, it may be that the afsA mutants lack only some of the A-factor-like compounds made by the parental strain, so y-butyrolactones might conceivably play a role in triggering antibiotic production in S. coelicolor. Alternatively, antibiotic production in S. coelicolor might perhaps have lost a requirement for such A-factor-like compounds, a situation reminiscent of the A-factor receptor-deficient mutants of S. griseus (Sect. 3.3.3); the failure to detect any A-factor-binding protein in cell extracts of S. coelicolor is consistent with this hypothesis.
3.3 Genetics of Antibiotic Production

3.3.1 Organization of Antibiotic Biosynthetic Genes and Clusters
In order to clarify discussion of the genetic regulation of antibiotic biosynthesis, we first briefly summarize the organization of the biosynthetic genes themselves. Genes specifically involved in the production of a particular antibiotic are invariably found clustered together, and only one set, for methylenomycin production (Sect. 3.3.4.9, is known to be plasmid-located rather than chromosomal. Thus, all of the act and red genes of S. coelicolor occur in chromosomal segments of ca. 23 kb and 35 kb, respectively. Transfer of these segments to Streptomyces parvulus, a host not known to make any structurally similar compounds, caused synthesis of the antibiotics (MALPARTIDA HOPWOOD,1984; MALand PARTIDA et al., 1990). Similarly, expression in surrogate hosts has led to the demonstration that the entire biosynthetic pathways for production of tetracenomycin by Streptomyces glaucescens and puromycin by Streptomyces alboniger are located in DNA segments of 12.6 kb and 15 kb, respectively (DECKER and 1993; LACALLEet al., 1992). HUTCHINSON, The biosynthetic genes generally seem to be organized into several transcription units of varying complexity. Pathway-specific regulatory genes have been identified in several of the clusters (Sect. 3.4), although there are exceptions that seem so far to lack such genes, e.g., the tetracenomycin pathway and the very large biosynthetic clusters (ca. 45 kb and 95 kb, respectively) for the macrolides erythromycin (made by Saccharopolyspora erythraea; DONADIO al., 1993, and references et therein) and avermectin (produced by Streptomyces avermitilis; MACNEIL et al., 1992). The biosynthetic clusters usually also contain one or more genes that confer immunity to the antibiotic, which vary considerably in their mode of action. For example, resistance to erythromycin in S. erythraea is accomplished by a 23s rRNA methylase that modifies the target of the antibiotic by N6 dimethy-
3.2.5 Summary
The various physiological factors that influence the onset of antibiotic production in streptomycetes do not fit into a simple unifying model. However, it seems reasonable to propose an overall regulatory influence of growth rate with superimposed pathway-specific regulatory effects influencing the production of individual antibiotics. Such effects could include responsiveness to catabolite repression or inhibition, imbalances in metabolism, and environmental signals and stresses. It is clear that low molecular-weight effectors, like A-factor, play essential roles for some antibiotics. Whether these are produced as a consequence of a reduction in growth rate, or in response to some extrinsic factor, or via some autoregulatory circuit such as that described for LuxR and OHHL remains to be determined, as does the potential role of ppGpp as an intracellular signaling molecule.
80
a
2 Regulation of Bacterial Antibiotic Production A-factor
Q o0
S. griseus
OH
OH
VB-D
S. bikiniensis and S. cyaneofuscatus
*AoH Factorl /1"4oH

n
w
OH
S. viridochromogenes
OH
IM-2
OH
Virginiae butanolides from S. virginiae
OH
Streptomyces sp. FRI-5
Fig. 16a, b. y-Butyrolactones produced by streptomycetes. a A-factor-like compounds in streptomycetes. b A-factor-like compounds from S. coelicolor. Structures taken (1992) and from HODGSON EFREMENKOVAal. et
(1985).
lation of a specific adenine residue (THOMPSON et al., 1982). Different types of efflux systems appear to confer resistance to different antibiotics; thus srmB, one of four spiramycin resistance genes in Streptomyces umbofuciens, appears to encode an ATP-dependent transport system for this macrolide antibiotic (SCHONER al., 1992), whereas the efflux et systems for tetracenomycin, methylenomycin, and probably actinorhodin encoded by the tcmA (GUILFOILE HUTCHINSON, and 1992a), mmr (NEAL and CHATER,1987) and uctZZ-
ORF2 (FERNANDEZ-MORENO al., 1991) et genes of S. gluucescens and S. coelicolor are presumed to be driven by transmembrane electrochemical gradients. In some cases, expression of the resistance gene appears to be induced by the corresponding antibiotic (Sect.
335. ..)
81
OH
OH
Q-2a
0aA&-2b
G-2c
02 Q d
OH
OH
Fig. 16b.
OH
3.3.2 Genes that Pleiotropically Affect Antibiotic Production in Streptornyces coelicolor: Introduction and Overview
Many genes have been identified that pleiotropically affect antibiotic production in S. coelicolor, and several of these are likely to play a global role in regulating the onset (and perhaps maintenance) of antibiotic synthesis. Mutants in about half of these pleiotropic genes also show deficiencies in morphological differentiation. These "bld" mutants are discussed in Sect. 3.3.2.6. Some of the work on regulatory genes has involved use of a very close relative of S. coelicolor, S. lividuns 66, which also contains act and red gene sets, but generally expresses them rather poorly. S. Zividuns is a slightly more convenient recipient strain for transformation than S. coelicolor.
3.3.2.1 afsB, afsR, and afsK - A Role for Protein Phosphorylation in Triggering the Onset of Antibiotic Production?
ufsB mutants resemble ufsA mutants (Sect. 3.2.4) in that they cannot induce streptomycin production and sporulation of A-factor-deficient mutants of S. griseus grown near them. However, unlike ufsA mutants, they are defective in actinorhodin and undecylprodigiosin synthesis (HARA al., 1983). (Production et of the other two antibiotics known to be made by S. coelicolor, methylenomycin and a calcium-dependent antibiotic (CDA), appears to be normal; ADAMIDIS and CHAMPNESS, 1992.) Northern analysis failed to detect transcripts corresponding to uctl, uctZZ (including actZZ-ORF4, the pathway-specific activator gene), uctZZZ, and uctVZ in the ufsB mutant BH5 (HORINOUCHI al., 1989a). Attempts et to clone ufsB, which was presumed to regulate production of the pigmented antibiotics and the A-factor-like compounds, have so far failed. However, they yielded ufsR, which was obtained by screening of a library of S. coeli-
82
color DNA for overproduction of undecylprodigiosin and actinorhodin in an ufsB-like mutant of S. lividuns (HH21) that could not cross-feed an A-factor-deficient mutant of S. griseus (HORINOUCHI al., 1983; STEINand et COHEN, 1989). At high copy number, a DNA fragment containing ufsR and ufsR2 (see below) suppresses the afsB phenotype in S. coeet licolor (HORINOUCHI al., 1983) and stimulates transcription of act genes in S. coelicolor and S. lividuns (HORINOUCHI al., 1989a). et ufsR encodes a protein of 933 amino acids that contains putative DNA-binding helixturn-helix motifs towards the C-terminus and potential ATP-binding sites towards the midet dle of the protein (HORINOUCHI al., 1986, 1990); site-directed mutagenesis of the latter resulted in a 4-fold reduction (though not an elimination) of stimulatory activity in S. lividuns (HORINOUCHI al., 1990). A similar reet duction was observed when fragments corresponding to the N-terminal264 or 510 amino acids of AfsR, rather than the entire coding region, were expressed in S. lividuns; even cloned fragments containing the C-terminal half of AfsR elicited a stimulatory effect in S. lividuns and restored actinorhodin, undecylprodigiosin, and A-factor production to afsB mutants of S. coelicolor (HORINOUCHI al., et 1983, 1990), although this effect could also be due to ufsR2, a recently identified small gene located immediately downstream of ufsR and also capable of stimulating actinorhodin and undecylprodigiosin production (VOGTLI et al., 1994; see below). Disruption of ufsR in S. coelicolor using DNA fragments from the 5, middle, and 3 regions of the ufsR coding sequence led to only a 4-fold reduction in actinorhodin production, which was also delayed (HORINOUCHI al., 1990), suggesting that et ufsR is not essential for antibiotic biosynthesis. AfsR is phosphorylated in vitro by the membrane-bound product of afsK which lies downstream of, and in the opposite orientation to, ufsR (HONGet al., 1991; HORINOUCHI, 1993). AfsK (799 amino acids) shows significant similarity to eukaryotic serinethreonine protein kinases, and phosphorylation of AfsR occurs at serine and threonine residues (MATSUMOTO al., 1994). Moreet over, phosphorylation of AfsR in vitro is severely reduced by K-252a and staurosporine,
inhibitors of eukaryotic protein kinases (HONGet al., 1993). Disruption of ufsK, like that of ufsR, resulted in reduced levels of actinorhodin production, though the reason for this is not clear, since AfsR could still undergo phosphorylation at serine and threonine et residues in the afsK mutant (MATSUMOTO al., 1994). Clearly some other protein kinase can phosphorylate AfsR, a conclusion consistent with emerging evidence of multiple protein kinases in S. coelicolor (WATERSet al., 1994). Interestingly, the N-terminal region of AfsR resembles the pathway-specific regulatory proteins RedD, ActII-ORF4, and DnrI (Sect. 3.3.4.1). Unlike redD and actll-ORF4, whose transcription increases dramatically during transition phase, transcription of ufsR occurs throughout exponential growth and declines slowly on entry into stationary phase (E. TAKANO and M. J. BIBB, unpublished). Thus, any major role of ufsR in activating genes expressed in stationary phase may depend on posttranscriptional regulation of ufsR, a deduction consistent with the modification of AfsR by phosphorylation. A previously undetected gene, ufsR2, has recently been identified in S. lividans (VOGTLI et al., 1994). afsR2, which appears to occur in a similar location in S. coelicolor (MATSUMOTO et al., 1994; VOGTLIet al., 1994), is transcribed in the same direction as ufsR but from its own promoter and encodes a 63 amino acid protein. When cloned at high copy number, afsR2 suppresses an ufsB mutation in S. lividuns and stimulates actinorhodin and undecylprodigiosin production in S. coelicolor. The stimulatory effect on actinorhodin production in both species appears to be mediated through transcription of uctll-ORF4, and is retained after deletion of most of the C-terminal domain of the chromosomal copy of ufsR in S. lividans. Fragments of S. coelicof lor DNA containing the C-terminal portion o afsR stimulated actinorhodin production (HORINOUCHI al., 1983, 1990; see above); et since these fragments were likely to have contained ufsR2, it is possible that the latter, rather than the C-terminus of AfsR, was responsible for the increase in antibiotic production. Since deletion of a chromosomal segment of S. lividuns that included afsR2 (and part of ufsR) did not completely block blue
83
pigment production, ufsR2 does not appear to play an essential role in actinorhodin production.
3.3.2.3 abaA Influences the Production of Three of the Four Antibiotics Made by Streptomyces coelicolor
ubuA of S. coelicolor was isolated by virtue of its ability to stimulate actinorhodin production in S. lividuns when cloned on a high-copy number plasmid; the effect on undecylprodigiosin production was not reported (FERNANDEZ-MORENO al., 1992). Sequencing of a et 2 kb Pstl fragment revealed five short ORFs, with ORFs A, B, and C transcribed divergently from ORFs D and E. ORFB and 137 nucleotides of downstream sequence were sufficient to give the same stimulatory phenotype in S. lividuns, and disruption of the chromosomal copy of ORFB in S. coelicolor resulted in loss of actinorhodin production, almost complete loss of undecylprodigiosin synthesis, a reduction in CDA production, but no effect on methylenomycin. When cloned at high copy number, ubuA was unable to confer actinorhodin production on a mutant deficient in the pathway-specific activator gene uctllORF4, consistent with a location higher up in any putative regulatory cascade.
3.3.2.2 afsQl and afsQ2 - A Two-Component Regulatory System that Can Influence Antibiotic Production
ufsQl and ufsQ2 were isolated in the same way as ufsR (ISHIZUKA al., 1992). Seet quence analysis of a 1.3 kb Kpnl-Pstl fragment of S. coelicolor DNA that stimulated actinorhodin, undecylprodigiosin, and A-factor production in S. lividuns HH21 identified ufsQl whose predicted product is homologous to bacterial response regulator genes (Fig. 9): AfsQl belongs to the OmpR subfamily (VOLZ, 1993). ufsQ2, which was subsequently discovered downstream of ufsQl, appears to be translationally coupled to it. AfsQ2 belongs to the family of sensory histidine protein kinases; thus the genes appear to constitute a two-component regulatory system (Fig. 9 STOCKet al., 1990 ALEXand SIMON,1994). AfsQ2 is presumed to be a membrane protein (it has putative membrane spanning domains towards its N-terminus) which is thought to be autophosphorylated at His294in response to an unknown signal; the phosphate group may then be transferred to Asp5*of AfsQl. AfsQl is, or interacts with, a transcriptional activator (ISHIZUKAet al., 1992). Evidence for this model of AfsQ2 action was obtained by changing His294 to G ~ uwhich resulted in a loss of stimulatory ~ ~ ~ , activity in S. lividuns. Cloned fragments containing only ufsQl gave the same level of stimulation as those containing both genes. However, disruption of both genes in S. coelicolor had no obvious phenotypic effect. Thus ufsQl and ufsQ2 either are inessential for antibiotic production or operate under as yet undefined physiological conditions. Alternatively, the stimulatory effects of ufsQl may reflect the ability of AfsQ1, when present at high levels, to substitute for AbsA, a response regulator that clearly does play a role in antibiotic production (Sect. 3.3.2.4).
3.3.2.4 absA and absB - Mutants Isolated on the Basis of a Pleiotropic Defect in Antibiotic Production
An extensive screen for UV-induced mutants deficient in both actinorhodin and undecylprodigiosin production led to the identification of ubsA and ubsB (ADAMIDIS al., et 1990; ADAMIDIS CHAMPNESS, and 1992); mutants of both classes were also defective in CDA and methylenomycin synthesis (though they expressed methylenomycin resistance). The rarity of ubsA mutants (5.10-6 per survivor) suggested that they may represent a particular allelic form (CHAMPNESS al., 1992). et Pseudorevertants of an ubsA mutant were obtained that fell into two classes: sub(1) pseudorevertants which overproduced actinorhodin and undecylprodigiosin, and sub( 11) which
84
made wild-type levels of antibiotic. Both classes contain suppressor mutations mapping close to the starting absA mutation, but recombining with it. Sequencing of a DNA fragment that complements the absA mutation showed that its product is homologous to bacterial sensory histidine kinases, and preliminary data suggest that a homolog of response regulators is located immediately downstream (P. BRIAN and W. CHAMPNESS, personal communication). Interestingly, additional copies of afsQl restored actinorhodin production to an absA (but not absB) mutant (ISHIZUKA et al., 1992), possibly indicating crosstalk between the two systems (Sect. 3.3.2.2). absB mutants sporulate less well than their progenitor and produce low levels of actinorhodin, undecylprodigiosin, and methylenomycin on some media. Attempts to clone absB on a low copy number vector by screening for restoration of actinorhodin production led to the isolation, in addition to actZZ-ORF4 and afsR, of a cloned fragment which fully complements the absB mutation (T. ADAMIDIS and W. CHAMPNESS, personal communication).
3.3.2.6 Genes that Affect Both Antibiotic Production and Morphological Differentiation
In surface-grown cultures of streptomycetes, antibiotic production generally coincides with the onset of morphological differentiation (Sect. 3.2). The isolation of bld mutants defective in both processes points to at least some common elements of genetic control. It is important to note that the morphological deficiencies of some classes of bld mutants can be suppressed nutritionally or in some cases by cross-stimulation by diffusible factors. Phenotypic suppression of the pleiotropic defect in antibiotic production has been observed in only a few cases and generally not under conditions that suppress the morphological deficiency (an exception to this is provided by bldH mutants; see below). The most extensive studies of bld mutants have been made in S. coelicolor in which at least 10, and perhaps 11, different classes of bld mutants with deficiencies in antibiotic production - bldA, B, D, E, F, G, H, Z, -17, -21, -830 - have been identified (reviewed by CHAMPNESS CHATER,1994). All mutant and 3.3.2.5 miu - Multicopy Inhibition classes except bldE and bldF (which both of Antibiotic Production produce abundant undecylprodigiosin) are deficient in both actinorhodin and undecylAttempts to clone absA on a high copy prodigiosin synthesis and most are also definumber plasmid led to the identification of S. cient in methylenomycin and CDA produccoelicolor DNA that inhibited the production tion. Antibiotic production is restored to of all four antibiotics. The DNA fragment, bldH mutants grown on mannitol instead of which was isolated repeatedly, had no inhibi- glucose, and undecylprodigiosin is produced tory effect at low copy number (CHAMPNESS by bldA mutants grown at low phosphate et al., 1992). Transcription of redD and actZZ- concentrations. Four of the bld genes ( A , B, O W 4 is undetectable in strains containing D, and G ) have been cloned, and detailed the fragment on a high copy number plasmid characterization has been reported for bldA. (W. CHAMPNESS, personal communication). Remarkably, bldA encodes the only tRNA in Subcloning localized the inhibitory function, S. coelicolor and S. lividans that can translate termed mia, to a 363 bp Sau3Al fragment the rare leucine codon UUA efficiently et et that does not appear to be protein-coding. It (LAWLOR al., 1987; LESKIW al., 1991a). is not known whether the inhibitory effect re- The lack of expression of the xylE reporter gene when fused to act, red, or mmy transults from the DNA itself or its transcript. scription units in bldA mutants suggests that the defect in antibiotic production reflects a failure to transcribe the biosynthetic structural genes even though bldA encodes a component of the translational apparatus (GUTHRIE
85
and CHATER,1990 BRUTON al., 1991; A. et WIETZORREK and K. F. CHATER, unpublished). An explanation for the failure to transcribe act genes is to be found in the presence of a TTA codon in the pathway-specific regulatory gene acrZZ-ORF4. If this codon is changed to the synonymous codon TTG, actinorhodin production takes place even in a bldA mutant (FERNANDEZ-MORENO al., et 1991). Phenotypically similar bldA mutants have also been isolated in the phylogenetically more distant S. griseus (MCCUE et al., 1992), suggesting that the role of bldA in secondary metabolism and differentiation is widespread among streptomycetes. The unimpaired vegetative growth of bldA mutants indicates that TTA codons are absent from genes essential for primary metabolism and growth, but l'TA codons have been found in several genes likely to be expressed late in growth. This, coupled with evidence that bldA-specific RNA is more abundant late in surface growth (LAWLOR al., 1987), proet vided support for the idea that bldA regulates antibiotic production by allowing the translation of UUA codon-containing mRNA only under appropriate conditions (LESKIW al., et 1991b). However, caution is necessary in assuming an active regulatory role for bldA, since one series of detailed experiments on liquid-grown cultures of S. coelicolor failed to reveal any limitation of translation of the acfll-ORF4 UUA codon during exponential growth (GRAMAJO et al., 1993). Together with the transition phase activation of acrZZORF4 transcription, these results were consistent with a more prosaic possibility that the absence of TTA codons from vegetatively expressed genes might reflect selection against codons that were inefficiently translated during growth, rather than a role for bfdA in the temporal regulation of actinorhodin production. On the other hand, the observations of LESKIW al. (1993) tend to support a regulaet tory role for bldA. Northern analysis of RNA from surface-grown cultures indicated that the amount of the bldA transcript increased with growth, and S1 nuclease protection assays revealed an increase in the level of the 5 ' end of the mature bldA transcript late in growth, both in rich liquid media (this was not observed by GRAMAJO al., 1993) and et
in surface-grown cultures; furthermore, the efficiency of translation of seven UUA codons of a heterologous reporter gene apparently increased in older cultures. The differences between the two sets of results may reflect the different growth conditions used possibly the liquid culture conditions of GRAMAIO et al. (1993) overrode a regulatory role of bldA adapted for surface growth.
3.3.2.7 An Outline Scheme for the Interactions of Pleiotropic Antibiotic Regulatory Genes in Streptomyces coelicolor
No satisfactory integrated model has yet emerged for the roles of the various pleiotropic regulatory genes in antibiotic production (probably because not enough of the pieces of the jigsaw are yet available), but here we summarize some of the key features that must be taken into account. For actinorhodin and undecylprodigiosin, expression of the pathway-specific activator genes actZZORF4 and redD, respectively, appears to play a major limiting role in determining the onset of antibiotic production (TAKANO et al., 1992; GRAMAJO al., 1993), so it is attractive et to propose that all the pleiotropic genes (ufs, aba, ubs, mia, bld) influence the synthesis of these two antibiotics via acrll-ORF and redD. The generalized and simplified scheme in Fig. 17 is built from the following observations and deductions. (1) Transcription of acrll-ORF4 is virtually undetectable in an afsB mutant, at least under certain culture conditions, suggesting that the (still uncharacterized) AfsB gene product may be higher than ActII-ORF4 in a transcriptional cascade (HORINOUCHIet al., 1989a). (It should be noted that afsB mutants are noticeably leaky in their actinorhodin deficiency on a variety of different media.) (2) absA and absB, whose mutant phenotype proves their importance for antibiotic biosynthesis, may perhaps play a role in maximizing expression of the pathway-specific activator genes, since the introduction of acfll-ORF4 and redD on high copy number plasmids res-
86
Fig. 17. Factors potentially determining the onset of antibiotic production in streptomycetes. Thinner lines present plausible interactions for which there is currently no direct evidence.
tores production of the relevant antibiotic in absA and absB mutants (T. ADAMIDIS and W. CHAMPNESS, personal communication). Alternatively, the abs genes may encode accessory elements normally required for actinorhodin and undecylprodigiosin synthesis which are rendered unnecessary by overproduction of ActII-ORF4 and RedD. (3) Multiple copies of afsR and afsR2 stimulate actinorhodin production, but appear to depend on actZZ-ORF4 for this effect (B. FLORIANO and M. J. BIBB, unpublished results; T. ADAMIDIS and W. M. CHAMPNESS, personal communication; VOGTLI et al., 1994), while multiple copies of segments encoding the C-terminal portion of AfsR, and in retrospect containing afsR2, confer actinorhodin and undecylprodigiosin production on absA and absB mutants (CHAMPNESS al., 1992). et (4) Taken together, observations (1)-(3) suggest a working model in which expression of afsR and afsR2 or the activities of their products, depend on absA and absB, and in which AfsR2 and AfsR (perhaps in its phosphorylated form) stimulate expression of actllORF4 (and possibly redD). ( 5 ) Extra copies of afsQl restore actinorhodin and undecylprodigiosin production to absA (but not absB) mutants (ISHIZUKA .al., et 1992), suggesting that afsQ may depend on absA for a role in enhancing acrll-ORF4 and redD activity, unless a high copy number of
afsQ takes the activity of its product above a threshold level. If afsQl has such a role, it is redundant in the wild-type strain under laboratory conditions; perhaps this role can also be filled by afsR, a question that should be resolved by isolating afsR and afsR afsQ nullmutants. It remains possible that the high copy number effects of afsQ result from artificially induced cross-talk between normally separated regulatory elements. (6) The recent sequence analysis of absA, together with the published data on afiRlafsK and afsQllafsQ2, strongly suggest that protein phosphorylation, and potentially phosphorylation cascades, play a role in triggering antibiotic production; presumably AfsK, AfsQ2, and AbsA sense external signals that cause phosphorylation of their regulatory counterparts (AfsR, AfsQ1, and the product of a gene located downstream of absA) which can then stimulate transcription of the antibiotic biosynthetic pathways, perhaps via the pathway-specific regulators. (7) None of the mutants described above are unconditionally and completely defective in production of all four antibiotics; even absA produces actinorhodin on some media (W. CHAMPNESS, unpublished data), and antibiotic production in several of the others shows media dependence. This may indicate a complex regulatory network in which there are several different routes to activation of a par-
87
ticular pathway. (Alternatively, the available specific regulators, nor whether they act in a linear cascade or by convergence. Furthermutants may not be truly null.) (8) Little is known of how abaA fits into this more, there is little information about the interactive scheme, but in affecting three of regulation of expression of the characterized the four antibiotics it differs from both the afs pleiotropic regulatory genes, and several releloci (which appear to affect only actinorhodin vant bfd genes remain uncharacterized. and undecylprodigiosin) and the abs loci (which affect all four). (9) bfdA is the only bfd gene whose mode of 3.3.3 Streptomyces griseus - The action is (partially) understood. bldA dependence of actinorhodin production appears to A-Factor Cascade be exerted entirely at the level of translation Although diffusible factors have been imof the unique UUA codon in the actll-ORF4 transcript. Undecylprodigiosin production, plicated in the production of several antibioexcept at low phosphate levels, also requires tics in streptomycetes (see Sect. 3.2.4), the bldA, though neither redD nor, apparently, role of A-factor in the production of streptoany of the red biosynthetic structural genes mycin in S. griseus is by far the best charactercontain TTA codons (NARVAand FEITEL- ized. A-factor was discovered by KHOKHLOV SON, 1990 GUTHRIE and CHATER,1990). In et al. (1967). It was found to be required for contrast to actll-ORF4 whose transcription is both streptomycin production and sporula1982) and also not bldA-dependent, transcription of redD tion in S. griseus (KHOKHLOV, could not be detected in a bldA mutant (J. for streptomycin resistance (HARAand BEPWHITEand M. J. BIBB,unpublished results). PU, 1982a). Subsequent genetic and molecular Presumably there is at least one other gene analyses have provided considerable insights required for redD transcription whose tran- into its mode of action. A-factor accumulates script does contain a UUA codon. The pwb to detectable levels in the culture medium just mutations which restore undecylprodigosin, before the onset of streptomycin production but not actinorhodin or aerial mycelium for- (HARAand BEPPU, 1982b). It appears to be mation, to bldA mutants and which may map freely diffusible across the cytoplasmic memand within the red cluster (E. P. GUTHRIE K. brane and binds to a cytoplasmic A-factorF. CHATER,unpublished) could be relevant binding protein of approximately 26 kDa with a stoichiometry of 1:1 and a dissociation conhere. (10) Antibiotic production is associated with stant of 0.7 nM, consistent with the similar reduced growth rate. One of the signals impli- low concentrations of A-factor required for cated is an increased ppGpp level. No other biological activity. The receptor protein is candidate signal, intracellular or extracellular, present at about 30-40 copies per genome. It has been described that might have pleiotrop- is thought that binding of A-factor prevents ic activity, though extracellular acidification the receptor protein from acting as a represcan activate methylenomycin production (see sor of a hypothetical gene (X) required for Sect. 3.2.2). Whatever the signals, evidence is both sporulation and streptomycin producgrowing that they may lead, directly or indi- tion (Fig. 18). A negative regulatory role for rectly, to phosphorylation of several regulato- the binding protein is indicated by the discovry proteins such as AfsR, AfsQ1, and AbsA ery that S. griseus mutants unable to make Aby specific kinases and thence, via pathway- factor can undergo further mutations that resspecific regulatory genes and their products, tore streptomycin production and sporulation to transcription of genes encoding antibiotic by eliminating A-factor-binding protein. Gene X is believed to activate transcription of pathways. (11) Model building is limited by major gaps a gene that in turn encodes an activator of in knowledge. It is not yet possible to deduce strR, a streptomycin pathway-specific activaat what level (transcriptional, translational, or tor gene (Sect. 3.3.4.3). In support of this, a posttranslational) the phosphorylated pleio- protein in extracts of A-factor-producing tropic regulators interact with the pathway- strains, but absent from A-factor-deficient
88
0
A-factor binding protein
A-factor
0
A-factor dependent protein
Sporuhtion
I
Streptomycin
tion; adp, regulatory gene encoding the A-factor-dependent protein that binds to the promoter region of strR, the pathway-specific activator gene for streptomycin production; aphD and strB encode a resistance determinant and biosynthetic enzyme, respectively; Sm, streptomycin. Redrawn from HORINOUCHI (1993).
Fig. 18. Model for the regulation of streptomycin biosynthesis in S. griseus. P, promoter; Protein X, regulatory protein derived from unidentified gene X required for both sporulation and streptomycin produc-
mutants, could bind to nucleotide sequences both processes; they also suggest that the Ajust upstream of strR. Little is known about factor-binding protein is present during early how A-factor synthesis takes place or is con- growth. In view of the positive autoregulation trolled. A putative A-factor biosynthetic of OHHL synthesis in Vibrio fischeri (see gene, @A, was cloned from S. griseus, but its Sect. 2.2), one might anticipate that the A-facpredicted translation product did not resem- tor-binding protein also represses - directly ble any other known protein (HORINOUCHIor indirectly - the expression of &A. et al., 1989b). Surprisingly, ufsA did not hybridize to DNA from some streptomycetes that produce structurally similar y-butyrolac- 3.3.4 Pathway-Specific Regulatory tones (HORINOUCHI al., 1984). The earlier et onset of streptomycin production and sporu- Genes lation in mutants lacking the A-factor-binding protein (MIYAKE al., 1990), and earlier et We have already made frequent references production of streptomycin on elevation of to the important role of pathway-specificregA-factor levels, either by exogenous addition ulatory genes. Here we review these genes (BEPPU, 1992) or by cloning ufsA on a multi and their (deduced) products in more detail. copy plasmid (HORINOUCHI al., 1984), are et clearly consistent with a role for ufsA and the y-butyrolactone in determining the timing of
89
3.3.4.1 The actll-ORF4, redD, and dnrl Family of Pathway-Specific Activator Genes
vators causes increased transcription of the corresponding biosynthetic structural genes and can be used to cause antibiotic production prematurely during rapid growth. The redD and actll-ORF4 genes are homoEarly genetic analyses identified putative logous to each other and to the positively-actpathway-specific activator genes for the unde- ing regulatory gene dnrl required for the procylprodigiosin (redD) and actinorhodin duction of daunorubicin in Streptomyces peuet al., (actll-ORF4) biosynthetic pathways of S. coe- cetius (STUTZMAN-ENGWALL 1992). Inficofor (reviewed by CHATER, 1992). The fail- sertional inactivation of dnrl blocks producure of redD and actll-ORF4 mutants to cotion of daunorubicin and all of its biosynthetsynthesize with representatives of any other ic intermediates and prevents transcription of red or act mutant class, the lack of expression putative operons containing daunorubicin of red and act biosynthetic structural genes in biosynthetic and resistance genes. The preredD and actll-ORF4 mutants, and the ability dicted redD, actll-ORF4, and dnrl gene prodof extra cloned copies of redD and actll- ucts show 33-37% amino acid sequence idenORF4 to elicit overproduction of undecyl- tity in pairwise alignments (Fig. 19; STUTZet prodigiosin and actinorhodin, respectively, all MAN-ENGWALL al., 1992). dnrl can comsuggest that redD and actll-ORF4 are path- plement mutations in actll-ORF4, and actllway-specific activator genes. Furthermore, ORF4 can stimulate daunorubicin production et the stationary-phase production of undecyl- in S. peucetius (STUTZMAN-ENGWALLal., prodigiosin and actinorhodin appears to re- 1992), but redD and actll-ORF4 do not show sult from transcriptional activation of redD cross-complementation. Since computer anal(TAKANO al., 1992) and actll-ORF4 (GRA- ysis using the algorithm of DODD and EGAN et (1990) failed to reveal likely helix-turn-helix MAJO et al., 1993), respectively. Production of the antibiotics in rapidly growing cultures ap- DNA-binding motifs in these proteins, they pears to be limited only by the absence of the may represent a novel family of DNA-bindrelevant pathway-specific activator protein, ing regulatory proteins. Perhaps more likely, because overproduction of the putative acti- they may need to interact with other proteins
Fig. 19. Alignment of the amino acid sequences of RedD, ActII-ORF4, DnrI, and the N-terminal region of
AfsR. The alignment was made using the PILEUP and PRETTYBOX programs contained in the UWG sequence analysis package (DEVEREUX al., 1984). et
90
to effect activation of biosynthetic structural may be binding sites for the 65 kDa putative gene promoters. transcriptional activator (SrmR) encoded by Intriguingly, the N-terminal region of srmR. SrmR shows no significant sequence AfsR, excluding the putative DNA-binding similarity to any other known protein. srmR motifs in the C-terminal region (HORINOU- homologs have not been reported in gene clusters for the production of other macrolide CHI et al., 1990), also shows significant identity to the RedD-ActII-ORF4-DnrI family antibiotics. The srmR gene contains a single (Fig. 19), raising the possibility that the stimu- TTA codon, so it may be a target for regulalation of actinorhodin and undecylprodigiosin tion by a bldA homolog (Sect. 3.3.2.6). production by multiple copies of afsR may reflect partial functional interchangeability of AfsR with ActII-ORF and RedD. However, since strong stimulatory effects were also ob- 3.3.4.3 strR Encodes a served with segments of AfsR (notably the Cterminal half) that are not homologous to DNA-Binding Protein that RedD and ActII-ORF4, models that rely Regulates at Least One of the solely on functional substitution are at best an Streptomycin Biosynthetic Genes oversimplification.
in Streptomyces griseus
3.3.4.2 srmR - A Regulatory Gene for Spiramycin Production in Streptomyces ambofaciens
Production of streptomycin and 5 '-hydroxy-streptomycin has been studied in S. grisew and S. glaucescens GLA.0, respectively. In S. griseus, strR appears to be a positive regulator of at least one of the biosynthetic structural genes, strBl, which encodes amidinoCloning and gene disruption revealed a pu- transferase I, and the strR homolog of S. gluutative regulatory gene, srmR, for spiramycin cescens GLA.0 is presumed to perform the production in Streptomyces ambofaciens. same function. StrR contains a potential hesrmR mutants fail to make spiramycin and do lix-turn-helix DNA-binding motif, and the not cosynthesize the antibiotic with srm mu- protein binds to at least two specific sites intants that accumulate intermediates in the side the str clusters of both species. Analysis biosynthetic pathway. srmR was required not of the binding sites reveals 11bp inverted reonly for transcription of srmG, which encodes peats separated by 11 bp. These results make the polyketide synthase that produces the it more likely that StrR acts as a conventional aglycone of spiramycin, but also for expres- transcriptional activator, rather than through sion of the resistance gene srmB (GEISTLICH transcriptional antitermination (RETZLAFF et et al., 1992). Lack of expression of srmB in al., 1993). The strR genes of both species consrmR mutants proved to be an indirect effect tain single TTA codons and bldA mutants of of the failure of srmR mutants to produce spi- S. griseus do not make streptomycin (MCCUE ramycin, which is an inducer of its own resist- et al., 1992). Single 7 T A codons are also presance gene. srmR was also required for the ent in strN, encoding a biosynthetic enzyme transcription of another flanking gene, srmX, of both species and in strA, the streptomycin that is also likely to play a role in spiramycin resistance gene of S. glaucescens (DISTLER et production. Multicopy cloning of srmR in the al., 1992). wild-type strain led to a 4-fold increase in spiramycin production. The srmG and srmX promoters are strikingly similar to each other, with three blocks of conserved sequences, centered at about position -39 (CCNGNCGTTCCT), -27 (CCCGGC), and - 10 (CTGTNN-GNT), one or more of which
91
3.3.4.4 brpA and dnrN Regulatory Genes for Bialaphos Production in Streptomyces hygroscopicus and for Daunorubicin Production in Streptomyces peucetius which Show Different Degrees of Similarity to Response Regulator Genes of Two-Component Systems
Bialaphos is made by Streptomyces hygroscopicus at the approach of stationary phase (HOLT et al., 1992). Early studies identified brpA as a likely pathway-specific activator gene for bialaphos production; brpA mutants were defective in at least 6 of the 13 steps leading to bialaphos production, lacked at least 7 of the bialaphos biosynthetic transcripts, and showed reduced levels of bialaphos resistance (ANZAI al., 1987). Furtheret more, a brpA mutant lacked 27 proteins implicated in bialaphos production (HOLTet al., 1992). brpA encodes a predicted product of 28 kDa whose C-terminal region resembles a region located towards the C-terminus of the response regulators of the UhpA subfamily of two-component regulatory systems (RAIBAUD et al., 1991; GROSSet al., 1989). This region includes a putative helix-turn-helix motif, but does not extend to the conserved region that includes the site of phosphorylation of the regulatory components. BrpA contains three hydrophobic regions towards its N-terminus, leading to suggestions that these might represent transmembrane domains or regions of hydrophobic interaction with other proteins (RAIBAUD al., 1991). brpA is tranet scribed from three promoters expressed at a low level early in exponential growth but more strongly during a pause in growth shortly before stationary phase, and the activity of one of them (brpAp3) continued to increase on entry into stationary phase. brpA contains a single TTA codon located towards the Cterminus of the coding region making it potentially bldA-dependent (bldA mutants of S. hygroscopicus have not been described). In addition to the actZZ-ORF4-like gene dnrl, the daunorubicin biosynthetic cluster of
S. peucetius encodes at least one other putative regulatory gene, dnrN. DnrN shows significant sequence similarity throughout its length to the UhpA subfamily of two-component response regulator proteins, including a likely site for phosphorylation, although dnrN does not appear to be closely linked to a sensory histidine protein kinase gene. Transcription of dnrZ is reduced in dnrN mutants (HUTCHINSON al., 1994), and this may be et the cause of their daunorubicin deficiency. Indeed, production is restored by adding extra copies of the cloned dnrl gene. (On the other hand, extra copies of dnrN do not restore production to a dnrl mutant.) A non-sporulating derivative (H6101) of Srreptomyces peuceticus var. caesius that is deficient in daunorubicin production has been described. Cloned copies of either dnrZ or dnrN restored daunorubicin production in H6101 suggesting that the mutant is defective in dnrN expression (presumably as a secondary result of the pleiotropic mutation) (HUTCHINSON al., et 1994; STUTZMAN-ENGWALL 1992). et al.,
3.3.4.5 Negative Regulation of Methylenomycin Production in Streptomyces coelicolor

The biosynthetic genes for methylenomycin production in S. coelicolor reside on the 350 kb linear plasmid SCPl (KINASHIet al., 1987). mmy genes were initially isolated by mutational cloning yielding over 20 kb of contiguous DNA (CHATERand BRUTON,1983, 1985). Insert-directed prophage insertions into the leftmost 3 kb of the cluster caused overproduction of methylenomycin. Sequence analysis of this region revealed a gene (mmyR) whose predicted product resembles the TetR family of repressor proteins (C. J. BRUTON and K. F. CHATER,unpublished results; Sect. 3.3.5.1). Disruption or deletion of mmyR resulted in overproduction of methylenomycin, providing the only known example of pathway-specific negative regulation of antibiotic production. The absence of methylenomycin production from various pleiotropic mutants (see Sect. 3.3.2) suggests that there may also be a positively acting pathway-spe-
92
cific regulatory gene, and the absence of mmy gene transcription in a bldA mutant (A. WIETZORREK K. F. CHATER,unpuband lished results) leads to the prediction that this gene should contain a TTA codon.
3.3.5 Induction of Antibiotic Resistance in Antibiotic Producing Streptomycetes - Antibiotics as Inducers of Gene Expression
Resistance towards an antibiotic made by a streptomycete often develops only at the onset of antibiotic production. In some cases, resistance may be a consequence of export alone (and may therefore be regarded as a late step in antibiotic production). In others, specific resistance mechanisms operate, in addition to efflux, to ensure continued viability. Below, we consider examples of resistance mechanisms that appear to be induced by their antibiotic substrates or by intermediates in the pathway.
3.3.5.1 actZZ-ORFU2 of Streptomyces coelicolor and tcmWA of Streptomyces glaucescens GLA.0 - Regulatory Cassettes for Antibiotic Export
al., 1991) demonstrated that ActII-ORF1 repressed transcription of ucfll-ORF2 and of itself (and indicated that both genes could be expressed in the absence of the pathway-specific activator ActII-ORF4; Sect. 3.3.4.1). Both promoters were most active in S. coelicolor cultures that were making actinorhodin. GUILFOILE and HUTCHINSON (1992a, b) showed that transcription of fcmA in S. gluucescens was induced by tetracenomycin C and that inactivation of fcmR resulted in constitutive tcmA expression; furthermore, in vitro binding of TcmR to the tcmWA intergenic region was inhibited in the presence of tetracenomycin C. It thus seems likely that export of tetracenomycin C is induced by the antibiotic once production begins (GUILFOILE and HUTCHINSON, 1992b). Resistance of S. coelicolor to methylenomycin is conferred by mmr which encodes a protein with significant sequence similarity to the same family of transporter proteins as ActII-ORE and TcmA (NEALand CHATER, 1987). HOBBSet al. (1992) found that transcripts corresponding to at least one of the methylenomycin biosynthetic genes appeared before that of mmr, suggesting that mmr expression might be induced by methylenomycin or by an intermediate in the pathway.
3.3.5.2 srmB of Streptomyces ambofaciens - A Probable ATP-Dependent Efflux System ucfZZ-ORF1/2 of S. coelicolor (CABALLERO Induced by its Antibiotic Substrate
et al., 1991) and fcmWA of S. gluucescens GLA.0 (GUILFOILEand HUTCHINSON, The srmB, flrC,and drrA gene products re1992a) are divergently transcribed gene pairs spectively involved in the export of spiramyfor export of actinorhodin and tetracenomycin from S. umbofuciens, tylosin from S. frucin C, respectively. Both ActII-ORE and diue, and daunorubicin from S. peucetius TcmA are similar to tetracycline transport show a high degree of amino acid sequence et proteins from several other gram-positive similarity (SCHONER al., 1992; GUILFOILE bacteria (including Tet347 of Sfrepfomycesri- and HUTCHINSON, 1991). The proteins each mosus) and gram-negative organisms, and possess a putative ATP-binding motif, sugActII-ORF1 and TcmR are clearly members gesting that they are components of ATP-deof the TetR family of repressor proteins. pendent efflux systems. Although details of Inactivation of ucfll-ORE appeared to pre- the regulation of flrC and drrA have not been vent the export of actinorhodin (FERNAN- published, expression of srmB in S. frudiue is DEZ-MORENO al., 1991). Studies of ucrZZ- induced by spiramycin (GEISTLICH al., et et ORF1/2 expression in E. coli (CABALLERO 1992; Sect. 3.3.4.2); increased levels of srmB et
93
transcription occurred on addition of spiramycin to mutants blocked in production of the antibiotic (no increase was observed with the wild-type strain).
3.3.5.3 Induction of tZrA in Streptomyces fradiae - A Role for Transcriptional Attenuation?

tlrA and tlrD are two of at least four S. fradiae genes that confer tylosin resistance. They cause di- and monomethylation, respectively, of residue A-2058 of 23s rRNA. Unlike tlrD, which is expressed constitutively, expression of rlrA is induced by tylosin or its biosynthetic intermediates (KELEMEN al., 1994). In the et absence of inducer, transcription terminates at the beginning of the coding region of tlrA through the adoption of a particular secondary structure in the RNA; the presence of an inducer is thought to cause sensitive ribosomes to stall in the non-translated leader region, preventing transcriptional termination and allowing production of the methylase.
3.3.5.4 Induction of Resistance to Novobiocin in Streptomyces sphaeroides and Streptomyces niveus - Roles for DNA Supercoiling and a Diffusible Signaling Molecule
Novobiocin, produced by Streptomyces sphaeroides and Streptomyces niveus, is an inhibitor of bacterial DNA gyrase. Gyrase exists as a tetramer (A2B2) in which the two subunits have different functions that can be blocked by different groups of antibiotics. The B subunit, encoded by gyrB, is the target for novobiocin. S. sphaeroides has two gyrB genes (THIARA and CUNDLIFFE, 1989,1993): one encoding a novobiocin-sensitive B subunit, GyrBS, that is produced constitutively, and the other encoding a resistant B subunit, GyrBR, that is produced in the presence of novobiocin. Transcriptional fusions showed
that the gyrBR promoter responds to changes in DNA supercoiling. Transcription of gyrBR increased when DNA gyrase was inhibited by novobiocin or ciprofloxacin (an inhibitor of the A subunit), i.e., under conditions that reduce negative supercoiling, and decreased during growth in a medium of high osmotic strength that should increase negative supercoiling (HIGGINS al., 1988). Thus resistance et to novobiocin in S. sphaeroides probably occurs, at least in part, by production of the resistant gyrase following the reduction in negative supercoiling which results from inhibition of the sensitive enzyme by the antibiotic. Two different, uncharacterized genes that confer novobiocin resistance on S. lividans have been isolated from S. niveus (HOGGARTH et al., 1994); one hybridizes with a second resistance determinant from S. sphaeroides (THIARA and CUNDLIFFE, 1988). The isolation of multiple resistance genes and a marked increase in the level of novobiocin resistance (from 25 to over 200 pg mL-') during growth of S. niveus suggest that resistance is determined by several mechanisms that may be subject to different regulatory controls. Recent studies on s. niveus (HOGGARTH et al., 1994) identified a diffusible 7butyrolactone signaling molecule that induces high-level resistance to novobiocin well before the onset of production.
3.3.5.5 Regulation of Isoforms of the Target for Pentalenolactone Inhibition in the Producing Organism
Pentalenolactone (PL) is a potent inhibitor of glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The producer, Streptomyces arenae, has two distinct isoforms of GAPDH: a PL-sensitive enzyme produced before antibiotic production and a PL-resistant form f produced on induction o PL synthesis. The sensitive isoform rapidly disappears when PL is produced (FROHLICH al., 1989). The two et isoforms are encoded by two distinct genes, but the mechanisms responsible for their regulation are unknown.
94
further example of this, beyond that described above for B. subfilis, regulates proStudies of the regulation of antibiotic production of three different secondary metabolduction in diverse prokaryotes are revealing ites (2,4-diacetyl phloroglucinol, HCN, and several common themes. Typically, antibiotic pyoluteorin) in Pseudomonas aeruginosa et biosynthetic genes appear to be regulated (LAVILLE al., 1992). Such two-component principally at the level of transcription, and systems are clearly important in Streptomyces this involves rather large numbers of appar- spp., but it is interesting that serine-threonine ent regulatory genes, most of them acting kinases, previously described only in eukamore or less globally on secondary metabol- ryotes, are also involved (there is also eviism. The regulatory genes might exert their dence of multiple tyrosine kinases in streptoeffects by direct interaction with promoters of mycetes, though their roles are unknown; antibiotic production genes, though it seems WATERSet al., 1994). more likely that they are mostly involved with While specialized minor u factors are often producing, detecting, transmitting, and inte- required for transcription of genes for antigrating information relevant to deciding the biotic production, there is no evidence that a appropriateness of commitment to produc- particular sub-branch of u factors is implition. Different organisms have different ecol- cated (Fig. 2b). Thus, the stationary-phase ogies, and because different antibiotics differ factor 2 in E. coli is sometimes involved, and in their biosynthetic origins and modes of ac- in Sfrepfomycesat least one antibiotic production, and thus in their potential effectiveness tion gene is transcribed in vifro by RNA polyin different ecological situations, activation of merase containing uE (G. H. JONESand M. J. any particular pathway of any particular or- BUTTNER, personal communication), a u facganism might be expected to require its own tor that was used as the paradigm of a new combination of signals; it is, therefore, not subfamily of u factors (the ECF family; LONsurprising to find great diversity in the sys- ETTO et al., 1994). Sfreptomyces spp. appear tems analyzed so far. Nevertheless, there are to contain a rather large number of different 1989 LONETTOet al., recurrent themes. One is the use of quorum u factors (BUTTNER, sensing. Lipid-soluble lactones play such a 1994), and Sfrepfomycespromoters are very role in widely different organisms, and in diverse in their sequences and complexity each case the receptor is cytoplasmic and cap- (STROHL, 1992), so it is quite likely that other able of direct interaction with DNA to in- minor u factors may prove to be involved in fluence gene expression. However, in B. sub- secondary metabolism. As knowledge of the regulation of antibiotfilis the most well-defined extracellular signaling is done by a small peptide probably mod- ic synthesis increases, new genetic approaches ified with a hydrophilic side chain (MAGNU- to industrial overproduction will probably be SON et al., 1994), and the receptor is a memdevised which may be especially useful when brane-located protein kinase that does not in- a new product is being developed. A particuteract with DNA directly but phosphorylates larly attractive approach is to add extra copies and thereby activates a transcription factor. of positive regulatory genes which are found This solution adopted by B. subtilis to the in many pathways (CHATER, 1990). This problem of quorum sensing resembles the could be done by self-cloning, but an alternamating pheromone systems of Enferococcus tive strategy is to collect a panel of cloned faecalis (CLEWELL, 1993). It is, therefore, in- DNA fragments (isolated from various speteresting that the B. subfilis pheromone also cies) capable of heterologous stimulation of controls the competence regulator, the func- production of some antibiotics, such as are tion of which (like mating) is to permit the in- being discovered by F. MALPARTIDA al. et gress of DNA. (personal communication) and to introduce Protein phosphorylation is widely impli- these into any interesting new species. A simcated in activating antibiotic synthesis, nota- ilar approach might also lead to the discovery bly involving members of the two component of new antibiotics in old strains (HOPWOOD histidine kinase-response regulator family. A et al., 1983).
4 Concluding Remarks
95
approach was used to clone the ppGpp syn. thetase gene (relA) of S coelicolor (CHARecently, several papers have been pub- KRABURTTY et al., 1996). The cloned gene lished on the regulation of antibiotic produc- was used to create a null-mutant that is totally . tion in streptomycetes, mostly in S coelicolor, deficient in ppGpp synthesis upon amino acid and are discussed briefly here. The absA lo- starvation (CHAKRABURTTY, 1996). The recus of S coelicolor (Sect. 3.3.2.4) has been sulting mutant grows at the same rate as the . shown to encode a two-component regulatory relA strain but fails to make Act or Red on system, absAIIA2, which acts as a negative some media, but does so on others; similar reregulator of antibiotic production (BRIAN et sults were obtained by MART~NEZ-COSTA et al., 1996). Disruption of absA results in early al. (1996). This indicates an obligatory role hyperproduction of both actinorhodin (Act) for ppGpp in antibiotic biosynthesis, and toand undecylprodigiosin (Red). All four pre- gether with the conditional phenotype of the viously isolated absA mutations lie in absAl afsR null-mutant (FLORIANOand BIBB, encoding the predicted sensor histidine ki- 1996), indicates the presence of multiple signase; these mutations may lock the kinase in nal transduction pathways for the activation an active conformation preventing the relief of antibiotic production. Further evidence for of the negative influence of the phosphory- this stems from the isolation and characterizalated form of AbsA2 on antibiotic synthesis. tion of an extracellular signaling molecule, a In a potentially similar fashion, the cutRS lo- novel y-butyrolactone, that elicits the precocus also acts to negatively regulate Act pro- cious production of both Act and Red when . . duction in S lividans and in S coelicolor added to the wild-type strain (E. TAKANO, T. and (CHANG al., 1996). Thus, protein phospho- NIHIRA, M. J. BIBB,unpublished results). et rylation mediated by absAIIA2 and cutRS The relA and afsR mutants produce, but do acts to negatively regulate antibiotic produc- not respond to, this factor. Genes encoding tion, in contrast to the positive effects of afsKI the binding proteins for the y-butyrolactones afsR (Sect. 3.3.2.1) and afsQlIQ2 (Sect. made by S griseus (A-factor) and S virginiae . . 3.3.2.2). Recent studies on afsR (FLORIANO (the virginae butanolides VB-A-E) (Sect. and BIBB, 1996) revealed that while it is ho- 3.2.4) have been cloned and sequenced (ONet mologous to actII-ORF4 and redD, pathway- AKA et al., 1995; OKAMOTO al., 1995). Anspecific regulatory genes for Act and Red other pleiotropic regulatory gene for antibiotproduction, respectively, it cannot substitute ic production in S coelicolor is afsB (Sect. . for them. Moreover, an in-frame deletion that 3.3.2.1). Attempts to complement afsB using removed most of the afsR coding sequence a genomic library made in a low copy-number resulted in loss of Act and Red production, plasmid led to the discovery that additional and a marked reduction in the synthesis of copies of hrdB, which encodes the major u . the calcium-dependent antibiotic (CDA), but factor of S coelicolor (BROWNet al., 1992), only under some (non-permissive) nutritional restored Red and Act production in afsB muconditions. Although additional copies of tants (WIETZORREK, 1996). The effect of afsR resulted in elevated levels of the actII- hrdB resembles a recent report in Pseudoet ORF4 and redD transcripts, transcription of monas fluorescens (SCHNIDER al., 1995), in the pathway-specific regulatory genes under which production of the antibiotics pyoluteonon-permissive conditions was unaffected by rin and 2,4-diacetylphloroglucinol,which are deletion of afsR. While afsR may operate in- made during stationary phase, was stimulated dependently of the pathway-specific regulato- by the presence of additional copies of rpoD, ry proteins to influence antibiotic production, which encodes the major and essential u facthe activity of ActII-ORF4 and of RedD un- tor of that organism. This may reflect a role der non-permissive conditions could depend for the major u factor of both organisms in on interaction with, or modification by, AfsR. the transcription of antibiotic biosynthetic To assess whether there might be a causal re- genes, or may result from an indirect effect lationship between ppGpp synthesis and anti- (e.g., provision of precursors). Consistent biotic production (Sect. 3.2.3), a PCR-based with the former notion, in vitro transcription
+
Note added in proof
96
of the pathway-specific regulatory gene redD was observed upon addition of a protein corto core responding in size to drdB RNA polymerase (FUJII et al., 1996). Recent studies have further elucidated the way in which bldA (Sect. 3.3.2.6) influences the activation of individual biosynthetic pathways. Analysis of the Pwb mutations (Sect. 3.3.2.7) has identified an additional regulatory gene, red2 (GUTHRIE, P., FLAXMAN, S., WHITE,J., E. C. HODGSON, A., BIBB,M. J., and CHATER, D. K. F., manuscript in preparation), and revealed a pathway-specific regulatory cascade. red2 is located approximately 4 kb downstream of redD and contains a single UUA codon. Disruption of red2 results in loss of Red production and loss of redD transcription, suggesting that RedZ is a transcriptional activator of redD (WHITEand BIBB,in press). RedZ shows end-to-end similarity to members of the response regulator family of proteins, and possesses a putative DNA-binding a-helix-turn-a-helix motif towards its C-terminus but lacks the charged amino acids normally essential for phosphorylation of a response regulator by its cognate sensory histidine protein kinase. The existence of two pathway-specific regulatory genes for Red production in S. coelicolor parallels the situation for daunorubicin synthesis in S. peuceticus, in which dnrl and dnrN are homologues of redD and red2, respectively. In S. peuceticus, transcription of dnrl depends on dnrN (Sect. 3.3.4.4), and DnrN has been shown recently to bind to the dnrl promoter region (FURUYA and HUTCHINSON, 1996). Moreover, DnrI has also been shown to bind to the promoters of daunorubicin biosynthetic structural genes (TANGet al., in press), providing the elusive evidence that this family of pathway-specific regulatory genes are indeed likely to act directly as transcriptional activators (Sect. 3.3.4.1).
lowed us to cite their unpublished results. We also thank MEREDYTH LIMBERG and ANNE WILLIAMS their patience in typing succesfor sive versions of the manuscript. Our laboratories' work in this area was funded by the Biotechnology and Biological Research Council and the European Community.
References
ADAMIDIS, CHAMPNESS, (1992), Genetic T., W. analysis of absB, a Streptomyces coelicolor locus involved in global antibiotic regulation, J. Bacterial. 114,46224628. ADAMIDIS, RIGGLE, CHAMPNESS, C. T., P., W. (1990), Mutations in a new Streptomyces coelicolor locus which globally block antibiotic biosynthesis but not sporulation, J. Bacteriol. 112, 2962-2969. ALEX,L. A., SIMON, I. (1994), Protein histidine M. kinases and signal transduction in prokaryotes and eukaryotes, Trends Genet. 10,133-138. AN, G., VINING, C. (1978), Intracellular levels of L. guanosine 5 '-diphosphate 3 '-diphosphate (ppGpp) and guanosine 5 '-triphosphate 3 '-diphosphate (pppGpp) in cultures of Streptomyces griseus producing streptomycin, Can. J. Microbiol. 24,502-511. ANISOVA, N., BLINOVA, N., EFREMENKOVA, L. I. 0. V., KOZ'MIN, Yu. P., ONOPRIENKO, V., V. G. SMIRNOVA, M., KHOKHLOV,A. S. (1984), Regulators of the development of Streptomyces coelicolor A3(2), Izv. Akad. Nauk SSSR, Ser. Biol. 1, 98-108. ANZAI, MURAKAMI, IMAI, SATOH, H., T., S., A., NAGAOKA, THOMPSON, J. (1987), TranK., C. scriptional regulation of bialaphos biosynthesis in Streptomyces hygroscopicus,J. Bacteriol. 169, 3482-3488. ASTURIAS, A., LIRAS, MARTiN, J. F. ( l m ) , J. P., Phosphate control of pabS gene transcription during candicidin biosynthesis, Gene 93, 79-84. BAINTON, J., BYCROFI-, W., CHHABRA, R., N. B. S. STEAD, GLEDHILL, HILL,P. J., REES,C. P., L., M. G. E. D., WINSON, K., SALMOND, P. C., STEWART, G . S. A. B., WILLIAMS, (1992a), A P. general role for the lux autoinducer in bacterial cell signalling: control of antibiotic biosynthesis Acknowledgements in Erwinia, Gene 116, 87-91. BAINTON, J., STEAD, CHHABRA, R., BYN. P., S. We are grateful to WENDY CHAMPNESS, CROFT, B. W., SALMOND, . P. C., STEWART, G ALAN GROSSMAN, DAVIDHOPWOOD, MIG. S. A. B., WILLIAMS, (1992b), N-(3-OxoP. CHIKO NAKANO, GEORGE SALMOND, and hexanoy1)-L-homoserine lactone regulates carPETERZUBERfor comments on parts of the bapenem antibiotic production in Erwinia carotovora, Biochem. J. 288,997-1004. manuscript, and to all those who have al-
5 References
97
BASCARAN, SANCHEZ,L., HARDISSON, V., C., B R A ~ A , F. (1991), Stringent response and inA. itiation of secondary metabolism in Streptomyces clavuligerus, J. Gen. Microbiol. 137, 1625-1634. BEPPU,T. (1992), Secondary metabolites as chemical signals for cellular differentiation, Gene 115, 159-1 65. B R A ~ A , F., M ~ N D E Z , DIAZ, L. A, MANZAA. C., NAL,M. B., HARDISSON, (1986), Glycogen C. and trehalose accumulation during colony development in Streptomyces antibioticus, J. Gen. Microbiol. 132, 1319-1326. BRIAN, RIGGLE, J., SANTOS, A., CHAMPP., P. R. NESS, W. C. (1996), Global negative regulation of Streptomyces coelicolor by an absA-encoded putative signal transduction system, J. Bacteriol. 178,3221-3231. BROWN, L., WOOD,S., BUTTNER, J. (1992), K. M. Isolation and characterization of the major vegetative RNA polymerase of Streptomyces coelicolor A3(2)-renaturation of a sigma-subunit using GroEL, Mol. Microbiol. 6, 1133-1139. BRUTON, J., GUTHRIE, P., CHATER,K. F. C. E. (1991), Phage vectors that allow monitoring of transcription of secondary metabolism genes in Streptomyces, Bio/Technology 9, 652-656. BUTTNER, J. (1989), RNA polymerase heteroM. geneity in Streptomyces coelicolor A3(2), Mol. Microbiol. 3, 1653-1659. CABALLERO, L., MALPARTIDA, HOPWOOD, J. F., D. A. (1991), Transcriptional organization and regulation of an antibiotic export complex in the producing Streptomyces culture, Mol. Gen. Genet. 228, 372-380. CAO, J. G., MEIGHEN, A. (1989), Purification E. and structural identification of an autoinducer for the luminescence system of Vibrio harveyi, J. Biol. Chem. 264,21670-21 676. CASHEL, RUDD,K. E. (1987), The stringent reM., sponse, in: Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology (NEIDHARDT, INGRAHAM,L., Low, K. B., F., J. MAGASANIK, SCHAECHTER, UMBARB., M., GER, H. E., Eds.), pp. 1410-1438. Washington, D C American Society for Microbiology. CHAKRABURTTY, N. (1996), The (p)ppGpp synR. thetase gene (relA) of Streptomyces coelicolor, A3(2). Ph D Thesis, University of East Anglia, Nonvich, UK. E., CHAKRABURTTY, WHITE, J., TAKANO, R., BIBB,M. J. (1996), Cloning, characterization and disruption of a (p)ppGpp synthetase gene (relA) from Streptomyces coelicolor A3(2), Mol. Microbiol. 19, 357-368. CHAMPNESS, C., CHATER, F. (1994), RegulaW. K. tion and integration of antibiotic production and morphological differentiation in Streptomyces
spp., in: Regulation of Bacterial Differentiation (PIGGOT,P., MORAN,C. P., YOUNGMAN, P., Eds.), pp. 61-93. Washington, DC: American Society for Microbiology. T., CHAMPNESS, C., RIGGLE,P., ADAMIDIS, W. VANDERVERE, (1992), Identification of StrepP. tomyces coelicolor genes involved in regulation of antibiotic synthesis, Gene 115, 55-60. CHANG, H.-M., SHIEH, Y.-T., BIBB, J., CHEN, M. C. W. (1996), The cutRS signal transduction system of Streptomyces lividans represses the biosynthesis of the polyketide antibiotic actinorhodin, Mol. Microbiol. 21, 1075-1085. CHATER, F. (1990), The improving prospects for K. yield increase by genetic engineering in antibiotic-producing streptomycetes, Bio/Technology 8, 115-121. CHATER,K. F. (1992), Genetic regulation of secondary metabolic pathways in Streptomyces, in: Secondary Metabolites. Their Function and EvoD. J., lution (CHADWICK, J., WHELAN, Eds.), pp. 144-162, Ciba Foundation Symposium 171. Chichester: Wiley. CHATER, F., BRUTON, J. (1983), Mutational K. C. cloning in Streptomyces and the isolation of antibiotic production genes, Gene 26, 67-78. CHATER, F., BRUTON, J. (1985), Resistance, K. C. regulatory and production genes for the antibiotic methylenomycin are clustered, EMBO J. 4, 1892-1893. CHATER,K. F., HOPWOOD, A (1993), StreptoD. myces, in: Bacillus subtilis and Other Gram-Positive Bacteria: Biochemistry, Physiology, and MoA. lecular Genetics (SONENSHEIN, L., HOCH,J. A., LOSICK,R., Eds.), pp. 83-89. Washington, D C American Society for Microbiology. CHATER,K. F., MERRICK, J. (1979), StreptoM. myces, in: Developmental Biology of Prokaryotes (PARISH, H., Ed.), pp. 93-114. Oxford: BlackJ. well. CLEWELL, B. (1993), Bacterial sex pheromoneD. induced plasmid transfer, Cell 73, 9-12. COSMINA, RODRIGUEZ, DE FERRA, F., P., F., GRANDI, PEREGO,M., VENEMA, VAN G., G., SINDEREN, (1993), Sequence and analysis of D. the genetic locus responsible for surfactin synthesis in Bacillus subtilis, Mol. Microbiol. 8,821831. DECKER,H., HUTCHINSON, R. (1993), TranC. scriptional analysis of the Streptomyces glaucescens tetracenomycin C biosynthesis gene cluster, J. Bacteriol. 175, 3887-3892. DEMAIN, L. (1992), Microbial secondary metabA. olism: a new theoretical frontier for academia, a new opportunity for industry, in: Secondary Metabolites. Their Function and Evolution (CHAD-
98

SHEVICH, Y. E. (1985), Regulators of differentiation in actinomycetes, Antibiotiki i Meditsinskaya Biotekhnologiya 9, 687-707. FERNANDEZ-MORENO, A, CABALLERO, L., M. J. HOPWOOD, A., MALPARTIDA, (1991), The D. F. act cluster contains regulatory and antibiotic export genes, direct targets for translational control by the bldA transfer RNA gene of Streptomyces, Cell 66, 769-780. FERNANDEZ-MORENO, A., MART~N-TRIANA, M. A. J., MARTINEZ, NIEMI, KIESER, M., E., J., H. HOPWOOD, A., MALPARTIDA, (1992), D. F. abaA, a new pleiotropic regulatory locus for antibiotic production in Streptomyces coelicolor, J. Bacteriol. 174, 2958-2967. FLORIANO, BIBB,M. J. (1996), afsR is a pleioB., tropic but conditionally required regulatory gene for antibiotic production in Streptomyces coelicolor A3(2), Mol. Microbiol. 21, 385-396. FROHLICH, U., WIEDMANN, LOTTSPEICH, K. M., F., MECKE, (1989), Substitution of a pentaleD. nolactone-sensitive glyceraldehyde-3-phosphate dehydrogenase by a genetically distinct resistant isoform accompanies pentalenolactone production in Streptomyces arenae, J. Bacteriol. 171, 6696-6702. FUJII,T., GRAMAJO, C., TAKANO, BIBB,M. H. E., J. (1996), redD and actII-ORF4, pathway-specific regulatory genes for antibiotic production in Streptomyces coelicolor A3(2), are transcribed in vitro by an RNA polymerase holoenzyme containing the non-essential u factor, ghrdD,J. Bacteriol. 178, 3402-3405. FUQUA, C., WINANS, C., GREENBERG, P. W. S. E. (1994), Quorum sensing in bacteria: the LuxRLux1 family of cell density-responsive transcriptional regulators, J. Bacteriol. 176, 269-275. FURUYA,K., HUTCHINSON, R. (1996), The C. DnrN protein of Streptomyces peuceticus, a pseudo-response regulator, is a DNA-binding protein involved in the regulation of daunorubicin biosynthesis, J. Bacteriol. 178, 6310-6318. GEISTLICH, LOSICK, TURNER, R., RAO, M., R., J. R. N. (1992), Characterization of a novel regulatory gene governing the expression of a polyketide synthase gene in Streptomyces ambofaciens, Mol. Microbiol. 6, 2019-2029. GENTRY, R., HERNANDEZ, J., NGUYEN, D. V. L. H., JENSEN, B., CASHEL, (1993), Synthesis D. M. of the stationary-phase sigma factor 2 is positively regulated by ppGpp, J. Bacteriol. 175, 7982-7989. GERBER, N., MCINNES,A. G., SMITH,D. G., N. J. WALTER,J. A., WRIGHT, L. C. (1978), Biosynthesis of prodiginines. 13C resonance assignments and enrichment patterns in nonyl-, cyclononyl-, methylcyclodecyl-, and butylcycloheptyl-
WICK, J., WHELAN, Eds.), pp. 3-23, Ciba D. J., Foundation Symposium 171. Chichester: Wiley. DEMAIN, L., AHARONOWITZ, MART~N, A. Y., J.-F. (1983), Metabolic control of secondary biosynthetic pathways, in: Biochemistry and Genetic Regulation of Commercially Important Antibiotics (VINING,L. C., Ed.), pp. 49-72. London: Addison-Wesley. DEVEREUX,J., HAEBERLI,P., SMITHIES, 0. (1984), A comprehensive set of sequence analysis programs for the VAX, Nucleic Acids Res. 12, 387-395. DfAZ-GUERRA, L., MORENO, SANMILLAN, F., J. L. (1989), appR gene product activates transcrip7 tion of microcin C plasmid genes, J. Bacteriol. 171,2906-2908. DISTLER, MANSOURI, MAYER,G., STOCKJ., K., MA, M., PIEPERSBERG, w. (1992), Streptomycin biosynthesis and its regulation in streptomycetes, Gene 115, 105-111. DODD,I. B., EGAN,J. B. (1990), Improved detection of helix-turn-helix DNA-binding motifs in protein sequences, Nucleic Acids Res. 18, 50195026. DONADIO, STASSI, MCALPINE, B., STAVS., D., J. ER,M. J., SHELDON, J., JACKSON, SWANP. M., SON,S. J., WENDT-PIENKOWSKI, YI-GUANG, E., W., JARVIS,B., HUTCHINSON, R., KATZ, L. C. (1993), Recent developments in the genetics of erythromycin formation, in: Industrial Microorganisms: Basic and Applied Molecular Genetics (BALTZ,R. H., HEGEMAN, D., SKATRUD, G. P. L., Eds.), pp. 257-265. Washington, DC: American Society for Microbiology. DSOUZA, NAKANO, M., CORBELL, ZuC., M. N., BER,P. (1993), Amino-acylation site mutations in amino acid-activating domains of surfactin synthetase: effects on surfactin production and competence development in Bacillus subtilis, J. Bacteriol. 175, 3502-3510. DSOUZA, NAKANO, M., ZUBER, (1994), C., M. P. Identification of comS, a gene of the srfA operon that regulates the establishment of genetic competence in Bacillus subtilis, Proc. Natl. Acad. Sci. USA 91,9397-9401. DUBNAU, (1993), Genetic exchange and homoD. logous recombination, in: Bacillus subtilk and Other Gram-Positive Bacteria: Biochemistry and Molecular Genetics (SONENSHEIN, L., HOCH, A. J. A., LOSICK, Eds.), pp. 555-584. WashingR., ton, DC: American Society for Microbiology. EBERHARD, BURLINGAME, L., KENYON, A., A. G. L., NEALSON,K. H., OPPENHEIMER, J. N. (1981), Structural identification of autoinducer of Photobacterium fischeri luciferase, Biochemistry 20, 2444-2449. EFREMENKOVA, V., ANISOVA, N., BARTO0. L.
5 References
99
prodiginine produced by actinomycete cultures supplemented with 13C-labeledacetate and "Nlabeled nitrate, Can. J. Chem. 56, 1155-1163. GRAFE, U., SCHADE, ERITT,I., FLECK, F. W., W. (1982), A new inducer of anthracycline biosynthesis from Streptomyces viridochromogenes, J. Antibiot. 35, 1722-1723. GRAFE,U., REINHARDT, SCHADE, ERITT, G., W., I., FLECK,W. F., RADICS,L. (1983), Interspecific inducers of cytodifferentiation and anthracycline biosynthesis from Streptomyces bikiniensis and Streptomyces cyaneofuscatus,Biotechnol. Lett. 5, 591-596. GRAMAJO, C., TAKANO, BIBB,M. J. (1993), H. E., Stationary-phase production of the antibiotic actinorhodin in Streptomyces coelicolor A3(2) is transcriptionally regulated, Mol. Microbiol. 7, 837-845. GROSS, ARICO, RAPPUOLI, (1989), FamiR., B., R. lies of bacterial signal-transducing proteins, Mol. Microbiol. 3, 1661-1667. GROSSMAN, H., TUCKMAN, ELLESTAD, T. M., S., OSBURNE, S. (1993), Isolation and characterM. ization of Bacillus subtilis genes involved in siderophore biosynthesis: relationship between B. subtilis sfp" and Escherichia coli entD genes, J. Bacteriol. 175, 6203-6211. GUILFOILE, G., HUTCHINSON, R. (1991), A P. C. bacterial analog of the mdr gene of mammalian tumor cells is present in Streptomyces peucetius, the producer of daunorubicin and doxorubicin, Proc. Natl. Acad. Sci. USA 88,8553-8557. GUILFOILE, G., HUTCHINSON, R. (1992a), SeP. C. quence and transcriptional analysis of the Streptomyces glaucescens tcmAR tetracenomycin-C resistance and repressor gene loci, J. Bacteriol. 174,3651-3658. GUILFOILE, G., HUTCHINSON, R. (1992b), P. C. The Streptomyces glaucescens TcmR protein represses transcription of the divergently oriented tcmR and tcmA genes by binding to an intergenic operator region, J. Bacteriol. 174, 3659-3666. GUTHRIE, P., CHATER, F. (1990), The level E. K. of a transcript required for production of a Streptomyces coelicolor antibiotic is conditionally dependent on a transfer RNA gene, J. Bacteri01. 172,61894193. HAHN, DUBNAU, (1991), Growth stage signal J., D. transduction and the requirements for srfA induction in development of competence, J. Bacteriol. 173, 7275-7282. HAMAGISHI, YOSHIMOTO, OKI,T., INUI, Y., A., T. (1980), Occurrence of guanosine-5 '-diphosphate-3 '-diphosphate and adenosine-5'-triphosphate-3 '-diphosphate in Streptomyces galilaeus, Agric. Biol. Chem. 44, 1003-1007.
HAMAGISHI, YOSHIMOTO, OKI, T. (1981), Y., A., Determination of guanosine tetraphosphate (ppGpp) and adenosine pentaphosphate (pppApp) in various microorganisms by radioimmunoassay, Arch. Microbiol. 130, 134137. HARA, BEPPU, (1982a), Induction of streptoO., T. mycin-inactivating enzyme by A-factor in Streptomyces griseus, J. Antibiot. 35, 1208-1215. HARA,O., BEPPU,T. (1982b), Mutants blocked in streptomycin production in Streptomyces griseus - the role of A-factor, J. Antibiot. 35, 349-358. S., T., HARA,O., HORINOUCHI, UOZUMI, BEPPU, T. (1983), Genetic analysis of A-factor synthesis in Streptomyces coelicolor A3(2) and Streptomyces griseus, J. Gen. Microbiol. l29, 29392944. J., I., R. HEALY, WEIR,J., SMITH, LOSICK, (1991), Post-transcriptional control of a sporulation regulatory gene encoding transcription factor # in Bacillus subtilis, Mol. Microbiol. 5, 477487. HECKER, SCHROETER, (1985), Synthese der M., A. alkalischen Phosphatase in einem stringent und relaxed kontrollierten Stamm von Escherichia coli nach Aminosauren- und Phosphatlimitation, J. Basic Microbiol. 25, 341-347. HENGGE-ARONIS, FISCHER, (1992), IdentifiR., D. cation and molecular analysis of glgS, a novel growth-phase regulated and rpoS-dependent gene involved in glycogen synthesis in Escherichia coli, Mol. Microbiol. 6, 1877-1886. HENNER, J., YANG,M., FERRARI, (1988), D. E. Localization of Bacillus subtilis sacU(hy) mutations to two linked genes with similarities to the conserved procaryotic family of two-component signalling systems, J. Bacteriol. 170, 5102-5109. HERNANDEZ, J., BREMER, (1990), GuanoV. H. sine tetraphosphate (ppGpp) dependence of the growth rate control of rrnA P1 promoter activity in Escherichia coli, J. Biol. Chem. 265, 1160511614. HERNANDEZ, J., BREMER, (1991), EscherichV. H. iu coli ppGpp synthetase I1 activity requires SPOT, Biol. Chem. 266,5991-5999. J. V. H. HERNANDEZ, J., BREMER, (1993), Characterization of RNA and DNA synthesis in Escherichia coli strains devoid of ppGpp, J. Biol. Chem. 268, 10851-10862. HIGGINS, F. (1992), ABC transporters: from miC. croorganisms to man, Annu. Rev. Cell Biol. 8, 67-113. HIGGINS, F., DORMAN, J., STIRLING, A, C. C. D. WADDELL, BOOTH,I. R., MAY, G., BREL., MER,E. (1988), A physiological role for DNA supercoiling in the osmotic regulation of gene expression in S. typhimurium and E. coli, Cell 52, 569-584.
100
S. HOBBS, OBANYE, I. C., PETTY,J., MASON, HORINOUCHI,(1993), Eukaryotic signal transG., A. duction systems in the bacterial genus StreptoJ. C., BARRATT, GARDNER, C. J., FLETT, E., D. myces, Actinomycetologica 7, 68-81. F., SMITH,C. P., BRODA,P., OLIVER,S. G. (1992), An integrated approach to studying reg- HORINOUCHI, BEPPU,T. (1992), AutoregulatoS., ulation of production of the antibiotic methylery factors and communication in actinomycetes. nomycin by Streptomyces coelicolor A3(2), J. Annu. Rev. Microbiol. 46,317-398. Bacteriol. 174, 1481-1494. HORINOUCHI, BEPPU,T. (1995), AutoregulaS., HOCH,J. A. (1993), sp00 genes, the phosphorelay, tors, in: Genetics and Biochemistry of Antibiotic and the initiation of sporulation, in: Bacillus subProduction (VINING, STUTTARD, Eds.), L., C., tilis and Other Gram-positive Bacteria: Biochempp. 103-119, Newton, MA: Butterworth-Heineistry, Physiology, and Molecular Genetics (SONmann. R., ENSHEIN, L., HOCH,J. A., LOSICK, Eds.), HORINOUCHI, HARA, O., BEPPU, T. (1983), A. S., pp. 747-755. Washington, D C American Society Cloning of a pleiotropic gene that positively confor Microbiology. trols biosynthesis of A-factor, actinorhodin, and HODGSON, A. (1992), Differentiation in actinoD. prodigiosin in Streptomyces coelicolor A3(2) and mycetes, in: Prokaryotic Structure and Function. Streptomyces lividans, J. Bacteriol. 155, 1238A New Perspective (MOHAN, DOW,C., COLE, S., 1248. J. A., Eds.), SGM Symposium Vol. 47, pp. 407HORINOUCHI, KUMADA, BEPPU, (1984), S., Y., T. 440. Cambridge: Society for General MicrobiolUnstable genetic determinant of A-factor bioogy. synthesis in streptomycin-producing organisms: HOGGARTH, H., CUSHING, E., MITCHELL, J. K. J. Cloning and characterization, J. Bacteriol. 158, I., RITCHIE,D. A. (1994), Induction of resist481-487. ance to novobiocin in the novobiocin-producing HORINOUCHI, SUZUKI, BEPPU,T. (1986), S., H., organism Streptomyces niveus, FEMS Microbiol. Nucleotide sequence of afsB, a pleiotropic gene Lett. 116, 131-136. involved in secondary metabolism in StreptoHOLT, T. G., CHANG,C., LAURENTWINTER, C., myces coelicolor A3(2) and Streptomyces liviJ. MURAKAMI, GARRELS,J. I., DAVIES, E., T., dam, J. Bacteriol. 168, 251-269. THOMPSON, J. (1992), Global changes in gene C. HORINOUCHI, MALPARTIDA, HOPWOOD, S., F., D. expression related to antibiotic synthesis in A., BEPPU, (1989a), afsB stimulates transcripT. Streptomyces hygroscopicus, Mol. Microbiol. 6, tion of the actinorhodin biosynthetic pathway in 969-980. Streptomyces coelicolor A3(2) and Streptomyces HONG, K, KITO,M., S. BEPPU, HORINOUCHI, T., S. lividans, Mol. Gen. Genet. 215, 355-351. (1991), Phosphorylation of the AfsR product, a HORINOUCHI, SUZUKI, S., H., NISHIYAMA, M., global regulatory protein for secondary-metaBEPPU,T. (1989b), Nucleotide sequence and bolite formation in Streptomyces coelicolor transcriptional analysis of the Streptomyces griA3(2), J. Bacteriol. 173,2311-2318. seus gene (afsA) responsible for A-factor bioA., S., HONG,S . K., MATSUMOTO, HORINOUCHI, synthesis, J. Bacteriol. 171, 1206-1210. BEPPU, C. (1993), Effect of protein kinase inT. S., M., hibitors on in vitro protein phosphorylation and HORINOUCHI, KITO, M., NISHIYAMA, FuRUYA, HONG,S. K., MIYAKE, BEPPU,T. K., K., cellular differentiation of Streptomyces griseus, (1990), Primary structure of AfsR, a global reguMol. Gen. Genet. 236,341-354. latory protein for secondary metabolite formaHOOD,D. W., HEIDSTRA, SWOBODA, K., R., U. tion in Streptomyces coelicolor A3(2), Gene 95, HODGSON,D. A. (1992), Molecular genetic 49-56. analysis of proline and tryptophan biosynthesis HUISMAN, W., KOLTER, (1994a), Regulation G. R. in Streptomyces coelicolor A3(2) - interaction of gene expression at the onset of stationary between primary and secondary metabolism - a phase in Escherichia coli, in: Regulation of Bacreview, Gene 115,5-12. terial Differentiation (PIGGOT, MORAN, P., P., C. HOPWOOD, A., BIBB, M. J., BRUTON,C. J., D. YOUNGMAN, Eds.), pp. 21-40. Washington, P., CHATER,K. F., FEITELSON, S., GIL, J. A. J. D C American Society for Microbiology. (1983), Cloning Streptomyces genes for antibiotHUISMAN, W., KOLTER, R. (1994b), Sensing G. ic production, Trends Biotechnol. 1, 4248. starvation: a homoserine lactone-dependent sigHOPWOOD, A., CHATER,K. F., BIBB, M. J. D. naling pathway in Escherichia coli, Science 265, (1995), Genetics of antibiotic production in 537-539. Streptomyces coelicolor A3(2), in: Genetics and C. H., K., Biochemistry of Antibiotic Production (VINING, HUTCHINSON, R., DECKER, MADDURI, L. OTTEN,S. L., TANG, (1994), Genetic control L., STUTTARD, Eds.), pp. 65-102. Newton, C., of polyketide biosynthesis in the genus StreptoMA: Butterworth-Heinemann.
5 References
101
myces, Anton. Leeuwenhoek Int. J. Gen. Microbiol. 64, 165-176. IRR, J. D. (1972), Control of nucleotide metabolism and ribosomal ribonucleic acid synthesis during nutrient starvation of Escherichia coli, J. Bacteri01. 110,554-561. ISHIZUKA, H., HORINOUCHI, KIESER, H. M., S., D. T. HOPWOOD, A., BEPPU, (1992), A putative two-component regulatory system involved in secondary metabolism in Streptomyces spp., J. Bacteriol. 174, 7585-7594. JONES, G. H. (1985), Regulation of phenoxazinone synthase expression in Streptomyces antibioticus, J. Bacteriol. 163, 1215-1221. KELEMEN,G. H., ZALACAIN, CULEBRAS, M., E., SENO,E. T., CUNDLIFFE, (1994), TranscripE. tional attenuation control of the tylosin-resistance gene tlrA in Streptomyces fradiae, Mol. Microbiol. 14, 833-842. KELLY,K. S., OCHI, JONES, G. H. (1991), PleioK., tropic effects of a relC mutation in Streptomyces antibioticus, J. Bacteriol. 173, 2297-2300. KHOKHLOV,A. S. (1982), Low molecular weight microbial bioregulators of secondary metabolism, in: Overproduction of Microbial Products (KRUMPHANZL, SIKYTA, VANEK,Z., V., B., Eds.), pp. 97-109. London: Academic Press. KHOKHLOV, S., TOVAROVA,I., BORISOVA, A. I. L. N., PLINER, A., SCHEVCHENKO,A., KORS. L. NITSKAYA, E. Y., IVKINA, N. S., RAPOPORT, I. A. (1967), A-factor responsible for the production of streptomycin by a mutant strain of Actinomyces streptomycini, Dokl. Akad. Nauk SSSR 177,232-235. KIESER, H. M., KIESER, T., HOPWOOD, A. D. (1992), A combined genetic and physical map of the Streptomyces coelicolor A3(2) chromosome, J. Bacteriol. 174, 5496-5507. KILLHAM, K., FIRESTONE, K. (1984a), Salt M. stress control of intracellular solutes in streptomycetes indigenous to saline soils, Appl. Environ. Microbiol. 47, 310-306. KILLHAM,K., FIRESTONE, K. (1984b), Proline M. transport increases growth efficiency in saltstressed Streptomyces griseus, Appl. Environ. Microbiol. 48, 239-241. KINASHI, SHIMAJI, SAKAI, (1987), Giant H., M., A. linear plasmids in Streptomyces which code for antibiotic biosynthesis genes, Nature 328, 454456. KOLTER, R., SIEGELE, A., TORMO, (1993), D. A. The stationary phase of the bacterial life cycle, Annu. Rev. Microbiol. 47, 855-874. KONG, L., DUBNAU, (1994), Regulation of comD. petence-specific gene expression by Mec-mediated protein-protein interaction in Bacillus
subtilis, Proc. Natl. Acad. Sci. USA 91, 57935797. LACALLE, A, TERCERO, A., JIMENEZ, A. R. J. (1992), Cloning of the complete biosynthetic gene cluster for an aminonucleoside antibiotic, puromycin, and its regulated expression in heterologous hosts, EMBO J. 11, 785-792. LANCE, HENGGE-ARONIS, R., R. (1991a), Identification of a central regulator of stationary-phase gene expression in Escherichia coli, Mol. Microbiol. 5, 49-59. LANGE, HENGGE-ARONIS, R., R. (1991b), Growth phase-regulated expression of bolA and morphology of stationary-phase Escherichia coli cells are controlled by the novel sigma factor us, J. Bacteriol. 173, 4474-4481. LANCE, HENGGE-ARONIS,(1994), The celR., R. lular concentration of the us subunit of RNA polymerase in Escherichia coli is controlled at the levels of transcription, translation, and protein stability, Genes Dev. 8, 1600-1612. LAVILLE, VOISARD, KEEL, C., MAURHOJ., C., FER, M., DBFAGO, HAAS,D. (1992), Global G., control in Pseudomonas fluorescens mediating antibiotic synthesis and suppression of black root rot of tobacco, Proc. Natl. Acad. Sci. USA 89, 1562-1566. LAWLOR, J., BAYLIS, A., CHATER, F. E. H. K. (1987), Pleiotropic morphological and antibiotic deficiencies result from mutations in a gene encoding a tRNA-like product in Streptomyces coelicolor A3(2), Genes Dev. 1, 1305-1310. LESKIW, K., LAWLOR, J., FERNANDEZ-ABAB. E. LOS, J. M., CHATER, F. (1991a), l T A codons K. in some genes prevent their expression in a class of developmental, antibiotic-negative, Streptomyces mutants, Proc. Natl. Acad. Sci. USA 88, 2461-2465. LESKIW, K., BIBB, J., CHATER, F. (1991b), B. M. K. The use of a rare codon specifically during development, Mol. Microbiol. 5, 2861-2867. LESKIW, K., MAH,R., LAWLOR, J., CHATER, B. E. K. F. (1993), Accumulation of bldA-specified transfer RNA is temporally regulated in Streptomyces coelicolor A3(2), J. Bacteriol. 175, 19952005. LILLEY, G., CLARK,A. E., LAWRENCE, C. G. (1981), Control of the production of cephamycin C and thienamycin by Streptomyces cattleya NRRL 8057, J. Chem. Tech. Biotechnol. 31,127134. LONEITO, M. A., BROWN, L., RUDD, K. E., K. BUTTNER, J. (1994), analysis of the StreptoM. myces coelicolor 9 gene reveals the existence of a subfamily of eubacterial RNA polymerase u factors involved in the regulation of extracyto-
102
scriptionally during carbon starvation, J. Bacteriplasmic functions, Proc. Natl. Acad. Sci. USA 91, 01. 175,2143-2149. 7573-7577. LOEWEN, c., VON OSSOWSKI, I., SWITALA, J., MCCUE,L. A., KWAK, BABCOCK, J., KENP. J., M. MULVEY, R. (1993), KatF (a")synthesis in M. DRICK, K. E. (1992), Molecular analysis of spoEscherichia coli is subject to posttranscriptional rulation in Streptomyces griseus, Gene 115, 173regulation, J. Bacteriol. 175, 2150-2153. 179. MACNEIL, J., OCCI,J. L., GEWAIN, M., MAC- MEIGHEN, A. (1991), Molecular biology of bacD. K. E. NEIL,T., GIBBONS, H., RUBY,C. L., DANIS, P. terial bioluminescence, Microbiol. Rev. 55, 123S. J. (1992), Complex organisation of the Strep142. tomyces avermitilis genes encoding the avermec- MENDEZ,C., BRARA,A. F., MANZANAL, B., M. tin polyketide synthase, Gene 115, 119-125. HARDISSON,(1985), Role of substrate mycelC. MAGNUSON, SOLOMON, GROSSMAN, D. R., J., A. ium in colony development in Streptomyces, (1994), Biochemical and genetic characterization Can. J. Microbiol. 31, 44-50. of a competence pheromone from B. subtilis, MIYAKE, HORINOUCHI, YOSHIDA, CHIK., S., M., Cell 77,207-216. BA, N., MORI, K., NOGAWA, MORIKAWA, N., MALPARTIDA, HOPWOOD, A. (1984), MolecF., D. N., BEPPU,T. (1989), Detection and properties ular cloning of the whole biosynthetic pathway of A-factor-binding protein from Streptomyces of a Streptomyces antibiotic and its expression in griseus, J. Bacteriol. 171, 4298-4302. a heterologous host, Nature 309,462464. MIYAKE, KUZUYAMA, HORINOUCHI, K., T., S., MALPARTIDA, NIEMI, J., NAVARRETE,R., F., BEPPU,T. (1990), The A-factor-binding protein HOPWOOD, A. (1990), Cloning and expresD. of Streptomyces griseus negatively controls strepsion in a heterologous host of the complete set tomycin production and sporulation, J. Bacteriol. of genes for biosynthesis of the Streptomyces 172,3003-3008. coelicolor antibiotic undecylprodigiosin, Gene MSADEK,T., KUNST,F., RAPOPORT,G. (1994), 93, 91-99. MecB of Bacillus subtilis, a member of the ClpC MARAHIEL, A., NAKANO, M., ZUBER,P. M. M. ATPase family, is a pleiotropic regulator con(1993), Regulation of peptide antibiotic productrolling competence gene expression and growth tion in Bacillus, Mol. Microbiol. 7, 631-636. at high temperature, Proc. Natl. Acad. Sci. USA MARTIN, J. F., GIL, J. A. (1984), Cloning and ex91,5788-5792. pression of antibiotic production genes, Bio/ MULVEY, R., LOEWEN, C. (1989), Nucleotide M. P. Technology 2,63-72. sequence of katF of Escherichia coli suggests MARTfN, J. F., LIRAS, (1989), Organization and P. KafF protein is a novel u transcription factor, expression of genes involved in the biosynthesis Nucleic Acids Res. 17, 9979-9991. of antibiotics and other secondary metabolites, NAKANO,M. M., ZUBER,P. (1993), Mutational Annu. Rev. Microbiol. 43,173-206. analysis of the regulatory region of the srfA opeL. MARTIN, J. F., MCDANIEL, E. (1975), Kinetics ron in Bacillus subtilis, J. Bacteriol. 175, 3188of biosynthesis of polyene macrolide antibiotics 3191. in batch cultures: cell maturation time, BiotechNAKANO, M., XIA,L., ZUBER, (1991), TranM. P. nol. Bioeng. 17, 925-938. scription initiation region of the srfA operon, H., P., N. MARTINEZ-COSTA, 0. ARIAS, ROMERO, M., PARRO, MELLADO, P., MALPARTIDA, which is controlled by the comP-comA signal V., R. transduction system in Bacillus subtilis, J. BacteF. (1996), A relAlspoT homologous gene from riol. 173,5487-5493. Streptomyces coelicolor A3(2) controls antibiotic M. N., J., biosynthetic genes, J. Biol. Chem. 271, 10627- NAKANO, M., CORBELL, BESSON, ZuBER, P. (1992), Isolation and characterization of 10634. sfp: a gene that functions in the production of MATSUMOTO, HONG,S . 4 , ISHIZUKA, HoA., H., the lipopeptide biosurfactant, surfactin, in BacilRINOUCHI, S., BEPPU,T. (1994), Phosphorylalus subtilis, Mol. Gen. Genet. 232, 313-321. tion of the AfsR protein involved in secondary K. J. metabolism in Streptomyces species by a euka- NARVA, E., FEITELSON, S. (1990), Nucleotide sequence and transcriptional analysis of the ryotic-type protein kinase, Gene 146, 47-56. redD locus of Streptomyces coelicolor A3(2), J. MCCANN, M. P., YIDWELL,J. P., MATIN, A. Bacteriol. 172, 326-333. (1991), The putative u factor KatF has a central role in development of starvation-mediated gen- NEAL,R. J., CHATER,K. F. (1987), Nucleotide seeral resistance in Escherichia coli, J. Bacteriol. quence analysis reveals similarities between pro173,4188-4194. teins determining methylenomycin A resistance in Streptomyces and tetracycline resistance in euMCCANN, P., FRALEY, D., MATIN, (1993), M. C. A. bacteria, Gene 58, 229-241. The putative ufactor KatF is regulated posttran-
5 References
103
NEIDHARDT, C., INGRAHAM,L., Low, K. B., F. J. MAGASANIK, SCHAECHTER, UMBARB., M., GER,H. E. (1987), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. Washington, DC: American Society for Microbiology. NGUYEN, H., JENSEN, B., THOMPSON, E., L. D. N. GENTRY, L., BURGESS, R. (1993), In vitro D. R. functional characterization of overproduced Escherichia coli katFlrpoS gene product, Biochemistry 32, 11 112-11 117. NISHINO, MURAO, (1981), Possible involveT., S. ment of plasmid in nucleotide pyrophosphokinase production and the relationship between this productivity and cellular accumulation of guanosine tetraphosphate (ppGpp) in Streptomycetes, Agric. Biol. Chem. 45, 199-208. NOVOA,M. A., DIAz-GUERRA, SANMILLAN, J. L., MORENO, (1986), Cloning and mapping of F. the genetic determinants for microcin (27 production and immunity, J. Bacteriol. 168, 13841391. OCHI,K. (1986), Occurrence of the stringent response in Streptomyces sp. and its significance for the inhibition of morphological and physiological differentiation, J. Gen. Microbiol. 132, 2621-2631. OCHI,K. (1987). A re1 mutation abolishes the enzyme induction needed for actinomycin synthesis by Streptomyces antibioticus, Agric. Biol. Chem. 51,829-835. OCHI,K. (1988), Nucleotide pools and stringent response in regulation of Streptomyces differentiation, in: Biology of Actinomycetes '88 (OKAMI, Y., BEPPU,T., OGAWARA, Eds.), pp. 330H., 337. Tokyo: Scientific Societies Press. OCHI,K. (1990a), Streptomyces relC mutants with an altered ribosomal protein ST-L11 and genetic analysis of a Streptomyces griseus relC mutant, J. Bacteriol. 172, 4008-4016. OCHI,K. (1990b), A relaxed (rel) mutant of Streptomyces coelicolor A3(2) with a missing ribosomal protein lacks the ability to accumulate ppGpp, A-factor and prodigiosin, J. Gen. Microbiol. 136, 2405-2412. OCONNOR, A., ZUSMAN, R. (1988), ReexK. D. amination of the role of autolysis in the development of Myxococcus xanthus, J. Bacteriol. 170, 41034112. OHASHI, ZHENG, H., NIHIRA, YAMADA, H., Y. T., Y. (1989), Distribution of virginiae butanolides in antibiotic-producingactinomycetes, and identification of the inducing factor from Streptomyces antibioticus as virginiae butanolide A, J. Antibiot. 42, 1191-1195. OKAMOTO, NAKAMURA,K., NIHIRA, T., YAS., MADA, Y. (1995), Virginiae butanolide binding
protein from Streptomyces virginiae, J. Biol. Chem. 270,12319-12326. OKAMOTO, NIHIRA, KATAOKA, SUZUKI, S., T., H., A., YAMADA, (1992), Purification and molecY. ular cloning of a butyrolactone autoregulator receptor from Streptomyces virginiae, J. Biol. Chem. 267, 1093-1098. OLUKOSHI, R., PACKTER, M. (1994), ImporE. N. tance of stored triacylglycerols in Streptomyces possible carbon source for antibiotics, Microbiology 140, 931-943. ONAKA, ANDO,N., NIHIRA, YAMADA, H., T., Y., BEPPU, T., HORINOUCHI, (1995), Cloning H. and characterization of the A-factor receptor gene from Streptomyces griseus, J. Bacteriol. 177, 6083-6092. POSTMA, W., LENGELER, W., JACOBSON, P. J. G. R. (1993), Phosphoenol pyruvate: carbohydrate phosphotransferase systems of bacteria, Microbiol. Rev. 57, 543-594. PU~TIKHUNT, OKAMOTO, NAKAMURA, C., S., T., T., Y. NIHIRA, YAMADA, (1993), Distribution in the genus Streptomyces of a homolog to nusG, a gene encoding a transcriptional anti-terminator, FEMS Microbiol. Lett. 110, 243-248. RAIBAUD, ZALACAIN, HOLT, T. G., TIA., M., ZARD, R., THOMPSON, J. (1991), Nucleotide C. sequence analysis reveals linked N-acetyl hydrolase, thioesterase, transport, and regulatory genes encoded by the bialaphos biosynthetic gene cluster of Streptomyces hygroscopicus, J. Bacteriol. 173, 4454-4463. RETZLAFF, MAYER, BEYER, AHLERT, L., G., S., J., S., J., W. VERSECK, DISTLER, PIEPERSBERG, (1993), Streptomycin production in streptomycetes: a progress report, in: Industrial Microrganisms: Basic and Applied Molecular Genetics (BALTZ,R. H., HEGEMAN, D., SKATRUD, G. P. L., Eds.), pp. 183-194. Washington, DC: American Society for Microbiology. ROGGIANO, DUBNAU, (1993), ComA, a M., D. phosphorylated response regulator protein of Bacillus subtilis, binds to the promoter region of srfA, J. Bacteriol. 175, 3182-3187. ROMEO, PREISS, (1989), Genetic regulation of T., J. glycogen biosynthesis in Escherichia coli: in vitro effects of cyclic AMP and guanosine 5'-diphosphate 3'-diphosphate and analysis of in vivo transcripts, J. Bacteriol. 171, 2773-2782. ROMEO, GONG, LIU,M. Y., BRUN-ZINKERT., M., NAGEL, M. (1993), Identification and molecuA. lar characterization of csrA, a pleiotropic gene from Escherichia coli that affects glycogen biosynthesis, gluconeogenesis, cell size, and surface properties, J. Bacteriol. 175, 47444755. RUDNER, Z., LE DEAUX,J. R., IRETON,K., D. GROSSMAN, D. (1991), The SPOOKlocus of A.
104
Bacillus subtilis is homologous to the oligopeptide permease locus and is required for sporulation and competence, J. Bacteriol. 173, 13881398. SAIER,M. H. (1989), Protein phosphorylation and allosteric control of inducer exclusion and catabolite repression by the bacterial phosphoenolpyruvate: sugar phosphotransferase system, Microbiol. Rev. 53, 109-120. SAIER,M. H. (1991), A multiplicity of potential carbon catabolite repression mechanisms in prokaryotic and eukaryotic microorganisms, New Biol. 3, 1137-1147. SARUBBI, RUDD, K. E., CASHEL,M. (1988), E., Basal ppGpp level adjustment shown by new spoT mutants affect steady state growth rates and rrnA ribosomal promoter regulation in Escherichia coli, Mol. Gen. Genet. 213, 214-222. SCHNIDER, KEEL,C., BLUMER, TROXLER, U., C., J., DEFAGO, HASS,D. (1995). Amplification G., of the housekeeping sigma factor in Pseudomonas fluorescens CHAO enhances antibiotic production and improves biocontrol abilities, J. Bacteriol. 121, 416-421. SCHONER, GEISTLICH, ROSTECK, RAO, B., M., P., R. N., SENO,E., REYNOLDS, Cox, K., BURP., C. GETT, S., HERSHBERGER, (1992), Sequence similarity between macrolide-resistance determinants and ATP-binding transport proteins, Gene 115,93-96. SCHREIBER, METZGER,S., AIZENMAN, G., E., ROZA, S., CASHEL,M., GLASER,G. (1991), Overexpression of the relA gene in Escherichia coli, J. Biol. Chem. 266, 3760-3767. SENO, T., BALTZ, H. (1989), Structural organE. R. ization and regulation of antibiotic biosynthesis and resistance genes in actinomycetes, in: Regulation of Secondary Metabolism in Actinomycetes (SHAPIRO, Ed.), pp. 2-48. Boca Raton, F L S., CRC Press. J., H. O., SIMUTH, HUDEC,J., CHAN, T., DANYI, ZELINKA, (1979), The synthesis of highly J. phosphorylated nucleotides, RNA and protein by Streptomyces aureofaciens,J. Antibiot. 32,5358. STASTNA, MIKULIK, (1981), Role of highly J., K. phosphorylated nucleotides and antibiotics in the development of streptomycetes, in: Proc. 4th Int. Symp. Actinomycete Biology, Cologne, pp. 481-486 (SCHAAL, P., PULVERER, Eds.). K. G. Stuttgart, New York: Gustav Fischer Verlag. STEIN, COHEN, N. (1989), A cloned regulatoD., S. ry gene of Streptomyces lividans can suppress the pigment deficiency phenotype of different developmental mutants, J. Bacteriol. 171, 2258-2261. STOCK,J. B., STOCK,A. M., MOTTONEN, M. J.
(199O), Signal transduction in bacteria, Nature 344,395-400. STRAUCH, TAKANO, BAYLIS, A., BIBB, E., E., H. M. J. (1991), The stringent response in Streptomyces coelicolor A3(2), Mol. Microbiol. 5, 289298. STROHL, R. (1992), Compilation and analysis of W. DNA sequences associated with apparent Streptomyces promoters, Nucleic Acids Res. 20, 961974. STUTZMAN-ENGWALL, OTTEN, HUTCHINK. J., S., SON,C. R. (1992), Regulation of secondary metabolism in Streptomyces spp. and overproduction of daunorubicin in Streptomyces peucetius, J. Bacteriol. 174, 144-154. SWIFT, WINSON, K., CHAN,P. F., BAINTON, S., M. M., P. N. J., BIRDSALL, REEVES, J., REES,C. E. S. J. D., CHHABRA, R., HILL,P. J., THROUP, P., BYCROFI-, W., SALMOND, P. C., WILB. G. LIAMS, STEWART, S. A. B. (1993), A novel P., G. strategy for the isolation of lux1 homologues: evidence for the widespread distribution of a LuxR: Lux1 superfamily in enteric bacteria, Mol. Microbiol. 10, 511-520. TAKANO, BIBB,M. J. (1994), The stringent reE., sponse, ppGpp and antibiotic production in Streptomyces coelicolor A3(2). Actinomycetologica 8, 1-10. TAKANO, GRAMAJO,H. C., STRAUCH, E., E., ANDRES,N., WHITE, J., BIBB, M. J. (1992), Transcriptional regulation of the redD transcriptional activator gene accounts for growth-phase dependent production of the antibiotic undecylprodigiosin in Streptomyces coelicolor A3(2), Mol. Microbiol. 6, 2797-2804. TAKAYANAGI, TAMAKA, TAKAHASHI, Y., K., H. (1994), Structure of the 5 ' upstream region and the regulation of the rpoS gene of Escherichia coli, Mol. Gen. Genet. 243, 525-531. TANAKA, TAKAYANAGI, FUJITA, ISHIK., Y., N., HAMA, A., TAKAHASHI, (1993), HeterogeneH. ity of the principal sigma factor in Escherichia coli: The rpoS gene product, d8, a second is principal sigma factor of RNA polymerase in stationary phase Escherichia coli, Proc. Natl. Acad. Sci. USA 90,3511-3515. TANG,L., GRIMM, ZHANG, A., Y.-X., HUTCHINSON, C. R. (in press), Purification of the DNAbinding protein DnrI, a transcriptional factor of daunorubicin biosynthesis in Streptomyces peuceticus, Mol. Microbiol. THIARA, S., CUNDLIFFE, (1988), Cloning and A. E. characterization of a DNA gyrase B gene from Streptomyces sphaeroides that confers resistance to novobiocin, EMBO J. 7,2255-2259. THIARA, S., CUNDLIFFE, (1989), Interplay of A. E. novobiocin-resistant and -sensitive DNA gyrase
5 References
105
activities in self-protection of the novobiocin producer, Streptomyces sphaeroides, Gene 81, 65-72. THIARA, S., CUNDLIFFE, (1993), Expression A. E. and analysis of two gyrB genes from the novobiocin producer, Streptomyces sphaeroides, Mol. Microbiol. 8, 495-506. R. J., THOMPSON, J., SKINNER, H., THOMPSON, C. WARD, M., HOPWOOD, A., CUNDLIFFE, J. D. E. (1982), Biochemical characterisation of resistance determinants cloned from antibiotic-producing streptomycetes, J. Bacteriol. 151, 678685. TSAO,S. W., RUDD,B. A. M., HE, X., CHANG, C., FLOSS,H. G. (1985), Identification of a red pigment from Streptomyces coelicolor A3(2) as a mixture of prodigiosin derivatives, J. Antibiot. 38,128-130. VAN SINDEREN, GALLI,G., COSMINA, DE D., P., FERRA, WITHOFF, VENEMA, GRAND], F., S., G., G. (1993), Characterization of the srfA locus of Bacillus subtilis: only the valine-activating domain of srfA is involved in the establishment of genetic competence, Mol. Microbiol. 8, 833841. VICENTE, KUSHNER, R., GARRIDO, ALM., S. T., DEA,M. (1991), The role of the gearbox in the transcription of essential genes, Mol. Microbiol. 5,2085-2091. VOGTLI,M., CHANG, P.-C., COHEN,S. N. (1994), afsR2 a previously undetected gene encoding a 63-amino-acid protein that stimulates antibiotic production in Streptomyces lividans, Mol. Microbiol. 14, 643-653. VOLZ, K. (1993), Structural conservation of the CheY superfamily, Biochemistry 32, 1174111753. WASSERMAN, H., SHAW,C. K, SYKES,R. J. H. (1974), The biosynthesis of metacycloprodigiosin and undecylprodigiosin, Tetrahedron Lett. 33, 2787-3091.
WATERS, VUJAKLIJA, GOLD, M. R., DAVB., D., IES,J. (1994), Protein tyrosine phosphorylation in streptomycetes, FEMS Microbiol. Lett. 120, 187-190. WEINRAUCH, GUILLEN, DUBNAU, A. Y., N., D. (1989), Sequence and transcription mapping of Bacillus subtilis competence genes comB and comA, one of which is related to a family of bacterial regulatory determinants, J. Bacteriol. 171, 5362-5375. WEINRAUCH, PENCHEV,R., DUBNAU,E., Y., SMITH, DUBNAU, (1990), A Bacillus subtilI., D. is regulatory gene product for genetic competence and sporulation resembles sensor protein members of the bacterial two-component signaltransduction systems, Genes Dev. 4, 860-872. WHITE,J., BIBB, M. J. (in press), The bldA-dependence of undecylprodigiosin production in Streptomyces coelicolor A3(2) involves a pathway-specific regulatory cascade, J. Bacteriol. WIETZORREK, (1996), Regulation of antibiotic A. production in Streptomyces coelicolor A3(2)analysis of the afsB mutant BH5. Ph D Thesis, University of East Anglia, Norwich, UK. WILLIAMS, (1994), Compromising bacterial comP. munication skills, J. Pharmacol. 46, 1-10. XIAO,H., KALMAN, IKEHARA, ZEMEL,S., M., K., GLASER, CASHEL, (1991), Residual guaG., M. nosine 3 ,5 -bispyrophosphate synthetic activity of relA null mutants can be eliminated by spoT null mutations, J. Biol. Chem. 266,5980-5990. YAMADA, SUGAMURA, KONDO,K, YAY., K., NAGIMOTO, OKADA,H. (1987), The strucM., ture of inducing factors for virginiamycin production in Streptomyces virginiae,J. Antibiot. 40, 496-504. ZAMBRANO, M., SIEGELE,D. A,, ALMIR6N, M. M., TORMO,A., KOLTER,R. (1993), Microbial competition: Escherichia coli mutants that take over stationary phase cultures, Science 259, 1757-1760.
Biotechnology
3 Screening of Novel Receptor-Active Compounds of Microbial Origin
HARUOTANAKA
OMURA SATOSHI
Tokyo, Japan
1 Introduction 109 2 Assay Methods for Screening of Receptor-Active Compounds 109 2.1 Assays Based on Physiological Activities in Animal Tissues and Cells 109 2.2 Assays Using Radio-Labeled Ligands 110 2.3 Functional Assays Using Recombinant Cells Transformed with a Receptor Gene 110 3 Receptor Antagonists 111 3.1 Antagonists of Low Molecular Weight Ligand Receptors 112 3.1.1 Muscarinic Acteylcholine Receptor Antagonists 112 3.1.2 Dopamine Receptor Antagonists 114 3.1.3 NMDA Receptor Antagonists 114 3.1.4 Leukotriene B4 Receptor Antagonists 114 3.1.5 PAF Receptor Antagonists 114 3.1.6 Fibrinogen Receptor Antagonists 115 3.1.7 Estrogen Receptor Antagonists 115 3.1.8 Androgen Receptor Antagonists 116 3.2 Antagonists of Peptide Ligand Receptors 116 3.2.1 Cholecystokinin Receptor Antagonists 116 3.2.2 Endothelin Receptor Antagonists 116 3.2.3 Substance P Receptor Antagonists 120 3.2.4 ANP Receptor Antagonists 120 3.2.5 Arginine - Vasopressin Receptor Antagonists 121 3.2.6 Oxytocin Receptor Antagonists 121
108
5 6 7 8
3.2.7 Complement C5a Receptor Antagonists 121 3.2.8 Devazepide - A Non-Peptide Peptide Ligand Antagonist under Development for Medical Use 121 Receptor Agonists 122 4.1 Motilides (Macrolides with Motilin Activity) 122 4.2 Other Agonists 123 Inhibitors of Virus Receptor Binding - gp12O-CD4 Binding Inhibitors 124 Current State and Future Perspectives 127 Concluding Remarks 127 References 128
2 Assay Methods for Screening of Receptor-Active Compounds
109
1 Introduction
Development of receptor agonists and antagonists as drugs had already begun before their receptors were characterized as substances. These drugs account for a fairly high percentage of the medications currently used. Most of them are produced by chemical synthesis. In recent years, many receptor-active compounds of microbial origin have been discovered, stimulating efforts for developing new drugs from these lead compounds. These are now providing an important tool for clarifying the functions of receptors and the relationship between receptor function and disease. Receptors are specific proteins, which recognize exogenous signaling molecules or physical stimuli, and induce cellular responses. Receptors located in the membrane or the cytoplasm function as the first window in the transmission of extracellular information into the cell. Some receptors, e.g., virus receptors and the LDL receptor, are involved in the uptake of components cells and not directly in signal transduction. Early in the 1980s, the nicotinic acetylcholine receptor gene was cloned and its primary structure was determined (NODAet al., 1983). Since then, knowledge regarding the receptor-constituting proteins has been cultivated, and cloning of many receptor genes has been performed, thus stimulating discussions about the relationship between receptor structure and function (PEROUTKA, 1994). However, before their characterization as substances, receptors had been recognized as a vague concept and had contributed greatly to the analysis of drug effects and to the development of new drugs. A number of derivatives and analogs of low molecular weight ligands have been synthesized. Through the study of the effects and the pharmacological actions of these substances on receptors, many agonists and antagonists have been developed and are currently used as drugs. In recent years, the development of radiolabeled ligands has simplified receptor binding experiments using tissue homogenates or cells. With such techniques, many synthetic compounds and microbial cultures have been
screened, resulting in the discovery of new subtances which act on receptors. They provide not only lead compounds for the development of new drugs but also important tools for clarifying the receptor functions. For the study of the individual functions of receptor subtypes, which have been discovered in recent years, the identification of substances selectively acting on receptor subtypes is needed now. This paper reviews the general considerations of receptor-active compounds of microbial origin, the screening methods, and the physiological and pharmacological activities of those discovered to date. In addition, future perspectives for this class of compounds are also discussed.
2 Assay Methods for Screening of Receptor-Active Compounds

Although animal experiments are the most reliable method of assessing pharmacological actions, they require much cost and labor and are not suitable for the examination of numerous samples. A common alternative is screening using radio-labeled ligands and receptor-containing cells or tissue homogenates. Various simple methods of screening have been devised, some of which are presented below.
2.1 Assays Based on Physiological Activities in Animal Tissues and Cells

The Magnus test, using blood vessels, has been employed for the screening of numerous samples, although the detection of ligands is not very easy with this technique. FUJIIet al. (1991) assessed the contraction inhibiting activity with the following technique: First, an
110
isolated tracheal specimen was suspended in a Magnus bath. Subsequently, bronchial contraction was induced by neurokinin A. Microbial cultures were then added to assess the contraction inhibitory activity. Using this technique, active substances from 10000 samples were selected. This led to the isolation of actinomycin D. This compound inhibited the contraction induced by neurokinin A (IC5,,= 1.8 x 10-6M), but did not inhibit that by substance P, acetylcholine, etc. (FUJI] et al., 1991). Methods based on cellular responses to ligands allow an easier examination of many samples than those using tissues. One example for this is the use of platelets. Platelet aggregation is known to be induced by collagen, ADP, arachidonic, acid, thrombin, PAF etc. Their antagonists can be isolated by selecting substances which inhibit platelet aggregation (NAKAGAWA, 1992). With such a method, OKAMOTO al. (1986a, b) discovered two et PAF receptor antagonists, i.e., FR-49175 produced by Penicillium terlikowskii and FR900452 produced by Streptomyces phaeofaciens. LAUERet al. (1991) screened thromboxane A2 receptor antagonists taking inhibition of platelet aggregation as an indicator. Another cell culture method is based on macrophage chemotaxis. TSUJIet al. (1992a), e.g., identified the leukotriene B4 antagonist WF11605 influencing chemotaxis of polymorphonuclear leukocytes. These methods examine the responses of tissues and cells and hence are advantageous in that they allow a distinction of agonists from antagonists. However, since all the various responses of tissues or cells are shown in which many receptors other than the target receptor are included, selectivity may not be very high with these methods. However, if the examiner is experienced and performs careful observation, the discovery of compounds with novel physiological actions can be expected.
commercially available. They can be used for experiments on receptor binding in cells, tissue homogenates and cell fractions (receptors for steroid hormones etc. are located in the cytoplasm). The amount of a radio-labeled ligand bound to the receptor is estimated from the amount of the ligand bound to the receptor in the presence of excess cold ligand to obtain the amount of specific receptor binding, which can be used for selecting substances which specifically inhibit receptor binding. The receptor binding inhibitors include both agonists and antagonists. Agonists can be distinguished from antagonists by examining the influence of receptor binding inhibitors on the physiological actions of ligands, e.g., using the Magnus method. These methods are estimated to have been employed in screening of ligands from microorganisms frequently. However, many publications lack descriptions of the screening methods. Following recent success in the cloning of many receptor genes, it is now possible to examine the binding of radio-labeled ligands to recombinant receptors by radioimmunoassay or ELISA.
2.3 Functional Assays Using Recombinant Cells Transformed with a Receptor Gene
A new type of assay for ligands using recombinant microorganisms has been reported. KING et al. (1990) constructed a recombinant yeast with which ligands of the &adrenergic receptors can be assessed by transfecting genes of the human &-adrenergic receptor and a G protein a-subunit into the yeast. Saccharomyces cerevisiae possesses the G protein but lacks the a-subunit which is necessary for intracellular signal transduction. Therefore, the human &-adrenergic receptor gene (hPAR) and the mammalian a-subunit gene (rat Gsa) were transfected into S. cerevisiae, resulting in coupling of P and ysubunits. Furthermore, a system for the identification of &-adrenergic receptor agonists by colorimetry was established by linking the &receptor G protein to the P-galactosidase gene (Fig. 1).
2.2 Assays Using Radio-Labeled Ligands

In recent years, a number of radio-labeled ligands (such as hormones, autacoids, cytokines and neurotransmitters) have become
3 Receptor Antagonists
11 1
Saccharomyces cerevisiae
Fig. 1 A new screening system for agonists and antagonists of G protein coupled receptors. .
Growth
Fig. 2. A new screening system for agonists and antagonists of steroid receptors. Saccharomyces cerevisiae ura3 was transformed with the URA3 gene by homologous recombination and the URA3 gene was expressed under control of a steroid-dependent promoter.
Agonist Agonistt Antagonist
MCDONELL constructed another recombinant yeast for the assessment of ligands for steroid hormone receptors, utilizing the similarities of the transcription factors between 1991). As steyeasts and mammals (ABBOTT, roid hormone receptors are a family of transcription factor, an in vivo transcription system could be established using inducible expression vector system containing steroid hormone receptor genes. The linking of this system to the URA3 gene in yeast resulted in the establishment of a system for measuring substances acting on the steroid hormone receptors in S. cerevisiae. As shown in Fig. 2, this system uses yeast growth as an indicator. These approaches to assay systems using recombinant microorganisms can also be applied to establish assay systems for many other receptor ligands. Since the distinction between agonists and antagonists, which could
not be made with binding experiments, is possible with such systems, they are expected to provide simpler assay systems if automation techniques are incorporated.
Ever since antibiotics were discovered from microorganisms, microorganisms have been regarded as a treasure-house of secondary metabolites. They serve as an important source of physiologically active substances, in addition to antibiotics. Plants and marine organisms have also been screened but the percentage of new substances discovered from microorganisms is much higher. Receptor-ac-
112
CI -
HO
Ergotamine
Muscarine
Zealarenone
(Gibberellazeae)
(Clavi=P Purpurea)
a-Receptor antagosist
(Amanita musearia)
Muscarinic acetylcholine receptor agonist
Estrogen receptor agonist
Fig. 3. Traditional receptor agonists and antagonists of microbial origin.
tive substances derived from microorganisms have a long history. It dates back to the discovery of ergot alkaloids (leading to the clinical use of ergotamine and ergometrine), muscarine (an acetylcholine receptor agonist) and zealarenone (an estrogen receptor agonist) (Fig. 3), although these substances were not found by screening for receptor-active compounds. There were more systematic approaches to discover receptor-active compounds of microbial origin after the identification of the cholecystokinin antagonist asperlicin in 1985. Receptor-active compounds discovered from microorganisms by such efforts are described below.
3.1 Antagonists of Low Molecular Weight Ligand Receptors

(Tab. 1, Fig. 4)
3.1.1 Muscarinic Acetylcholine Receptor Antagonists

TAKESAKO al. (1988) screened microoret ganisms for muscarinic acetylcholine receptor antagonists in order to develop anticholinergic agents useful as anti-ulcer agents. Taking guinea pig brain homogenates as receptors and [3H]quinuclidinyl benzilate as ligands,
Tab. 1 Antagonists of Microbial Origin of Low Molecular Weight Ligand Receptors .
Ligand Acetylcholine
Antagonist IJ2702-I and -11
Producer Soil isolate
Reference
TAKESAKO al. (1988), et UENOet al. (1988) Actinoplanes sp. Dopamine Sch42029 HEDGE al. (1991) et Verticillium sp. NMDA ES-242-1 to -8 TOKIet al. (1992a, b) TSUJI al. (1992a), et Fungus Leukotriene Ba WF11605 SHIGEMATSU (1992) et al. Novobiocin Streptomyces sp. TSUJI al. (1992b) et PAF FR-49175 Penicillium terlikowskii OKAMOTO al. (1986a) et FR-900452 S. phaeofaciens OKAMOTO al. (1986b) et Phomatins (A, B, A,, B,) SUGANO al. (1991) et Phoma sp. Fibrinogen Tetrafibricin S. neyagawaensis KAMIYAMA al. (1993a, b) et Estrogen R1128 A, B, C, D HORIet al. (1993a, b, c) Streptomyces sp. Napiradiomycin A and B1 Streptomyces sp. HORIet al. (1993e) 3-Chloro-4-(2-amino-3-chloroPseudomonas sp. Androgen HORIet al. (1993d) pheny1)-pyrrole WS9761 A and B Streptomyces sp. HORI al. (1993f) et
113
IJ 2702-1 (R: CH f CHCHSH3) IJ 2702-11 (R:CH&H&H&H3)
42029
I I O C b OH
ES242-1 (R1= H, R2 = OCOCH3) ES242-2 (R1 , R2 = OCOCb) ES242-3 (R1= OH, & --OCOCH3) ES242-4 (R1, R2 =OH) ES-242-5 ( R1= H, R2 = OH)
FR-49175
S-cb
WF 11605 C b FR-W)0452
Phomatin A
Phomatin B
Phomalin81
Phomatin82
OH
OH
Tetrafibricin
OH OH OH OH OH OH C b
of antagonists of Fig. 4. Structures low molecular weight ligand receptors of microbial origin.
HO
J p "W q
OH
OH
n "
OH
&C
OH
R1128A: R = CH&H&& R1128B: R = CH&H&H&b R1128C: R = CH&H&H(CH& R1128D: R = CH&H&H&H&b
WS9761 A: R = CH3 WS9761 B: R = CH@H
114
they assayed inhibitory effects on the receptor ligand binding of cultures of soil isolated actinomycetes, fungi and bacteria. The cultures showing inhibition were examined for anticholinergic activity, using isolated guinea pig ileum. In this way, two new compounds, IJ2702-I and IJ2702-I1 (IC50=0.3 resp. 0.6 pg/mL), were isolated from the culture of an actinomycete strain, I52702 (TAKESAKO et al., 1988; UENOet al., 1988).
3.1.4 Leukotriene B4 Receptor Antagonists
Leukotriene B, (LTB4) is an autacoid which promotes aggregation, degranulation and chemotaxis of polymorphonuclear leukocytes. LTB4 is thought to be involved in inflammatory reactions. TSUJIet al. (1992a) examined microorganisms for substances that inhibit the LTBCinduced chemotaxis of rat polymorphonuclear leukocytes, leading to the discovery of WF11605 produced by a fungus. 3.1.2 Dopamine Receptor WF11605 is a new compound with a triterAntagonists pene glucoside structure (SHIGEMATSU al., et 1992). WF11605 was found to inhibit not only HEDGE al. (1991) searched for dopamine chemotaxis (IC50=1.7 x lo-' M) but also the et D1 receptor-active compounds of microbial binding of [3H]LTB4 to the membrane fracorigin, using [3H]Sch23390 and rat striatum. tion of polymorphonuclear leukocytes In their study, Sch42029 (2,5-dihydroxyace- (IC5,, = 5.6 x 10 M) and the LTBCinduced toanilide) was found to be a D1 receptor-spe- degranulation of polymorphonuclear leukocific ligand. Many of the drugs for the treat- cytes (ICS0=3.0x lo-' M). These results inment of Parkinson's disease, which is related dicate that WF11605 is an LTB4 receptor anto abnormal dopamine metabolism in the tagonist. Its LD,,, in mice was 1.0 g/kg or brain, are D2 receptor-specific. Sch42029 is more (i.p.). During screening for LTB4 antagonists, the first natural substance specific to the D1 TSUJIet al. (1992b) recently found that novoreceptor (Ki 0.6 pM). = biocin, an antibiotic in clinical use, acts as an antagonist of LTB4 receptors. This compound inhibited the binding of [3H]LTB4 to 3.1.3 NMDA Receptor the membrane fraction of polymorphonuclear Antagonists M). leukocytes (ICs0= 1.0 x 10 -' Since leukotrienes are known to be involved in the onset TOKI al. (1992a, b) carried out screening of ear edema in mice, the investigators examet work with [3H]TCP [l-(l-(2-thienyl)cyclohex- ined the effect of novobiocin on ear edema in ylpiperidine] and a rat brain membrane frac- mice induced by arachidonic acid, and found tion. They identified new compounds, ES- that local treatment with this antibiotic 242-1 through ES-242-8, which are produced (0.1 pg or more per ear) suppressed the forby Verticilliurnsp. and serve as NMDA recep- mation of edema. The compound was effector antagonists. These compounds inhibit the tive even when administered orally binding of [3H]TCP to the synaptic mem- (ED50 =220 pg/kg). brane (IC50:0.1 pM for ES-242-1) but not the binding of [3H]kainic acid. Although MK801 and ketamine are also known as synthetic compounds of this type, the ES-242 series are 3.1.5 PAF Receptor Antagonists the first new compounds of microbial origin Platelet activating factor (PAF) is produced (TOKIet al., 1992d). They have recently been used in experiments to clarify the pharmaco- by various cells and tissues. Even very small logical actions at the molecular level of amounts of PAF exert various biological actions. It is a new physiologically active subNMDA receptors. stance involved in different diseases such as bronchial asthma, inflammatory reactions,
-'
115
renal disease, collagen disease, and anaphy- conformational changes within the molecule. laxis. Therefore, PAF antagonists are ex- Fibrinogen binding to the receptors on the pected not only to clarify the physiological ac- surface of platelets is a prerequisite for platetions and pathophysiological roles of PAF but let aggregation. Thus, fibrinogen receptor analso to provide an effective therapeutic agent tagonism is a good target for a platelet aggregation inhibitor. for the treatment of these diseases. In the course of their screening program OKAMOTO al. (1986a, b) examined miet croorganisms for PAF antagonists using inhi- for fibrinogen binding antagonists, KAMIYAbition of PAF-induced platelet aggregation as MA et al. (1993a, b) isolated a non-peptide anan indicator. They found FR-49175 and FR- tagonist, tetrafibricin, from the culture broth 900452; FR-49175 was identified as bisdes- of an actinomycete. Tetrafibricin strongly inthiobis(methylthio)gliotoxin, while FR- hibited the binding of fibrinogen to its recepof 900452 was a new compound. The ICSO the tors with an ICSO 46nM. It also inhibited of platelet aggregation inhibiting effect of FR- ADP-, collagen-, and thrombin-induced agof 49175 was 8.5 pM. Intravenous injection gregation of human platelets with an ICSO (0.1 mglkg) to guinea pigs inhibited PAF-in- 5.6 pM, 11.0 pM, and 7.6 pM, respectively. et duced bronchial stenosis (OKAMOTO al., Tetrafibricin is a novel non-peptide anta1986a) FR-900452 is a compound with a gonist of the fibrinogen receptor. unique structure, including piperidine and indolinone. It was able to inhibit PAFinduced rabbit platelet aggregation 3.1.7 Estrogen Receptor (ICS0=3.7 x lo- M), while its inhibitory effect on platelet aggregation induced by col- Antagonists lagen, arachidonic acid or ADP was much Non-steroidal estrogen receptor antagoweaker. The compound markedly suppressed PAF-induced bronchial stenosis, hypotension nists, e.g., tamoxifen, have been used successand elevation in vascular permeability in gui- fully in the therapy of advanced breast cannea pigs when it was administered intrave- cer, especially estrogen receptor positive nously, even in low doses below 10 pg/kg breast cancer. Although this therapy results in remarkable improvements for breast can(OKAMOTO al., 1986b). et SUGANO al. (1991) examined marine mi- cer patients, the development of tamoxifen et croorganisms for secondary metabolites and resistance frequently occurs and most patients isolated the phomatins A, B, B,, and B2 from eventually relapse. One potential method to Phoma sp., a species of Fungi Zmperfecfi liv- overcome the resistance is the use of estrogen ing upon crabshells. These four compounds receptor antagonists with a new chemical inhibited PAF-induced platelet aggregation, structure different from tamoxifen and re1 with an ICSOof 1 . 0 ~ l O - ~ M ,. 7 ~ 1 O - ~ M , lated compounds, containing the triphenyl M, and 1.6 x M, respectively. ethylene moiety. 9.8 x Based on such considerations HORIet al. screened microbial products for new non-steroidal estrogen receptor antagonists without 3.1.6 Fibrinogen Receptor the triphenyl ethylene moiety. They found Antagonists new non-steroidal estrogen receptor antagonists - R1128 A, B, C, and D - from the culPlatelet aggregation plays a key role in nor- ture broth of Streptomyces sp. No. 1128 mal hemostasis and thrombosis. Platelets first (HORIet al., 1993a, b, c). These compounds adhere and spread onto the thrombogenic inhibited estrogen binding to its receptor. The components of the vascular subendothelium ICSOvalues of R1128 A, B, C, and D for at the sites of vascular lesions. When stimu- partially purified rat uterine cytosol reM, 1.2 x lo- M, lated by an agonist, such as ADP, collagen or ceptor were 1.1 x M, respectively. thrombin, the fibrinogen receptors acquire 2.6 x l o p 7M, and 2.7 x the ability to bind fibrinogen through some R1128B was a competitive inhibitor of es-
116
trogen receptor binding and inhibited the growth of estrogen-responsive human mammary adenocarcinoma MCF-7 cells in soft agar. This inhibition was reversed by addition o estradiol to the culture medium. R1128B f showed antitumor activities against MCF-7 when xenografted to nude mice by implantation into the subrenal capsule of mice (SRC assay). The potency of R1128 B was about 8fold lower than that of tamoxifen both i v i m n and i vivo (HORIet al., 1993~). recent n A study by HORI al. (1993d) revealed that naet piradiomycin A and B1, which have been known to possess antimicrobial activities, are estrogen receptor antagonists M and 3.5 x M, re(ICS0=4.2x spectively.
3.2 Antagonists of Peptide Ligand Receptors

(Tab. 2, Fig. 5)
3.2.1 Cholecystokinin Receptor Antagonists
Cholecystokinin (CCK) is a digestive hormone which promotes lipid degradation and absorption by stimulating gallbladder contraction, pancreatic juice secretion and small bowel motility. Its involvement in the central regulation of appetite and pain has recently been noted. Known CCK receptors include CCK-A, primarily located in the periphery, and CCK-B, primarily located centrally. After the discovery of the CCK-A antagonists asperlicins (CHANGet al., 1985; GOETZ et al., 3.1.8 Androgen Receptor 1985, 1988; LIESCHet al., 1985, 1988), the Antagonists CCK-B antagonist tetronothiodin (ICSO against the CCK-B receptor = 3.6nM) was et Androgen plays an important role in the found (OHTSUKA al., 1992, 1993a, b; WAprostatic growth including benign prostatic TANABE et al., 1993). Both compounds are hyperplasia and prostate cancer. Androgen non-peptide antagonists. Using asperlicin as a actions are thought to be mediated through lead compound, devazepide was synthesized et binding to its own receptor. Therefore, an- (GOETZet al., 1985; EVANS al., 1986) and drogen receptor antagonists can be used in is under development now as an oral agent the treatment for androgen-responsive dis- for the treatment of pancreatitis etc., as mentioned below (see Sect. 3.2.8). Recently, aneases. et During the course of search for non-steroi- thramycin (KUBOTA al., 1989) and virginiadal androgen receptor binding inhibitors, mycin M1 (LAMet al., 1991) were found to be HORIet al. (1993d) found that 3-chloro-4-(2- CCK-B antagonists. Anthramycin has a benamino-3-chlorophenyl)-pyrrole (WB2838), a zodiazepin moiety like asperlicin, but it binds known antifungal antibiotic, is a non-steroidal to the CCK-B receptor unlike asperlicin. androgen receptor antagonist. More recently, HORIet al. (1993f) discovered the novel androgen receptor antagonists WS9761 A and 3.2.2 Endothelin Receptor B. WS9761 A and B inhibited androgen receptor binding with ICs0 values of Antagonists 8.6 x M and 4.5 x M, respectively, Endothelin (ET) was discovered in 1988 as and showed weak inhibitory activity against a new peptide with potent activity to induce estrogen receptor binding. vascular contraction. During the subsequent five years, three isopeptides of ET (ET-1, -2, and -3) were found and there are at least two receptors (ETA and ETB) for ET. Studies of the agonists and antagonists of ET have also been carried out. Following the discovery of cyclic peptide antagonists of microbial origin (BE-18257A and B) (IHARAet al., 1991; Ko-
3 Receptor Antagonists Tab. 2. Antagonists of Microbial Origin of Peptide Ligand Receptors Ligand Subtype Antagonist Asperlicin A, B, C, D, E" Anthramycin" Tetronothiodin a Producer
Aspergillus alliaceus
117
Reference CHANG al. (1985), et GOETZet al. (1985, 1988), LIESCH al. et (1985, 1988) KUBOTA al. (1989) et OHTSUKA al. et (l992,1993a, b), WATANABE (1993) LAMet al. (1991)
Cholecystokinin A
B B
S. spadicogriceus
Streptomyces sp.
Gastrinkholecystokinin Endothelin
Verginiamycin MI L-156586, L-156587, L-156588, L-156906 BE-18257 A and B
S. olivaceus
Substance P
ANP
Arginine-vasopressin Oxytocin C5a
IHARA al. (1991), et KOJIRI al. (1991), et NAKAJIMA al. et (1991) A MIYATA al. et WS-7338 C and D Streptomyces sp. (1992a, b, c) A WS-OOO9 A and B" Streptomyces sp. MIYATA al. et (1992d, e) Cochinmicin I, 11, 111 Microbispora sp. A and B LAMet al. (1992), ZINKet al. (1992) A Asterric acid" Aspergillus sp. OHASHI al. (1992) et NK-1 and 2 WS-9326 A S. violaceusniger HAYASHI al. et (1992) NK-1 Actinomycin D Streptomyces sp. FUJII al. (1991) et NK-1 Fiscalin A, B, C" Neosartorya fischeri WONGet al. (1993b) NK-1 Anthrotainin" Gliocladium catenulatum WONGet al. (1993a) NK-1 and 2 WIN 64821" SEDLOCK al. et Aspergillus sp. (1994) Anantin S. coerulescens WEBERet al. (1991), WYSSet al. (1991) HS-142-1" Aureobasidium pullulans MORISHITA al. et (1991a, b) Fischerella sp. Hapalindolinone A SCHWARTZ al. et (1987) and B" S. silvensis PETTIBONE al. et L-156373 (1989) Streptomyces sp. HENSENS al. et L-156602 = PD 124966 (1991), HURLEY et al. (1986)
S. misakiensis
" Nonpeptide antagonist.
JIRI
et al., 1991; NAKAJIMAet al., 1991) and WS7338C and D (MIYATAet al., 1992a,b, c), non-peptide antagonists such as WS009A and B (MIYATA et al., 1992e, d), cochinmicins (LAMet al., 1992; ZINK al., 1992) and asteret
et ric acid (OHASHI al., 1992) have been identified. The cyclic peptide BE-18257B exhibited IC,, values of 1.4 and 0.8 mM against [ '251]ET-1 binding to aortic smooth muscle tissue and to ventricle membranes from pig,
118
HO HN
O=d\
0
Telronothiodin
I ,c*O H, NH
BE-18%7A (R : H) BE-182578 (R: C w
Cochinmidn OH
HO I II 111
*
R s
H CI CI
SCH&HNHCOCH3 WSOOSA ( R H ) 'COOH WS OOSB (R: OH)
Anantin
GIG+ g1+6
G I C 4 OH n~0,3 - 0
(cap),
m=5-15
HS142-1 Glc; Dglucose, Cap; capronic add
$ y
H
Hapalindolinone A (R: Cl) Hapalindolinone B (R: H)
119
Asperlidn
L-156373
Fiscalin A
Fiscalin B
Fiscalin C
Anthrotainin
WIN 64821
Fig. 5. Structures of antagonists of peptide ligand receptors of microbial origin.
which are ETA-rich tissues. It did not inhibit ['251]ET-1binding to ETB-rich tissues. In isolated rabbit iliac arteries, BE-18257B antagonized ET-1-induced vasoconstriction. Thus, itwas found that BE-18257B is an ETArecep-
tor antagonist (IHARAet al., 1991). On the other hand, the peptolide cochinmicin 1 is a nonselective antagonist for ETA and ETB sites (LAMet al., 1992). Studies using these antagonists are expected to clarify the physi-
120
ological and pathophysiological significance of ET isopeptides and their receptors and to provide new therapeutic agents.
3.2.4 ANP Receptor Antagonists
Atrial natriuretic peptide (ANP) is a peptide hormone involved in the regulation of body water, electrolytes and blood pressure. It has potent diuretic and vasoconstrictive ac3.2.3 Substance P Receptor tions. ANP and two other peptides with difAntagonists ferent amino acid sequences (RNP and CNP) constitute a natriuretic peptide family. Two Neuropeptides such as substance P and types of ANP receptors are known, one carryneurokinin A, which markedly induce airway ing a guanylate cyclase domain inside the constriction and promote mucosal secretion, cells and another without any guanylate cyhave been investigated because of their rela- clase domain. The receptors with a guanylate tionship to respiratory diseases such as asth- cyclase domain may be involved in the ANPma. HAYASHI al. (1992) searched for inhi- induced elevation of the intracellular cGMP et bitors of [3H] substance P binding to guinea level, while the receptor without this domain pig lung membrane fractions and discovered is thought to be involved in the ANP metabWS9326A, a cyclic depsipeptide produced by olism. However, the exact physiological roles an actinomycete. WS9326A exhibits an ICSO of ANP and its pathophysiological signifiM in the above assay and cance have not yet been fully clarified. For value of 3.6 x acts as a tachykinin antagonist in various this reason, the development of ANP recepfunctional assays. Its tetrahydro derivative, tor antagonists is needed. WEBER et al. (1991) examined microbial FK224, was more potent than WS9326A (ICSO= x lo-' M) and antagonizes both metabolites for those substances which inhibit 1.0 the neurokinin 1receptor (involved in airway the binding of [ '251]-labeled rat ANP, i.e., edema) and the neurokinin 2 receptor (in- [ '251]rANP to the bovine adrenocortical volved in airway constriction) (HASHIMOTO membrane. They discovered the peptide anet al., 1992). This compound is under devel- tagonist anantin produced by an actinomyopment for clinical use. In addition, non-pep- cete. This is a cyclic peptide composed of 17 tide inhibitors, termed fiscalins, with moder- amino acids (WYSSet al., 1991). The comate neurokinin 1 binding activity have also pound was suggested to be an ANP receptor been reported (WONG et al., 1993b). There antagonist because it inhibited both the bind1.0 has also been a report of a tetracyclic coming of [ lZI]rANP to receptors (ICSO= p,M) pound, anthrotainin, with neurokinin 1 activi- and the ANP-induced intracellular cGMP acty (WONGet al., 1993a), but this compound cumulation in bovine aortic smooth muscle was found to be a noncompetitive substance cells, and it did not show any agonist effect (WEBERet al., 1991). P antagonist. MORISHITA al. (1991a, b) examined miet More recently, WIN64821, a non-peptide secondary metabolite produced by Aspergil- crobial cultures for substances inhibiting the lus sp. was found to inhibit radiolabeled sub- binding of ['"IIrANP to the rabbit renal corstance P binding in a variety of tissues with tical membrane. They isolated HS-142-1 from Ki values ranging from 0.24 p,M in human the culture broth of a fungus belonging to the astrocytoma U-373 MG cells to 7.89 FM genus Aureobasidiurn. HS-142-1 is a new poin submaxillary membranes. Additionally, lysaccharide composed of a linear pl,6-gluWIN64821 was found to inhibit [ '2SI]-neuro- a x e chain conjugated to caproic acid. In the kinin A binding to the neurokinin 2 receptor target tissues and cells of ANP, i.e., in the boin human tissue at a concentration equivalent vine adrenocortical membrane fraction (Moto its neurokinin 1 activity (0.26 p,M). RISHITA et al., 1992; ODAet al., 1992), bovine WIN64821 was shown to be a functional an- vascular smooth muscle cells (IMURAet al., tagonist of neurokinin 1 and neurokinin 2 re- 1992), rat renal glomerulus (SANO et al., 1992a, b, c), LLC-Pkl cells (TANAKAet al., ceptors (OLEYNEK al., 1994). et 1992), and PC12 cells (TOKIet al., 1992c), this
121
compound specifically antagonized the binding of [ 1251]rANP ANP receptors containto ing guanylate cyclase (ICSOfor rabbit renal cortical membranes = 0.3 p,g/mL), and inhibited the ANP-induced elevation in cGMP level. Although anesthetized rats treated intravenously with HS-142-1 alone showed no reaction, the diuretic response of the animals to exogenous or endogenous ANP was not seen after pretreatment with HS-142-1 (SANO et al., 1992b, c). HS-142-1 thus seems to be a new non-peptide ANP antagonists useful for the analysis of the physiological and pathophysiological role of ANP. In the future, this compound often will be used in cardiovascular studies.
stance from an actinomycete (L-156373) which inhibited the binding of [3H]oxytocin to the rat uterine membrane fraction. The Ki value of this compound was 150 p,M. Its affinity for oxytocin receptors was more than 20 times higher than that for the arginine-vasopressin receptors (AVP-V, and AVP-V2). Its derivative L-365209, produced by dehydroxylation of the N-hydroxyleucine unit and oxidation of the piperazic acid residues of the L156373, was 20 times as potent as L-156373 and had a K i of 7.3 p,M. This derivative was highly selective and antagonized the oxytocin action to the rat uterus (IDS0=460 p,g/kg).
3.2.5 Arginine-Vasopressin Receptor Antagonists

Arginine-vasopressin (AVP) is a peptide composed of 9 amino acids. It is a hormone secreted from the posterior lobe of the pituitary gland and possesses antidiuretic and hypertensive properties. The renal AVP-V2 receptor is involved in the antidiuretic, the AVP-Al receptor of the cardiac smooth muscle in the hypertensive action. SCHWARTZ al. (1987) isolated substances et from a microorganism of the genus Fischerella that inhibited the binding of [3H]AVP to the renal tissue containing V2 receptors. They called the substances hapalindolinone A and B. These compounds are non-peptide antagonists which carry cyclopropane and indolinone skeletons. They inhibit not only the = binding of [3H]AVP to renal tissue 37.5 p,M) but also the AVP-induced activation of adenylate cyclase (ICSO= 44.6 p,M).
3.2.7 Complement C5a Receptor Antagonists

C5a is thought to be involved in the aggravation of various inflammatory allergic diseases. HENSENS al. (1991) isolated a subet stance from actinomycete metabolites (L156602) that inhibited the binding of human polymorphonuclear leukocytes. This compound was considered to be identical to PD124966 which had been discovered as an antitumor antibiotic. The structure of this compound shown in Fig. 5, was proposed by HENSENS et al. (1991) and has never been determined.
3.2.8 Devazepide- A Non-Peptide Peptide Ligand Antagonist under Development for Medical Use
In 1985, CHANG al. (1985) isolated the et CCK receptor antagonist asperlicin from a microorganism (Aspergillus alliaceus), which is the first non-peptide antagonist of peptide ligand receptors. Asperlicin does not act on the CCK-B receptors primarily located in the center of the body, but selectively acts on the CCK-A receptors mainly located in the periphery. Its discovery confirmed the presence of CCK receptor subtypes and made it possible to distinguish receptor A from receptor B. CCK is known to be involved in diseases such as pancreatitis. Asperlicin was initially ex-
3.2.6 Oxytocin Receptor Antagonists

Oxytocin is a peptide composed of 9 amino acids. It is a hormone secreted from the posterior lobe of the pituitary gland and induces uterine contraction and milk secretion. PEITIBONE et al. (1989) isolated a sub-
122
Diare-pm
Fig. 6. Structures and activities of the CCK-A antagonist asperlicin and its analog derazepide.
pected to be useful as a therapeutic agent, but, because of low solubility, it is ineffective when administered orally. In 1983, KUBOTA al. (1983, 1985) found et in an experiment with peripheral tissue and brain that diazepam, %ananti-anxiety drug which can be administered orally, antagonized CCK. In in v i m binding experiments, however, it did not antagonize CCK (GOETZ et al., 1985). Since asperlicin has a benzodiazepine skeleton like diazepam, efforts have been started to synthesize highly soluble, orally applicable derivatives. EVANSet al. (1986) synthesized a number of derivatives and analogs containing benzodiazepine and indole because the asperlicin molecule contains benzodiazepine and L-tryptophan. Devazepide which contains D-tryptophan, carrying a 2-indolyl bond to benzodiazepine as shown in Fig. 6, is more than loo0 times more potent than asperlicin. It is highly soluble in water while retaining selective activity. Thus devazepide was selected as an excellent candidate for development (EVANS al., 1986). et At present, devazepide is under development for treatment of acute pancreatitis, biliary colic, abdominal pain, and anorexia (EVANS, 1989).
4 Receptor Agonists
4.1 Motilides (Macrolides with Motilin Activity)
ITOH et al. (1985) found that the side effect of erythromycin causing diarrhea etc., is similar to the effect of motilin which is a hormone promoting gastrointestinal motility. Subseand et quently, OMURA coworkers (OMURA et et al., 1987; TSUZUKI al., 1989; SUNAZUKA al., 1989) synthesized a number of erythromycin derivatives which exert only motilin activity and no antibacterial activity. EM536, one of these derivatives, showed 2890 times higher motilin activity than erythromycin A. Quaternary ammonium derivatives of the desosamine moiety of 6,9-hemiketal erythromycin A possess the most potent gastrointestinal motor stimulating activity. However, these ionized derivatives have low permeability. Consequently, EM523 (18 times as potent as erythromycin A) and EM574 (248 times as potent) were selected from the tertiary ammonium derivatives as candidates for medical use (Fig. 7, Tab. 3). At present, these two derivatives (EM523 and EM574) are under de-
4 Receptor Agonists
123
HO
OH
Erythromycin A (EMA)
EM201 R = C & EM523 R=CHfi& EM574 R = CH(CH&
X-
HO
HO
0
EM485 EM491 EM511 EM536 R=CH3 R=CH&H3 R = CH&H=C& R = CH&&H
OH
0
EM502 R=CH3 EM506 R = CH&H=C& EM507 R = CH&ICH
OH
Fig. 7. Structures of erythromycin A and its derivatives.
velopment to be used as gastrointestinal motility regulators when administered either by injection or orally. The macrolides exerting motilin activity were called motilides by TSUZUKI al. (1989). KONDO et al. (1988) et demonstrated that motilides are agonists of the motilin receptor. Because motilin is a peptide hormone composed of 22 amino acids, motilin itself is ineffective when administered orally while the abovementioned motilide EM574 (a non-peptide agonist) is effective. At present, morphine, an enkephalin agonist, is the only non-peptide agonist of peptide ligand receptors in clinical use, motilide EM374 is expected to be used as the second non-peptide agonist.
4.2 Other Agonists

The traditional receptor agonists muscarine and zearalenone (see Fig. 3) are well known to be isolated from microorganisms. Muscarine is an alkaloid from the red variety of Amanita muscaria, a poisonous mushroom (KUEHL et al., 1955; KOGL et al., 1957). It is an acetylcholine receptor agonist and used as a important biochemical reagent. Zearalenone is an estrogen receptor agonist isolated from the mycelia of the fungus Gibberella zeae (Fusarium graminearum) (STOBet al., 1962; URRYet al., 1966). No agonists of peptide ligand receptors of microbial origin, except erythromycin A (see
124
Tab. 3. Antimicrobial Activities and Gastrointestinal Motor Stimulating Activities of Erythromycin A and
its Derivatives Compound Antimicrobial Activity (MIC, pg/mL) SA EMA EM201 EM523 EM574 EM485 EM491 EM511 EM536 EM502 EM506 EM507 0.2 50 >loo >loo BS
0.1 25 >loo >loo
Gastrointestinal Motor Stimulating Activity (relative activity) KP 6.25

1 10 18 248 21 111 256 2890 65 115 202
BC 0.1 25 >loo
EC
>loo
12.5
>loo
>loo loo 100 >loo 100
>loo
100 >loo 100 >loo
>loo
>loo
>loo >loo
>loo >loo loo
>loo >loo >loo >loo >loo
>lo0 >loo
>loo >loo >loo

>loo >loo
>loo >loo >loo
>loo
>loo loo
>loo
>loo
>loo >loo >loo
SA Staphylococcus aureus ATCC6358P BS Bacillus subtilis ATCC6633; BC Bacillus cereus IF03001; EC Escherichia coli NIHJ; KP Klebsiella pneumoniae ATCC10031.
Sect. 4.1), have been reported to date. Their discovery is desired for the development of new orally available drugs to replace peptide hormones and cytokines - in the same way as motilides are being developed as orally available gastrointestinal motor stimulating drugs.
5 Inhibitors of Virus
Receptor Binding gp120-CD4 Binding Inhibitors
In the screening program for new inhibitors of gp120-CD4 binding from microorganisms, OMURA al. (1993) discovered the novel inet hibitors isochromophilone I and I1 (Fig. 8) from the culture borth of Penicillium sp. FO2338, and chloropeptin I and I1 (Fig. 9) from Sfrepfomyces sp. WK-3419 (OMURAet al., unpublished data). Chloropeptin 11, however, was identified with complestatin (KANEKO et al., 1989). The inhibitory activities against gp120CD4 binding were determined by enzymelinked immunosorbent assay (ELISA) using recombinant soluble CD4 and recombinant gp120 as described by GILBERT al. (1991). et Isochromophilone I and I1 inhibited gp120CD4 binding with ICs0 values of 6.6 p M and 3.9 pM, respectively. The IC,, values for chloropeptin I and I1 were 2.0 pM and The entry of viruses needs their specific 3.3 pM, respectively. binding to a receptor of the susceptible cell. Anti-HIV activity was assayed as follows. Human immunodeficiency virus (HIV) entry Peripheral human lymphocytes were isolated begins with the highly specific binding of the by density gradient centrifugation. After stimHIV gp120 envelope glycoprotein with a CD4 ulation by a mitogen, the cells were infected molecule on the surface of most susceptible with a standardized preparation of HIV-1. cells (MCDOUGAL al., 1986; SADROSKI et et Subsequently, the infected cells were cultured al., 1986; LIFSONet al., 1986). Blocking of in the presence of the agent for 4 days. The HIV entry is one of the most important tar- amount of viral core protein p24 synthesized gets for HIV therapy (JOHNSTON HOTH, and released by the infected cells was deterand mined by the capture-ELISA technique on 1993).
5 Inhibitors of Virus Receptor Binding - gp120-CD4 Binding Inhibitors
125
CH3
lsochromophiloneI
Fig. 8. Structures of the gp120-CD4 binding inhibitors isochromophilone I and 11.
lsochromophiloneII
Tab. 4. Inhibition of HIV Replication on the Viral Core Protein Level (for the assay method, see text)
Sample Viral Core Protein p24 Synthesized (ng/mL) Day2 None Isochromophilone I1 Chloropeptin I 0 0 0 Day3 97.3 0 0 Day4 129.6 13.5 7.3
days 2, 3, and 4. By comparing with a standard preparation, the amount of protein (p24) produced by the virus infected cells was calculated. As shown in Tab. 4 isochromophilone I1 and chloropeptin I significantly inhibited HIV replication at 25 p M and 7.5 pM, respectively. The inhibition of HIV replication by isochromophilone I1 and chloropeptin I is considered to be due to blocking of HIV entry into the cells. Isochromophilone I and I1 are the first novel non-peptide compounds to inhibit gp120-CD4 binding. Isochromophilones and chloropeptins are expected to provide the lead compounds for development of HIV therapy.
126
Chloropeptin I
Chloropeptin II (complestatin)
Fig. 9. Structures of the gp120-CD4 binding inhibitors chloropeptin I and 11.
127
6 Current State and Future Perspectives

As described above, substances acting on receptors (i.e., agonists and antagonists) have been synthesized before the exact nature of receptors was clarified. These substances have contributed greatly not only to treatment of diseases but also to advances in studies in the field of cellular biology and pharmacology. Following recent commercialization of radio-labeled peptide ligands, screening of receptor-active compounds in various specimens such as microbial cultures has been performed in experiments involving the binding of these ligands to tissue or cells. In this way, new substances affecting receptors of peptide ligands have been discovered. At present, derivatives of these substances (i.e., devazepide, FK-224, and motilide) are under development for clinical use. This class of drugs will further increase in the future. It is speculated that orally administered non-peptide agonists and antagonists will bring about an epochal reform of drug therapy. To date, however, no low molecular weight compound acting on the receptors of macromolecular peptide ligands (e.g., ligands with 100 or more amino acid residues) has been reported although many natural peptide ligands or their analogs have been clinically used by injection. If the three-dimensional structure of ligands and their receptors is identified and the mode of the ligand-receptor binding is clarified in detail, the development of new drugs by computerized information processing will be possible. Although the primary structure of many receptors has been clarified to date, the three-dimensional structure of a receptor is not known until the crystalline structure of the growth hormone receptor complex (see below) has been determined. The first analysis of the crystalline structure of a macromolecular peptide ligand receptor complex was reported in 1992 (DEVOSet al., 1992). The analysis of the three-dimensional structure of the complex formed between human growth hormone and the extracellular
domain of its receptor revealed that this complex is composed of one hormone and two receptor molecules. The hormone has four helical structures with abnormal topology, and the receptor bound to it has two different binding domains. The two receptor molecules bind through the same amino acid residues in each domain to two structurally distinct sites of the hormone. At their C-terminal domains distant from the binding sites the two receptor molecules are in contact to each other. This contact may play a crucial role in intracellular signal transduction. Since the three-dimensional features of the mode of binding between growth hormone and its receptor has been clarified, molecular designing of new agonists and antagonists using computer graphics technology will advance in the future. However, because crystallization of the ligand-receptor complex usually is not easy, search for new receptoractive compounds and subsequent chemical modification of the thus discovered compounds will, for the time being, continue to play a principal role in the development of new receptor-active compounds.
This chapter provides a general review of receptor-active compounds. Studies of substances acting on receptors of peptide ligands still have only a very short history. We expect more simple assay techniques to be developed in the future, facilitating the discovery of many receptor-active compounds. These compounds will help to clarify the function of cells and elucidate the physiological and pathophysiological functions of receptors. As studies on receptor-active compounds of microbial origin have been advancing, some known substances have been highlighted because of their additional action on receptors. Considerably small differences in the chemical structures of ligands or receptors often reflect quite different actions and it is quite likely that the same compound can have two or more target molecules. This indicates that it is not easy to discover a highly selective
128
Fermentation, isolation and biological properties, J. Antibiot. 38, 1633-1637. GOETZ,M. A., MONAGHAN, L., CHANG, S. R. R. L., ONDEYKA, CHEN, T. B., LOTTI, V. J. J., (1988), Novel cholecystokinin antagonists from Aspergillus alliaceus. I. Fermentation, isolation, and biological properties. J. Antibiot. 41, 875Acknowledgements 877. HASHIMOTO, HAYASHI, MURAI, FUJII, M., K., M., The authors are indebted to Dr. H. KLEINT., NISHIKAWA, KIYOTO, OKUHARA, M., S., M., KAUF for providing the opportunity of this KOHSAKA, IMANAKA (1992), WS9326A, M., H. presentation, and also to Drs. S. TAKAMATSU a novel tachykinin antagonist isolated from Streptomyces violaceusniger No. 9326.11. Biologand J. INOKOSHIand Mr. K. MATSUZAKIfor ical characterization and pharmacological chartheir useful help in preparation of this manuacterization of WS9326A and tetrahydroscript. WS9326A (FK224), J. Antibiot. 45,1064-1070. K., M., N., HAYASHI, HASHIMOTO, SHIGEMATSU, NISHIKAWA, EZAKI,M., YAMASHITA, M., M., KIYOTO, S., OKUHARA,M., KOHSAKA,M., IMANAKA, (1992), WS9326A, a novel tachyH. kinin antagonist isolated from Streptomyces violaceusniger No. 9326. I. Taxonomy, fermentation, isolation, physico-chemical properties and ABBOIT, A. (1991), Receptor screens detect bioacbiological activities, J. Antibiot. 45, 1055-1062. tivity, Biotechnology 9, 694. R. CHANG, S. L., LOTTI,V. J., MONAGHAM, L., HEDGE, V. R., PATEL, M. G., HORAN,A. C., R. SCHWARTZ, L., HART,R., PUOR,M. S., GULJ. J., E. BIRNBAUM, STAPLAY, O., GOETZ,M. A., S. LO,V. P., IYENGAR, (1991), Sch42029, a natuALBERS-SCHONBERG, PATCHEIT, A. A., G., rally produced dopamine rceptor ligand taxonoJ. 0. J. LIESCH, M., HENSENS, D., SPRINGER, P. my, fermentation, isolation and structure, J. Ind. (1985), A potent nonpeptide cholecystokinin anMicrobiol. 8, 187-192. tagonist selective for peripheral tissues isolated 0. HENSENS, D., BORRIS,R. P., KOUPAL,L. R., from Aspergillus alliaceus, Science 230,177-179. S. A. CALDWELL, G., CURRIE, A., HAIDRI, C. DEVOS, A. M., ULTSCH, M., KOSSIAKOFF, A. A. A., HOMIK,C. F., HONEYCUTT, S., LINDENS. (1992), Human growth hormone and extracelluMAYER, M., SCHWARTZ, D., WEISSBERS. C. lar domain of its receptor: crystal structure of GER, B. A., WOODRUFF, B., ZINK, L., ZIH. D. the complex, Science 255, 306-312. TANO, L., FIELDHOUSE,M., ROLLINS,T., J. EVANS, E. (1989), Recent developments in choB. SPRINGER, S., SPRINGER, P. (1991), LM. J. lecystokinin antagonist research, Drugs of the 156,602, A C5a antagonist with a novel cyclic Future, 14,971-979. hexadepsipeptide structure from Streptomyces EVANS, E., BOCK,M. G., RIITLE, K. E., DIPARB. sp. MA6348. Fermentation, isolation and strucDO,R.M., WHIITER,W. L., VEBER,D. F., ANture determination, J. Antibiot. 44,249-254. DERSON, P. s., FREIDINGER, M. (1986), DeR. sign of potent, orally effective, nonpeptidal an- HORI,Y., ABE, Y., EZAKI,M., GOTO,T., OKUHARA, M., KOHSAKA,M. (1993a), R1128 subtagonists of the peptide hormone cholecystokinstances, novel non-steroidal estrogen-receptor in, Proc. Natl. Acad. Sci. USA 83,4918-4922. antagonists produced by a Streptomyces. I. TaxH., FUJII, T., MURAI,M., MORIMOTO, NISHIKAonomy, fermentation, isolation and biological S. WA,M., KIYOTO, (1991), Effects of actinomyproperties, J. Antibiot. 46, 1055-1062. cin D on airway constriction induced by tachyS., N., kinin and capsaicin in guinea-pigs, Eur. J. Phar- HORI,Y., TAKASE, SHIGEMATSU, GOTO,T., OKUHARA, KOHSAKA, (1993b), R1128 M., M. macol. 194,183-188. substances, novel non-steroidal estrogen-recepGILBERT, KIRIHARA, MILLS,J. (1991), EnM., J., tor antagonists produced by a Streptomyces. 11. zyme-linked immunoassay for human immunoPhysico-chemical properties and structure deterdeficiency virus type I envelope glycoprotein mination. J. Antibiot. 46, 1063-1068. 120, J. Clin. Microbiol. 29,142-147. R. GOETZ, M. A,, LOPEZ, M., MONAGHAN, L., HORI, Y., ABE, Y., NISHIMURA, GOTO, T., M., CHANG,R. S. L., LOTTI, V. J., CHEN, T. B. OKUHARA, M.,KOHSAKA, (1993c), R1128 M. (1985), Asperlicin, a novel non-peptidal cholesubstances, novel non-steroidal estrogen-recepcystokinin antagonist from Aspergillus alliaceus. tor antagonists produced by a Streptomyces. 111.
compound and that it is even possible to modify a compound so that its minor action replaces the major action - as seen in the case of motilide.
8 References
8 References
Pharmacological properties and antitumor activities, J. Antibiot. 46, 1069-1075. S., HORI, Y., ABE, Y., NAKAJIMA, TAKASE, H., FUJITA, GOTO, T., OKUHARA, KOHSAT., M., KA, M. (1993d), WB2838 [3-chloro-4-(2-amino3-~hlorophenyl)-pyrrole]:Non-steriodal androgen-receptor antagonist produced by a Peudomonas, J. Antibiot. 46, 1327-1333. HORI,Y., ABE, Y., SHIGEMATSU, S., GOTO, T. N. OKUHARA, KOHSAKA, (1993e), NapyraM., M. diomycins A and B1: Non-steroidal androgenreceptor antagonists produced by a Streptomyces, J. Antibiot. 46, 1890-1894. HORI, Y., ABE, Y., NAKAJIMA, SHIGEMATSU, H., S., M., N., TAKASE, GOTO, T., OKUHARA, KOHSAKA,M. (1993f), WS9761A and B: New nonsteroidal androgen-receptor antagonists produced by a Streptomyces, J. Antibiot. 46, 19011903. HURLEY,T. R., BUNGE,R. H., WILLMER, E., N. HOKANSON,G. C., FRENCH, J. C. (1986), PD124,895 and PD124,966, two new antitumor antibiotics, J. Antibiot. 39, 1651-1656. T., T., IHARA,M., FUKURODA, SAEKI, NISHIKIBE, M., KOJIRI,K., SUDA,H., YANO,M. (1991), An endothelin receptor (ETA) antagonist isolated from Streptomyces misakiensis, Biochem. Biophys. Res. Commun. 178,132-137. IMURA,R., SANO, T., GOTO, J., YAMADA,K., MATSUDA,Y. (1992), Inhibition by HS-142-1, a novel nonpeptide atrial natriuretic peptide antagonist of microbial origin, of atrial natriuretic peptide-induced relaxation of isolated rabbit aorta through the blockade of guanylyl cyclaselinked receptors, Mol. Pharmacol. 42,982-990. ITOH, Z., SUZUKI,T., NAKAYA, M., INOUE,M., ARAI,H., WAKABAYASHI, (1985), StructureK. activity relation among macrolide antibiotics in intitiation of interdigestive migrating contractions in the canine gastrointestinal tract, Am. J. Physiol. 248, G320-525. JOHNSTON, I., HOTH,G. F. (1993), Present staM. tus and future prospects for HIV therapies, Science 260,1286-1293. KAMIYAMA, UMINO, FUJISAKI, FUJIMOT., T., N., RI, K., SATOH,T., YAMASHITA, OHSHIMA, Y., S., WATANABE, YOKOSE,K. (1993a), TetrafiJ., bricin, a novel fibrinogen receptor antagonist. I. Taxonomy, fermentation, isolation, characterization and biological activities, J. Antibiot. 46, 1039-1046. KAMIYAMA, ITEZONO,Y., UMINO, SATOH, T., T., T., NAKAYAMA, YOKOSE,K. (1993b), TetraN., fibricin, a novel fibrinogen receptor antagonist. 11. Structural elucidation, J. Antibiot. 46, 10471054. KANEKO, I., KAMOSHITA,K., TAKAHASHI, S.
129
(1989), Complestatin, a potent anti-complement substance produced by Streptomyces lavendulae. I. Fermentation, isolation and biological characterization, J. Antibiot. 42,236-241. KING,K., DOHLMAN, G., THORNER, CARON, H. J., M. G., LEFKOWITZ,R. J. (1990), Control of yeast mating signal transduction by a mammalian a-adrenergic receptor and Gs a subunit, Science 250,121-123. KOGL, F., Cox, H. C., SALEMINK, A. (1957), C. Synthese eines Gemischs von Muscarin und seinen stereoisomeren Formen, Experienta 13, 137-138. KOJIRI,K., IHARA,M., NAKAJIMA, KAWAMUS., RA, K., FUNAISHI, YANO, M., SUDA, H. K., (1991), Endothelial-binding inhibitors, BE18257A and BE-18257B. 1. Taxonomy, fermentation, isolation and characterization, J. Antibiot. 4 , 1342-1347. 4 KONDO, Y., TORII, K., OMURA, ITOH, Z. S., (1988), Erythromycin and its derivatives with motilin-like biological activities inhibit the specific binding of lZsII-motilinto duodenal muscle, Biochem. Biophys. Res. Commun. 150,877-882. KUBOTA,K., SUNAGANE, SUGAYA, MATN., K., SUOKA, URUNO,T. (1983), Competitive anY., tagonists of cholecystokinin and some benzodiazepines at cholecystokinin receptors of smooth muscle, Jpn. J. Pharmacol. 33, 87. KUBOTA,K., SUGAYA, SUNGANE, MATSUK., N., DA, I., URANO,T. (1985), Cholecystokinin antagonism by benzodiazepines in the contractile response of the isolated guinea-pig gallbladder, Eur. J. Pharmacol. 110,225-231. Y., M. KUBOTA, SUGAYA, KOIZUMI, TODA, K., K., (1989), Cholecystokinin antagonism by anthramycin, a benzodiazepine antibiotic, in the central nervous system in mice, Brain Res. 485, 6266. KUEHL, F. A., JR., LEBEL, N., RICHTER,J. W. (1955), Isolation and characterization studies on muscarine, J. Am. Chem. SOC.77,6663-6665. LAM, Y. K. T., BOGEN,D., CHANG, S., FAUST, R. K. A., HENSENS,0. D., ZINK, D. L., SCHWARTZ,C. D., ZITANO,L., GARRITY, G. M., GAGLIARDI, M., CURRIE,S. A., WOODM. RUFF, H. B. (1991), Novel and potent gastrin and brain cholecystokinin antagonists from Streptomyces olivaceus. Taxonomy, fermentation, isolation, chemical conversion, and physico-chemical and biological properties, J. Anti4 biot. 4 , 613-625. D. J. LAM, Y. K. T., WILLIAMS, L. JR., SIGMUND, M., SANCHEZ,M., GENILLOUD, KONG, Y. O., S., L., STEVENS-MILES, HUANG,L., GARRITY, G. M. (1992), Cochinmicins, novel and potent cyclodepsipeptide endothelin antagonists from a
130
3 Screening of Novel Receptor-Active Compounds of Microbial Origin MIYATA, HASHIMOYO, FUJIE,K., SHOUHO, S., M., M., M., SOGABE,K., KIYOTO, S., OKUHARA, KOHSAKA, (1992e), WS009A and B, new enM. dothelin receptor antagonists isolated from Streptomyces sp. No. 89009. 11. Biological characterization and pharmacological characterization of WS009A and B, J. Antibiot. 45, 10411046. MORI, Y., ABE, Y., SHIGEMATSU, GOTO, T., N., OKUHARA, KOHSAKA, (1993c), NeoiraM., M. diomycins A and B1: non-steroidal estrogen-receptor antagonists produced by a Streptomyces, J. Antibiot. 46, 1890-1893. MORISHITA, SANO,T., ANDO,K., SAITOH,Y., Y., KASE,H., YAMADA, MATSUDA, (1991a), K., Y. Microbial polysaccharide, HS-142-1, competitively and selectively inhibits ANP binding to its guanylyl cyclase-containing receptor, Biochem. Biophys. Res. Commun. 176,949-957. MORISHITA, TAKAHASHI, SANO,T., KAY., M., WAMOTO, ANDO,K., SANO,H., SAITOH,Y., I., KASE, H., MATSUDA,Y. (1991b), Isolation and purification of HS-142-1, a novel nonpeptide antagonist for the atrial natriuretic peptide receptor form Aureobasidium sp., Agric. Biol. Chem. 55,3017-3025. MORISHITA, SANO,T., KASE,H., YAMADA, Y., K., T., Y. INAGAMI, MATSUDA, (1992), HS-142-1, a novel nonpeptide atrial natriuretic peptide (ANP) antagonist, blocks ANP-induced renal responses through a specific interaction with guanylyl cyclase-linked receptors, Eur. J. Pharmacol. 225,203-207. NAKAGAWA, (1992), Vasoactive substances, in: A. The Search for Bioactive Compounds from Microorganisms (OMURA,S., Ed.), pp. 198-212. Berlin-Heidelberg-New York: Springer-Verlag. S., K., NAKAJIMA, NIIYAMA, IHARA,M., KOJIRI, K., SUDA,H. (1991), Endothelin-binding inhibitors, BE-18257A and BE-18257B. 11. Structure determination, J. Antibiot. 44, 1348-1356. NODA,M., TAKAHASHI, TANABE, TOYOSAH., T., TO, M., KIKYOTANI, FURUTANI, HIROSE, S., Y., S., T., T., TAKASHIMA, INAYAMA, MIYATA, H., NUMA,S. (1983), Structural homology of Torpedo californica acetylcholine receptor subunits, Nature 302,528-532. ODA, S., SANO,T., MORISHITA, MATSUDA, Y., Y. (1992), Pharmacological profile of HS-142-1, a novel nonpeptide atrial natriuretic peptide (ANP) antagonist of microbial origin. 11. Restoration by HS-142-1 of ANP-induced inhibition of aldosterone production in adrenal glomerulosa cells, J. Pharmacol. Exp. Ther. 263,241-245. OHASHI,H., AKIYAMA, NISHIKORI, MoH., K., CHIZUKI, (1992), Asterric acid, a new endoJ.
Microbispora sp. I. Production, isolation and characterization. J. Antibiot. 45, 1709-1716. LAUER,U., ANKE,T., HANSSKE, (1991), AntiF. biotics from basidiomycetes XXXVIII. 2-Methoxy-5-methyl-1,4-benzoquinone, thromboxane a A2 receptor antagonist from Lentinus adhaerens, J. Antibiot. 44,5945. J. LIESCH,J. M., HENSENS,0. D., SPRINGER, P., R. CHANG, S. L., LOTH, V. J. (1985), Asperlicin, a novel non-peptidal cholecystokinin antagonist from Aspergillus alliaceus. Structure elucidation, J. Antibiot. 38,1638-1641. LIESCH, J. M., HENSENS,0. D., ZINK, D. L., GOETZ,M. A. (1988), Novel cholecystokinin antagonists from Aspergillus alliaceus. 11. Structure determination of asperlicins B, C, D, and E, J. Antibiot. 41, 878-881. LIFSON, D., FEINBERG, B., REYES,G. R., RAJ. M. BIN, L., BANAPOUR,B., CHAKRABART, S., Moss, B., WONG-STALL, STEIMER,K. S., F., ENGELMAN, G. (1986), Induction of CD4-deE. pendent cell fusion by the HTLV-IIIILAV envelope glycoprotein, Nature 323,725-728. MCDOUGAL, S., KENNEDY, S., SLIGH,J. M., J. M. J. CORT, S. P., MAWLE,C. A., NICHOLSON, K. A. (1986), Binding of HTLV-III/LAV to T4+ cells by a complex of the llOK viral protein and the T4 molecule, Science 231, 382-385. S., M., MIYATA, HASHIMOTO, MASUI,Y., EZAKI, M., M., TAKASE,S., NISHIKAWA, KIYOTO,S., OKUHARA,M., KOHSAKA,M. (1992a), WS7338, new endothelin receptor antagonists isolated from Streptomyces sp. No. 7338. I. Taxonomy, fermentation, isolation, physico-chemical properties and biological activities, J. Antibiot. 45,7682. MIYATA, HASHIMOTO, FUJIE, NISHIKAS., M., K., S., M., WA,M., KIYOTO, OKUHARA, KOHSAKA, M. (1992b), WS-7338, new endothelin receptor antagonists isolated from Streptomyces sp. No. 7338.11. Biological characterization and pharmacological characterization of WS-7338B, J. Antibiot. 45,83-87. MIYATA,S., FUKAMI, NEYA,M., TAKASE,S., N., KIYOTO,S. (1992c), WS-7338, new endothelin receptor antagonists isolated from Streptomyces sp. No. 7338. 111. Structures of WS-7338A, B, C, and D and total synthesis of WS-7338B, J. Antibiot. 45,788-791. S., H., MIYATA, OHHATA, MURAI,H., MASUI,Y., TAKASE, NISHIKAWA, KIYOS., M., EZAKI,M., M. TO, S., KOHSAKA, (1992d), WS009A and B, new endothelin receptor antagonists isolated from Streptomyces sp. No. 89009. I. Taxonomy, fermentation, isolation, physico-chemical properties and biological activities, J. Antibiot. 45, 1029-1040.
I
8 References
thelin binding inhibitor, J. Antibiot. 45, 16841685. OHTSUKA, KUDOH,T., SHIMMA, KOTAKI, T., N., H., NAKAYAMA, ITEZONO,Y., FUJISAWA, N., N., WATANABE,J., YOKOSE, K., SETO, H. (1992), Tetronothiodin, a novel cholecystokinin type-B receptor antagonist produced by Streptomyces sp., J. Antibiot. 45, 140-143. OHTSUKA,T., KOTAKI,H., NAKAYAMA, ITEN., N., ZONO,Y., SHIMMA, KUDO,T., KUWAHARA, T., ARISAWA, M., YOKOSE, K., SETO, H. (1993a), Tetronothiodin, a novel cholecystokinin type-B receptor antagonist produced by Streptomyces sp. NR0489. 11. Isolation, characterization and biological activities. J. Antibiot. 46, 11-17. OHTSUKA,T., NAKAYAMA, N., ITEZONO, Y., SHIMMA,N., KUWAHARA, T., YOKOSE, K., SETO,H. (1993b), Tetronothiodin a novel cholecystokinin type-B receptor antagonist produced by Streptomyces sp. NR0489. 111. Structural elucidation, J. Antibiot. 46, 18-24. OKAMOTO, YOSHIDA,K., UCHIDA,I., NISHIM., KAWA, M., KHOSAKA,M., AOKI, H. (1986a), Studies of platelet activating factor (PAF) antagonists from microbial products. I. Bisdethiobis(methy1thio)gliotoxin and its derivatives, Chem. Pharm. Bull. 34,340-344. OKAMOTO,M., YOSHIDA, K., NISHIKAWA, M., ANDO,T., IWAMI, KHOSAKA, AOKI,H. M., M., (1986b), FR-900452, a specific antagonist of platelet activating factor (PAF) produced by Streptomyces phaeofaciens. I. Taxonomy, fermentation, isolation and physico-chemical and biological characteristics, J. Antibiot. 39, 198204. J. D. OLEYNEK, J., SEDLOCK, M., BARROW,C. J., APPELL, K. C., CASIANO,F., HAYCOCK,D., WARD, S. J., KAPLITA, P., GILLUM,A. M. (1994), .WIN64821, a novel neurokinin antagonist produced by an Aspergillus sp. 11. Biological activity, J. Antibiot. 47, 399-410. OMURA,S., TSUZUKI, SUNAZUKA, MARUI, K., T., S., TOYODA,H., INATOMI, ITOH, Z. (1987), N., Macrolides with gastrointestinal motor stimulat0 ing activity, J. Med. Chem. 3 , 1941-1943. OMURA, TANAKA, MATSUZAKI, IKEDA, S., H., K., H., MASUMA,R. (1993), Isochromophilones I and 11, novel inhibitors against gp120-CD4 binding from Penicillium sp., J. Antibiot. 46, 19081911. PEROUTKA,S. J. (1994), Handbook of Receptors and Channels. Boca Raton, FL: CRC Press. PETTIBONE, D. J., CLINESCHMIDT, V., ANDERB. SON,P. S., FREIDINGER, M., LUNDELL, F., R. G. KOUPAL, L. R., SCHWARTS, D., WILLIAMC. SON, J. M., GOETZ, M. A., HENSENS,0. D., LIESCH,J. M., SPRINGER, P. (1989), A strucJ.
131
turally unique, potent, and selective oxytocin antagonist derived from Streptomyces silvensis, Endocrinology 125,217-222. SADROSKI, GOH, W. C., ROSEN,C., CAMPBELL, J., K., HASELTIME, A. (1986), Role of HTLVW. III/LAV envelope in syncytium formation and cytopathicity, Nature 322, 47W74. SANO,T., IMURA, MORISHITA, MATSUDA, R., Y., Y., YAMADA, (1992a), HS-142-1, a novel poK. lysaccharide of microbial origin, specifically recognizes guanylyl cyclase-linked ANP receptor in rat glomeruli, Life Sci. 51, 1445-1451. SANO, T., MORISHITA, MATSUDA, Y., YAY., MADA, K. (1992b), Pharmacological profile of HS-142-1, a novel nonpeptide atrial natriuretic peptide (ANP) antagonist of microbial origin. I. Selective inhibition of the actions of natriuretic peptides in anesthetized rats, J. Pharmacol. Exp. Ther. 260,825-831. SANO,T., MORISHITA, YAMADA, MATSUY., K., DA, Y. (1992c), Effects of HS-142-1, a novel non-peptide ANP antagonist, on diuresis and natriuresis induced by acute volume expansion in anesthetized rats, Biochem. Biophys. Res. Commun. 182,824-829. SCHWARTZ, E., HIRSCH, F., SPRINGER, P., R. C. J. D. PETTIBONE, J., ZINK,D. L. (1987), Unusual cyclopropane-containing hapalindolinones from a cultured Cyanobacterium, J. Org. Chem. 52, 3704-3706. SEDLOCK, M., BARROW,C. J., BROWNELL, D. J. E., HONG, A., GILLUM, M., HOUCK,D. R. A. (1994), WIN64821, a novel neurokinin antagonist produced by an Aspergillus sp. I. Fermentation and isolation, J. Antibiot. 47, 391-398. SHIGEMATSU, TUJI, E., KAYAKIRI, TAN., N., KASE, s., TANAKA, H., TADA, T. (1992), WF11605, an antagonist of leukotriene B4 produced by a fungus. 11. Structure determination, J. Antibiot. 45, 704-708. STOB, M., BALDWIN, S., TUITE,J., ANDREWS, R. F. N., GILLETTE, K. G. (1962), Isolation of an anabolic, uterotrophic compound from corn in1318. fected with Gibberella zeae, Nature 1%, SUGANO, M., SATO, A., IIJIMA,Y., OSHIMA, T., H., T., FURUYA, KUWANO, HATA, HANZAK., WA,H. (1991), Phomatin A: a novel PAF antagonist from a marine fungus Phoma sp., J. Am. Chem. SOC.113,5463-5464. SUNAZUKA, TSUZUKI, MARUI,S., TOYODA, T., K., N., H., OMURA,S., INATOMI, ITOH, Z. (1989), Motilides, macrolides with gastrointestinal motor stimulating activity. 11. Quaternary N-substituted derivatives of 8,9-anhydroerythromycin A 6,9-hemiacetal and 9,9-dihydroerythromycin A 6,9-epoxide, Chem. Pharm. Bull. 37,2701-2709.
132
3 Screening of Novel Receptor-Active Compounds of Microbial Origin Motilides, macrolides with gastrointestinal motor stimulating activity. I. 0-Substituted and tertiary N-substituted derivatives of 8,9-anhydroerythromycin A 6,9-hemiacetal, Chem. Pharm. Bull. 37,2687-2700. UENO,M., TAKESAKO, IKAI,K., KATAYAMA, K., K., SHIMANAKA, NAKAMURA, (1988), K., T. Isolation of novel muscarine antagonists, IJ 2702 and PF6766, Nippon Nogei Kagaku Kaishi 62, 339. URRY,W. H., WEHRMEISTER, L., HODGE, E. H. B., HIDY,P. H. (1966), The structure of zearalenone, Tetrahedron Lett. 27,3109-3114. J., N., K., WATANABE, FUJISAKI, FUJIMORI, ANZAI, Y., OSHIMA,S., SANO,T., OHTSUKA, T., WATANABE, OKUDA,T. (1993), TetronoK., thiodin, a novel cholecystokinin type-B receptor antagonist produced by Streptomyces sp. NR0489. I. Taxonomy, yield improvement and fermentation, J. Antibiot. 46,l-10. WEBER,W., FISCHLI, HOCHULI, KUPFER, W., E., E., WEIBEL,E. K. (1991), Anantin - a peptide antagonist of the atrial natriuretic factor (ANF). I. Producing organism, fermentation, isolation 4 and biological activity, J. Antibiot. 4 , 164-171. WONG,S.-M., KULLNIG, DEDINAS, APPELL, R., J., K. C., KYDD,G. C., GILLUM, M., COOPER, A. R., MOORE,R.(1993a), Anthrotainin, an inhibitor of substance P binding produced by Gliocladium catenulatum, J. Antibiot. 46,214-221. WONG,S.-M., MUSZA,L. L., KYDD,G. C., KULLNIG,R., GILLUM, M., COOPER,R. (1993b), A. Fiscalins: New substance-P inhibitors produced by the fungus Neosartorya fischeri. Taxonomy, fermentation, structures, and biological activities, J. Antibiot. 46, 545-553. WYSS, D. F., LAHM, H. W., MANNEBERG, M., LABHARDT, M. (1991), Anantin - a peptide A. antagonist of the atrial natriuretic factor (ANF). 11. Determination of the primary sequence by NMR on the basis of proton assignments, J. Antibiot. 44, 172-180. ZINK,D., HENSENS, D., LAM,Y. K. T., REAM0. J. ER,R., LIESCH, M. (1992), Cochinmicins, novel and potent cyclodepsipeptide endothelin antagonists from a Microbispora sp. 11. Structure determination, J. Antibiot. 45, 1717-1722.
TANAKA, ICHIMURA, NAKAJO, SNAJT., M., S., Y., DAR, R. M., MORISHITA, SANO, T., YAMADA, INAGAMI, MATSUDA, (1992), K., T., Y. HS-142-1, a novel non-peptide antagonist for atrial natriuretic peptide receptor, selectively inhibits particulate guanylyl cyclase and lower cyclic GMP in LLC-PKI cells, Biosci. Biotech. Biochem. 56,1041-1045. TASKESAKO, KURODA, UENO,M., SAITO, K., H., H., YAMAMOTO, NAKAMURA, (1988) J., T. Screening of muscarine antagonists by radioreceptor assay, Nippon Nogei Kagaku Kaishi 62, 338. M., I., TOKI,S., ANDO,K., YOSHIDA, KAWAMOTO, SANO,H., MATSUDA, (1992a), ES-242-1, a Y. novel compound from Verticilliumsp., binds to a site on N-methyl-D-aspartate receptor that is 5 coupled to the channel domain, J. Antibiot. 4 , 88-93. I., TOKI, S., ANDO, K., KAWAMOTO, SANO,H., YOSHIDA, MATSUDA, (1992b), ES-242-2, M., Y. -3, -4, -5, -6, -7, and -8, novel bioxanthracenes produced by Verticilliumsp., which act on the Nmethyl-D-aspartate receptor, J. Antibiot. 45, 1047-1054. TOKI,S., MORISHITA, SANO, MATSUDA, Y., T., Y. (1992c), HS-142-1, a novel non-peptide ANP antagonist, blocks the cyclic GMP production elicited by natriuretic peptides in PC12 and NG 108-5 cells, Neurosci. Lett. 135, 117-120. E., M., TOKI,S., TSUKUDA, NOZAWA, NONAKA, H., YOSHIDA, MATSUDA, (1992d), The M., Y. ES-242s, novel N-methyl-D-aspartate antagonists of microbial origin, interact with both the neurotransmitter recognition site and the ion channel domain, J. Biol. Chem. 267,14884-14892. TSUJI, E., TSURUMI, MIYATA,S., FUJIE, K., Y., KAWAKAMI, OKAMOTO, OKUHARA, A., M., M. (1992a), WF11605, an antagonist of leukotriene B4 produced by a fungus. I. Producing strain, fermentation, isolation and biological activity, J. Antibiot. 45,698-703. TSUJI, SHIGEMATSU, HATANAKA, YAE., N., H., M., M., OKUHARA, M. MASHITA, OKAMOTO, (1992b), Novobiocin, an antagonist of leukotriene B4, J. Antibiot. 45, 1958-1960. TSUZUKI, SUNAZUKA, MARUI, TOYODA, K., T., S., H., OMURA,S., INATOMI, ITOH, Z. (1989), N.,
Biotechnology
4 Microbial Lipids
COLINRATLEDGE
Hull, United Kingdom
1 Introduction 135 1.1 Lipid Nomenclature and Major Lipid Types 138 2 Accumulation of Lipid 139 2.1 Patterns of Accumulation 139 2.2 Efficiency of Accumulation 142 2.3 Biochemistry of Accumulation 143 3 Triacylglycerols and Fatty Acids 146 3.1 Bacteria 147 3.1.1 Polyunsaturated Fatty Acids in Bacteria 147 3.2 Yeasts 148 3.2.1 Production of a Cocoa Butter Equivalent Yeast Fat 148 3.2.1.1 Direct Feeding of Stearic Acid 150 3.2.1.2 Inhibition of Stearoyl Desaturase 151 3.2.1.3 Mutation 152 3.2.1.4 Metabolic Manipulation 154 3.2.1.5 Conclusions 154 3.3 Molds 155 3.3.1 y-Linolenic Acid (GLA, 183 w-6) 159 3.3.2 Dihomo-y-Linolenic Acid (DHGLA, 20:3 w-6) 160 3.3.3 Arachidonic Acid (ARA, 20:4 w-6) 161 3.3.4 Eicosapentaenoic Acid (EPA, 20: 5 w-3) 162 3.3.5 Docosahexaenoic Acid (DHA, 22 :6 w-3) 162 3.3.6 Eicosatrienoic Acid (ETA, 20:3 w-9, Mead Acid) 163 3.3.7 Conclusions 163 3.4 Algae 164 3.4.1 y-Linolenic Acid (GLA, 18:3 w-6) 167 3.4.2 Arachidonic Acid (ARA, 20:4 w-6) 167 3.4.3 Eicosapentaenoic Acid (EPA, 20: 50-3) 168 3.4.4 Docosahexaenoic Acid (DHA, 22:6 0-3) 169 3.4.5 Conclusions 170
134
4 Microbial Lipids
4 Sterols, Carotenoids, and Polyprenes 170 4.1 Sterols 170 4.2 Carotenoids 172 4.3 Polyprenoids 175 5 Wax Esters and Polyesters 176 5.1 Wax Esters 176 5.2 Polyesters - Poly-P-Hydroxyalkanoates 177 6 Other Lipids 180 6.1 Biosurfactants 180 6.2 Ether (Archaebacterial) Lipids 181 6.3 Phospholipids and Sphingolipids 183 6.4 Prostanoid-Type Lipids 184 7 Conclusions 185 8 References 186
1 Introduction
135
low- (or zero-) erucic acid (20:l) oil which is then a permitted oil for food manufacture. Overall production of plant and animal oils Since the publication of the 1st Edition of is increasing at about 3% per annum; producBiotechnology and the earlier chapter on tion in 1992193 was about 85-106t and is exthe biotechnology of lipids in 1986, a consid- pected to reach 105*106t by the year 2000 1992). Pricing of these materials reerable number of developments have taken (MIELKE, place in this field. Some microbial lipid prod- mains highly competitive as most products usucts have now been produced commercially ing oils can switch between the various types and prospects for other developments appear according to the price of the day. The average to be not too far away. In some cases, as e.g. price index for the major commodity oils is with the bacterial lipid poly-p-hydroxybuty- about US$ 500-550 per t though groundnut rate, no counterpart exists from plant or ani- oil, e.g., is always significantly higher than the mal sources and consequently the economics average at $800-850 per t. The highest priced of producing this product lie outside the nor- commodity oil, excluding the speciality matemal oils and fats domain. With most other mi- rials, is always olive oil at $ 1,500-2,000 per t. crobial lipids, these are the equivalent in com- Its price depends on its quality which includes position to plant-derived oils and consequent- minor, but very important, flavor compoly must compete against these in any potential nents. Animal fats (tallow and lard) have market place. Only the highest valued oils steadily declined in consumption over the have any chance of being produced by bio- past decade and are likely to fall even further technological means as it is impossible for mi- to about 20% of the total market by 2001 1994). Their prices are therefore croorganisms to produce oils and fats as (SHUKLA, cheaply as the main commodity oils are pro- usually at or below the average index level. The trends in world oil and fats supplies duced from plant and animal sources. However, there is always the possibility of producing are under constant surveillance and are frea microbial oil as an adjunct to some waste quently reviewed in various publications: the treatment process in a way similar to that of- extensive reviews by SHUKLA(1994) and ten used to produce microbial proteins (SCP MIELKE(1992) can be recommended though - single cell protein) for animal feed from for current information journals such as Lipid some unwanted substrate. Microbial oils - Technology (P. T. Barnes & Associates), Oils which could then be referred to as single cell and Fats International (Chase Webb, St Ives oils (SCO) - would have the double advan- PLC), INFORM (American Oil Chemists tage over SCP in that they could probably sell Society, Illinois) provide invaluable and confor a higher price than SCP and, moreover, tinuously up-dated information in most areas. could be used for a technical purpose should There are, in addition, a number of specialthe nature of the substrate prevent the prod- ized trade reviews that provide weekly prices of the traded oils. uct being returned into the food chain. The fatty acid composition of the major The major commercial plant oils continue to be dominated by soybean oil (current 1993 commercial oils is given in Tab. 1. The nomproduction is about 18*106t); palm oil, enclature of lipids is given in Sect. 1.1. It will though, continues to be the fastest growing be appreciated that, unlike say animal feed market with 14-106 t now being produced protein, the composition of the fats varies compared to 6.106 t in 1983. If the present considerably from species to species. In all rate of expansion in palm oil production con- cases, however, the oil or fat is composed altinues in Malaysia and Indonesia (BASIRON most entirely ( > 98%) of triacylglycerols and IBRAHIM, 1994; LEONARD, 1994), then (formerly known as triglycerides) - Sect. 1.1. palm oil will overtake soybean oil production For edible purposes, the oil or fat is retained by the end of this decade. Rapeseed oil (now in this form though the individual fatty acyl 9 -lo6 t in 1993) is also expanding mainly due groups on the glycerol can be modified to increased cultivation in Europe and Cana- usually by chemical means - without affecting da. The variety now under cultivation is the the triacylglycerol structure per se. Some en-
1 Introduction
Tab. 1. Fatty Acid Composition of Fats and Oils of Animal and Plant Origin
Relative Proportion of Fatty Acyl Groups [h(w/w)]
FatslOils
4:O-10:0 1 : 2O
1 0
-
1: 4O 11 3 2 18 1 6
8
1: 6O 27 24 2 6 9
-
1: 6l 2 4 3 1 2 1 9 14
3
1: 8O
1: 8l 29 43 44
18:2 2 3 1 0
18:3
-
20:O 1
2: 0l
Others
Animal Fats Butterfat Beef tallow
Lard
15:0+17:0, 3 % 15:0+17:0, 2%; 14:1+17:1,2 %
1 5 8
47 48
1 -
1 1 0
1 1
1 1
-
1 1 7 1 1
22:0+24:0,5%
2 6 1 3 4 11 7 11 11 22
-
3 3 5 3 2 2 5 4 2 3
11
-
Plant Oils Coconut oil Palm kernel oil Cocoa butter Olive oil Rapeseed oil Groundnut oil Sunflower oil Soybean oil Corn oil Cotton seed oil Exotic Plant Oils Borage seed oil
6 1 5 3 5 7 1 62 48 1 9 24 28 1 9 4 1 6
9
2 2 3 10 22 32 6 8 54 58 54 3 9
22b
10
-
45 .
8 6
22:1, 24:0,
25 .% 15 .%
Evening primrose seed oil Blackcurrant seed oil
2 1
70 48
9 b 17b
(Y-18~3,1 % 3
Also known as peanut oil. 83 2. yLinolenic acid, 1 : (6, 9,1)
I Introduction
137
zymatic reformulation of triacylglycerols occurs on an industrial scale using stereospecific lipases to transesterify palm oil fractions into the much more expensive cocoa butter-like triacylglycerols (OWUSU-ANSAH, 1993). With some technical applications of oils, it is the fatty acid that is required: consequently saponification (hydrolysis) of the triacylglycerol is carried out and the fatty acid used either as such, e.g., with soap manufacture, or is modified to an appropriate derivative which is then used in a multitude of products: from detergents to adhesives. The aim of all biotechnological processes is to produce products that are either cheaper than can be obtained from other sources, including possible chemical synthesis, or are not available by any other means. Within the field of lipids, the opportunities to produce triacylglycerol lipids are limited to the highest valued materials. The highest priced bulk (commodity) oil is cocoa butter whose price has varied between $ 8,000 to $ 3,000 per t over the past decade. At the higher price level, the prospects of producing a cocoa butter equivalent oil by yeast technology have looked favorable. This topic is specifically reviewed later (see Sect. 3.2.1). Other very high valued oils are those in the health care market and which have had various claims made on their behalf for the amelioration of various diseases and conditions. Of current interest are oils containing the polyunsaturated fatty acids: y-linolenic acid, 18:3 (0-6); arachidonic acid, 20:4 (w-6); eicosapentaenoic acid, 20: 5 (w-3); and docosahexaenoic acid, 22 :6 (w-3). Oils containing such fatty acids are found in a number of microorganisms and are reviewed in Sects. 3.3 and 3.4. The very highest priced lipids though are probably the prostanoid compounds encompassing the prostaglandins, leukotrienes, and thromboxanes. These are mainly used for treatment of uncommon disorders or for experimental purposes. Consequently, the amounts required per annum are probably at the kilogram stage rather than the ton (or kiloton) stage with other lipid products. Prospects for producing such materials are briefly mentioned in Sect. 6.4. Thus, if we view microorganisms as a po-
tential source of the widest types of lipids then it is possible to identify a number of potentially attractive products. For the purposes of this article, I have therefore used the broad definition of a lipid as any material that is derived from a (micro)organism, is directly soluble in organic solvents, and is essentially a water-insoluble material. However, as there is still considerable interest in the manner in which microorganisms synthesize large quantities of lipids, much of the review will be taken up with the more conventional types of oils and fats that they produce. The entire subject of microbial lipids, encompassing all aspects and not just biotechnology, has been the above subject of a two-volume monograph by RATLEDGE WILKINSON and (1988a, 1989). The industrial applications of microbial lipids have also been the subject of a monograph edited by KYLE and RATLEDGE (1992). Details concerning the degradation of fats, oils, and fatty acids, including the action of lipases and phospholipases, which are not covered here have been recently reviewed elsewhere by the author (RATLEDGE, 1993). It will be appreciated, of course, that although microorganisms remain a potential source of oils and fats, there is considerable effort being put into the production of oils and fats from conventional plant sources. Such efforts include the modification of peanut oils (groundnut oil) to produce changes in the fatty acid composition so that the more desirable oils can be produced more cheaply. The application of genetic engineering is now gathering pace as a means of producing tailor-made oils and fats in plants and is likely to supersede the traditional plant breeding approach as a means of creating what is wanted more quickly and with greater certainty. This review, however, will not include any detailed review of the current developments in plant genetic engineering as applied to the commodity oils and fats. Readers should though be aware that such advances are now likely to be a major influence in the availability of improved oils for everyday use and will undoubtedly ensure that these materials remain highly competitively priced for many years to come. The recent reviews by HARWOOD (1994a, b), MURPHY (1994a) and RATTRAY(1994) and the monographs
138
4 Microbial Lipids
Tab. 2. Developments in Genetic Engineering (by in vitro Mutagenesis and Gene Transfer) Being Applied for the Modification of Plant Oils (adapted from RAITRAY,1994)
Plant Target Soybean and rapeseed
Fatty Acid
16:O 16:O 18:O 18:l ~~-18:3
Objective increase decrease increase increase decrease increase increase decrease increase increase
Application margarine edible oil margarine; cocoa butter substitute(?) improved edible oil improved stability and odor erucic acid for oleochemicals olive oil substitute oleochemicals improved edible oil
Rapeseed Sunflower Linseed Groundnut
22: 1
18:l 18:2+ 18:3 18:3 18:l
edited by RATTRAY (1991) and by MURPHY (1994b) will be found particularly useful in this respect. Tab. 2 summarizes some of the current developments that are now taking place in this area. The opportunities for microorganisms to produce oils and fats of commercial value for the bulk markets remain doubtful but where the product cannot be obtained from elsewhere then this provides a much better opportunity for a microbial oil than attempting to replicate what is already available from plant sources. Some opportunities nevertheless do exist but they have to be identified with some care. Hopefully, some of the following material may indicate to the astute reader where such opportunities may lie.
1.1 Lipid Nomenclature and Major Lipid Types

Fatty acids are long chain aliphatic acids (alkanoic acids) varying in chain length from, normally, CI2 to CZ2though both longer and shorter chain-length acids are known. In most cells (microbial, plant, and animal), the predominant chain lengths are 16 and 18. Fatty
acids may be saturated or unsaturated with one or more double bonds which are usually in the cis (or 2) form. The structure of a fatty acid is represented by a simple notation system -X: Y , where X i s the total number of C atoms and Y is the number of double bonds. Thus, 18:0 is octadecanoic acid, that is stearic acid; 16:1 is hexadecenoic acid, that is palmitoleic acid, with one double bond, and 18:3 would represent octadecatrienoic acid, a C18 acid with three double bonds. The position of the double bond(s) is indicated by designating the number of the C atom, starting from the COOH terminus, from which the double bond starts: oleic acid is thus 18: 1 (9) signifying the bond is from the 9th (to the 10th) C atom. If it is necessary to specify the isomer, this is added as c (for cis=Z) or t for trans = E). In this review, the cisltrans system is used. Thus, cis, cis-linoleic acid is 18:2 (c 9, c 12). Most naturally-occurring unsaturated fatty acids are in the cis configuration and, unless it is stated otherwise, this configuration may be assumed. With polyunsaturated fatty acids (PUFA), the double bonds are normally methylene-interrupted: -CH= CH -CH2- CH= CH- . Thus, once the position of one bond is specified all the others are also indicated. In num-
2 Accumulation of Lipid
139
bering PUFAs, a reverse system is used where the position of only the last double bond is given by the number of carbon atoms it is from the CH3 terminus. To denote the counting back system is in operation the notation is given as 0-x or n-x; thus the two main isomers of linolenic acid (18:3) are given as 18:3 (0-3) and 18:3 (w-6) or as 18:3 (n-3) and 18:3 (n-6). In some systems, the minus sign may be omitted giving, for example, 18:3 (w3) or 18:3 (n3) and 18:3 (&) and 18:3 (n6). Respectively, these two isomers are:
in
17
16
With respect to the monoacylglycerols, there are obviously three possible isomers and similarly for the diacylglycerols. Where different acyl groups are attached to the glycerol moiety, these can then be individually given. For example, 1-stearoyl-Zoleoyl3-palmitoyl-sn-glycerol is the major triacylglycerol of cocoa butter with stearic, oleic, and palmitic acids on the three OH positions. Phospholipids possess two fatty acyl groups at the sn-1 and sn-2 positions of glycerol with a phospho group at sn-3 which is also linked
10
CH3-CH2-CH=CH-CH2-CH=CH-CH2-CH=CH
2 3
4
IS
14
13
12
I1
HOOC-H~C-H~C-H~C-H~C-HZC-H~C-H~C
S
CH3-CH2-CH2w w-1 w-2 n n-1 n-2
CH2- CH2-CH=CH-CHz-CH=CH 0-3 0 4 w-5 w-6 n-3 n-4 n-5 n-6
HOOC-H2C-HzC-H2C-H2C-HC=HC-H2C The w-x system will be used in this review. When fatty acids are esterified to glycerol, they give a series of esters: mono-, di-, and triacylglycerols (1, 11, 111). This is the preferred nomenclature to the older mono-, di-, and triglycerides.
CHp-0-CO-R
to a polar head group: choline, serine, ethanolamine, and inositol are the common ones. For the nomenclature and naming of phospholipids and other microbial lipids, the multiauthored treatise Microbial Lipids, edited by RATLEDGE WILKINSON and (1988a, 1989),
I CH-OH I
CHP-OH
I CH -0-COR I
CHZ-OH
CH2-O-COR
CHp-0-COR CH -0-COR CHZ-0-COR I11
I I
I1
where R is a long alkyl chain and RCO- is, therefore, the fatty acyl group. As various isomeric forms are possible, the position of attached acyl group must be specified in most cases. For this, the stereospecific numbering (sn-) system is used so that the two prochiral positions of glycerol (IV)can be distinguished as sn-1 and sn-3.
may be helpful though there are numerous text books on lipids that provide similar information.
2.1 Patterns of Accumulation
Not all microorganisms can be considered as abundant sources of oils and fats, though, like all living cells, microorganisms always contain lipids for the essential functioning of
CH20H
2
LHOH CH20H
IV
140
4 Microbial Lipids
membranes and membranous structures. Those microorganisms that do produce a high content of lipid may be termed oleaginous in parallel with the designation given to oilbearing plant seeds. Of the some 600 different yeast species, only 25 or so are able to accumulate more than 20% lipid; of the 60,OOO fungal species fewer than 50 accumulate more than 25% lipid (RATLEDGE, 1989a). The lipid which accumulates in oleaginous microorganisms is mainly triacylglycerol (see Sect. 1.1). If lipids other than this type are required then considerations other than those expressed here might have to be taken into account to optimize their production. With few exceptions, oleaginous microorganisms are eukaryotes and thus representative species include algae, yeasts, and molds. Bacteria do not usually accumulate significant amounts of triacylglycerol but many do accumulate waxes and polyesters (see Sects. 5.1 and 5.2) which are now of commercial interest. The process of lipid accumulation in yeasts and molds growing in batch culture was elucidated in the 1930s and 1940s (see WOODBINE, 1959, for a review of the early literature, and RATLEDGE, 1982, for an updated review of these aspects). A typical growth pattern is shown in Fig. 1. This pattern is also found with the accumulation of polyester material in bacteria (see Sect. 5 ) . The key to lipid accumulation lies in allowing the amount of nitrogen supplied to the
15
Fig. 2. Electron micrograph of Cryptococcus curvatus ( = Candida curvata =Apiotrichum curvatum) strain D grown for 2 days on nitrogen-limiting medium (viz. Fig. 1) showing presence of multiple lipid droplets. Total lipid content approx. 40%, market al., er bar: 1 km (from HOLDSWORTH 1988).
60
20 40 60 Culture time ( h l
80
Fig. 1 Typical lipid accumulation pattern for a . yeast (Rhodotorula glutinis = R. gracilis) growing on a high C:N ratio medium in batch culture. Biomass M, % lipid content 0, NH: in medium 0 (from YOONet al., 1982).
culture to become exhausted within about 2 4 4 8 h. Exhaustion of nutrients other than nitrogen can also lead to the onset of lipid accumulation (see GRANGER al., 1993, for a et recent reference) but, in practice, cell proliferation is most easily effected by using a limiting amount of N (usually NH4+ or urea) in the medium. The excess carbon which is available to the culture after N exhaustion continues to be assimilated by the cells and, by virtue of the oleaginous organism possessing the requisite enzymes (see below), is converted directly into lipid. The essential mechanism which operates is that the organism is unable to synthesize essential cell materials - protein, nucleic acids, etc. - because of nutrient deprivation and thus cannot continue to produce new cells. Because of the continued uptake of carbon and its conversion to lipid, the cells can then be seen to become engorged with lipid droplets (Fig. 2). It is important to appreciate, however, that the specific rate of lipid biosynthesis does not increase; the cells fatten because other processes slow down or cease altogether and, as lipid biosynthesis is not
141
linked to growth, this may continue unabated. The process of lipid accumulation (Fig. 1) can be seen as a two-phase batch system: the first phase consists of balanced growth with all nutrients being available; the subsequent fattening or lipogenic stage occurs after the exhaustion of a key nutrient other than carbon and, of course 02. role of O2 during The lipid formation was discussed briefly in the 1st Edition of Biotechnology (RATLEDGE, 1986). Accumulation of lipid has also been achieved in single stage continuous culture et (RATLEDGE al., 1984) and a typical accumulation profile dependent upon the dilution rate (growth rate) is shown in Fig. 3. As with batch cultivation, the medium has to be formulated with a high carbon-to-nitrogen ratio, usually about 50:l. The culture must be grown at a rate which is about 25-30% of the maximum. Under this condition, the concentration of nitrogen in the medium is virtually nil and the organism then has sufficient residence time within the chemostat to assimilate the excess carbon and convert it into lipid. The rate of lipid production (i.e., g L h-) is usually faster in continuous cultures than in batch ones (EVANSand RATLEDGE,1983; FLOETENMEYER al., 1985). et
7
6
-- 5
VI v)
ZL
g 3
60
. .
E
c C W
L
LO
20
2
1
0025
005
0075
01
Ollution rate (h-)
Fig. 3. Typical lipid accumulation pattern for a yeast (Rhodotorula glutinis) growing on nitrogen limiting medium in continuous culture. Biomass W, % lipid content 0 (from YOONand RHEE, 1983).
The exact ratio of C to N chosen for the medium was originally considered to be of little consequence provided N was the limiting nutrient and sufficient carbon remained to ensure good lipid accumulation. However, YKEMA al. (1986) showed that a range of et lipid yields in an oleaginous yeast, Apiotrichum curvatum (originally Candida curvata but now Cryptococcus curvatus; see BARNETT et al., 1990) were traversed in continuous culture by varying the C:N ratio of the growth medium. There was a hyperbolic relationship between the C:N ratio and the maximum growth (dilution) rate that the organism could attain: the lowest growth rate was at the highest C:N ratio of 50:l and this, in turn, controlled the amount of lipid produced and the efficiency of yield (g lipid per g glucose used) with which it was produced. Although the highest lipid contents of the cell (50% w/w) were obtained with a C:N ratio of 50:l or over, the optimum ratio for maximum productivity (g L- h- lipid) was at a ratio of 25: 1with glucose (YKEMA al., 1986) and at et 30-35 :1 when whey permeates were used with same yeast (YKEMA al., 1988). Similar et results for describing the optimum C:N ratio for lipid accumulation have been developed by GRANGER al. (1993) using Rhodotorula et glutinis. Interestingly, YKEMA et al. (1986) commented that Apiotrichum curvatum simultaneously accumulated about 20% carbohydrate in the cells along with the 50% lipid. Such a phenomenon of carbohydrate formation had been conjectured by BOULTON and RATLEDGE (1983a) to be a likely event to account for an observed delay in lipid synthesis after glucose assimilation had been initiated. This carbohydrate was also recognized independently by HOLDSWORTH al. (1988) in et the same yeast and was considered to be glyet cogen. As YKEMA al. (1986) pointed out, if the biosynthesis of the polysaccharide which, like lipid, is a reserve storage material, could be prevented then this would enhance the total amount of lipid producible with a cell. Although most studies on microbial lipid accumulation have been conducted using batch cultivation and, for accuracy, in continuous culture, other growth systems have also been explored. In particular, fed-batch cul-
142
4 Microbial Lipids
ture has proved effective in increasing both the cell density and lipid contents of oleaginous yeasts: YAMAUCHI al. (1983) used et ethanol as substrate with Lipomyces starkeyi and achieved a biomass density of 150 g L- with a lipid content of 54%. Similarly, PAN and RHEE(1986) achieved 185 g (dry wt.) of Rhodotorula glutinis per liter with a lipid content of 43% using glucose as the fed-batch substrate. In this latter case, Oz-enriched air (40% Oz + 60% air) had to be used to sustain the cells. At the density recorded, the packed cell volume was 75% of the total volume of the fermentation medium. Without using additional OZrit seems likely that cell densities of up to 1OOgL- could be achieved with most oleaginous yeasts (see, e.g., YKEMA al., 1988) though filamentous et molds may pose other problems. Economic considerations, however, would probably be against the use of OZ-enriched air for any commercial process. Interestingly, it is suggested that higher rates of lipid formation may occur with fed-batch techniques than with batch- or continuous-culture approaches (YKEMA al., 1988). et At the end of the lipid accumulation phase (see Fig. l), it is essential that the cells are promptly harvested and processed. If glucose, or other substrate, has become exhausted on the end of the fermentation, then the organism will begin to utilize the lipid as the role of the accumulated material is to act as a reserve store of carbon, energy, and possibly even and (1988) water. HOLDSWORTH RATLEDGE showed with a number of oleaginous yeasts that after carbon exhaustion following lipid accumulation, the lipid began to be utilized within 1.5 h thus indicating the dynamic state of storage lipids in these organisms.
2.2 Efficiency of Accumulation

The efficacy of conversion of substrate to lipid has been examined in some detail in both batch and continuous culture. In general, the latter technique offers the better means of attaining maximum conversions as the cells are operating under steady state conditions and carbon is not used with different efficiencies at each stage of the growth cycle.
Conversions of glucose and other carbohydrates including lactose and starch, to lipid up to 22% (w/w) have been recorded with a variety of yeasts (RATLEDGE,1982; YKEMAet al., 1988; DAVIESand HOLDSWORTH, 1992; HASSAN al., 1993) which compares favoraet bly with the theoretical maximum of about 1988). Somewhat lower 31-33% (RATLEDGE, yields appear to pertain with molds (WOODBINE, 1959; WEETE, 1980). The reason for this difference is not obvious though it may be due to a somewhat slower growth rate of molds than yeasts. It should be said, however, that there has not been the same amount of detailed work carried out with molds as with yeasts. Claims that microorganisms have achieved higher conversions of glucose or other sugars to lipid should be treated with caution: either there will be found to be additional carbon within the medium and not taken into the mass balance or, as may occasionally happen, the lipid has been improperly extracted and may contain non-lipid material. However, if experimental data are calculated so that the yield of lipid or fatty acids can be based on the fraction of glucose being used solely for lipid biosynthesis, then values close to the theoretical value have been attained in practice (GRANGER al., 1993). et When ethanol is used as substrate, the theoretical yield of lipid is 54% (w/w) (RATLEDGE, 1988). Though only a 21% conversion of ethanol to lipid was recorded by YAMAUCHI et al. (1983) in the fed-batch culture of Lipomyces starkeyi, higher conversions were and recorded by EROSHIN KRYLOVA (1983) also using a fed-batch system for the cultivation of yeasts on ethanol: conversions of 26%, 27%, and 31% were obtained using, respectively, Zygolipomyces lactosus (Lipomyces tetrasporus) and two strains of Cryptococcus albidus var. aerius. The reason for these very high values lies in the efficiency by which ethanol can be converted to acetyl- CoA, the starting substrate for lipid biosynthesis (see below). With glucose, the maximum yield of acetyl-CoA can only be 2 mol per mol utilized whereas with ethanol the yield is 1 mol per mol. On a weight-to-weight basis, therefore, ethanol (MW 46) is almost twice as efficient as glucose (MW 180) in providing Cz units. GRANGER al. (1993) have recorded direct et
143
2.3 Biochemistry of Accumulation

The pathway of synthesis of any microbial product, or indeed plant or animal product, needs to be understood so that it then becomes possible to manipulate the cell to enhance the formation of that product. The formation of lipids is not different from any other product in this general concept. The pathway of triacylglycerol formation from glucose is given in Fig. 4. The pathway accounts for lipid formation in oleaginous microorganisms and, with the accompanying regulatory control mechanisms, can explain how lipid accu-
conversions of ethanol to lipid (that is excluding ethanol being converted to non-lipid biomass) of 42% (wlw) with Rhodotorula glutinis. While most oleaginous microorganisms accumulate lipid equally well from a number of different carbon sources (see, e.g., EVANS and RATLEDGE,1983; YooN et al., 1982; DAVIES and HOLDSWORTH, 1992; HAMMOND et al., 1990), and do so without regard to the source of nitrogen used in the medium, a few yeasts are known which only accumulate lipid when an organic source of nitrogen, such as urea, glutamate, or aspartate, is used (WITTERet al., 1974; EVANS RATLEDGE, and 1984a, b). These yeasts appear to be mainly confined to strains of Trichosporon (Endomycopsis) pullulans and Rhodosporidium toruloides. A biochemical explanation for this has been advanced based on nitrogen catabolite inhibition of phosphofructokinase (see Fig. 4) which is a key enzyme controlling the rate of flux of glucose (as carbon substrate) to acetyl-CoA (EVANS and RATLEDGE, 1984~). The effect, though, is not explicable in terms of a greater pH decrease with inorganic NH4+ salts than with glutamate or asparagine, as has been suggested elsewhere (MORETON, 1988a). The same effect was produced in Rhodotorula gracilis using ammonium tartrate as was produced with NH4Cl where the former salt caused little downward pH drift but, without changing the biomass yield, increased the lipid content of the cells from 18% to over 50% (EVANS and RATLEDGE, 1984a, c).
mulation occurs in this group of organisms. In non-oleaginous organisms, one of the key enzymes, ATP-citrate lyase, does not occur and consequently the formation of acetyl-CoA occurs by another route (see SHERIDAN al., et 1989) and does not lead to a lipogenic state being created. In oleaginous microorganisms, the following sequence of events is considered to happen to cause lipid accumulation: (1) When the culture has consumed all available N from the medium, a nitrogen-scavenging process is initiated. This takes the form, but there are likely to be other examples, of deaminating AMP via the enzyme AMPdeaminase (see Reaction 1) which becomes activated at the point of N exhaustion (EVANS RATLEDGE, and 1985~). (1) As a result of the activity of AMP deaminase, some NH3 is provided for the cell to help maintain protein and nucleic acid synthesis but, simultaneously, the concentration of AMP drops rapidly (BOULTONand RATLEDGE, 1983a) and this is then the first major trigger in the lipogenic cascade mechanism. (2) AMP is required as an activator of the enzyme isocitrate dehydrogenase (Reaction 2) operating in the mitochondrion (EVANS et al., 1983). As a consequence of the absence of AMP the reaction is unable to proceed as part of the tricarboxylic acid cycle. Isocitrate NADP -,2-Oxoglutarate NADPH + CO2 (2) (3) With the cessation or slowing down of isocitrate dehydrogenase, isocitrate cannot be further metabolized and both isocitrate and citrate begin to accumulate. The equilibrium lies in favor of citrate so as the assimilation of glucose continues unabated by the N-limited cells, (BOTHAMand RATLEDGE,1979), citrate becomes a major product of its metabolism (BOULTON and RATLEDGE, 1983a). (4) Citrate now exits from the mitochondrion in a malate-mediated citrate translocase reaction (see Fig. 4). Citrate is then cleaved by ATP-citrate lyase (Reaction 3), an enzyme which appears to be uniquely associated with the lipogenic process. The dependency of isocitrate dehydrogenase (Reaction 2) upon AMP + IMP
+ NH3
144
4 Microbial Lipids
Fig. 4. Pathway of biosynthesis of triacylglycerols from glucose in oleaginous microorganisms. Numbers in parentheses indicate approximate stoichiometry for conversion of glucose to fatty acyl-CoA as elucidated in oleaginous yeasts (see also RATLEDGE, 1988 RATLEDGE EVANS,1989). and G3P Glycerol 3-phosphate (this is derived from 0.5 mol glucose during glycolysis)
145
AMP is; though, another metabolic feature, possibly peculiar to the oleaginous organisms (EVANS and RATLEDGE, 1985b, 1986). (3) (5) The acetyl-CoA from Reaction 3 serves as the primer for fatty acid synthesis. However, in addition to a supply of C2 units, the cell must also provide NADPH as reductant for fatty acid synthesis. This is provided from the subsequent metabolism of oxaloacetate, first to malate via malate dehydrogenase, and then to pyruvate via malic enzyme (Reaction 4): Malate NADP+ Pyruvate C 0 2 (4) NADPH Some NADPH may also be supplied by metabolism of glucose via the pentose phosphate pathway. The malic acid for the latter reaction is presumed to be by it leaving the mitochondrion in exchange for pyruvate in a series of coupled transport reactions across the mitochondrial membrane (see Fig. 4). The overall flux of glucose to fatty acylCoA and then into triacylglycerol is given in Fig. 4. The overall stoichiometry is approximately:
+
Citrate + ATP + CoA AcetylCoA Oxaloacetate ADP Pi
15 Glucose
Triacylglycerol
+ 36C02
Therefore, if glucose is not used for the synthesis of any other product the yield of lipid is approximately 32 g per 100 g glucose. The role of ATP-citrate lyase in lipid accumulation appears to be central. The enzyme itself has been purified and partially characet terized from Rhodotorula gracilis (SHASHI al., 1990). It has a M,of approx. 520 kDa and comprises four identical subunits each about 120-130 kDa in size. Other properties of the enzyme to establish its involvement in lipid accumulation have been presented by BoTHAM and RATLEDGE (1979), BOULTON and and RATLEDGE (l981,1983b), and by EVANS RATLEDGE (1985a, c). The enzyme appears similar in overall size and structure to mammalian ATP-citrate lyase (HOUSTON and NIMMO, 1984, 1985). Microorganisms lacking ATP-citrate lyase generally do not accumulate lipid above 10-15%. However, it is important to state that the corollary, that microorganisms with ATP-citrate lyase will accumulate lipid, does not necessarily follow because other enzymes that function at key regulatory points such as AMP deaminase (Reaction l), the AMP-dependent isocitrate dehydrogenase (Reaction 2) and malic enzyme (Reaction 3) may be lacking or be under alternative control mechanisms. Malic enzyme (ME) itself is absent in some
I Enzymes: PC: Pyruvate carboxylase (Pyruvate + C 0 2+ ATP + Oxaloacetate + ADP + Pi F'yruvate dehydrogenase (Pyruvate + CoA + NAD + -+ Acetyl-CoA+ C 0 2+ NADH) PD: cs: Citrate synthase (Acetyl-CoA + Oxaloacetate + Citrate + CoA) ACL: ATP:citrate lyase MDH: Malate dehydrogenase ME: Malic enzyme A CC: Acetyl-CoA carboxylase FAS: Fatty acid synthase PIMEX: Pyruvate/malate exchange (triple-linked system) ICD H: Isocitrate dehydrogenase (NAD - requiring, AMP-dependent; see text Reaction 5) CT: Citrate translocase TC: Tricarboxylic acid cycle reactions 7. .. unknown (unspecified) reactions Overall conversion (assuming the 9 mol of NADH required for the MDH reaction will be either from glycolysis or the PD reaction): 4 5 Glucose + CoA + 9 NAD + + 7 NADPH + 17 ATP C18-fattyacyl-CoA 9 C 0 2+9NADH . +7 NADP++ 17 ADP+17 P,) The production of 1 mol triacylglycerol, therefore, requires approx. 5 mol glucose as an additional 0.5 mol glucose is needed to provide glycerol 3-phosphate (G3P) for the triacylglycerol and a further mol of glucose must be oxidized to provide additional energy (ATP) and reducing power (NADPH).
+
146
4 Microbial Lipids
oleaginous yeasts, notably Lipomyces spp. (BOULTON,1982), and here the supply of NADPH is presumably by reactions of the pentose phosphate cycle. ME activity has also been found in oleaginous fungi and has also been implicated as the provider of NADPH for the desaturation of fatty acids (KENDRICK 1992c) in Mucor circinelloand RATLEDGE, ides. Here, a second form of ME is associated with the membranes of endoplasmic reticulum - the microsomal fraction of the cell that then serves to provide NADPH directly within the membranes. The transfer of reducing equivalents from the NADPH to the 02dependent desaturation reaction (Reaction 5 ) may not be linked via cytochrome bs which is normally associated with such reactions in other tissues (see Reaction 5).
fungi; CARMAN HENRY and (1989) which covers phospholipid biosynthesis in yeast; PIERINGER (1989) covering the biosynthesis of non-terpenoid lipids from fatty acids; and (1989) reviewCOOLBEAR and THRELFALL ing the biosynthesis of terpenoid lipids.
3 Triacylglycerols and Fatty Acids
Fatty acids do not occur as such in living cells because of their inherent toxicity. Consequently, fatty acids are esterified, usually as triacylglyerols (see Sect. 1.1, Structure 111), or occasionally as wax esters (see Sect. 5 ) . Triacylglycerols (TAG) are produced as the ma- CH, - CH,n . n e p - m & desaturase Fatty acid jor storage product of most oleaginous yeasts Malate NADP+and molds. Their proportion of the total lipid - CH = CH is usually over 80% (see RATLEDGE, 1986) 2 H,O and can be over 90%. Clearly, the proportion (5) of TAG will depend upon a number of factors So far there have been no other reports of which would include the total amount of lipid the occurrence of malic enzyme in microso- in the oleaginous cell (the more accumulated lipid, the greater the amount of TAG) and ma1 membranes in other systems. In non-oleaginous microorganisms, where also the method used for lipid extraction. ATP-citrate lyase does not occur, acetyl-CoA More severe methods of extraction will refor fatty acid biosynthesis is generated from move both free and bound lipids whereas the intramitochondrial pool by transfer of the gentler methods will extract only the freely acetyl group via carnitine acetyl transferase soluble lipid which will be predominantly (CAT). Oleaginous microorganisms also pos- TAGS. Thus, comparisons of lipid analysis, sess this enzyme activity (RATLEDGEand carried out by different researchers using difGILBERT, 1985; HOLDSWORTH al., 1988) ferent methods with different organisms, et though it is not immediately obvious why two must be treated with some caution. however, routes to acetyl-CoA generation are neces- from a biotechnological viewpoint, the TAG sary though it does emphasize that without content of the cell is usually of paramount imATP-citrate lyase cells cannot presumably portance as this is the form in which an oil provide sufficient acetyl-CoA under N-limit- will eventually be offered for sale. Should the ing growth conditions to keep lipid biosynthe- TAG content of a cell be low, but the fatty sis fully primed. Non-oleaginous microorgan- acids of interest, then total extraction of all isms, therefore, tend to accumulate carbohy- fatty acyl lipids will have to be carried out drate reserves when placed under the same and the fatty acids then recovered after hydrolysis. In these cases, the fatty acyl profile growth conditions. The mechanism of fatty acid biosynthesis of the total lipid becomes the major determiitself from acetyl-CoA is well documented in nant of the potential usefulness of lipid. The most standard text books. Recent reviews fatty acids can be esterified (usually methyl or that may be consulted on this topic include ethyl derivatives) and then offered for sale. Lipid extraction is fraught with a number those by SCHWEIZER (1989) which covers fatty acid biosynthesis in bacteria, yeasts, and of difficulties and pitfalls for the unwary. The
147
article by RATLEDGE and WILKINSON (DHA). EPA has now been identified in the (1988b) discussed these problems at some lipids (presumably phospholipids) of a numlength and suggested methods that could be ber of species of marine bacteria: Alteromons used to minimize post-harvest changes in the a , Shewanella, Flexibacter, and Vibrio (RINet lipid composition of microbial cells. The pres- GO et al., 1992; AKIMOTO al., 1990, 1991; et ence of partial acylglycerols (mono- and di- YAZAWAet al., 1988, 1992; HENDERSON acylglycerols) dong with free fatty acids in al., 1993), though it was first recognized in the the extracts is indicative of faulty extraction lipids of Flexibacter polymorphus some years and PERRY(1977). YAZAWA procedures that have failed to subdue the la- ago by JOHNS tent activity of lipases and phospholipases of et al. (1992) carried out an extensive survey of the cells. These enzymes function successfully some 24,000 bacteria isolated from various in the very solvent systems that are used for fish and marine mammals. One isolate, which lipid extraction and require rapid inactivation approximated taxonomically to Shewanella (usually by heating) to ensure that the lipid putrefaciens, was found that produced EPA at remains unchanged. Fractionation of ex- up to 40% of the total fatty acids; the lipid tracted lipids that show the present of signifi- content of the bacterium was between 10 to cant amounts of these hydrolysis products, 15% and the organism grew readily in the laand especially free fatty acids, are therefore boratory achieving 15 g (dry wt.) L - ' in 12seldom worth reporting and, as there are so 18 h. The content of EPA in dry cells was apfew good examples with oleaginous yeasts prox. 2%. Total hydrolysis of the extracted and molds (see RATLEDGE, 1986), the erudite lipids was needed to release the EPA from reader is therefore referred to the more ex- the phospholipid fraction. EPA was found altensive reviews of RATTRAY(1988) and LO- most exclusively at the sn-2 position of phosSEL (1988) who detail the lipid analyses of a phatidylethanolamine and phosphatidylglycelarge number of yeasts and molds. This pres- rol. It was the only PUFA that was recogent chapter, however, is concerned solely with nized in the total lipids; other unsaturated fatthe biotechnological potential of microorgan- ty acids were 16:l (12%), 18:l (oleic+cisisms and consequently it is only those species vaccenic) (2%) and 17:l (6%). HENDERSON that are prolific in the production of desirable et al. (1993), using a marine Vibrio isolate, oils or fatty acids that will be considered showed that EPA formation was higher (at here. 9% of the total fatty acids) when the cells were grown at 5C than 20"C, whereas YAZAWA et al. (1992) had only used 10-15C for 3.1 Bacteria the growth of their bacterium. The gene(s) responsible for EPA production have been Although bacteria do not produce triacyl- transferred into E. coli with the resultant proglycerols and are usually considered as duction of EPA in this bacterium (WATAsources of novel lipids (see Sect. 5), they nev- NABE and YAZAWA, 1992). ertheless are of current interest in that some DHA was first recognized in several maspecies are known that produce polyunsatu- rine bacteria by DELONG and YAYANOS rated fatty acids (PUFA) of dietary or even (1986). Like EPA, DHA was exclusively assopharmaceutical importance. ciated with the phospholipids in all cases. However, DHA was predominant in bacteria grown at 2C and at very high pressures (- lo5 Pa) making them unlikely candidates 3.1.1 Polyunsaturated Fatty Acids for biotechnological exploitation. More rein Bacteria cently, YANOet al. (1994) have reported similar findings with a further five deep-sea bacSeveral bacterial species have been recently terial isolates. Again there was a strong deisolated from marine sources that synthesize pendency on high pressures and low temperaeither eicosapentaenoic acid (20: 5 0-3) tures for DHA formation though, for the first (EPA) and docosahexaenoic acid (22:6 0-3) time, both DHA and EPA, together with
148
4 Microbial Lipids
traces of 20:4 (0-3) and 20:3 (w3), were ly, the opportunities for exploiting yeast oils as a commercial possibility are limited and noted. The commercial values of EPA and DHA other more expensive targets have had to be as individual fatty acids are unclear as they identified. One major target that has been are both readily obtainable as a mixture in identified is to produce a yeast oil as a cocoa fish oils which are relatively cheap. It is clear, butter equivalent (CBE). The fatty acid comhowever, that the bacterium described above position of cocoa butter is given in Tab. 1. by YAZAWA al. (1992) - Shewunellu putreet fuciens - could represent a valuable source of EPA but the commercial potential for EPA 3.2.1 Production of a Cocoa and DHA would both be considerably enhanced if it should be shown that single fatty Butter Equivalent Yeast Fat acids were required for medical purposes. CBE fats are used extensively in the conWATANABE al. (1994) have recently shown et that isolated DHA (and by inference EPA fectionery business: in the manufacture of would act similarly) can be readily incorpo- cooking chocolate and chocolate-type materirated into bacterial phospholipids, including als and it can also be included, to an agreed those of E. coli and also Rhodopseudomonus percentage - usually 5% - of cocoa butter itcupsulutu which is currently used for the pro- self, in chocolate manufacture in several duction of larvae of marine fish. In this way countries including UK, Ireland, and Den1992). CBE materials have DHA could be delivered to the developing mark (KERNON, larvae and potentiate their growth rate. DHA to have similar physical characteristics to cothough at present must be obtained from fish coa butter itself and currently they are manuoils though mold and algae sources of it are factured from palm oil by fractional crystalliknown (see Sects. 3.3.5 and 3.4.4). If recombi- zation though they can also be produced by nant DNA technology could be used to in- enzymic transesterification of palm oil and et crease the levels of DHA and EPA in more stearic acid or its esters (WILLNER al., 1993). easily cultivatable bacteria (see WATANABE 1993; OWUSU-ANSAH, The requirements for a satisfactory CBE is and YAZAWA, 1992), or even yeasts, this could open up new and exciting horizons for that there should be approximately equal these PUFAs. At the moment, however, EPA amounts of stearic acid (18:0), oleic acid production by bacteria is someway off and (18: l), and palmitic acid (16:O) attached to that of DHA would appear almost impossi- the glycerol moiety with the unsaturated acid ble. Other sources of EPA, however, include being in the sn-2 position. This is known as both molds and algae (see Sects. 3.3 and the POSt-TAG (1-palmitoyl-2-oleoyl-3-stearoyl-sn-glycerol). The price of cocoa butter it3.4). self has varied from $8600 per t in the mid 1980s (MORETON, 1988b;SMIT al., 1992) to et 3.2 Yeasts about $3OOO per t in 1994. As CBEs command a price of about 80% of cocoa butter, it The number of oleaginous yeasts (Tab. 3) is can be appreciated that at the higher price relatively small in comparison to the total level, a yeast CBE would be an attractive ecoet number of species - 590 (see BARNETT al., nomical proposition. Not surprisingly, a num1990). Besides producing triacylglycerols, ber of developments took place in the mid these yeasts also usually produce between 2- 1980s using appropriate yeast technology with 10% of their total lipid as phospholipids with this objective in mind. Simultaneously, the atsmaller amounts of sterol and sterol esters tractive price of cocoa butter and of CBEs (RATLEDGE, 1986). The fatty acid profile of stimulated the current interest in lipase-catathe yeasts lipids (Tab. 3) is typically that of lyzed transformation of oils and fats. Howevseveral commercial plant oils (see Tab. 1) er, with the present low price of commercial and, as such, would not command a high cocoa butter and CBEs, the yeast route has price: say, maximally $600 per t. Consequent- been deemed to be non-profitable and even
3 Triacylglycerols and Fatty Acids Tab. 3. Lipid Contents and Fatty Acid Profiles of Oleaginous Yeastsa Yeast Speciesb Maximum Lipid Content ["A (w/w)] 37 42 65 65 58 51 32 28 50 28 64 63 67 66 72 36 28 45 20 65 36 Major Fatty Acyl Residues [Relative % (w/w)] 16:O 16:l 3 5 1 trace 18:O 5 8 3 3 15 12 8 1 18:l
~ ~
149
Others 18:2
~~~
18:3 5
1
Candida diddensiae Candida sp. 107 Cryptococcus albidus var. aerius C. albidus var. albidus C. curvatus D' C. curvatus R' C. laurentii Endomyces (Endomycopsis) magnusiid Galactomyces geotrichume Williopsis saturnusf Waltomyces lipoferg Lipomyces starkeyi L. tetrasporus (Zygolipomyces lactosus) Rhodosporidium toruloides Rhodotorula glutinis R. graminis R. mucilaginosa Trichosporon .beigeliih T.fermentans' T. pullulans' Yarrowia lipolyticak
a
44
12 16 32 31 25 17
19
45 31 73 56 51 49 36
17 9 12
18:4 (1%)
8 6 17 25
16 3 3 6
-
44
1 19
no record' 16 16 37 4 34 6 31 4 18 37 30 12 17 15 11 3 1 2
7 5 15 3 3 12 22 4 2 1
45 48 51 43 66
1
-
1 6
47 8 36 15 no record 50 12 42 34 57 24 28 51
23:O (3%) 24:O (6%)
trace 1 1
A lower lipd content of cells for inclusion in the list has been taken as 20%. Data taken from RATLEDGE
and EVANS(1989) and RATLEDGE(1989b) except for Cryptococcus albidus var. aerius where original et al. (1989) - is used. Not included in the data of ZVYAGINSTEVA (1975) - see RATLEDGEand EVANS table and of uncertain oleaginicity are: Candida (Pichia) guilliermondii (22% lipid), C. methylica (20%), C. stellatoidea (C. albicans) (20%), C. tropicalis (23%), Cryptococcus (Filobasidiella) neoformis (22%), Hansenula (Pichia)ciferri (22%), Lipomyces spp. (two unspecified species: 59 and 67% lipid), Schwanniomyces occidentalis (23%), and Trigonopsis variabilis (an unusual yeast which only accumulates lipid - up to 40% - if grown with methionine in the medium). Nomenclature given accordingly to BARNETT al. (1990). et Although Candida curvata strains D and R are often referred to as Apiotrichum curvatum, this name is not recognized as such. BARNETTet al. (1990) consider C. curvata to be a synonym of Cryptococcus curvatus which is now the recommended name. Endomyces magnusii. This name is of dubious status and may be the imperfect stage of Trichosporon pullulans (q.v.)according to KREGER-VAN (1984). It is not listed by BARNETTet al. (1990). RIJ ' Formerly Geotrichum candidum. Although this organism was quoted as containing up to 50% lipid in the 1930s (see HESSE, 1949; WOODBINE, 1959) there have been no recent reports of such levels being repeated. The original oleaginous strains may, therefore, be lost. Formerly Hansenula saturnus. g Formerly Lipomyces lipofer. h Formerly Trichosporon cutaneum. ' Formerly Trichosporon fermentans. 1 Trichosporon pullulans possibly includes Endomycopsis vernalis (see KREGER-VAN RIJ, 1984) though yeast correspond to this latter description appears to be no longer available. Yarrowia lipolytica is variously referred to as Candida lipolytica and Saccharomycopsis lipolytica amongst other names. Only a few strains may be oleaginous.
150
4 Microbial Lipids
the lipase route must also be under severe financial pressure. Nevertheless, the approach used to achieving a yeast CBE is illustrative of what can be achieved given limited resources and manpower in trying to produce a marketable biotechnological bulk product. Four different approaches have been tried to produce a satisfactory yeast CBE. The main difficulty to be overcome has been how to increase the inherently low content of stearic acid (maximally from 10 to 12%) in oleaginous yeasts (see Tab. 3) up to the required 30% stearate or more.
3.2.1.1 Direct Feeding of Stearic Acid

The simplest way to increase the stearic acid content of a yeast oil is to feed stearic acid or its ester to the yeast. The fatty acid or its ester is then taken up by the yeast and, although some may be degraded, the bulk seems to be directly esterified into the storage triacylglycerols in the cell. Typical results of such efforts are shown in Tab. 4. Principal commercial concerns that attempted this route were Fuji Oil Co. Ltd. (1979, 1981) and CPC International Inc. (1979, 1982a, b). Best results were obtained by feeding both palmitic acid and stearic acid, as their methyl esters, to Torufopsis ATC C 20507 (Fuji Oil Co. Ltd., 1979, 1981). However, even this oil required fractionation to produce a satisfactory CBE and this then significantly added to the costs. The obvious problem with this approach is the cost of the stearic acid or the stearate palmitate mixture. Stearic acid itself is usually produced by chemical hydrogenation of oleic acid, which in turn is usually derived most cheaply from animal fats. Unfortunately, this origin of stearic acid then negates any claim that may be made for the yeast CBE being wholly derived from non-animal sources and would make it unacceptable for vegetarians and some religious groups. This situation has now, however, changed with the advent of high-oleic acid sunflower oil (see Tab. 2). For the first time, it is therefore possible to produce stearic acid from oleic acid at a relative-
0"0"?!
22222

DS-9 DS-12 b 181 (9) 182(9,12) 180 stearic oleic acid linoleic acid acid 18:2 (6,9) octadecadienoic DS-15
151
im4 4
DS-5
183(9,12,15) a -1inolenic acid DS-6
DS-6 183 (6,9,12) Y-linolenic acid
184 (6,9,12,15) octadecatetraenoic
202 (8,ll) eicosadienoic acid
203 (8,11,14) dihomo-y-linolenic acid (DHGLA)
204 (8,11,14,17) eicosatetraenoic acid
203 (5,8,11) eicosatrienoic acid (ETA) (Mead Acid)
DS-17 DS-5 204 (5,8,11,14) 205 (5,8,11,14,17) arachidonic eicosapentaenoic acid acid (AM) (El
Fig. 5. Formation of polyunsaturated fatty acids in fungi. DS-n: fatty acyl desaturase acting at nth C atom of fatty acid; EL: elongase using acetyl-CoA. The DS-17 is only found in certain filamentous fungi; it is not found in animals. It could though conceivably also convert DHGLA (dihomo- ylinolenic acid) to eicosatetraenoic acid.
225 (7,10,13,16,19) docosapentaenoic acid (DPA)
226 (4,7,10,13,16,19) docosahexaeneoic acid (DHA)
10- 3 series
EL
0-9 series
0-6 series
ly low cost and still being classed as a plant oil. In spite of the development of a cheap, plant source of stearic acid, this approach to a yeast CBE appears to have been abandoned though it did establish that yeast oils could be produced with an unprecedented high content of saturated fatty acids. The remaining three approaches to produce a yeast CBE have all sought to limit the conversion of stearic acid to oleic acid within the yeast cell. This reaction (see Reaction 5, Sect. 2.3), functions at the level of the coenzyme A derivatives of the fatty acids and requires molecular O2 as well as a supply of reducing equivalents (NADPH). It is catalyzed by A 9-stearoyl desaturase. The subsequent desaturations (see Fig. 5) are carried out with
fatty acyl group being detached to a phospholipid (KENDRICK and RATLEDGE 1992c, d)
3.2.1.2 Inhibition of Stearoyl Desaturase

The naturally occurring sterculic acid, cis9,10-methyleneoctadecenoic acid (V), which is found in the seed oil of sterculia and kapok plants is an effective inhibitor of the A 9 desaturase.
CH3-(CH2)7-C=C-(CH2)7COOH
\ /
Sterculic acid
CH2
V
152
4 Microbial Lipids
Tab. 5. Effect of A9- and Al2-Cyclopropene Fatty Acids on the Fatty Acyl Composition of Rhodosporidium toruloides IF0 0559 (from MORETON, 1988b) Relative Fatty Acyl Composition ["h (w/w)] 14:O 16:O 16:l 18:O 18:l 18:2 18:3 Control (no additions) Sterculia oil"0.3 mL L-' A12-Cyclopropeneb0.4 mL L-' Sterculia oil" + 0.3 mL L-' A12-Cyclopropeneb0.3 mL L-' Al2-Cyclopropene c18:1 1.7 0.7 1.0 27.8 0.3 14.9 36.0 0.3 19.5 7.3 48.3 8.3 46.9 40.0 17.5 4.6 18.5 9.1 3.7 44.7 1.8 21.8 4.7 1.5
" Contains 50% (w/w) the A9-cyclopropene c 81 of 1:
MORETON(1985) successfully demonstrated that as little as 100 mg sterculic acid per L of growth medium was effective inhibiting the reaction in a number of yeasts: Cundidu sp. 107, Rhodosporidium toruloides, and Trichosporon cutuneum. Stearic acid then accumulated up to 48% of the total fatty acids (see Tab.'5). However, the effect of the inhibitor was extremely specific and did not affect the subsequent conversion of oleic acid to linoleic acid (18:2) via the action of the A 12 desaturase enzyme. Consequently, the proportion of 18:2 in the yeast oil was unaffected by sterculic acid, and this was detrimental to the properties required of the CBE lipid. MORETON (1988b) subsequently showed that when the cis-A 12 analog of sterculic acid which had to be chemically synthesized, was added to the yeasts this now had the desired effect of decreasing the linoleic acid content (see Tab. 5 ) . In the presence of both sterculic acid and cis-12,13-methyleneoctadecenoic acid, the oil of R. toruloides (the best of the yeasts examined) was now almost exactly as required: the three principal fatty acids, palmitic, oleic, and stearic, were present at a ratio of and CLODE,1985). 1:1:2 (MORETON Although this approach of MORETON (1985, 1988b) was scientifically very successful in meeting its objectives, the costs of the cyclopropene inhibitors were to prove beyond what the economics of the process could accommodate. Further, the acceptability of using known metabolic inhibitors in a
biotechnology process designed to produce an edible oil was very uncertain. Clearly, regulatory authorities would be extremely cautious in allowing a yeast CBE to be used in foods that had some possibility of containing any residual inhibitor. This approach, which had been pioneered by Cadbury-Schweppes plc, the large UK-based chocolate manufacturer, was abandoned in 1986.
3.2.1.3 Mutation
Mutation and genetic manipulation of bacteria are now commonplace; haploid yeasts and molds are also similarly mutatable though, of course, many yeasts are diploid or aneuploid and would thus be not amenable to simple mutational strategies. Nevertheless and without ever apparently assessing whether their chosen yeast was haploid, aneuploid, diploid, or polypoid SMITet al. at the Free University of Amsterdam embarked upon an ambitious project to delete the stearoyl desaturase from the yeast Cryptococcus curvutus (YKEMA al., 1989, 1990; VERWOERT al., et et 1989; SMITet al., 1992). (This yeast was originally isolated by HAMMOND al. from dairy et wastes as a yeast that would readily grow on lactose, the principal carbohydrate of whey, with whey being judged to be a potential cheap and plentiful substrate (MOON et al., 1978). Initially, the yeast was named as Cundidu curvucu and then reclassified as Apiotri-
153
chum curvatum which is now recognized as Cryptococcus curvatus (BARNETT et al.,
1990).) In the work of SMITet al., C. curvata was treated with chemical mutagens and a number of auxotrophic mutants isolated that required oleic acid for growth. Such mutants would thus be unable to produce unsaturated fatty acyl groups for incorporation into their phospholipid membrane structures and would be unable to maintain membrane fluidity and growth. It was presumed, but never shown by direct enzyme assay, that these mutants would have been affected in the gene coding for the A 9 desaturase. One of these mutants, Ufa 33, now contained 50% stearic acid (see Tab. 6). However, as the mutant required the addition of oleic acid to the medium, it was obvious from the fatty acid analysis that the 412- and Al5-desaturases (forming linoleic and then linolenic acids, respectively) were unaffected in this mutant. Although small amounts of the 18:2 and 18:3 fatty acids can be tolerated in CBE preparations their amounts need to be very low in order to
maintain the sharp melting point transition of a CBE. To decrease the content of 18:2 and 18:3 in the yeast oil, a further mutation program would have been required but this was never carried out. From the initial oleate-auxotroph, Ufa 33, a number of partial revertants and hybrids were subsequently produced (YKEMA al., 1990 et VERWOERT al., 1989) that no longer reet quired oleic acid to be added to the growth medium: in other words the A9-desaturase was now partially functional allowing a small amount of oleic acid to be synthesized inside the cell. The fatty acid profile of these yeasts (Tab. 6) all contained substantial amounts of stearic acid and, in some cases, notably that of the hybrid F33.10, with excellent similarities to cocoa butter itself. Similar mutational programs have been reported by BEAVAN al. (1992), working for et Diversified Research Laboratories as a subsidiary company of G. Weston Ltd., Canada, and by HASSAN al. (1993, 1994a) working et in the group of GERARDGOMA,Toulouse, France. BEAVAN al. (1992) isolated 6,725 et
Tab. 6. Cocoa Butter Fatty Acids and the Fatty Acyl Compsition of Triacylglycerols from Cryptococcus curvatus wild type (WT) strain, two unsaturated fatty acid auxotrophic mutants (Ufa 33 and Ufa M3), a revertant mutant (R22.72), and a hybrid (F33.10), both derived from Ufa 33 compared to the best results obtained with the wild-type strain grown with limited O2 supply do diminish desaturase activity Major Fatty Acyl Groups [Relative % (w/w)]
16:O 18:O 32-37 12 50 37 43 31 24 18:l 30-37 18:2 2-4 18:3
-
24:O
Cocoa butter Yeast WT Ufa 33" Ufa M3b R22.72' F33.10 WT-NZ'
a
23-30 17 20 26 16 24 18
trace
1 4 2 2 4 2
55 6 22 27 30 48
8 11 8 7 6 3
2 4 4 1 1
' '
Grown with 0.2 g L - ' oleic acid; from YKEMAet al. (1990). et Grown with 0.2 g L-' oleic acid; from HASSAN al. (1993, 1994a). Grown without oleic acid; from YKEMAet al. (1990). Grown without oleic acid; from VERWOERT al. (1989). et Wild type grown on whey lactose in a 500 L bubble column fermenter with a restricted O2 supply; from DAVIESet al. (1990).
154
4 Microbial Lipids
yeasts (which would include many identical organisms) of which one organism, DRLD221, was taken as the best with respect to the rate of lactose utilization, lipid production, yield, and composition. The organism was tentatively identified as Trichosporon cutuneum which was originally suggested as an identification of C. curvutu. A mutant of this yeast, DRL-JF34, contained 19% 16:0, 34% et 18:0, and 23% 18:1. The work of HASSAN al. (1993,1994a) used the same yeast as SMIT et al.; their results are also shown in Tab. 6. Although all three groups engaged on this work produced greatly increased proportions of stearic acid in the yeast oils, without diminution of the lipid content of the cells and without significant decrease in the overall growth performance of the yeasts, none of the mutants have yet been used in commercial trials to produce CBE oils. Large-scale trials with mutants are, however, essential as the characteristics of these organisms are often not fully revealed until they are grown in large fermenters. Thus, the use of mutants to produce yeast oil CBEs remains a potential, rather than a proven, approach to this goal.
position (see Tab. 6). Significantly and interestingly, this approach by DAVIES et al. (1990) also served to decrease the contents of the polyunsaturated fatty acids (18:2 and 18:3) which, of course, also require O2 in their formation. Thus by a single and obvious metabolic manipulation, the goal of a highstearate yeast oil was almost perfectly achieved.
3.2.1.5 Conclusions
The pursuit of yeast oil as a CBE has been now ceased after a decade of intensive activity. DAVIES (1984) was the first to appreciate the potential of this as a commercial target and to carry out a sustained program to this end. Of key significance was the appreciation that a cheap substrate would be essential for success as, from the previous sections (Sect. 2.2), approximately 5 t of glucose or equivalent carbohydrate are needed to produce 1 t of oil. DAVIES,by working in New Zealand, quickly realized that the waste whey generated from the extensive dairy industry of that country could represent such a source of fermentable carbohydrate. The earlier discovery 3.2.1.4 Metabolic Manipulation of the yeast C. curvutu (MOONet al., 1978) that could be readily grown on lactose, which The final approach to increasing the stearic is the major carbohydrate of whey, quickly inacid content of yeast oils has been to control dicated that it was probably the most likely et the amount of O2 entering the fermentation. one to use in this process. DAVIES al. (see As mentioned above (see Reaction 5, Sect. DAVIES, 1988, 1992a, b, c; DAVIES and 1992) then developed a pro2.3), O2 is an essential co-reactant in the stea- HOLDSWORTH, royl desaturase reaction. Accordingly, DAV- cess up to a pilot-plant level of 200 m3 using a IES et al. (1990) carried out a series of ferbubble column fermenter with Cundidu curmentation runs with C. curvufu in which the vatu and casein (milk) whey as its substrate. oxygen uptake rate was progressively de- From those extensive trials, DAVIES able was creased by the simple expedient of restricting to calculate the probable economics of a yeast the air supply (Tab. 6). Best results were ob- CBE process. This was based on the use of tained with a 500 L fermenter in which the O2 six, similar sized reactors which would opersupply could be effectively regulated. (The ef- ate continuously and with a cell recycling fect of O2 deprivation is extremely difficult to mode of operation to allow yeast densities of demonstrate with conventional 5 L laboratory 50 g dry cells per L to be achieved. Recovery fermenters as the amount of air that needs to of the oil from the yeast was also examined be supplied to produce the effect is so small and the economics of this process taken into that the usual air control systems are unsuita- account in the final calculations. A summary ble; C. RATLEDGE D. GRANTHAM, and unof DAVIES costings is given in Tab. 7. The published work.) The highest levels of stearic assumption is made that the yeast CBE prodacid so obtained by DAVIES al. (1990) were uct would sell for approximately 80% of the et not far from the required cocoa butter com- price of cocoa butter and that no significant
155
Tab. 7. Yeast CBE Process: Estimated Operating Budget (from DAVIES and HOLDWORTH, 1992)
Basis of process Available whey Lactose content Cost of whey (in New Zealand) CBE yield CBE production Value of CBE Total sales value of CBE
200,000 m3/a-' 39% (WIV) $ 0.5 per m3 0.16 kg kg-' lactose 1,250 t a - ' $ 2,400 t -' $3,000,000
costs
Direct manufacturing costs (untilities, substrate costs, downstream processing, wages, etc.) Manufacturing overheads (laboratory costs, site charges, effluent disposal, maintenance, insurance, service overheads, etc.) Finance and sales (distribution, research, and development) Plant depreciation over 10 years on capital of $ 5.4 M Interest at 12% Total costs Total sales Profit
$l,OoO,000
$46O,OOo
$300,000
$540,000 $650,000 = $ 2.95 M = $ 3.0 M = $ 50,000
costs would be involved in toxicological trials before the CBE could be offered for sale. DAVIES (1992a) and DAVIES and HOLDSWORTH (1992) assumed that a fermentation plant would have to be built specifically for the process. Clearly, if a fermentation plant already existed and could be used without major modification, then this would significantly decrease the indirect costs and the whole process might then show a significant annual profit. Although DAVIES'cost analysis will obviously change from country to country and will be heavily influenced by the cost of substrate, the overall conclusion is that this yeast CBE process is a process that is waiting for its day to come rather than being another uneconomic biotechnology pipe dream. Clearly, a number of other substrates can be used besides whey (see BEDNARSKI al., 1986; et GLATZet al., 1985, VEGAet al., 1988; FALL et al., 1984; GUERZONI al., 1985; DOSTAet LEK, 1986; HASSAN al., 1994b) though it is et essential that these be available throughout
the year to enable continual operation of the plant. Costs of labor and utilities are obviously cheaper in some countries than in others and, consequently, it would not be surprising if this yeast CBE process or a similar one was not adopted somewhere in the world over the next decade. Only the continuing low price of cocoa butter will prevent it from becoming an operating reality. It has recently been reported by Roux et al. (1994) that some species of Mucor, especially M . circinelloides, will simultaneously produce a CBE-SCO as well as producing a valuable polyunsaturated fatty acid (y-linolenic acid). This is described later in Sect. 3.3.1.
3.3 Molds
A list of oleaginous species of mold is given below.
Lipid contents of oleaginous molds grown in
156
4 Microbial Lipids
the vegetative mycelial state are given in parentheses (taken from RATLEDGE, 1986 1989a, 1993; LOSEL, 1988 and additional references as indicated).
Zygomycetes Entomophthorales Conidiobolus nanodes (26); Entomophthora conica (38); E. coronata (43); E. obscura (34); E. thaxteriana (32); E. virulenta (26); Glomus caledonius (72) Mucorales Absidia corymbifera (27); A. spinosa (28); Blakesleea trispora (37); Cunninghamella echinulata (45); C. japonica (60); C. elegans (44, 56); C. homothallica (38); Mortierella alpina (33) (TOTANIet al., 1992); M. elongata (34) (BAJPAIet al., 1992); M. isabellina (63-86); M. pusilla (59); M. vinacea (66); Mucor albo-ater (42); M. circinelloides (65); M. hiemalis (42) (KENNEDY al., 1993); M. miehei et (25); M. mucedo (51); M. plumbeus (63); M. pusillus (26); M. ramanniana (56); M. spinosus (47); M. vinacea (25) (HANSSON und DASTALEK, 1988); Phycomyces. blakesleeanus (33); Rhizopus arrhizus (32-57); R. delemar (32-45); R. oryzae (57); Zygorhynchus moelleri (40) Peronosporales Pythium irregulare (42); P. ultimum (48) Ascomycotina Ascomycetes Aspergillusfisheri (53); A.flavus (35); A. minutus (35); A. nidulans (51, 25); A. ochraceus (48); A. oryzae (37-57); A. terreus (57); A. ustus (28); Chaetomiumglobosum (54); Fusarium bulbigenum (50); F. equiseti (48); F. graminearum (31); F. lini (35); F. lycopersicum (35); F. oxysporum (29,34); Fusarium sp. NII (39); Gibberella fujikuroi (F. moniliforme) (48); Humicola lanuginosa (75); Penicillium gladioli (32); P. javanicum (39); P. lilacinum (51,56); P. soppii (40); P. spinulosum (64); Stilbella thermophila (38) Clavicipitacae Claviceps purpurea (31-60) Tulasuellales Pellicularia practicola (39) Hyphomycetes Cladosporium herbarum (49); Malbran-
chea pulchella (27); Myrothecium sp. (30); Sclerotium bataticola (46) Hymenomycetes Lepista (Tricholoma)nuda (48) Ustilaginomycetes Ustilaginales Sphacelotheca reiliana (41); Tilletia controversa (35); Tolyposporium ehrenbergii (41); Ustilago zeae (59) Uredmiomycetes Cronartium fusiforme (28); Puccinia coronata (37)
For inclusion in the list, the lower cut-off of a lipid content of 25% has been used. There are numerous molds that could have been listed if the limit had been dropped to 20% and, indeed, the commercial value of mold lipids, as will be explained below, could still be high even with molds having lipid contents of less than 20%. Commercialization of mold oils, therefore, unlike yeast oils, does not depend so much on the amount of oil that a mold produces but on the quality of that oil. The quality of the oil is determined by its fatty acid profile (Tab. 8): some fatty acids, particularly the polyunsaturated ones (PUFA) that have nutritional and some medical importance, are therefore select targets for current developments in this field. The number of oleaginous species of mold is greater than the number of oleaginous yeasts (cf. Tab. 3 and the list given above) but this is hardly surprising considering that there are some 100 times more molds (60,000 approx.) than yeasts. Although most of the 590 species of yeast have probably been assessed for oleaginicity, it is likely that most molds have not. Therefore, one may expect the list of oleaginous molds will be considerably increased as further work continues to be carried out. Extensive reviews of fungal lipids have been prepared by LOSEL (1988) and WEETE(1980). As a generalization, molds produce higher levels of the polyunsaturated acids 18:2 and 18: 3, than yeasts. Although there are various isomeric possibilities, the fatty acids in yeasts, molds, and algae appear to be the same as those found in plants. With linolenic acid (18:3), two isomers occur in both plants and microorganisms. The more common isomer is
157
Tab. 8. Fatty Acid Analyses of Lipid from Selected Molds (data taken from RATLEDGE (1986,1989a) and LOSEL(1988) and additional references indicated)
Organism Major Fatty Acyl Groups of [Relative % (w/w)]
14:O 16:O 23 9 37 24 16 19 7 29 18 15 18:O 15 2 7 7 14 8 2 3 6 7 18:l 25 14
4
Others
18:2 1 2 trace
8 14 9
18:3 4' 1' trace 10' 8' 8' 25 20:l (13%) 22:l (8%) 20:4 (4%) 12:o (40%) 1 2 0 (41%)
-
Entomophthorales Conidiobolus nanodes"

Entomophthora coronata E. obscura Mucorales Absidia corymbifera Cunninghamella japonica Mortierella alphab
1 31 8 1 trace
46 48 28
M.elongata'
M. isabellina Rhizopus arrhizus Mucor alpina-peyron"
18
55 22 30
12 3 10 9
'
20:3 20:4 20:4 205

-
(7%) (21%) (16%) (150/)
1 19 10
3' 12' If
20:o (8%) 20:3 (6%) 20:4 (5%) 20:l (4%) 20:4 (15%) 205 (12%) 20:1 (5%) 20:4 (11'3'0) 20:5 (14%)
-
Peronosporales Pythium ultimumd
15
20
16
1*
-
P. irregulare'
17
14
18
Ascomycetes Aspergillus terreus Fusarium oxysporum Pellicularia practicola Pennicillium spinulosum Hyphomycetes Cladosporium herbarum Ustilaginales Tolyposporium ehrenbergii Calvicipitacae Claviceps purpurea
a
2 trace trace
-
23 17 8 15 31 7 23
trace 8 2 7
12 5 2
14 20 11 42 35
81
40 46 72 31 18 2
8
21 5 2 1
trace
1
1
-
trace
19
12-HO18:l (42%)
'
KENRICKand RATLEDGE (1992a) YAMADA al. (1992) et BAJPAIet al. (1992) GANDHI WEETE(1991) and O'BRIENet al. (1993) y-linolenic acid, 18:3 (0-6)
158
4 Microbial Lipids
Tab. 9. Fatty Acid Profiles of a Commercial Fungal Oil Product Compared with Evening Primrose Oil, Borage Oil, and Blackcurrant Oils Containing y-Linolenic Acid Mucor circinelloides"
Evening Primrose Seed Oil

16
Borage Seed Oil

30
Blackcurrant Seed Oil

30
Oil content ["h (w/w)] Fatty acid 16:O 16:1 18:O 18:l 18:2 y-18:3 (~-18:3 20: 1 22:l 24:O
a
20
22-25 0.5-1.5 5-8 3841 10-12 15-18 0.2
6-10
9-13
1.5-3.5 6-12 65-75 8-12 0.2 0
3-5 15-1 7 3741 19-25 0.5 4.5 2.5 1.5
1 10 48 17 13
Production organism used by J & E Sturge Ltd., Selby, Yorkshire, UK, from 1985-1990.
a-linolenic acid, 18:3 (c 9, c 12, c 15), or 18:3 of this acid. Epoxy and dihydroxy fatty acids (03), - see Sect. 1.1- which is found in most are also found in relative abundance in the plants seed oils, yeasts, and the majority of lipids extracted from the spores of some bamolds. The more unusual isomer is y-linole- sidiomycetes. Some acetylenic acids may also nicacid-18:3 ( c 6 , ~ 9 , ~ 1 2 ) o r 1 8 : 3 ( ~ 6 ) occur (LOSEL, 1988). The occurrence of hywhich occurs in the seed oils of Oenothera droxy fatty acids in fungi has been recently (evening primrose), Ribes (blackcurrant, red- reviewed by VAN DYKet al. (1994). currant, etc.) and the borage family (BoragiAlthough molds contain an exceptional naceae) (see Tab. 9). It also occurs through- diversity of fatty acids, current commercial inout the lower fungi, also known as phycomy- terest centers principally upon the formation cetes (see Tab. 8). Both isomers occur simul- of particular PUFAs that have dietary or taneously in some algae, though not in molds medical applications. The role of such nor with Oenothera and borage plants. In PUFAs, in both healthy and dysfunctional paRibes, however, both isomers occur in equal tients, has been the subject of considerable amounts. Longer chain polyunsaturated fatty research and investigation. For a more than acids, up to 22:6, have been detected in the adequate statement of the current status of lipids of many species of the phycomycetes this work, the reader is referred to a recent and these are discussed separately below. Congress whose proceedings are given in a Unusual fatty acids such as hydroxy fatty volume edited by SINCLAIR and GIBSON acids or branched fatty acids are found, re- (1992). A discussion, or even prkcis, of the spectively, in a few species of Claviceps and various conditions and aliments that seemingConidiobolus (TYRRELLand WEATHER- ly benefit by a dietary intake of PUFA is STONE, 1976). The high content of ricinoleic beyond the scope of this review but consultaacid, 12-hydroxyoleic acid, in Claviceps spp. tion of this symposium volume should give (see Tab. 8) occurs only in the sclerotial tissue adequate information on most topics. The folof the fungus and is absent from the vegeta- lowing sections review the current work being tive mycelium (see LOSEL, 1988). It has, carried out to produce PUFAs from fungal therefore, no potential as an alternative sources. The formation of PUFAs is set out in source of castor oil which is the major source Fig. 5.
159
Co. Ltd. of Japan (see RATLEDGE,1989b) which used Mortierella isabellina. However, a recent report from NAKAJIMA I z u (1992) and of that company makes no mention of any Lower fungi, that is the phycomycetes, previous large-scale industrial process for invariably produce the y-isomer of 18:3 in- GLA production and, indeed, highlights only stead of the more common a-isomer. SHAW strains of Mucor circinelloides as potential (1966) was the first to point out this distinc- candidate organisms. The largest scale of option amongst fungi, though the occurrence of eration was only 30 L but, optimistically, the GLA in fungal lipids was first reported by authors consider that scale-up to 200 m3 was BERNHARD ALBRECHT and (1948) who had now a feasible proposition, but no reports of examined the lipid from Phycomyces blakes- work at this level were given. One strain of the mold was found that could produce 25% leeanus. Interest in oils containing this acid has a GLA in the total fatty acid with, however, long history and the oil from the seeds of only a 6% oil content of cells. The highest oil evening primrose (Oenothera biennis) have producer (30% of the cells) only produced been used for many centuries, being de- 10% GLA in the fatty acid. This reverse corscribed as the Kings Cure-All, as a remedy relation between oil and GLA content has for a number of disorders. The efficacy of been previously noted by RATLEDGE (1989b) evening primrose oil has been attributed to its and now been documented in some detail by content of GLA (about &lo% of the total KENNEDY al. (1993). et fatty acids). GLA itself has been reported as KENNEDY al. (1993) showed that even in et suppressing acute and chronic inflammations, a single organism (they used Mortierella radecreasing blood cholesterol concentrations, manniana, Mucor hiemalis, and Mucor circiand improving atopic eczema (HORROBIN, nelloides) the GLA content of the oil could 1992), however, SCHAFERand KRAGBALLE range from 5%-32% with oil contents from (1991) found no clear evidence in support of 43%-4% with Mucor hiemalis with a narrowGLA being an efficacious treatment for atop- er range for Mortierella rammaniana. Maxiic dermatitis. Alternative plant sources to mum productivity of GLA (that is g L- h-) evening primrose have been identified more was calculated for Mucor hiemalis with a recently and include borage (Borago offici- GLA content of the oil at &lo%. For Mucor nalis) and Ribes spp. (see Tab. 7). circinelloides, maximum productivity was with In view of the known occurrence of GLA a GLA content of 14-16% which is close to in fungi, steps to develop a biotechnological the commercial process outlined above and route to GLA were first initiated in the au- shown in Tab. 9. thors laboratory in 1976. A strain of Mucor High proportions of GLA of 20-26% in the circinelloides (M. javanicus) was identified total fatty acids, together with oil contents of from a large screening program as being a 15-25%, appear to be attainable by a number suitable production organism. The first sales of species of Mucor and Mortierella (HANSof the GLA-rich oil were in 1985, the process SON and DOSTALEK,1988; DAVIES,1992a; being run by J & E Sturge Ltd. (now Har- KENNEDY al., 1993; R o u x et al., 1994). It et mann & Reimer) of Selby, Yorkshire, UK, at is, therefore, a matter of simple screening to the 220 m3 level. The process ran until 1990 identify a likely candidate for large-scale prowhen production ceased following transfer of duction. However, because of the high costs the company to its present owners who are of fermentation processing, other parameters part of the Bayer industrial group. have to be taken into account besides GLA The specifications of the fungal oil were content: these include (1) the density to which equal or better than evening primrose oil in the cells can be grown - values of up to almost every respect including its much high- 2 5 g L - are not uncommon (see KENNEDY and er content of GLA (see Tab. 9). A similar et al., 1993; NAKAJIMA Izu, 1992) but are process to the Sturge one, was considered to still far from optimal (about 40-80 g L- have been developed by Idemitsu Petroleum should be attainable); (2) the rate of growth -
3.3.1 y-Linolenic Acid (GLA, 18:3 w-6)
160
4 Microbial Lipids
full growth and lipid accumulation should be attained within 72-96 h; and (3) the ability to extract the oil from the cells. The oil should obviously be free from any deleterious material, including free fatty acids and partial acylglycerols, but, as toxicological trials may be called for prior to the release of the oil for human consumption, it is an obvious advantage if the organism being used already has an established record of safe usage in foods. For these many reasons, Mucor circinelloides seems to have been an excellent choice for a GLA-production organism. Roux et al. (1994) have recently reported that some Mucor spp. when grown on acetic acid also produce a high content of stearic acid (18:O) besides GLA (up to 38 mg per g dry biomass was attained) and have suggested that, by appropriate fractionation, it should be possible to produce both a GLA-rich oil and a cocoa butter equivalent fat. Although some strains of Mucor produced a high content of stearic acid when grown in glucose (27% of the total fatty acids with M.flavus), the highest combined yields of stearic acid and GLA were with M . circinelloides grown in a pH-stat, fed-batch culture using acetic acid as sole carbon source. Stearic acid was up to 19% of the neutral lipid with GLA at 8%.
3.3.2 Dihomo- y-Linolenic Acid (DHGLA, 20:3 0-6)

DHGLA is produced biosynthetically from GLA by chain elongation (see Fig. 5). It is the precursor of the Group 1 of prostaglandins and thromboxanes and is often a minor component of lipids from fungi and algae but is also found in animals. There is probably no large market for DHGLA. Nevertheless, various attempts have been made to develop a process for its production as, undoubtedly, small amounts of oils containing high amounts of DHGLA would command a very high price if only for exploratory trials and for experimental laboratory work. Small amounts of DHGLA, up to 5% of the total fatty acids have been found in oils from Conidiobolus spp. (NAKAJIMA IZU, and
1992) and in Mortierella spp. of the subgenus Mortierella (AMANO al., 1992). The highest et amounts occurred with M. alpina strain 1S-4 which had been grown in the presence of sesame oil (SHIMIZU al., 1989a; YAMADAet et al., 1992). The concept behind adding the sesame oil to the fungal culture had been to see if exogenous oils could be taken up by the cells and then desaturated to particular fatty acids. With sesame oil there was an apparent inhibition in the formation of arachidonic acid (20: 4), which being produced directly from DHGLA (see Fig. 5), then led to the accumulation of DHGLA. The inhibitor was identified as a minor component of sesame oil, sesamin, which acted specifically against the A5 desaturase (SHIMIZU al., 1991). et DHGLA was produced up to 23% of the total fatty acids and at a yield of 2.2 g L-'. NAKAJIMA Izu (1992) have similarly and shown that a number of anisole derivatives when presented to Conidiobolus nanodes also led to the accumulation of DHGLA. Like sesame seedoil with M. alpina, these compounds had only a minor effect on cell growth and lipid accumulation. Maximum effect was tert-butylhydroxyanisole produced with (BHA) giving 18% DHGLA in the total lipid fraction which was about 35% of the cells. As both BHA and sesamin appeared to act as specific inhibitors of the A5 desaturase, the next logical step was to delete this enzyme by mutational techniques. JAREONKITMONGKOL et al. (1992a, c) reported the results of such a study using M. alpina and succeeded in isolating a mutant that produced 3.2gL-' DHGLA, that is 123 mg per g cells, and accounting for 23% of the total fatty acids. In comparison, the wild type produced less than a quarter of this amount. Other mutants of the same organism have been reported that accumulated increased amounts of other PUFAs (see Sect. 3.3.6, eicosatrienoic acid). The approach of deleting various desaturases at the genetic level that are involved in conversion of the PUFAs is obviously a very powerful one. Simple mutational techniques coupled with extensive screening for the correct mutants can clearly pay handsome commercial rewards.
161
3.3.3 Arachidonic Acid (ARA, 20:4 0-6)

Arachidonic acid (ARA) and eicosapentaenoic acid (EPA, see below) are intermediates in the formation of several key prostaglandins and leukotrienes which exert profound physiological control over various bodily functions and are the subject of much nutritional and medical research (see SINCLAIR and GIBSON, 1992). ARA has a much more restricted distribution than GLA and it is clear that many molds do not synthesize fatty acids beyond CI8in length (SHAW,1966). However, phycomycetes molds of the subdivision Mastigomycotina synthesize fatty acids up to Cz2and formation of ARA has been recorded, for example, in several Pythium spp. (GANDHIand WEETE, 1991), Saprolegnia parasitica (GELLERMAN and SCHLENK, 1979), in several Conidiobolus and Entomophthora spp. (TYRRELL, 1967, 1968,1971; KENDRICK RATand LEDGE, 1992a, b, C NAKAJIMA ; and IZU, 1992) and in a number of species of Mortierella subgenus Mortierella (Totani and OBA, 1987, 1988; YAMADA al., 1987a; SHIMIZU et et al., 1988a; SHINMEN al., 1989 AMANO et et al., 1992). The subject has been recently reviewed by RADWAN (1991) and BAJPAIand BAJPAI(1992), the latter recorded the ARA contents in 27 species of phycomycetes fungi as well as in 42 species of marine algae (see also below). The more general review of LoSEL (1988) records a number of fungi that contain ARA, though without regard to any biotechnological potential. All researchers and reviewers to date have been generally agreed that the family of w-6 PUFA, which includes ARA as well as GLA, are confined to the lower fungi or phycomycetes. RADWANand SOLIMAN(1988), however, reported that they had found ARA in the lipids of a number of ascomycetes (or higher) fungi: Aspergillus versicolor, A. niger, A. oryzae, A. ustus, A. fumigatus, Paecilomyces lilacinus, Penicillium sp., Fusarium oxysporum, and another Fusarium sp. In all cases, the fungi had been cultivated on single, shorter chain fatty acids, either saturated or monounsaturated, i.e., 14:0, 16:0, 18:0, or
18:l. As the identity of the ARA was not confirmed by capillary GC or by CG-MS, only by argentation TLC and by packed column GC, there must be grave doubts about the authenticity of these claims. Biochemically, it would be an unprecedented reaction that could convert a fatty acid of the w-3 series, which are invariably produced by these fungi (LOSEL,1988), into the 0-6 series (see Fig. 5) as this would involve a saturation reaction of a double bond at the 0 3 position which has never yet been recorded in any aerobicallygrowing microorganism. It is, therefore, more than likely that the 20:4 PUFA reported by RADWANand SOLIMAN(1988) as arachidonic acid was not the w-6 isomer (i.e., ARA) but was the 0-3 isomer, that is 20:4 (8,11, 14, 17) which would have behaved in both the GC and TLC analyses as ARA. Nevertheless, it is still quite exceptional for these higher fungi to be recorded as producing any fatty acid beyond CI8 in length (LOSEL,1988) and one can only conclude that it was the cultivation of these fungi on shorter chain fatty acids that led to this most unusual result, a result, though, which has yet to be confirmed in another laboratory. The highest ARA contents have been recorded in Mortierella alpina with up to 79% ARA in the total fatty acids which represented 26% of the cell dry weight (TOTANI and OBA, 1987). Further work with this organism has been developed up to the 300L scale (TOTANIet al., 1992) with some slight diminution of yield. Significantly, exceptionally long fermentation times up to 16 d were needed to produce the greatest yields. Such lengthy times would invariably increase the costs of any large-scale process. Both YAMADA al. (1992) and BAJPAIet et al. (1991~) have reported some possibly interesting developments in which the mold of choice is first grown for up to 10 d then allowed to stand at a lower temperature without further aeration. Under these conditions, the content of ARA (and EPA) increased up to 70% of the total fatty acids. However, mass balances were not carried out with these stored cells so it is not immediately apparent where the extra ARA or lipid might have originated; although the lipid content of M. alpinu increased from 1415% to 3345% during
162
4 Microbial Lipids
the aging of the cells for 6 d (BAJPAIet al., 1991c) it is not clear from where this extra lipid could have arisen. One simple explanation is that biomass other than lipid was self-utilized. Thus the total amount of lipid may not have changed but only appeared to increase as the remainder of the biomass was consumed.
tion (28C) (SHIMIZU al., 1988b). Highest et yields were attained with Mortierella alpina and M . hygrophila which produced 29 and 41 mg EPA per g cells. Subsequent work showed that exogenously added a-linolenic acid (18:3 w-3) was converted by M. alpina into EPA eventually pushing up the yield to 67 mg per g dry cells (YAMADA al., 1992; et SHIMIZU al., 1989b). This was an unusual et finding as most fungi will not modify exogenously added fatty acids (RATLEDGE, 3.3.4 Eicosapentaenoic Acid 1989b). (EPA, 20 :5 w-3) Other groups have not been as successful as the Japanese in finding productive fungi EPA exerts a number of physiological ef- for EPA. BAJPAI et al., (1992) found the fects when fed to experimental animals in- highest yields with Mortierella elongata were cluding lowering of blood triacylglycerol con- 15 mg per g dry cells and when a-linolenic centration. Consequently, this would decrease acid was added this increased to only 36 mg the potential for a coronary heart attack and, per g. In Pythium ultimum the maximum contherefore, the consumption of EPA is encour- tent was 34 mg EPA per g dry weight which aged by many advocates (SIMOPOULOS, was attained only after careful selection of the and 1989). EPA, like ARA is the precursor of strain and its culture conditions (GANDHI prostaglandins and leukotrienes. Currently, WEETE, 1991). OBRIEN et al. (1993) rethe major source of EPA is fish oil where it ported a maximum yield of 25 mg EPA per g occurs, usually in low concentrations along dry cells using Pythium irregulare and have with docosahexaenoic acid (DHA) (see be- recently described a pilot-plant process using a colloid mill for the recovery of EPA at 96% low). and In molds, EPA frequently occurs along yield from this fungus (OBRIEN SENSKE, with ARA. The conversion of ARA to EPA 1994). occurs directly in fungi (see Fig. 5) but not in animals (GELLERMAN SCHLENK, and 1979); the necessary 417 desaturase for this conver- 3.3.5 Docosahexaenoic Acid sion is thus apparently unique to fungi. Small amounts (usually < 10% of the fatty acids) of (DHA, 22:6 0-3) EPA have been recorded in a number of lowDHA is abundant in the phospholipids of er fungi (LOSEL, 1988) as well as in the marine fungi Thraustochytrium and Schizochy- retina and brain tissues and is usually retrium spp. (ELLENBOGEN al., 1969) though garded as an essential fatty acid for humans et these latter fungi also contain higher amounts and other animals (THOMASand HOLUB, 1994). It occurs in the oils of many fish where, of DHA (see below). SHIMIZU al. (1988a) were the first re- along with EPA, it may account for over 50% et 1994; NIsearch group to search specifically for the oc- of the total fatty acids (ACKMAN, currence of EPA at high levels in fungi CHOLS et al., 1994). However, fish do not synthough its presence had long been known as thesize either EPA or DHA but acquire them minor component amongst many of arachi- by ingestion of planktonic algae. DHA is condonic acid-containing fungi (SHAW, 1966). sidered important in the development of SHIMIZUet al. (1988a, b) having earlier brain tissue of babies but conclusive evidence screened fungi for their ARA contents (YA- for the nutritional role of DHA is lacking as it is always administered during feeding trials MADA et al., 1987a, b) observed that many of these fungi showed enhanced contents of along with EPA as fish oil is always used in EPA if they were grown at a lower tempera- such studies. A number of different nutritionture (12C) than that used for ARA forma- al roles for DHA have been proposed (THO-

MAS and HOLUB,1993; see also YADWAD et al., 1991). Although algae are clearly a potential source of DHA (see below), there are several fungi that have been considered as possible candidate organisms for its production though it ,must be said that prospects for a biotechnological route to DHA via fungi seems remote at this stage. The presence of DHA has been recorded in small amounts in the lipids of Conidiobolus and Entomophthora spp. (TYRRELL,1967, 1968, 1971) but more abundantly in the lipids of the marine fungi Thraustochytrium and Schizochytrium (ELLENBOGEN al., 1969), et in which the high content of DHA and EPA were considered to have a role in maintaining membrane fluidity in the organism whilst at low temperatures and in saline conditions. BAJPAIet al. (1991a, b) and KENDRICK and RATLEDGE(1992b) both independently reexamined the marine fungi as potential sources of DHA. All these fungi grew slowly and generally to low growth yields giving only low contents of lipid. None contained the key enzyme for oleaginicity, ATP-citrate lyase (see Sect. 2.3) and KENDRICK and RATLEDGE, (1992b) concluded that these fungi would be extremely difficult to exploit directly for the production of DHA. Maximum amounts of DHA were produced by T au. reum ATCC 34304 at up to 50% of its total fatty acids but with a lipid content of less than 15% (BAJPAIet al., 1991a, b) and usually not more than 10% (KENDRICK RATLEDGE, and 1992b). About two-thirds of the lipid was neutral lipid (triacylglycerols) and this still had a high content (30%) of DHA. However, the greatest difficulty with the exploitation of this fungus was its low growth yield: KENDRICK and RATLEDGE (1992b) obtained only 4 g biomass per L over 72 h under conditions which yielded up to 12 g L- of other fungi.
163
first described by MEADand SLATON (1956). Its nutritional status still remains unclear though it may be converted to the 12-hydroxy derivative which can affect blood platelet aggregration (LAGARDE al., 1985). It is preet sumably produced because animal tissues are unable to desaturate fatty acids between the existing double bond, in this case at the A9 position, and the terminal CH3 group. There has only been one report of the formation of ETA in fungi: JAREONKITMONGKOL et al. (1992b), following their work on the deletion of various fatty acid desaturases in Mortierella alpina (see Sects. 3.3.2 and 3.3.3) which led to increased production of DHGLA and ARA, found another mutant that was no longer able to convert oleic acid (18:l w-9) to linoleic acid (18:2 w-6). This mutant accumulated several fatty acids of the 0-9 series: 18:2 (6,9), 20:2 (8,ll) and also ETA. Under optimal conditions, in a fedbatch submerged cultivation for 10 d at 2 0 T , ETA reached 56 mg per g dry biomass = O.SgL-. Growth of the mutant itself reached about 15 g L- indicating that a small scale process might be possibly developable to produce this acid if sufficient commercial demand for it was forthcoming.
3.3.7 Conclusions
Although molds contain a large number of fatty acids, their biotechnological potential lies in their ability to produce a few selected polyunsaturated fatty acids (PUFA) in some quantity. Already we have seen the commercial production of ylinolenic acid (GLA) during the 1980s and it is likely that the production of arachidonic acid (ARA) may not be far off. A number of industrial companies are known to be developing processes for this PUFA and at least one company (Martek Corp. Inc, Maryland, USA) now offers a triacylglycerol oil containing 48% ARA for sale at $2,OOO per kg. Free ARA itself at 80% purity is offered at $20 per g (i.e., $20,000 per kg). Demand may increase for this particular fatty acid if it can be unequivocally demonstrated that it is beneficial when added to milk destined for neonatal babies. The absence of ARA, DHA, and GLA in cows milk
3.3.6 Eicosatrienoic Acid (ETA, 20:3 0-9, Mead Acid)

This comparatively rare PUFA occurs in small amounts in lipids of animal tissues and arises by direct elongation and desaturation of oleic acid (18:l w-9) (see Fig. 5). It was
164
4 Microbial Lipids
(but not mothers milk) has suggested that these acids may fulfil important nutritional roles in the development of the early brain of children. Current sources of ARA are from animals and as long as there is continuing and developing concern over the presence of undetected viruses and prions in animal products, the greater will be the driving force to identify alternative and safe sources of these acids. PUFAs derived from molds by fermentation technology are, of course, free from such infectious agents and can be produced to higher levels of quality control than plant oils. Furthermore, they do not contain herbicide or pesticide residues that would occur in oils derived from plant crops that have been treated with these chemical agents as part of the usual agricultural regimen of routine crop spraying. Experience with GLA from Mucor circinelloides has indicated it to be a very high quality oil free from all deleterious substances. Once a high market demand for an ARA-rich oil has been developed, the price should fall dramatically from the level quoted above. Production costs for a finished mold oil should lie in the region of $25-35 per kg. This price would include refinement (removal of non-TAG lipids) and decolorization or deodorization if needed by passage through charcoal which are standard procedures for producing all high quality oils. Demand for PUFAs other than GLA and ARA from molds is less certain although DHGLA and ETA (Mead acid) could be produced if needed but these tend to be regarded as rare PUFAs which are probably only required in small amounts (say 10 kg annually) for experimental purposes. With EPA and DHA, mold sources are not as good as algae or bacteria for these acids: EPA rarely exceeds 20% of the total fatty acids in a mold oil and is often much less though the recent description of a pilot-plant extraction process for the recovery of EPA from Pythiurn irregulure (OBRIENand SENSKE, 1994) may indicate possible future developments in this area. Although DHA can occur at up to 50% of the total fatty acids in a few molds, these species are slow-growing and may be difficult to develop on a large scale though their exploitation appears under active consideration.
As with all biotechnological products, what is produced will be dictated by market forces. Increased demand for a commodity invariably pushes up the price: should demand begin to increase for any of the PUFAs then rapid exploitation of molds could then be anticipated. The high quality of the oils ensures that molds are realistic alternative sources to either plant or animal products.
3.4 Algae
The term algae covers 14 distinct biological groups and includes both the macro- and microalgae. The macroalgae are the seaweeds and related families; the microalgae are the equivalent of eukaryotic microorganisms but also include the cyanobacteria, formerly termed the blue-green algae, which are part of the prokaryotic eubacteria. It is the microalgae that are the subject of most research for the production of designated lipids and will therefore be covered here. Microalgae have long been used as sources of protein for use in animal and human foods. Their potential as sources of biomass and fine chemicals has been the subject of several recent major monographs (BECKER, 1993; CRESSWELL al., 1989; BOROWITZKA et and BOROWITZKA, 1988; STADLER al., 1988). et Prospects for the production of oils and fatty acids by algae biotechnology have been reviewed, in general by KYLE (1991), YONGMANITCHAI and WARD (1989), VOLKMANN (1989), and BOROWITZKA (1988), and, with respect to the production of specific fatty acids by KYLE (1992), SETO et al. (1992), COHENand HEIMER (1992), BOSWELL al. et (1992), and KYLEet al. (1992). The main problem in the biotechnological exploitation of algae is their cultivation. Most algae will only grow photosynthetically and, therefore, require illumination and, although their carbon source, being C02, is regarded as free, growth is usually limited by the C 0 2 content of the air. When algae are grown outdoors in ponds or lagoons, a warm ambient temperature is needed besides high illumination which limits their geographical development but many algae are susceptible to contamination by other algae or even bacteria or
165
may be attacked by predatory protozoa. Consequently, to maintain a pure monoculture of an alga, high-cost illuminated fermenters may be needed which become impractical because of cost for large-scale growth. Although numerous devices, including clear plastic tubular fermenters, have been suggested for photosynthetic algae growth (see, e.g., LEE, 1986), no satisfactory system has yet been developed. Those commercial algae units that do exist, mainly for the production of carotenoids (see Sect. 4.2), use robust algae that have a particular nutritional advantage that prevents contaminants or predators affecting the culture. For example, Dunaliella salina grows in hypersaline ponds or lakes and little else can survive in such environments. However, an alternative to autotrophic growth (sunlight and C02) of algae may be possible in some cases. Thus KYLE(1991, 1992) has described the heterotrophic growth (darkness and glucose) of several algae that continue to produce high amounts of lipid in the absence of light. These examples are described below (see Sects. 3.4.3 and 3.4.4). Current attention on microalgae as sources of lipids is mainly because of their high contents of PUFAs. However, there is also interest in algae as potential sources of essential fatty acids for marine animals, particularly by developing fish larvae, mollusks, and crusta1989). In these cases, the cea (VOLKMANN, actual microalgae itself is of importance as the whole cells, not the isolated lipid, becomes the feed. With PUFA production, the choice of algae is less critical as the lipid itself will be extracted and used as the source of fatty acids. This, though, highlights a second major problem with algal lipids. Unlike oleaginous yeasts and molds, algae produce a large number of lipids many of which have functional roles in connection with the photosynthetic process. Only a relatively small proportion (1040%) of the total lipid may be triacylglycerol (see RATLEDGE, 1986) with the remainder being phospholipids, other polar lipids, and a variety of glycolipids. In the context of algae being used for animal feeding, the type of lipid does not matter as long as it is accessible and digestible. With oils for human consumption, the commercial emphasis is on producing an oil
which is acceptable in appearance: a clear, pale yellow oil with no taste, or aftertaste, is usually needed. This means that algae lipids may have to be fractionated or alternatively the constituent fatty acids removed from the total lipid by hydrolysis, either chemically or enzymatically, and then re-esterified to ethanol or glycerol. Ethyl esters of PUFAs are acceptable alternatives to the natural TAGS. Such additional processing though increases the costs of the final product quite considerably. A list of oleaginous microalgae is given below. Maximum reported lipid contents as % biomass dry weight are given in parentheses (from ROESSLER, 1990 and RATLEDGE, 1989a). Prokaryota (Cyanobacteria, blue-green algae) Anabaena cylindrica (9); Calothrix castelli (10); Nostoc sp. (8); Oscillatoria ssp. (18); Spirulina maxima (2); S. platensis (17). Eukaryota Amphiprora pyalina (30); Ankistrodesmus sp. (40); Biddulphia aurita (40); Botryococcus braunii (53-70); Chlorella minutissima (23); C. pyrenoidosa (36, 72); C. vulgaris (40); Chlamydomonas applanta (33); Chrysochromulina ssp. (3348); Crypthecodinium cohnii (25); Cyclotella cryptica (37); Dunaliella bardawil (D. salina) (47); Euglena gracilis (1420); Isochyrysis galbana (22); Monalanthus salina (72); Nannochloris sp. (48,55); N. oculata (42); Navicula acceptata (38); N. pelliculosa (22-45); Neochloris oleoabundans (35-54); Nitzschia palea (40); Ochromonas danica (3971); Oocystis polymorpha (35); Ourococcus sp. (50); Peridinium cinctum (36); Phaeodactylum tricornutum (14); Porphyridium cruentum (14, 22); Prymnesium parvum (22-38); Radiosphaera negevensis (43); Scenedesmus acutus (26); S. dimorphus (1640); S. obliquus (49); Scotiella sp. (16-35). The fatty acid profiles of selected species of microalgae are given in Tab. 10. The cyanobacteria (blue-green algae) tend to have low lipid contents and do not produce fatty acids longer than CIS. Although some species contain y-linolenic acid (see below), the amount is too low to warrant recovery though the
Tab. 10. Fatty Acid Profiles of Selected Microalgae Major Fatty Acyl Residues in Lipids [Relative YO] 14:O 16:O 16:l 18:l 18:2 18:3 18:3 20:3 20:4 20:5 22:6 (0-7) (0-9) (0-6) (0-6) (0-3) (0-6) (0-6) (0-3) (0-3) 8 6 3
1 26
-
Others
Reference
Prokaryota 5 5
1 58
8
2 23 3 25 10 21 24
12
HUDSON and KARIS (1974)
Spirulina maxima S. platensis 41
3
11
TORN ABENE et al. (1985) COHENet al. (1992 4 14 14
S. platensis (SRS-lh) Eukaryota Chlorella minutissima Chlorella vulgaris Chlorella NKG042401 12 13 - 1 6 <1 22
<1 35 <1
1
21 2 3
2 9 28
44
5 7
<1
Chlorella CHLOR-1 47 12
10
11
3 2
-
c1
25 9
11
Crypthecodium cohrii
19
Zsochrysis galbana
4 1 5 5 1 4 - 1 0
-
22 21 21 30 2229 5
3 4 1
3 4
1
<1
1 2 6 5 3
1
-
Nannochloropsis oculata (849/1) Nannochloropsis sp. Phaeodactylum tricornutum
7 3 8 1 3 3
<1
4 16
1
-
Porphyridium cruentum (SRP-7) Unspecified Martek isolate MK 8908
5 -
SETOet al. (1984) SHIFRIN (1984) HIRANO al. et (1990) GUCKERT and (1990) COOKSEY 12:0,l6% HENDERSON et al. (1988) 18:4, 11% MOLINA GRIMA et al. (1993) HODGSON et al. (1991) SETOet al. (1992) YONGM ANITCH AI and WARD(1992) COHENet al. (1992 BOSWELL et al. (1992)
167
presence of this PUFA in dry biomass may found that above pH 11 the triacylglycerol have a marginal nutritional benefit if algae (TAG) fraction of the total lipid increased to are used as a dietary supplement. However, over 20% whereas below this pH it was less the cyanobacteria are not readily digested than 3%. GLA was only about 10% of the toand may possess some toxicity (TORNABENE tal fatty acids in the TAG fraction but was et al., 1985). Spirulina spp. have though been over 40% in the glycolipid fraction which was used as a source of supplementary dietary between 50 and 60% of the total lipids. Thus no readily recognizable, useful source protein in both Mexico and Chad where blooms of the algae occur on Lake Texcoco of GLA has been found in any microalgae and Lake Chad (RATLEDGE, 1989a). These whether prokaryotic or eukaryotic. The plespecies, therefore, may be presumed to have a thora of lipid types (see, e.g., GUCKERT and 1990) means that direct producsafe health record. Most attention on microal- COOKSEY, gae has, though, focussed on the eukaryotic tion of a GLA-TAG oil from algae is an uneconomic proposition. species as sources of PUFA.
3.4.1 y-Linolenic Acid (GLA, 18:3 w-6)
3.4.2 Arachidonic Acid (ARA, 20 :4 0-6)
Surveys of the lipids of macro- and microalSpirulina platensis and S. maxima have and been known to contain GLA for some time gae (WOOD, 1988; YONGMANITCHAI 1990) indicate that (NICHOLS WOOD, 1968) and have occa- WARD,1989; ROESSLER, and sionally been considered as potential sources the red alga (Rhodophyceae) Porphyridium of this PUFA. HIRANO al. (1990), who ap- cruentum is superior to all other species for et pear to be the last research group to look at the formation of ARA. Small amounts of this source of GLA in any seriousness, were ARA do, though, occur in many of the maonly able to achieve a maximum content of rine microalgae but only in P. cruentum does GLA in dry biomass of S. platensis of 12 mg its content exceed 30% of the total fatty acids. per g and this was after heterotrophic cultiva- Very high amounts of ARA in the alga were et tion for 7 d at 30C. The total fatty acid con- originally reported by AHERN al. (1983) at and tent of the cells was less than 5%. These val- up to 60% of the total fatty acids. COHEN ues are not substantially different from those HEIMER(1992) have confirmed the potential recorded by previous workers (see RAT- of this alga for large-scale production of LEDGE,1989a) using these and other cyano- ARA which can be grown satisfactorily in et bacteria. HIRANO al. (1990) also screened a large outdoor ponds (VONSHAK al., 1985). et large number (>300) of marine eukaryotic However, under such conditions, the alga algae for GLA formation and found that the produces mainly a reserve polysaccharide at highest production was with a Chlorella sp. 40% of the biomass; ARA is only 1.5% of the (NKG 042401) that contained about 10% to- cell dry weight. According to how this alga is cultivated, it may also contain equal amounts tal fatty acids with a 10% content of GLA. In an attempt to improve GLA production of EPA (qv.) to ARA (COHENet al., 1988) in S. platensis, COHENet al. (1992) isolated a and may, therefore, be considered as a potennumber of cell lines that were resistant to the tial source of either or both PUFAs. A methherbicide known as SAN 9785 which is a sub- od for fractionating the glycolipid fraction of stituted pyridazinone that selectively inhibits fatty acids, which is the major lipid fraction, fatty acid desaturation. Slight increases in to- has been described and has yielded ARA at tal fatty acid contents ( 4 4 % dry wt.) were 80% purity and EPA at 97% purity (COHEN obtained in the resistant cells with GLA in- and COHEN,1991). Improvements to this alga are still being creasing from 21.5%-23.5%. GUCKERT and COOKSEY (1990) used an alkaline stress culti- sought by the selection of herbicide-resistant vation regimen with a Chlorella isolate and cell lines so that they will have higher propor-
168
4 Microbial Lipids
Tab. 1 .Cultivation of Porphyridiurn curenturn in Open (Outdoor) Ponds for the Production of ARA and 1 1992) EPA (from COHENand HEIMER, Cell Concentration ARA ph dry wt.=] EPA [% dry wt."] Output Rates [g m-' d-'1 Biomass Winterb Low Medium High Summer' Low Medium High
a
ARA 0.03 0.04 0.02 0.14 0.28 0.17
EPA 0.08 0.12 0.05 0.25 0.28 0.17
0.76 0.75 0.73 0.71 1.2 1.3
2.2 2.3 2.2 1.3 1.2 1.o
3.6 5.0 2.1 19.7 24.0 13.0
Ash-free dry wt. Maximum daily temperature 1618C. Maximum daily temperature 28-31 "C.
tions of desirable PUFAs (COHENand HEIMER,1992). The best results so far achieved in outdoor cultivation (see Tab. l l ) , which is the only route available for large-scale production, still indicate that there is little chance of this being an economic route to ARA, or EPA, production. Contents of ARA at less than 2% of the cell dry weight compare unfavorably with yields attained with molds which can achieve 5% and possibly 6% ARA contents (4.v.). Moreover, molds produce TAG oils requiring the minimum of downstream processing and can be easily grown to produce zero EPA. Thus fractionation and separation of the w-6 and 0-3 fatty acids would be unnecessary using mold technology.
3.4.3 Eicosapentaenoic Acid (EPA, 20:5 w-3)

SETO et al. (1984) reported that the high contents of EPA in Chlorellu minutissima made this alga an attractive source of this acid for both health foods and for the pharmaceutical industry. Maximum contents of EPA of the total fatty acids reached 45% but were only 2.7% of the dry biomass. Cell yields, moreover, were very low at 3OOmgL-l (compared to molds which reach cell densities of over 50 g L -I). Nevertheless, interest in
this and other alga as a source of EPA has continued (YONGMANITCHAI and WARD, 1991). Tab. 12 summarizes the principal photosynthetically-grown algal species that have been considered of some potential in the production of EPA. Although there are other algae with higher contents of EPA in their total fatty acids (see YONGMANITCHAI WARD, and 1989, 1991), the lipid content of these algae may not be very high and furthermore, they are mostly macroscopic algae which are not easily cultivatable under controlled conditions. A separate approach to PUFA production by algae has been developed by the Martek Corp. Inc. (Maryland, USA) in which selected algae are grown heterotrophically using glucose as carbon and energy source. Thus, photobioreactors are unnecessary and the organism behaves, and performs, as a eukaryotic yeast or mold (KYLE,1992). Of several thousand algae that were screened for PUFA production, one - MK8909 - was selected for EPA production. (Another isolate was subsequently exploited for DHA production, see below.) This isolate has been described as a apochlorotic diatom, that is not possessing chlorophyll and, therefore, incapable of photosynthesis. The alga may be related to Nuviculu suprophillu which KYLE et
169
Tab. U.Photosynthetically-Grown Algae as Potential Sources of EPA

Alga Lipid Content of Biomassa
[To 1
EPA in Fatty Acids
Reference
[%I
35
45 26 41 45
Phaeodacty lum tricornutum Nannochloropsis oculata Isochrysis galbana Porphyridium cruentum Chlorella minutissima
a
15 14 22
5.6b
15
YONGMANITCHAI and WARD(1992) SETOet al. (1992) MOLINA GRIMA al. et (1992, 1993, 1994) COHEN al. et (1992) SETOet al. (1984)
The lipids are usually composed of glycolipids, phospho- and other polar lipids with triacylglycerol as a minority component. The total fatty acid content of the lipids may be less than 50%. Total fatty acids.
al. (1988) had earlier described as having some potential for PUFA production. The content of EPA in MK8909 was only 5% of the total fatty acids but the oil content was 50% (w/w) of the cells. The organism grew readily and yielded 50 g dry cells per L in 3 d. As the alga produced an easily extractable oil, the EPA was considered of potential commercial value as there was a complete absence of any other PUFA: 18 :2 was present at only 3% and 18: 1 was at 26%. Thus enrichment of EPA from the oil is a relatively easy proposition and such oils are now offerred for sale by Martek. The overall problem with EPA production from any microbial source has already been discussed (Sect. 3.3.4) in that EPA is usually easily obtainable from fish oils though as a mixture with DHA. Unless, and until, someone demonstrates the clear need for EPA (or DHA) as a single PUFA, there seems little prospect that EPA will be required for anything more than experimental work.
3.4.4 Docosahexaenoic Acid (DHA, 22:6 w-6)

DHA has a very limited distribution in most algae though it occurs throughout the family of dinoflagellated algae (Dinophyceae)
along with EPA and also 18:4 (YONGMANITCHAI and WARD, 1989). Y~NGMANITCHAI and WARD (1989) highlighted three species that could warrant further attention for DHA production: Crypthecodinium cohnii, Gonyaulax catenella, and Gymnodinium nelsonii as all contained over 30% DHA in their total fatty acids. HENDERSON al. (1988) have exet amined the first algae in some details growing the cells non-photosynthetically. The harvested biomass had a total lipid content of 25% of which the triacyglycerol fraction constituted over 50% though this fraction only contained 9% DHA. The major DHA-containing component was phosphatidylcholine (PC) having 57% DHA in its total fatty acids and PC itself constituted 18% of the lipid. The complete fatty acid profile is given in Tab. 9. Of the other photosynthetic algae, only the species of Zsochrysis have been reported as containing more than a small amount of DHA. This marine alga has been recently examined by MOLINA GRIMA al. (1992,1993, et 1994) for its potential to produce DHA and EPA. Following screening of a number of isolates, one was selected for further work (LoPEZ ALONSO al., 1992). This strain, when et grown in chemostat culture with illumination, produced up to 2% of the cell dry weight as DHA (MOLINAGRIMAet al., 1993). EPA though reached up to 6% of the cells (see
170
4 Microbial Lipids
above and Tab. 12). As a source of DHA, the alga is, therefore, no better than many fish oils; DHA would be difficult to purify as a single PUFA and its presence in all lipid fractions (MOLINA GRIMA al., 1994) at about et 10% of the total fatty would not be regarded as particularly encouraging. On the other hand, the concept developed by the Martek Corp. Inc. (see Sect. 3.4.3) of using algae growing heterotrophically has succeeded in identifying an unknown (i.e., undeclared) phytoplankton species (MK 8805) that produced DHA as the sole PUFA at 50% of its total fatty acids (KYLEet al., 1992). The only other unsaturated fatty acids in the algae oil were 18:l (at 11%) and 16:l (at 2%). The only drawback with this alga was its low lipid content at between 10-15%. Nevertheless, it could be grown to high cell densities (about 25 g L-') in 84 h in a conventional (non-illuminable) fermenter. The oil and DHA-enriched fractions of it with up to 80% purity are now offered for sale by Martek. This oil is, therefore, the only one which is commercially available that does not contain EPA.
them heterotrophically: treating them as yeasts or lower fungi and growing them without illumination in conventional bioreactors. Martek Corp. Inc. in the USA have pioneered this approach and since 1992 have on sale oils containing EPA and DHA as single PUFA not containing any other PUFA or even diunsaturated fatty acids. It is understood, however, that demand for these oils has been very modest principally because there is, as yet, no clear dietary or clinical indication that one acid or the other would be useful in the treatment of any disorder or nutritional imbalance. In view of the current high costs of these oils, not unnaturally the public have preferred to buy various fish oils and PUFA concentrates derived from fish oils that contain both EPA and DHA together. The possible inclusion of DHA in infant food formulations (see Sect. 3.3.7), however, is likely to stimulate interest to find an alternative source to fish oil for this PUFA. The algae and fungi already mentioned will be the principal candidates.
3.4.5 Conclusions
Algae are undoubtedly a rich source of PUFA. They are the sources of PUFA in fish, which do not carry out de novo synthesis of these acids, but rely on the ingestion of algae for them. Nevertheless, algae are not exceptional sources of PUFA. ARA is possibly producible from Porphyridium cruentum being grown in outdoor ponds but the economics do not look as attractive as obtaining ARA from fungi. EPA and DHA may occur together in algae in which case the lipid would be no better than fish oil as a source of either acid. Some algae though only produce EPA and, therefore, might be useful sources of this acid if it were not for the fact that EPA is distributed throughout all the many lipid classes that occur in algae. This plethora of lipid classes necessitates complete hydrolysis of the total lipid and recovery of the EPA as the free acid. The more realistic commercial approach to PUFA production by algae has been to grow
4 Sterols, Carotenoids, and Polyprenes

4.1 Sterols
Although steroids, including their precursor squalene, have been reported occasionally from the prokaryotic microorganisms, these are only of academic interest as the quantities involved are extremely small. Sterols, however, have been produced commercially from eukaryotic microorganisms though only as a by-product by the extraction of either spent fungal mycelium recovered from a fermentation process or from spent brewer's yeast. Sterols for commercial use are usually obtained by extraction of plant materials which is not only a cheaper route than from microorganisms but also provides the correct substituents on the molecule for easy transformation into the commercially lucrative steroid hormone market. Most microbial processes
171
for steroid formation now appear to be defunct but some extraction of brewers yeast for the production of ergosterol (VI) does appear to continue in a few locations though the exact number and scale of the various operations is not generally revealed by industrialists. Ergosterol (VI) is the commonest microbial sterol and it occurs both in the free form and as its fatty acyl ester in algae, yeasts, and molds. Highest levels of sterol are found in
the yeasts, especially in Saccharomyces, Kluyveromyces, Metschnikowia, Pichia, and Torulaspora (RATTRAY, 1988). While the major steroid is ergosterol, over 14 other sterols have also been identified (WEETE,1980), the principal ones of which are lanosterol (VII), zymosterol (VIII), and ergosta-5,7,22,24(28)tetraen-3/3-01 (IX). Although the usual range of extractable sterols (free sterols plus sterol esters) in yeasts is generally in the range of 0.034% of
Ergosterol
Lanosterol
vn
Zymosterol VIII
Ergosta-5,7,22,24(28)tetraen-3@-01
IX
Cholecalciferol
7-Dehydrocholesterol
XI
172
NEY
4 Microbial
Lipids
the cell dry weight (RATTRAY, 1988), DULAet al., (1954) found several strains of S. cerevisiae which produced ergosterol up to 10% of the biomass. When the yeasts were grown in submerged culture yields of sterol of up to 4 g L - were obtained. A process for the production of ergosterol was patented by DULANEY (1957). Other workers have been also able to achieve similar results: e.g., ELREFAI and EL-KADY (1968a, b; 1969) recorded up to 23% sterol contents in Saccharomyces fermentati (now Torulaspora delbrueckii) if grown in the presence of potassium persulphate, hydroquinone, or indigocarmine. Increased accumulation of sterols in S. cerevisiae occurs at low specific growth rates under nitrogen-limited, aerobic growth conditions (NOVOTNY al., 1988) and by exploitet ing such information BEHALOVA VORIand SEK (1988) increased the total sterol content of this yeast to 7.2% of the dry biomass. The total lipid content of the cells, including the sterols, reached 31% making the yeast a candidate for inclusion in the list of oleaginous yeasts (see Tab. 3). However, such lipid contents in S. cerevisiae are quite exceptional indicating that these authors may be using an atypical strain. Ergosterol is not of major economic significance though it has some commercial value. MARGALITH (1989) has outlined the various attempts to produce it using yeasts. Its main application is that of an analog for cholecalciferol, vitamin D3, (X), which arises from 7dehydrocholesterol (XI) by the action of ultraviolet light; likewise ergosterol (VI) is converted to ergocalciferol, vitamin D2 (XII). However, the latter is only 10% as nutritionally effective in chickens as is vitamin D3.The prospects of being able to produce mutants of S. cerevisiae which would accumulate 7-dehydrocholesterol rather than ergosterol are, therefore, extremely attractive (PARKSet al., 1984). In molds, a large range of sterols has been recognized (WEETE, 1980; LOSEL, 1988), though once more ergosterol is the major constituent in many, though not every, species. Cholesterol (XIII) has been noted as the major sterol in a number of Mucorales and also, unexpectedly, in Penicillium funiculo-
sum. The highest level of sterol reported would appear to be that found by OSMAN et al. (1969) for Aspergillus fumigatus which contained 5% of its dry biomass as ergosterol. SHIMIZU al. (1992) have recently reported et the occurrence of a novel sterol, 24,25- methylenecholest-5-en-3p-01(XIV) in Mortierella alpina being grown for arachidonic acid production (Sect. 3.3.3). This sterol contains a cyclopropane ring in its side chain. Similar sterols occur in sponges but this is the first example of a sterol with a cyclopropane in this particular position in any organism.
4.2 Carotenoids
Carotenoids occur through the whole of the microbial world and, of course, throughout nature. Their occurrence and structure has been comprehensively reviewed by GOODWIN (1980, 1983). The review of LOSEL (1988) on fungal lipids also includes coverage of the carotenoids of fungi. A useful monograph on the various pigments of microorganisms has been prepared (MARGALITH, 1992) and contains a short but informative chapter on carotenoids, their chemistry, biochemistry, and functions. The biosynthesis of carotenoids has been described by HARRISON (1986), BRAMLEY (1985), and BRAMLEY and MACKENZIE (1988). Although all algae contain carotenoids, these are part of the chloroplast photosynthetic apparatus and, therefore, do not usually form a major constituent of the biomass. In the marine brown macroalgae (the Phaeophyta), the total annual biosynthesis of carotenoids through the oceans of the world has been calculated as 1.2-10 t (JENSEN, 1966) but, of course, none of this is harvested for the carotenoids. The major carotenoids, in order of abundance, in these seaweeds are fucoxanthin (XV), violaxanthin (XVI), and p carotene (XVII). Amounts of carotene in microalgae do not usually exceed a few mg per g dry weight (GOODWIN, 1980). However, Spirulina platensis, which is used as a source of food by inhabitants around Lake Chad in Africa and, which will be grown on brackish water, contains about 0.4% of its dry weight as pcaro-
173
HO Ergocalciferol
&&
HO Cholesterol
XI11 XI1
24,25-methylenecholest-5en-3~-ol
XTV
Fucoxanthin
xv
Violaxanthin
XVI
p-Carotene
XVII
tene and other xanthophylls (CLEMENT, RON,1980). PCarotene may be up to 10% of 1975). The halophilic Dunaliella salina or the dry weight of D . bardawil (MARGALITH, sometimes referred to as D . barduwil, which 1992) and with such yields has been grown can grow in waters of high salinity and pro- commercially in Australia by Western Bioduce glycerol in some abundance (DUBINSKYtechnology Ltd. at Hutt Lagoon in Western et al., 1978), can produce up to 400 mg p- Australia, in Israel (Nature Beta Technolocarotene m-' d-' (BEN-AMOTZand Av- gies), and in the USA (Microbio Resources).
174
4 Microbial Lipids
All systems use outdoor lagoons: that in Aus- viewpoint, the most important carotenoid is tralia, e.g., has a coverage of 50 ha (10x5 ha pcarotene (XVII) which, besides being an ponds) and was scheduled to add a further important foodstuffs colorant, has provitamin 25 ha pond in 1994. (The author is grateful to A activity (vitamin A: XVIII). It has also Dr. M. A. BOROWITZKA this informa- been suggested that pcarotene may also act for tion.) Full descriptions of this process have through its role as an antioxidant as a tumorbeen provided by BOROWITZKA BORO- suppressing agent and be useful in chemoand 1992). WITZKA (1989, 1990) and BOROWITZKA prevention of cancer (MARGALITH, (1992). In Israel, Dunaliella is grown in a sec- However it is the p-cis isomer which appears to be effective rather than the all-trans, chemtion of the Dead Sea. Algal pcarotene is comprised of 60% 9-cis ically-produced pcarotene. Demand may isomer (BEN-AMOTZ and AVRON, 1983; consequently shift towards the more expenBEN-AMOTZ al., 1988). This appears to be sive natural p-carotene if these claims for its et assimilated by experimental animals more therapeutic effectiveness are confirmed. pCarotene is also the predominant caroterapidly than the all-trans isomer (BENAMOTZ et al., 1989). From a commercial noid in many fungi and is a minor constituent
Vitamin A
XVIII
Torulene
Astaxanthin (3R. 3 R isomer)
xx
Botryococcene
XXI
175
in most others, though not all fungi do contain carotenoids. The most abundant production of p-carotene has been achieved with the phycomycete fungi, Blakesleea trispora and Phycomyces blukesleeanus (MURILLO al., et 1978; NINETand RENAUT, 1979; CERDA-OLMEDO, 1989) where yields of carotene in mutant strains of up to 2% of the biomass have been recorded. Commercial processes for the production of &carotene using B. trispora have been developed and at least one, in the Ukraine, is still in operation. SHLOMAI al. et (1991) have recently re-examined the p-carotene produced by P. blakesleeanus and, contrary to the original supposition that the alltrans isomer would be found (BRAMLEY and MACKENZIE, 1988), 15% of the total p-carotene was the 9-cis isomer. The remaining 85% was though the all-trans form. Nevertheless, this result was considered of considerable interest as further research should be able to increase the proportion of 9-cis-p-carotene. Such formation of the highly desirable isomer for cancer prevention (see above) could now re-stimulate interest in the exploitation of fungi for &carotene production. Cultivation of heterotrophic fungi appears to be a better commercial proposition than having to grow algae autotrophically where the costs of land, harvesting and the necessity for high light intensities and high ambient temperatures pose many severe limitations (MARGALITH, 1992). Torulene (XIX), which has only half the provitamin A capability as p-carotene, is the major carotenoid pigment of the red Rhodotorula and Rhodosporidium yeasts, of which several species are oleaginous (see Tab. 3). Other carotenoids also occur in these species (GOODWIN, 1980). However, the amounts are far less than needed for any commercial interest and these carotenoids remain of academic interest. The pink yeast, Phaffiu rhodozyma produces astaxanthin (XX) which is the carotenoid giving the characteristic pink color to salmon, crabs, lobsters, and other crustaceans (and indirectly flamingos and other birds living off these life forms). It is now produced commercially by Gist-Brocades, Netherlands, for use in feed formulations for poultry but mainly for pen-reared salmonids. The demand for astaxanthin for fish feed may ex-
ceed 100t a-' by the end of this century (JOHNSON and AN, 1991). The yeast carotenoid has, however, the opposite chirality at 3R and 3R' (i.e., the hydroxyl group on the cyclohexene end groups) to the lobster astaxanthin (see GOODWIN, 1980; 1983) but this does not affect its acceptability. Synthetic astaxanthin which is the all-trans isomer, is produced by Hoffman-La Roche Ltd. but awaits approval from the FDA for use in the USA (JOHNSON and AN, 1991). Consequently the emphasis is now on P. rhodozymu as a potential worldwide source. Initial yields of astaxanthin by P. rhodozyet mu were less than 500 Fg g-' (JOHNSON al., 1977; 1980). Whilst some increase is possible by careful selection of the growth medium and conditions (HAARD, 1988; NELISand DE LEENHEER,1989), high yields can only be achieved after mutagenesis (AN et al., 1989; MEYER et al., 1993). Even these yields though can be enhanced by selecting individual cells by a cell sorter using the fluorescence of astaxanthin as the indicator (AN et al., 1991). Yields of astaxanthin of 2.5 mg per g dry cells have been achieved by using a mutant and carefully selected growth conditions (MEYERet al., 1993). However, it is considered that commercial production probably needs to reach 4 to 5 mg g-' in order to be economic. An alternative microbial source of astaxanthin is the freshwater green alga, Huemufococcus pluviulis (BOROWITZKA, 1992). Outdoor cultivation of this alga so far has proved unreliable and commercialization seems unlikely at the moment even though astaxanthin may reach up to 20 mg per g cell dry weight.
4.3 Polyprenoids
The only polyprenoid of possible biotechnological significance is botryococcene (XXI) which occurs as the principal lipid component of the alga Botryococcus bruunii. This alga has been reported as containing up to 85% of its biomass as a mixture of polyprene hydrocarbons when it has been isolated from coal deposits, which appears to be its natural ecological niche. The alga can be grown in the laboratory, though yields of botryococcene
176
4 Microbial Lipids
and related C,, hydrocarbons are then only dual purpose: sewage cleanup and hydrocarfrom 24 to 45% of the biomass (HILLEN al., bon production. As the hydrocarbon would et 1982; YAMAGUCHI al., 1987; CASADEVALL be used as a fuel and not as a food suppleet et al., 1985). The hydrocarbons can be ther- ment, there is obviously no restriction on mally cracked to give gasoline and other fuels what the alga may be grown on as none of it of commercial value (HILLEN al., 1982) and would be returned into the food chain. et thus have been suggested as possible sources of energy. However, it seems unlikely that this would represent a commercially exploitable source of fuel hydrocarbons particularly in view of the slow rate of hydrocarbon production at about 0.15 g L-' d-' (CASADEVALL et al., 1985) though a slight increase occurs when the cells are immobilized (BAILLIEZ et al., 1985; 1986) and used in an air lift, illuminated bioreactor (BAILLIEZet al., 5.1 Wax Esters 1988). Wax esters of the type RCOOR' where R The impracticality of scaling-up such a system would, though, preclude any serious bio- and R' are long alkyl chains occur in bacteet technological application but SAWAYAMA ria, algae, and yeast. Their route of biosyntheal. (1992) have recently reported that B. bruu- sis is given in Fig. 6. The fatty acid and alconii can be usefully cultivated on secondarily hol moieties may be saturated or unsaturated. treated sewage from domestic wastewaters so The diunsaturated wax ester is desirable as a that it not only removes N and P but still pro- substitute for jojoba oil. With algae, it is the duces about 50% of its biomass weight as the protozoan, Euglena grucilis, that has been the and hydrocarbon. Thus, the only hope for com- most studied (see KAWABATA KANEYAmercial take-up would be to use the alga for a NA, 1989) for wax ester production but the
5 Wax Esters and

Polyesters
Feedstock
Hydrocarbon
Fatty alcohol
-I
I
lntracellular activities
Hydrocarbon Oxidation
Fatty'alcohol Oxidationw Unsaturated fatty alcohol
Oxidation
Fatty a&
___j
Fatty acid
(Fatty acyl-CoAester)
Degradation Biosynthesis
Acetyl-CoA
lt it
Reduction
Diunsaturated monoester
~
Oxidation
Unsaturated fatty acyl CoA
acetic acid, etc.
Fig. 6. Pathways to a diunsaturated wax ester (jojoba oil type substitute) from various substrates.
5 Wax Esters and Polyesters
177
amounts are less than 10% of the cell biomass (about 100 kg per lo6 cells). With yeasts, wax esters are not usual components of the lipid fraction but SEKULA (1992) has reported formation of them in several yeasts when grown in fatty alcohols. The amounts formed were not disclosed but appeared, from TLC evidence, to be equal in amount to the triacylglycerol fraction. Evidence was presented that the fatty acid moiety of the ester may be oxidized to a e l keto fatty acid which could then be incorporated into the wax ester (see Fig. 6). The greatest amounts of monoester appear to occur in bacteria of the genera Acinetobacter, Micrococcus, Nocardia, Mycobacteria, and Corynebacterzurn. In the latter three genera, the fatty acids involved may be longchained ( > &) and/or methyl branched; the waxes may also be associated with virulence and thus biotechnological exploitation is unlikely. Acinetobacter and Micrococcus, which are now probably synonymous at least as far as the wax-producing strains are concerned, have been studied by various groups including that of Cetus Corp., California, USA and numerous patents (see RATLEDGE, 1986) taken out. The aim of this work had been to achieve production of a jojoba oil-like material which is essentially a 20: 1-20: 1 fatty acid/ fatty alcohol ester. Yields, however, have remained low (< 1g L-') and no take-up of the
process has occurred. Indeed, prospects for a microbial route to wax ester production would now seem to have receded even further with the recent description of how oleoyl oleate can be chemically synthesized using oleic acid, oleoyl alcohol, and a zeolite catalyst (SANCHEZ al., 1992). et
5.2 Polyesters Poly-P-Hydroxyalkanoates

The major microbial polyesters of commercial interest are the poly-p-hydroxy alkanoates (PHA) of which poly-p-hydroxybutyrate (PHB, R=CH3 in XXII) is of major importance. PHB and PHA are found principally in bacteria though related molecules have been found in small amounts in the memand branes of eukaryotic cells (ANDERSON DAWES,1990). The subject has been the topic of a number of monographs and international symposia (DAWES,1990 DOI, 1990; SCHLEGEL and STEINBUCHEL, 1992; VERT et al., 1992) and more are to follow, see below. A typical electron micrograph of a PHB-containing cell is given in Fig 7. Present interest in PHB/PHA arises from its use as a biodegradable plastic. PHB itself is considered too brittle to be conveniently molded into appropriate shapes and so is conr
1
x = 10 000 to 20 000
R = -CH3 for poly-Ehydroxybutyrate (PHB) R = - C2H5 for poly-P-hydroxyvalerate (PHV) R = - CnHh-1 for poly-P-hydroxyalkanoates(PHA) up to n = 9
Poly-p-hydroxybutyrate and alkanoates XXII
178
4 Microbial Lipids
cose. PHB and the co-polymer, PHB/V, are produced by Zeneca plc, UK, (formerly ICI plc) and uses Alcaligenes eutrophus as production organism. Yields are up to 80% of the total cell biomass. For commercial purposes the molecular weight of the polymer should be as high as possible and certainly in excess of lo6 Da. The process has been described, with perhaps understandable perfunctoriness, by BYROM(1990, 1992). The commercial product is sold under the trade name of Biopol@. A rival industrial process operated by Chemie Linz GmbH, Austria, has been described by HRABAK (1992) but it is uncertain whether this operates other than as a demonstration unit. Accumulation of PHB in bacteria is faFig. 7. Electron micrograph of poly-p-hydroxybutyvored by much the same environmental facrate granules in Alcaligenes eutrophus; marker bar: 1 pm (photograph kindly supplied by Dr. A. J. AN- tors that are needed for triacylglycerol accumulation in yeasts and molds: that is a depleDERSON, University of Hull, UK). tion of N (or other nutrient) from the culture with the provision of excess carbon to ensure continued formation of the polymer (see Sect. sequently produced as a heteropolymer along 2.1). As bacteria do not readily synthesize with Phydroxyvalerate (V, R = -C2H5 in triacylglycerols, PHB and PHA may be reXXII) as the other monomeric unit to P-hy- garded as the bacterial equivalent to triacyldroxybutyrate. Whilst Phydroxybutyrate is glycerols serving the same metabolic funcsynthesized from acetate-acetate condensa- tions as a (chemically) reduced storage comtion, P-hydroxyvalerate is produced from ace- pound; that is, it is accumulated under conditate-propionate condensation. This requires tions of nutrient deficiency and utilized durpropionic acid to be presented to the bacteri- ing periods of carbon starvation. The pathway al cultures as a cosubstrate along with glu- of biosynthesis of PHB (Fig. 8) is less com-
Acetyl-CoA
& Acetoacetyl-CoA --D-3-Hydroxybutyryl-CoA

(3-ketobutyryl-CoA) 3-Ketovaleryl-CoA
COA-SH
C
PHB/V
Fig. 8. Biosynthesis of poly-/3-hydroxybutyrate (PHB) and the copolymer of poly-(p-hydroxybutyrate/@ hydroxyvalerate) (PHBN). PHB, R = CH3 in Fig. 10 PHV, R = C2H5in Fig. 10. Enzymes: A: 3-ketothiolase; B: acetoacetyl-(3-ketoacyl-)CoAreductase (NADPH-dependent);C: PHB synthase (or polymerase); l x : one mol; 2 x : two mol.
5 Wax Esters and Polyesters
179
plex than that of triacylglycerols. Indeed, the biosynthesis requires only three additional enzymes to those already present for fatty acid biosynthesis: 3-ketothiolase, acetoacetylCoA reductase, and PHB (PHA) synthase or polymerase. All three enzymes have been studied in some detail (see ANDERSON and DAWES, 1990) and the genes for each of them have been identified, sequenced (STEINBUCHEL et al., 1992), and now cloned and expressed in plants (POIRIER al., 1992). (The et PHA polymerase is probably a different enzyme from that of PHB polymerase and the corresponding gene awaits to be described.) This latter work has considerable commercial potential. Whilst the economics of PHB/ V production by large-scale bacterial fermentation are regarded as only just favorable for commercial exploitation, the economics of production would be considerably enhanced if the same production could be achieved in plants. Just as plants produce triacylglycerol oils and fats at a tenth, or even less, of the cost of producing the same materials biotechnologically, so the costs of producing PHB would be considerably decreased by switching production into plants. For the work to be successful, yields of PHB will have to equal those currently achieved by oilseed crops for the production of triacylglycerol oils, i.e., up to at least 30% of the harvested crop. Present results indicate that PHB formation is only about 100 pg in the hybrid plants (POIRIER et al., 1992) which places the work as being still in the preliminary experimental stages. For successful production of PHB in plants, the most suitable plant would appear to be one which has the essential machinery for accumulation of a (plant) product already in place. Moreover, not only will the genes for PHB biosynthesis have to be inserted but the genes for producing the existing product will have to be deleted so that the flux of carbon can then be diverted wholly into PHB: COz Old - Sugar / PHBproduct (oil or fat)
As PHB is synthesized by direct condensation of acetyl-CoA units (Fig. S), it is fairly obvious that the ideal plant for PHB produc-
tion would be an oil-producing one as such a plant would already have the necessary biochemical machinery present to produce acetyl-CoA in some abundance. Thus, if fatty acid biosynthesis could be prevented, say by deletion of acetyl-CoA carboxylase or even impairment of the fatty acid synthase complex, then, with the three PHB genes being successfully introduced and fully expressed in the plant, carbon should now flow into PHB production. Such scenarios are now being developed for oilseed rape by Zeneca Seeds in the UK (SMITHet al., 1994) and by the Carnegie Institution in conjunction with Procter & Gamble in the USA using Arabidopsis as as a model plant system (POIRIERet al., 1992). Alternative plants to oilseed rape could clearly include sunflower, which is clearly amenable to genetic modification (see Tab. 2), and possibly even the palm oil tree. Yields, however, at the moment are very far from any practical value, and considerable technical work, if not scientific innovation, will be necessary to achieve commercially viable yields. The future demand for PHB and related molecules is currently seen to depend on their value as a biodegradable plastic. Its environmentally friendly nature has been repeatedly stressed. However, progress to producing other non-PHB, biodegradable plastics using chemical synthesis is now proceeding apace (VERTet al., 1992). Should large-scale chemical synthesis of an alternative, but still environmentally friendly polymer be achieved in the near future, this would have serious consequences for the economic viability of PHB/ V. Conceivably though, if production of PHB could be achieved in plants to the same yield that they now produce oils, then this could remain an alternative route to production. Long term prospects for the production of PHB by microbial fermentation would appear uncertain. However, just as microorganisms have been dismissed as alternative producers of oils and fats that are already commercially available, the way forward for microbial PHBs may be to identify higher valued products for niche markets. The range of polyalkanoates produced by microorganisms extends far beyond the simple PHB and PHV polymers. Other potential-
180
4 Microbial Lipids
ly useful polymers could include the poly-phydroxyalkanoates where the side chain (R in XXII) could be an alkyl chain up to C9. These are known as the medium chain length PHAs. Their occurrence has been described mainly et in Pseudomonas oleovorans (WITHOLT al., 1990 ANDERSON DAWES, and 1990). Formation is considerably enhanced by growing the bacteria on long chain alkanes, fatty alcohols, and fatty acids. However, growing the bacterium on a long chain fatty acid, such as oleic acid (18:l ) , only induces formation of the CI2 monomer (R=C9 in XXII) though, by using it, PHAs are produced with an unsaturated side chain (WITHOLT al., 1990) but still no et longer than C,. The properties of these PHAs 1995) and have been described (DE KONING, have indicated some potential for producing, after chemical modification, a rubber latex material that still retained its biodegradability. Considerably further work with the PHAs may be anticipated over the next few years. Although there is currently only one industrial producer of a PHA (see above), a recent (1994) conference held in Montreal, Canada, attracted over 300 delegates. The proceedings of this meeting were published in the Canadian Journal of Microbiology, probably during 1995.
may be a minor component of apparently only academic curiosity but then in the hands of an appropriate industrial company is scaled up into being a significant product. In the following some examples of the range of microbial lipids are given; in some cases commercial exploitation has occurred, in others the lipids remain academic curiosities.
6.1 Biosurfactants
Most microorganisms produce a range of surfactants that seemingly allow them to become attached to water-insoluble substrates or to surfaces of leaves and other parts of plants or soils. Production of surfactants may be enhanced by growing the organism on hydrocarbons or vegetable oils and, for a while, this was considered to be a prerequisite for growth to take place. It is now not certain if this is the case as some surfactants are produced in some abundance even when watersoluble substrates are used. A wide diversity of chemical types is known ranging from glycolipids which usually involve one or more fatty acyl residues attached glycosidically to a mono- or di-saccharide. Examples include the sophorolipid from Candidu bombicolu (XXIII) and the rhamnolipids from Pseudomonus aeruginosa (XXIV). In some cases, macromolecular surfactants are formed with a M , of up to lo6 Da in which the lipid component may not be the major one. An extensive monograph on the entire subject has been recently published (KOSARIC,1993). Readers are therefore reThe range of microbial lipids is, of course, ferred to this book for full details of the range extensive (see RATLEDGEand WILKINSON, and potential of these molecules. 1988a, 1989). The biotechnological exploitation of any particular lipid will depend upon the perception of the purpose to which that lipid can be put. In some instances the formation of a particular lipid may have been known for many years but how it may be of commercial benefit is not so obvious. An example of this would be the formation of the Ladone form sophorolipids by several Candida spp. which, R=HorCH3 although excellent surfactants, have not proved sufficiently superior to chemicallySphorolipid produced materials to warrant full scale production (see below). In other cases, the lipid M(m
6 Other Lipids
6 Other Lipids
181
ness of these organisms from conventional bacteria, contain a plethora of novel lipid types not seen elsewhere in either prokaryotic or eukaryotic organisms. The organisms which are regarded as the most ancient of all life forms, are amongst the most resistant of all organisms to the extremes of environment. They therefore include the thermoacidophiles, the extreme halophiles, and the methanogens. R = H or - CH-CHz - COOH The lipids of these bacteria are characterI (CH2)6 ized by being ether, rather than ester, derivaI tives of glycerol. The alkyl groups, however, CH3 are not derived from fatty acids but are formed by the mevalonate pathway which leads to the formation of the isopentenyl lipRhamnolipid ids. (The isopentenyl-derived lipids include carotenoids, sterols, and other terpenoid lipXXIV ids which are of ubiquitous distribution but the archaebacterial ether lipids are confined to the Archaea.) Some typical examples of these lipids are given in Fig. 9. Although these lipids also use glycerol as the polyol unit to which the acyl groups are attached, they are attached to the sn-2 and sn-3 positions (c.f. Sect. 1.1) rather than the sn-1 and sn-2 positions of the conventional diacylglycCOOH erols and phospholipids. The ether lipids may be classed as di- or tetraethers depending on the number of ether linkages that are present Spiculisporic acid in a particular molecule. Excellent reviews of these highly unusual XXV lipids have been written by DE ROSA and GAMBACORTA (1988), SMITH(1988), KATES et Of potential commercial relevance in this (1992, 1993), and by GAMBACORTA al., mono- (1995). A comprehensive monograph has also field but not included in KOSARIC'S graph, is spiculisporic acid (XXV) which is recently been published (KATES et al., 1993). produced at up to 110 g L-' by Penicillium The much greater chemical stability of these spiculisporum and whose possible uses have lipids than the acylated glycerol ester lipids of included acting as a surfactant or as a synthet- Eubacteria and the Eukaryota (the other two ic intermediate for such materials (ISHIGAMI, domains of the microbial world) has sug1993). Other fatty acid derivatives may also gested the means whereby the archaebacteria have a similar potential: the review by ISHI- are able to withstand temperatures of over 110C, salinities approaching that of saturated GAMI (1993) may, therefore, prove helpful in salt solution, and pH values of 1 or even less. delineation of these different molecules. The biotechnological applications of the lipids are not immediately apparent though they 6.2 Ether (Archaebacterial) Lipids may present interesting ideas to chemists for synthesis of novel compounds. The biotechThe newly-created bacterial domain of Ar- nological applications of the Archaea themchaea, which has arisen from the appreciation selves, however, are the subject of much curof the chemical and biochemical distinctive- rent speculative research; this area has been
182
4 Microbial Lipids
VH-0
' dJ J r / v d
I
a
~CH,-O
b
CH,OH
CH,OH
Fig. 9. Isoprenoid lipids of Archaea. a 2,3-di-O-phytanyl-sn-glycerol (archaeol) found in several genera; b 2,3-di-O-biphytanyl-sn-glycerol, a macrocyclic diether found in the thermophilic methanogen, Methanococcus, jannadschii; c Glycerol-dialkyl-glycerol tetraether (caldarchaeol) found in Sulfobolus and other genera; d Tetracyclized glycerol dialkylnonitol tetraether (cyclized nonitolcaldarchaeol) found in thermoacidophiles and some methanogens.
6 Other Lipids
183
recently reviewed by VENTOSAand NIETO Very few novel biotechnological applications (1995) and LEUSCHNER and ANTRANIKANhave been identified as the amounts are (1995) amongst others (see AGUILAR, 1995). usually less than 5% of the dry microbial biomass and growth of a microorganism specifically to produce a phospholipid would be clearly uneconomic. An attempt though was 6.3 Phospholipids and made in the early 1980s to extract the phosSphingolipids pholipid from bacteria being grown on methanol as a source of single cell protein Phospholipids, specially sn-1,2-diacyl glyc- (SMITH,1981). Some yeast lecithin (which erol-3-phosphorylated compounds, are found is the unfractionated phospholipid fraction of in all living cells save for the Archaea where which phosphatidylcholine is the predomialternative phospholipids occur (see Fig. 14). nant type) is produced for the health food The range of phospholipids in microorgan- market by extraction of spent brewers yeast. In most cases, large-scale commercial deisms is extensive and has been reviewed by the present author (RATLEDGE, 1987,1989b). mand for phospholipids is satisfied by using
R - CH - CH - CH20-Y I I OH NH I X I. Sphingosinebases
X=H; Y=H;R=
Cm(CH2)12CH=CHCH3(CH2)14CH~(CH~)I~CH(OH)Sphingosine Dihydrosphingosine Phytosphingosine Dehydrophytosphingosine
IIa. Ceramide phosphates
CH~(CH~)SCH=CH(CH&!CH(OH)1 . Ceramides 1
R=asinI; Y=H; X = CHNH2)nCH3(CH2)n-ICH(OH)CO(n=10 to 24)

111. Sphingomyelins
R=asinI;X=asinII; Y =
0
- 0 - P - OH
I 0
IV. Cerebrosides R=asinI; X=asinII; Y =
R = a s i n I ; X=asinII; Y =
0
Fig. 1 . Structures and nomenclature 0 of sphingolipids found in yeasts and fungi.
0-
II I
- CH2CH2k(CH3)3
- sugar (glucosyl or galactosyl) - phosphoinositol - phosphoinositol-rnannose
184
4 Microbial Lipids
plant-derived materials, which are described loosely as lecithin, and are recovered as byproducts following the refining of plant seed oils (see SZUHAJ,1989, for their applications in the food industry). However, phospholipOH ids also have important applications in the OH pharmaceutical industries and can be used for a number of functions (HANIN and ANSELL, Stearoylphytosphingone 1987). Such functions include the use of highly purified phosphatidylcholine as a lung surXXVI factant for neonatal children (BANGHAM, 1992) and its uses to improve biocompatibility of various medical devices including contact amides in which the base is either phytosphinand lenses, implants, and disposable devices used gosine or dehydrosphingosine. KULMACZ in human health care. It is now sold by a SCHROEPFER (1978) observed that addition to the growth number of commercial companies. The mate- of pentadecanoic acid rial may be produced by large-scale chroma- medium of Pichia ciferri increased the protographic purification but is more likely to be duction of the extracellular materials. This synthesized from glycero-sn-3-phosphocho- technique is now used to produce a number line, which can be produced from plant leci- of ceramides with this yeast. Patent applicathin, being reacted with ethyl esters of highly tions have been filed by Gist-Brocades (Nepurified individual fatty acids in the presence therlands) with respect to the production of of appropriate phospholipases (see RAT- N-stearoylphytosphingosine(XXVI) and othLEDGE,1994). Other uses for phospholipids er related compounds. The ceramides, after could include preparation of artificial lipo- purification, are used in cosmetic industry for somes which may be used as a means of de- the controlled release of dermatologically aclivery of chemotherapeutic agents to selected tive (or beneficial) compounds into the epitissues of a patient (HANINand ANSELL, dermal, dermal, and subcutaneous tissues of 1987). If such liposomes require the presence the skin. The yeast ceramides are regarded as of specific acyl groups or polar head groups identical to the ceramides that occur naturally then it may be possible to identify these in in the human skin; this includes the correct certain microorganisms: however, for reasons chirality at the three chiral centers in the molalready given it is unlikely that such lipids ecule (see XXVI). This ensures optimal percould be produced cost-effectively as the sole formance in various skin-care formulations. (I of microbial product from a biotechnological am grateful to Dr. H. STREEKSTRA GistBrocades for supplying the above informaprocess. Sphingolipids (see Fig. 10) are usually only tion.) minor components in microbial lipids. They occur in bacteria (WILKINSON,1988), in yeasts (RATTRAY, 1988; RATLEDGE and 6.4 Prostanoid-Type Lipids EVANS,1989), and in some fungi ( L ~ S E L , Prostaglandins, and the related leuco1988). All types of sphingolipid shown in Fig. 5, have been recognized in some microorgan- trienes and thromboxanes, exert unique physism. Their contents in cells are usually less iological control over many key metabolic sethan 0.5% of the dry biomass. The four sphin- quences in animals, including humans (SIN1992). Such materials are gosine bases (see Fig. 5) do not usually occur CLAIR and GIBSON, free but usually as acylated derivatives, or used clinically for induction of childbirth and ceramides. are also required by a large number of mediThe major microorganism of interest in this cal research groups. Prospects for producing area is Hansenula ciferri (now known as Pich- such materials microbiologically may be reia ciferri). This yeast appears to be unique in mote but some indication has been given that producing both intra- and extracellular cer- the yeast Dipodascopsis uninucleata may be
7 Conclusions
185
Fig. 11. ARA metabolites produced by Dipodascopsis uninucleata; a a-pentanor PGF2, y-lactone; b 3-HETE 3-hydroxyarachidonic acid (from VAN DYKet al., 1994).
Ho
OH
able to convert exogenously supplied arachidonic acid to a compound or compounds (see Fig. 11) showing prostaglandin-like activities when administered to experimental animals (KOCK et al., 1991; VAN DYK et al., 1991; KOCK and RATLEDGE, 1993). Current ongoing research should be able to determine within the next two or three years if such prospects of producing prostanoid lipids in this way are realistic. A recent review (VAN DYKet al., 1994) has indicated that the wellcharacterized oxygenase systems that occur in plants and animals for the formation of hydroxy PUFAs, which become the immediate precursors of the prostaglandins, may be found in some lower fungi (Saprolegniales and Lagenidiales) as well as species of Dipodascopsis.
7 Conclusions
Microbial lipids seemingly offer an almost bewildering array of possibilities for biotechnological exploitation. However, careful examination reveals that in many cases the oil or fat that is produced by a microorganism is not essentially different from that found in a plant oil, or occasionally, an animal fat. In these circumstances, the cost of the microbial route of production is likely to be many times that of the existing route. The only way that such processes could, therefore, be economically viable would be if the microbial oil was being produced as an adjunct to some waste disposal process. Such concepts were developed widely in the 1960s and 70s for the production of single cell proteins (SCP) which, although just selling, literally, as chicken feed, nevertheless produced a positive income to
offset the cost of waste disposal via a biotechnological route. With respect to a microbial oil - or single cell oil (SCO) - this too could be similarly produced. There are two obvious additional advantages with this alternative strategy to producing SCP. Firstly, the oil would, if carefully selected, be worth considerably more than just chicken-feed SCP. This is nicely illustrated by the attempt to produce a cocoa butter equivalent (CBE) in New Zealand using deproteinized whey as substrate (see Sect. 3.2.1.5). Even here though, given the comparatively high price of a CBE, there was still insufficient profit margin to proceed against rival plant products selling for about $2,000 per t. The second potential advantage of SCO over SCP is that the SCO need not be used for food or feed. SCP must be put back with the food chain by being fed to animals which, in turn, will be consumed, in whole or in part, by ourselves. This then requires that the substrate, or feedstock, is of an acceptable foodgrade quality or poses no long-term toxicological problems. SCO can, of course, when produced from food-acceptable substrates also be used in food materials (see, e.g., the yeast SCO-CBE product already cited), but an SCO could also be used for some technical purpose. In this case, the requirement for the substrate to be of food-grade quality no longer applies. In the extreme case, SCO could be produced from any fermentable substrate if it were subsequently put to some non-food use such as incorporation into paints, lubricants, detergents, plasticizers, etc. However, the cost-effectiveness of this scenario would have to rely almost entirely on the biological process being used in some form of waste removal or environmental cleanup process as the selling price of the non-food SCO would probably be quite low. The advantage would be
186
4 Microbial Lipids
that the SCO-producing microorganism achieves environmental cleanup and simultaneously produces a saleable end product more valuable than SCP. As it appears at the present time, the future of microbial oils probably lies outside these areas and the best chance for producing a commercially viable product is probably with high valued materials that are difficult to produce from plant or animal sources. This approach began in the 1980s with the use of fungi to produce polyunsaturated fatty acids (PUFA) and especially y-linolenic acid (GLA) that was available from only one or two plant sources at exceedingly high prices. However, as with any high-priced material which appears to generate considerable profit for one or two producers, other producers quickly began rival processes of the GLAproducing plant crops so that in a few short years the price of GLA-rich oils was halved and fell even lower. Profit margins which were sufficient to maintain commercial interest in producing SCO-GLA in the mid to late 1980s then disappeared under this downward commercial pressure. The scene in PUFAs was now moved on beyond GLA with other PUFAs, especially ARA and DHA, becoming the principal targets for microbial production. Whether these other PUFAs will follow the example of GLA and become more abundant, and thus less expensive, from other non-microbial sources will be a risk that the biotechnologist must, therefore, try to assess. In the longer term, though, it is probably these selected, very high-priced fatty acids, or derivatives from them, that will become the next SCOs to enter the marketplace. Certain lipids, besides those based on the triacylglycerols, are already being produced biotechnologically where there is a market opportunity at the correct (i.e., profitable) price level. Carotenoids, such as pcarotene and astaxanthin, are produced commercially (see Sect. 4.2) but need to be marketed very shrewdly to convince the purchaser that these microbial products are superior to the chemically-produced, and thus cheaper, products. As microorganisms tend to produce only one chiral isomer - and usually this is equivalent to the existing plant or animal material - then
this gives the microbial product a significant edge over the chemical product that rarely has the right chirality. Thus where stereospecificity is an important attribute of a product, a biotechnological route will usually be found to be superior to a chemical process. This is also seen in the formation of ceramides by yeast technology (Sect. 6.3). The range of microbial lipids is enormous. Which are of potential commercial value and which are of academic interest is hard to assess. Insight of a particular field may bring an appreciation to one person that a microbial lipid is of value but this may not be apparent to most others. Therefore, the perusal of the range of types of lipid molecules that are available from microbial sources could bring rich rewards to the shrewd reader. However, let me conclude by saying I have touched on only some of the microbial lipids; there are many others that have not been included directly here but may just have been referred to obliquely or en passant. Nevertheless, I hope that I have given sufficient references for the reader to pursue some of these more esoteric opportunities for themselves: fortune will always favor the prepared mind.
8 References
ACKMAN, (1994), Animal and marine lipids, in: R. Technological Advances in Improved and AlterB. native Sources of Lipids (KAMEL, s., KAKUDA, Y., Eds.), pp. 292-328. London: Blackie. AGUILAR, (Ed.) (1995), Special Review Topic: A. Biotechnology of Extremophile Microorganisms, World J. Microbiol. Biotechnol. 11, issue NO. 1, pp. 7-131. T. E. AHERN, J., KATOH,S., SADA, (1983), Arachidonic acid production by the red alga Porphyridium cruentum, Physiol. Plant. 16,36-643. K., AKIMOTO, ISHII,T., YAMAGAKI, OHTAGUM., CHI, K., KOIDE, K., YAZAWA, (1990), ProK. duction of eicosapentaenoic acid by a bacterium isolated from mackerel intestines, J. Am. Oil Chem. SOC. 67,911-915. K., AKIMOTO, ISHII,T., YAMAGAKI, OHTAGUM., CHI, K., KOIDE, K., YAZAWA, (1991), Metal K. salts requisite for the production of eicosapentaenoic acid by a marine bacterium isolated from mackerel intestines, J. Am. Oil Chem. SOC. 68, 504-508.
8 References
AMANO, SHINMEN, AKIMOTO, KAWAN., Y., K., H., T., S., SHIMA, AMACHI, SHIMIZU, YAMADA, H. (1992), Chemotaxonomic significance of fatty acid composition in the genus Mortierella (Zygomycetes, Mortierellaceae), Mycotaxon 44, 257265. AN, G.-H., SCHUMAN, B., JOHNSON, A. D. E. (1989), Isolation of Phafla rhodozyma mutants with increased astaxanthin content, Appl. Environ. Microbiol. 55, 116-123. AN, G. H., BIELICH, AUERBACH, JOHNSON, J., R., E. A. (1991), Isolation and characterization of carotenoid hyperproducing mutants of yeast by flow cytometry and cell sorting, Biotechnology 9, 70-73. A. ANDERSON, J., DAWES,E. A. (1990), Occurrence, metabolism, metabolic role, and industrial uses of bacterial polyhydroxyalkanoates, Microbiol. Rev. 54, 450-472. BAILLIEZ,C., LARGEAN, CASADEVALL, C., E. (1985), Growth and hydrocarbon production of Botryococcus braunii immobilized in calcium alginate gel, Appl. Microbiol. Biotechnol. 23, 99105. BAILLIEZ, LARGEAN,C.,BERKALOFF, C., C., CASADEVALL, (1986), Immobilization of BotryoE. coccus braunii in alginate: influence on chlorophyll content, photosynthetic activity and degeneration during batch cultures, Appl. Microbiol. Biotechnol. 23,361-366. BAILLIEZ, LARGEAN, CASADEVALL, C., C., E., YANG, W., BERKALOFF, (1988), PhotosynL. C. thesis, growth and hydrocarbon production of Botryococcus braunii immobilized by entrapment and adsorption in polyurethane foams, Appl. Microbiol. Biotechnol. 29, 141-147. BAJPAI, BAJPAI, K. (1992), Arachidonic acid P., P. production by microorganisms, Biotechnol. Appl. Biochem. 15, 1-10. BAJPAI, K., BAJPAI, WARD,0. P. (1991a), P. P., Optimization of production of docosahexaenoic acid (DHA) by Thraustochytrium aureum ATCC 34304, J. Am. Oil Chem. SOC. 68, 509514. BAJPAI, BAJPAI, K., WARD,0. P. (1991b), P., P. Production of docasohexaenoic acid by Thraustochytrium aureum, Appl. Microbiol. Biotechnol. 35, 706-710. BAJPAI, BAJPAI, K., WARD,0. P. (1991c), P., P. Effects of ageing Mortierella mycelium on production of arachidonic and eicosapentaenoic acids, J. Am. Oil Chem. SOC.68,775-780. BAJPAI, K., BAJPAI, WARD,0. P. (1992), OpP. P., timization of culture conditions for production of eicosapentaenoic acid by Mortierella elongata NRRL 5513, J. Ind. Microbiol. 9, 11-18.
187
BANGHAM, (1992), Surface tension in lungs, A. Nature (London) 359, 110. BARNETT,J. A., PAYNE,R. W., YARROW,D. (1990) Yeasts; Characteristics and Identification, 2nd Edn. Cambridge: University Press. BASIRON, IBRAHIM, (1994), Palm oil: old Y., A. myths new facts, INFORM 5, 977-980. BEAVAN,M., KLIGERMAN, DRONIUK, A., R., DROUIN, GOLDENBERG, EFFIO,A., Yu, C., B., P., GIULIANY, FEIN, (1992), Production of B., J. microbial cocoa butter equivalents, in: Industrial Applications of Single Cell Oils (KYLE, D. J., RATLEDGE, Eds.), pp. 156-184. Champaign, C., I L American Oil Chemists Society. BECKER, W. (1993), Microalgae: Biotechnology E. and Microbiology. Cambridge: University Press. BEDNARSKI, LEMAN, TOMASIK, (1986), W., J., J. Utilization of beet molasses and whey for fat biosynthesis by a yeast, Agric. Wastes 18, 19-26. BEHALOVA, VORISEK, (1988), Increased steB., J. rol formation in Saccharomyces cerevisiae. Analysis of cell components and ultrastructure of vacuoles, Folia Microbiol. 33, 292-297. A., AVRON,M. (1980), Glycerol, BEN-AMOTZ, beta-carotene and dry algal meal production by commercial cultivation of Dunaliella, in Algae G., C. Biomass (SHELEF, SOEDER, J., Eds.), pp. 603-610. Amsterdam: Elsevier. BEN-AMOTZ, AVRON, (1983), On factors A., M. which determine massive p-carotene accumulation in the halotolerant alga Dunaliella bardawil, Plant Physiol. 72, 593-597. BEN-AMOTZ, LERS, A., AVRON, A., M. (1988), Steroisomers of pcarotene and phytoene in the alga Dunaliella bardawil, Plant Physiol. 86, 1286-1291. BEN-AMOTZ, MOKADY, EDELSTEIN, AvA,, S., S., RON, M. (1989), Bioavailability of natural isomer mixture as compared with all-trans-p-carotene in rats and chicks, J. Nutr. 119, 1013-1019. K., H. BERNHARD, ALBRECHT, (1948), Die Lipide aus Phycomyces blakesleeanus, Helv. Chim. Acta 31,977-988. BOROWITZKA, A. (1988), Fats, oils and hydroM. carbon, in: Micro-Algal Biotechnology (BOROWITZKA, A., BOROWITZKA, J., Eds.), pp. M. L. 257-287. Cambridge: University Press. BOROWITZKA, A. (1992), Comparing carotenoM. genesis in Dunalliella and Haematococcus: implications for commercial production strategies, in: Profiles on Biotechnology (VILLA, T . G., ABALDE, Eds.), pp. 301-310. Servicio de PubJ., lications, Universidade de Santiago, Spain. BOROWITZKA, A., BOROWITZKA, J. (Eds.) M. L. (1988), Micro-Algal Biotechnology, Cambridge: University Press. BOROWITZKA, J., BOROWITZKA, A. (1989), L. M.
188
4 Microbial Lipids
PCarotene (provitamin A) production with algae, in: Biotechnology of Vitamins, Pigments and E. Growth Factors (VANDAMME, J., Ed.), pp. 15-26. London: Elsevier Applied Science. BOROWITZKA, J., BOROWITZKA, A. (1990), L. M. Commercial production of p-carotene by Dunaliella salina in open ponds, Bull. Marine Sci. 41, 244-252. BOSWELL, D. V., GLADUE,R., PRIMA,B., K. KYLE,D. J. (1992), SCO production by fermentative microalgae, in: Industrial Applications of Single Cell Oils (KYLE,D. J., RATLEDGE, C., Eds.), pp. 274-286. Champaign, IL: American Oil Chemists Society. BOTHAM, A., RATLEDGE, (1979), A biochemP. C. ical explanation for lipid accumulation in Candida 107 and other oleaginous micro-organisms,J. Gen. Microbiol. 114, 351-375. BOULTON, A. (1982), The biochemistry of lipid C. accumulation in oleaginous yeasts, Ph. D. Thesis, University of Hull, UK. BOULTON, A., RATLEDGE, (1981), CorrelaC. C. tion of lipid concentration in yeasts with possession of ATP:citrate lyase, J. Gen. Microbiol. 127, 169-176. BOULTON, A., RATLEDGE, (1983a). Use of C. C. transition studies in continuous cultures of Lipomyces starkeyi, an oleaginous yeast, to investigate the physiology of lipid accumulation, J. Gen. Microbiol. 129, 2871-2876. BOULTON, A., RATLEDGE, (1983b), Partial C. C. purification and some properties of ATP:citrate lyase from the oleaginous yeast Lipomyces starkeyi, J. Gen. Microbiol. 129,2863-2869. BRAMLEY, M. (1985), The in vitro biosynthesis P. of carotenoids, Adv. Lipid Res. 21,243-279. BRAMLEY, M., MACKENZIE, (1988), RegulaP. A. tion of carotenoid biosynthesis, Curr. Top. Cell. Regul. 29,291-343. BYROM, (1990), Industrial production of copoD. lymer from Alcaligenes eutrophus, in: Novel Biodegradable Microbial Polymers (DAWES, E. A., Ed.), pp. 113-117 (NATO AS1 Series No 186). Dordrecht: Kluwer. BYROM, (1992), Production of poly-p-hydroxyD. butyrate: poly-p-valerate copolymers, FEMS Microbiol. Rev. 103, 247-250. G. S. CARMAN, M., HENRY, A. (1989), Phospholipid biosynthesis in yeast, Annu. Rev. Biochem. 58, 635-669. CASADEVALL, DIF, D., LARGEAU, GUDIN, E., C., C., CHANMON, DESANTI, (1985), Studies D., 0. on batch and continuous cultures of Botryococcus braunii, hydrocarbon production in relation to physiological state, cell ultrastructure and phosphate nutrition, Biotechnol. Bioeng. 27, 286-295.
CERDA-OLMEDO, (1989), Production of caroteE. noids by fungi, in: Biotechnology of Vitamins, Pigments and Growth Factors (VANDAMME, E. J., Ed.), pp. 27-42. London: Elsevier Applied Science. O., 0. CIFERRI, TIBONI, (1985), The biochemistry and industrial potential of Spirulina, Ann. Rev. Microbiol. 39, 503-526. G. CLEMENT, (1975), Producing Spirulina with C02, in: Single Cell Protein, Vol. 2 (TANNENS. BAUM, R., WANG,D. I. C., Eds.), pp. 79-89. Cambridge, MA: M.I.T. Press. COHEN, COHEN, (1991), Preparation of eicoZ., S. sapentaenoic acid (EPA) concentrate from Porphyridium cruentum, J. Am. Oil Chem. SOC. 68, 16-19. COHEN,Z., HEIMER, M. (1992), Production of Y. polyunsaturated fatty acid (EPA, ARA and GLA) by the microalgae Porphyrium and Spirulina, in: Industrial Applications of Single Cell Oils (KYLE, D. J., RATLEDGE, Eds.), pp. C., 243-273, Champaign, I L American Oil Chemists Society. COHEN, VONSHAK, RICHMOND, (1988), Z., A., A. Effect of environmental conditions on fatty acid composition of the red alga Porphyridium cruenrum: correlation to growth rate, J. Phycol. 24, 328-332. COHEN, DIDI,S., HEIMER, M. (1992), OverZ., Y. production of y-linolenic acid and eicosapentaenoic acids by algae, Plant Physiol. 98, 569-572. COOLBEAR, THRELFALL, R. (1989), BiosynT., D. thesis of terpenoid lipids, in: Microbial Lipids, Vol. 2 (RATLEDGE, WILKINSON,G., Eds.), C., S. pp. 115-254, London: Academic Press. CPC International INC. (1979), Process for the microbiological production of oil, Eur. Patent 0005 277. CPC International INC. (1982a), Multistage process for the preparation of fats and oils, Brit. Patent 2091285A. CPC International INC. (1982b), Preparation of fats and oils, Brit. Patent 2091286A. CRESSWELL, C., REES, T. A., SHAH, (Eds.) R. N. (1989), Algal and Cyanobacterial Technology. London: Longmans. DAVIES, (1984), Oil from whey, Food Technol. R. N. 2. (June), 33-37. DAVIES, J. (1988), Yeast oil from cheese whey R. process development, in: Single Cell Oil (MORETON, R. S., Ed.), pp. 99-145. Harlow: Longman. DAVIES, J. (1992a), Advances in lipid processing R. in New Zealand, Lipid Technol. 4, 6-13. DAVIES, J. (1992b), Scale up of yeast oil techR. nology, in: Industrial Applications of Single Cell
8 References
Oils (KYLE,D. J., RATLEDGE, Eds.), pp. C., 196-218. Champaign, I L American Oil Chemists Society. DAVIES, J., HOLDSWORTH,E. (1992), SyntheR. J. sis of lipids in yeasts: biochemistry, physiology, production, Adv. Appl. Lipid Res. 1, 119-159. DAVIES, J., HOLDSWORTH,E., READER, L. R. J. S. (1990), The effect of low oxygen uptake rate on the fatty acid profile of the oleaginous yeast, Apiotrichum curvatum, Appl. Microbiol. Biotechnol. 33, 569-573 (erratum: ibid. 1991, 34, 832-833). DAWES,E. A. (Ed.) (1990), Novel Biodegradable Microbial Polymers, (NATO AS1 Series No. 186). Dordrecht: Kluwer. DE KONING, (1995), Physical properties of bacG. terial poly((R)-3-hydroxyalkanoates), Can. J. Microbiol. 41 (Suppl. l), 303-309. DE LONG,E. F., YAYANOS, A. (1986), BioA. chemical function and ecological significance of novel bacterial lipids in deep-sea prokaryotes, Appl. Environ. Microbiol. 51, 730-737. DE ROSA,M., GAMBACORTA, (1988), The lipA. ids of Archaebacteria, Prog. Lipid Res. 27, 153175. DELONG, F., YAYANOS, A. (1986), BiochemE. A. ical function and ecological significance of novel bacterial lipids in deep-sea prokaryotes, Appl. Environ. Microbiol, 51, 730-737. Dot, Y. (1990), Microbial Polyesters. Weinheim: VCH. DOSTALEK, (1986), Production of lipid from M. starch by a nitrogen-controlled mixed culture of Saccharomyces fibuliger and Rhodosporidium toruloides, Appl. Microbiol. Biotechnol. 24, 1923. DUBINSKY, BERNER, AARONSON, (1978), Z., T., S. Potential of large-scale algal culture for biomass and lipid production in arid lands, Biotechnol. Bioeng. Symp. 8, 51-68. DULANEY, L. (1957), Preparation of ergosterol E. containing yeasts, US.Patent 2817624. DULANEY, L., STAPLEY,E. O., SIMPF, K. E. (1954), Studies on ergosterol production by yeasts, Appl. Microbiol. 2,371-379. EL-REFAI,A. E. M., EL-KADY, A. (1968a), SteI. rol production of yeast strains, Z. Allg. Mikrobwl. 8,355-360. EL-REFAI, E. M., EL-KADY, A. (1968b), SteA. I. rol biosynthesis in Saccharomyces fermentati, Z. A&. Mikrobiol. 8, 361-366. I. EL-REFAI, E. M., EL-KADY, A. (1%9), UtiliA. zation of some industrial by-products for the microbiological production of sterols from Saccharomyces fermentati, J. Bot. U.A.R. 12, 55-66. ELLENBOGEN, A., AARONSON, GOLDSTEIN, B. S.,
189
S., BELSKY,M. (1969), Polyunsaturated fatty acids of aquatic fungi: possible phytogenetic significance, Comp. Biochem. Physiol. 29, 805-811. EROSHIN, K., KRYLOVA, I. (1983), Efficiency V. N. of lipid synthesis by yeasts, Biotechnol. Bioeng. 25,1693-1700. EVANS, T. (1983), The control of metabolism in C. lipid-accumulating yeasts, Ph. D. Thesis, University of Hull, UK. EVANS,C. T., RATLEDGE, (1983), A compariC. son of the oleaginous yeast, Candida curvata, grown on different carbon sources in continuous and batch culture, Lipids 18,623-629. EVANS, T., RATLEDGE, (1984a), Effect of niC. C. trogen source on lipid accumulation in oleaginous yeasts, J. Gen. Microbiol. 130,1693-1704. EVANS, T., RATLEDGE, (1984b), Influence of C. C. nitrogen metabolism on lipid accumulation by Rhodosporidium toruloides CBS 14, J. Gen. Microbiol. 130, 1705-1710. EVANS,C. T., RATLEDGE, (1984c), PhosphoC. fructokinase and the regulation of the flux of carbon from glucose to lipid in the oleaginous yeast Rhodosporidium toruloides, J. Gen. Microbiol. 130,3251-3264. EVANS, T., RATLEDGE, (1985a), Partial puriC. C. fication and properties of pyruvate kinase and its regulatory role during lipid accumulation by the oleaginous yeast Rhodosporidium toruloides CBS 14, Can. J. Microbiol. 31, 479-484. EVANS, T., RATLEDGE, (1985b), The role of C. C. the mitochondria1 NAD :isocitrate dehydrogenase in lipid accumulation by the oleaginous yeast Rhodosporidium toruloides CBS 14, Can. J. Microbiol. 31, 845-850. EVANS,C. T., RATLEDGE, C. (1985c), Possible regulatory roles of ATP:citrate lyase, malic enzyme and AMP deaminase in lipid accumulation by Rhodosporidium toruloides CBS 14, Can. J. Microbiol. 31, 1000-1005. EVANS,C. T., RATLEDGE, (1986), The physioC. logical significance of citric acid in the control of metabolism in lipid-accumulating yeasts, Biotechnol. Gen. Eng. Rev. 3,349-375. EVANS,C. T., SCRAGG,A. H., RATLEDGE, C. (1983), A comparative study of citrate efflux from mitochondria of oleaginous and non-oleaginous yeasts, Eur. J. Biochem. 130, 195-204. FALL,R., PHELPS,P., SPINDLER, (1984), BioD. conversion of xylan to triglycerides by oil-rich yeasts, Appl. Environ. Microbiol. 47, 11301134. FLOETENMEYER, D., GLATZ, B. A., HAMM. MOND, G. (1985), Continuous culture fermenE. tation of whey permeate to produce microbial oil, J. Dairy Sci. 68, 633-637.
+
190
4 Microbial Lipids HANSSON, DOSTALEK, (1989), Effect of culL., M. ture conditions on mycelial growth and production of ylinolenic acid by the fungus Mucor ramanniana, Appl. Microbiol. Biotechnol. 28,240246. HARRISON, M. (1986), The biosynthesis of caroD. tenoids, Nut. Prod. Rep. 3, 205-215. HARWOOD, L. (1994a), Environmental effects on J. plant lipid metabolism, INFORM, 5, 835-839. HARWOOD, L. (1994b), Environmental factors J. which can alter lipid metabolism, Prog. Lipid Res. 33, 193-202. HASSAN, BLANC, J., GRANGER, M., PARM., P. L. EILLEUX, GOMA,G. (1993), Lipid producA., tion by an unsaturated fatty acid auxotroph of the oleaginous yeast Apiotrichum curvatum grown in single-stage continuous culture, Appl. 0 Microbiol. Biotechnol. 4 ,483-488. HASSAN,M., BLANC, P. J., PAREILLEUX, A., GOMA,G. (1994a), Selection of fatty acid auxotrophs from the oleaginous yeast Cryptococcus curvatus and production of cocoa butter equivalents in batch culture, Biotechnol. Lett. 16, 819824. HASSAN,M., BLANC, P. J., PAREILLEUX, A., GOMA,G. (1994b), Production of single cell oil from prickly pear juice fermentation by Cryptococcus curvatus grown in batch culture, World J. Microbiol. Biotechnol. 10, 534-537. HENDERSON, J., LEFTLEY,J. W., SARGENT, R. J. R. (1988), Lipid composition and biosynthesis in the marine dinoflagellate Crypthecodium cohnei, Phytochemistry 27, 1679-1683. HENDERSON, J., MILLAR,R. M., SARGENT, R. J. R., JOSTENSEN, P. (1993), Trans-neonoenoic J. and polyunsaturated fatty acids in phospholipids of a Vibrio species in relation to growth conditions, Lipids, 28, 389-396. HESSE,A. (1949) Industrial biosyntheses, part I. fats, Adv. Entymol. 9, 653-704. HILLEN, L. W., POLLARD,G., WAKE, L. V., WHITE,N. (1982), Hydrocarbon of the oils of Bdtryococcus braunii to transport fuels, Biotechnol. Bioeng. 24, 193-205. HIRANO, MORI,H., MIURA,Y., MATSUNAGA, M., N., NAKAMURA, MATSUNAGA, (1990), yN., T. Linolenic acid production by microalgae, Appl. Biochem. Biotechnol. 24/25, 183-191. HODGSON, A., HENDERSON, J., SARGENT, P. K. J. R ,LEFTLEY, W. (1991), Patterns of variation . J. in the lipid class and fatty acid composition of Nannochloropsis oculata (Eustigmatophyceae) during batch culture, I. The growth cycle, J. Appl. PhyCOl. 3, 169-181. HOLDSWORTH, E., RATLEDGE, (1988), Lipid J. C. turnover in oleaginous yeasts, J. Gen. Microbiol. 134,339-346.
Fuji Oil CO. LTD. (1979), Method for producing cacao butter substitute, Brit. Patent 1555OOO. Fuji Oil CO. LTD. (1981), Method for producing cacao butter substitute, U.S. Patent 4268527. GAMBACORTA, GLIOZZI, DE ROSA, A. A,, A., (1995), Archaeal lipids and their biotechnological applications, World J. Microbial Biotechnol. U, 115-131. GANDHI, R., WEETE, D. (1991), Production of S. J. the polyunsaturated fatty acids arachidonic acid and eicosapentaenoic acid by the fungus Phytium ultimum, J. Gen. Microbiol. 137, 18251830. GELLERMAN, L., SCHLENK, (1979), MethylJ. H. directed desaturation of arachidonic to eicosapentaenoic acid in the fungus Saprolegnia parasitica, Biochim. Biophys. Acta 573,23-30. GLATZ, B. A., FLOETENMEYER, D., HAMM. MOND, G. (1985), Fermentation of bananas E. and other food wastes to produce microbial lipid, J. Food Prot. 48,574-577. GOODWIN, W. (1980), The Biochemistry of the T. Carotenoids, 2nd Edn., Vol. 1. London: Chapman & Hall. GOODWIN, W. (1983), The Biochemistry of the T. Carotenoids, 2nd Edn., Vol. 2. London: Chapman & Hall. GRANGER, M., PERLOT, GOMA, PAREILL. P., G., LEUX, (1993), Effect of various nutrient limiA. tations on fatty acid production by Rhodotorula glutinis, Appl. Microbiol. Biotechnol. 38, 784789. GUCKERT, B., COOKSEY, E. (1990), TriglycerJ. K. ide accumulation and fatty acid profile changes in Chlorella (Chlorophyta) during high pH-induced cell cycle inhibition, J. Phycol. 26, 72-79. GUERZONI, E., LAMBERTINI, LERCKER, M. P., G., MARCHEITI,R. (1989, Technological potential of some starch degrading yeasts, StarcWStarke 37,52-57. HAARD,N. F. (1988), Astaxanthin formation by the yeast Phafia rhodozyma on molasses, Biotechnol. Lett. 10, 609-614. HAMMOND, G., GLATZ,B. A. (1990), BiotechE. nology applied to fats and oils, in: Food Technology, Vol. 2 (KING,R. D., CHEETHAM, S. J., P. Eds.), pp. 173-217. London: Elsevier Applied Science. HANIN, ANSELL, B. (Eds.) (1987), Lecithin: I., G. Technological, Biological and Therapeutic Aspects. New York: Plenum Press. HANSSON, DOSTALEK, (1988), Effect of culL., M. ture conditions on mycelial growth and production of ylinolenic acid by the fungus Mortierella rarnanniana, Appl. Microbiol. Biotechnol. 28, 240-246.
8 References
HOLDSWORTH, E., VEENHUIS, RATLEDGE, J. M., C. (1988), Enzyme activities in oleaginous yeasts accumulating and utilizing exogenous or endogenous lipids, J. Gen. Microbiol. 134, 2907-2915. HORROBIN, F. (1992), Clinical applications of D. n-6 essential fatty acids: atopic eczema and inflammation, diabetic neuropathy and retinopathy, breast pain and viral infections, in: Essential Fatty Acids and Eicosanoids, 3rd Int. Congr. (SINCLAIR, GIBSON, Eds.), pp. 367-372. A., R., Champaign: IL: American Oil Chemists SocietY. HOUSTON, NIMMO, G. (1984), Purification B., H. and some kinetic properties of rat liver ATP:citrate lyase, Biochem. J. 224, 437443. HOUSTON, NIMMO, H. G. (1985), Effects of B., phosphorylation on the kinetic properties of rat liver ATP:citrate lyase, Biochim. Biophys. Acta 884,233-239. HRABAK, (1992), Industrial production of poly0. bhydroxybutyrate, FEMS Microb. Rev. 103, 251-256. HUDSON, J. F., KARIS, G. (1974), The lipids of B. I. the alga Spirulina, J. Sci. Food Agric. 25, 759763. ISHIGAMI, (1993), Biosurfactants face increasing Y. interest, INFORM 4, 1156-1165. JAREONKITMONGKOL,KAWASHIMA, SHIS., H., S., H. RASAKA, SHIMIZU, YAMADA, (1992a), N., Production of dihomo-y-linolenic acid by a A5desaturase-defective mutant of Mortierella alpinu 1s-4, Appl. Environ. Microbiol. 58, 21962200. JAREONKITMONGKOL, S., KAWASHIMA, SHIMIH., H. zu, S., YAMADA, (1992b), Production of 5,8,11-cis-eicosatrienoic acid by a A12-desaturase-defective mutant of Mortierella alpina 1s-4, J. Amer. Oil Chem. SOC. 69, 939-944. JAREONKITMONGKOL, S., SHIMIZU, YAMADA, S., H. (1992c), Fatty acid desaturation-defective mutants of an arachidonic acid-producing fungus, Mortierella alpina 1s-4, J. Gen. Microbiol. 138,997-1002. JENSEN, (1966), Report No. 31, Norwegian InstiA. tute of Seaweed Research. G. JOHNS,R. B., PERRY, J. (1977), Lipids of the marine bacterium Flexibacter polymorphus, Arch. Microbiol. 114, 267-271. JOHNSON, A., AN, G. H. (1991), Astaxanthin E. from microbial sources, CRC Crit. Rev. Biotechnol. 11,297-321. JOHNSON, A., CONKLIN, E., LEWIS,M. J. E. D. (1977), The yeast Phaffia rhodozyma as a dietary pigment source for salmonids and crustaceans, J. Fish. Res. Board Can. 34,2417-2421. JOHNSON, A., VILLA, G., LEWIS, J. (1980), E. T. M.
191
Phaffia rhodozyma as an astaxanthin source in salmonid diets, Aquaculture 20, 123-134. KATES,M. (1992), Archaebacterial lipids: structure, biosynthesis and function, Biochem. SOC. Symp. 58,51-72. KATES, M. (1993), Membrane lipids of extreme halophytes: biosynthesis, function and evolutionary significance, Experientia 49, 1027-1036. KATES,M., KUSHNER, J., MATHESON, T. D. A. (Eds.) (1993), The Biochemistry of Archaea (Archaebacteria). Amsterdam: Elsevier Science Publishers B.V. KAWABATA, A., KANEYAMA, (1989), The efM. fect of growth temperature on wax ester content and composition of Euglena gracilis, J. Gen. Microbiol. 135, 1461-1467. KENDRICK, RATLEDGE, (1992a), Lipids of A., C. selected molds grown for the production of n-3 and n-6 polyunsaturated fatty acids, Lipids 27, 15-20. KENDRICK, RATLEDGE, (1992b), Lipid forA., C. mation in the oleaginous mold Entomophthora exitalis grown in continuous culture: effects of growth rate, temperature and dissolved oxygen tension on polyunsaturated fatty acids, Appl. Microbiol. Biotechnol. 37, 18-22. KENDRICK, RATLEDGE, (1992c), DesaturaA., C. tion of polyunsaturated fatty acids in Mucor circinelloides and the involvement of a novel membrane-bound malic enzyme, Eur. J. Biochem. 209,667-673. KENDRICK, RATLEDGE, (1992d), PhosphoA., C. lipid fatty acyl distribution of three fungi indicates positional specificity for n-6 vs. n-3 fatty acids, Lipids 27, 505-508. KENNEDY, J., READER,S. L., DAVIES, J. M. R. (1993), Fatty acid production characteristics of fungi with particular emphasis on y-linolenic acid, Biotechnol. Bioeng. 42, 625-634. KERNON, (1992), Vegetable fats in chocolate J. EC and US legislation issues, Lipid Technol. 4, 113-114. KOCK,J. L. F., RATLEDGE, (1993), Changes in C. lipid composition and arachidonic turnover during the life cycle of the yeast Dipodascopsis uninucleata, J. Gen. Microbiol. 139, 459464. D. KOCK,J. L. F., COETZEE, J., VAN DYK,M. S., TRUSCOTT, CLOETE, C., VAN WYK,V., M., F. 0. AUGUSTYN, P. H. (1991), Evidence for pharmacologically active prostaglandin in yeasts, South Afr.J. Sci. 87, 73-76. KORITALA, (1989), Microbiologial synthesis of S. wax esters by Euglena gracilis, J. Am. Oil Chem. SOC. 66,133-134. KOSARIC, (Ed.) (1993), Biosurfactants: ProducN. tion, Properties, Applications. New York Marcel Dekker.
192
4 Microbial Lipids
KREGER-VAN RIJ, N. J. W. (Ed.) (1984), The Yeasts. A Taxonomic Study, 3rd Edn. Amsterdam: Elsevier Science Publishers. KULMACZ, J., SCHROEPFER, J. (1978), SphinR. G. golipid base metabolism, J. Am. Chem. Soc. 100, 3963-3964. KYLE, D. J. (1991), Speciality oils from microalgae: new perspectives, in: Biotechnology of Plant Fats J. and Oils (RATTRAY, Ed.), pp. 130-143. Champaign, IL: American Oil Chemists Society. KYLE,D. J. (1992), Production and use of lipids from microalgae, Lipid Technol. 4, 59-64. KYLE,D. J., RATLEDGE, (1992), Industrial ApC. plications of Single Cell Oils, Champaign, IL: American Oil Chemists Society. P., S., KYLE,D. J., BEHRENS, BINGHAM, ARNETT, K., LIEBERMAN, (1988), Microalgae as a D. source of EPA-containing oils, in: Biotechnology T. for the Fats and Oils Industry (APPLEWHITE, H., Ed.), pp. 117-121. Champaign, IL: American Oil Chemists Society. V. KYLE,D. J., SICOTTE, J., REEB,S. E. (1992), Bioproduction of docosahexaenoic acid (DHA) by microalgae, in: Industrial Applications of SinC., gle Cell Oils (KYLE,D. J., RATLEDGE, Eds.), pp. 287-300. Champaign, I L American Oil Chemists Society. LAGARDE, BURTIN,M., RIGAND,M., SPREM., CHER, H., DECHAVANNE, M., RENAND,S. (1985), Prostaglandin E2 - like activity of 2 0 3n-9 platelet lipoxygenase end-product, FEBS Lett. 181, 53-56. LEE, Y. K. (1986), Enclosed bioreactors for the mass cultivation of photosynthetic microorganisms: the future trend, Trends Biotechnol. 4,186189. LEONARD, C. (1994), Salim group - major factor E. in oil palm trade, INFORM 5, 987-992. LEUSCHNER, ANTRANIKAN, (1995), HeatC., G. stable enzymes from extreme thermophilic and hyperthermophilic microorganisms, World J. Microbiol. Biotechnol. l l , 95-114. LOPEZ ALONSO,D., MOLINAGRIMA,E., SANCHEZ PEREZ,J. A., GARCIA SANCHEZ, L., J. GARCIA CAMACHO, (1992), Fatty acid variaF. tion among different isolates of a single strain of Isochrysis galbana, Phytochemistry 31, 39013904. LOSEL, D. M. (1988), Fungal lipids, in: Microbial C., S. Lipids, Vol. 2 (RATLEDGE, WILKINSON, G. Eds.), pp. 699-806. London: Academic Press. MARGALITH, (1989), Vitamin D: the biotechnolP. ogy of ergosterol, in: Biotechnology of Vitamins, E. Pigments and Growth Factors (VANDAMME, J., Ed.), pp. 81-104. London: Elsevier Applied Science.
MARGALITH, Z. (1992), Pigment Microbiology. P. London: Chapman & Hall. W. MEAD,J. F., SLATON, H. (1956), Metabolism of essential fatty acids. 111. Isolation of 5,8,11-eicosatrieonic acid from fat-deficient rats, J. Biol. Chem. 219,705-709. MEYER,P. S., DU PREEZ, J. C., KILIAN, G. S. (1993), Selection and evaluation of astaxanthinoverproducing mutants of Phaffia rhodozyma, World J. Microbiol. Biotechnol. 9, 514-520. S. MIELKE, (1992), Trends in supply, consumption and prices, in: Oils and Fats in the Nineties (SHUKLA, K. S., GUNSTONE, D., Eds.), pp. V. F. 10-22. Lystrup, Denmark: International Food Science Centre AIS. MILLER, KOCK, J. L. F., BOTES,P. J. (1989a), M., The significance of long-chain fatty acid compositions and other phenotypic characteristics in the yeast Pichia Hansen emend. Kurtzman, Syst. Appl. Microbiol. 12, 70-79. MOLINA GRIMA, SANCHEZ E., PEREZ,J. A., GARCIA SANCHEZ, L. (1992), EPA from Isochrysis J. galbana. Growth conditions and productivity, Process Biochem. 21, 299-305. MOLINA GRIMA, SANCHEZ E., PEREZ,J. A., GARCIA CAMACHO, F., GARCIASANCHEZ, L., J. LOPEZALONSO, (1993), n-3 PUFA producD. tivity in chemostat cultures of microalgae, Appl. Microbiol. Biotechnol. 38, 599-605. MOLINA GRIMA, SANCHEZ E., PEREZ,J. A., GARCIA CAMACHO, FERNANDEZ F., SEVILLA, M., J. ACIEN FERNANDEZ, G. (1994), Effect of F. growth rate on the eicosapentaenoic acid and docasahexaenoic acid content of Isochrysis galbana in chemostat culture, Appl. Microbiol. Biotechnol. 41,23-27. E. MOON, N. J., HAMMOND, G., GLATZ,B. A. (1978), Conversion of cheese whey and whey permeate to oil and single-cell protein, J. Dairy Sci. 61, 1537-1547. MORETON, S. (1985), Modification of fatty acid R. composition of lipid accumulating yeasts with cyclopropene fatty acid desaturase inhibitors, Appl. Microbiol. Biotechnol. 33,4145. MORETON, S. (1988a), Single Cell Oil. Harlow: R. Longman. MORETON, S. (1988b), The physiology of lipid R. accumulating yeasts, in: Single Cell Oil (MORETON,R. S., Ed.), pp. 1-32. Harlow: Longman. MORETON, S., CLODE,D. M. (1985), Microbial R. desaturase enzyme inhibitors and their use in a method of producing lipids. Brit. Patent 8407 195. F. I. I., MURILLO, J., CALDERON, L., LOPEZ-DIAZ, CERDA-OLMENDO, (1978), Carotene-superE. producing strains of Phycomyces, Appl. Environ. Microbiol, 36. 477-479.
8 References
MURPHY, J. (1994a), Transgenic plants - a fuD. ture source of novel edible and industrial oils, Lipid Technol. 6, 84-91. MURPHY, J. (Ed.) (1994b), Designer Crop Oils. D. Weinheim: VCH. NAKAHARA, YOKOCKI, KAMISAKA, SuT., T., Y., ZUKI, 0.(1992), Gamma-linolenic acid from genus Mortierella, in: Industrial Applications of C., Single Cell Oils (KYLE,D. J., RATLEDGE, Eds.), pp. 61-97. Champaign, I L American Oil Chemists Society. NAKAJIMA, Izu, S. (1992), Microbial producT., tion and purification of w-6 polyunsaturated fatty acids, in: Essential Fatty Acids and EicosaA., R., noids (SINCLAIR, GIBSON, Eds.), pp. 5764. Champaign, I L American Oil Chemists Society. NEIDLEMAN, HUNTER-CEVERA,(1992), Wax S., J. ester production by Acinetobacter sp. HOl-N, in: Industrial Applications of Single Cell Oils (KYLE,D. J., RATLEDGE, Eds.), pp. 16-28. C., Champaign, IL: American Oil Chemists Society. A. NELIS,H. J., DE LEENHEER, P. (1989), Microbial production of carotenoids other than @arotene, in: Biotechnology of Vitamins, Pigments E. and Growth Factors (VANDAMME, J., Ed.), pp. 43-80. London, New York: Elsevier Applied Science. NICHOLS, W., WOOD,B. J. B. (1968), The occurB. rence of biosynthesis of ylinolenic acid in a blue-green alga, Spirulina platensis, Lipids 3,4650. NICHOLS,P. D., NICHOLS, S., BAKES,M. J. D. (1994), Marine oil products in Australia, INFORM 5,254-261. NINET, RENAUT, (1979), Carotenoids, in: MiL., J. crobial Technology, 2nd Edn., Vol. l (PEPPLER, D., H. J., PERLMAN, Eds.), pp. 529-544. San Diego: Academic Press. Y., H. NOGUCHI, KAME,M., IWAMOTO, (1982), Studies on lipid production by yeasts. Fatty acid composition of lipid from glucose and fatty acid esters by Rhodotorula sp. and Candida sp., Yukagawa 31, 431437. NOVOTNY, BEHALOVA, STRUZINSKY, C., B., R., NOVAK, M., ZAJICEK, (1988), Sterol composiJ. tion of a AS, 7-sterol-rich strain of Saccharomyces cerevisiae during batch growth, Folia Microbiol. 33, 377-385. OBRIEN, J., SENSKE, E. (1994), Recovery of D. G. eicosapentaenoic acid from fungal mycelia by solvent extraction, J. Am. Oil Chem. SOC. 71, 947-950. M. R. OBRIEN,D. J., KURANTZ, J., KWOCZAK, (1993), Production of eicosapentaenoic acid by
193
the filamentous fungus Pythium irregulare, Appl. Microbiol. Biotechnol. 40,211-214. OLEARY, M., WILKINSON, G. (1988), GramW. S. positive bacteria, in: Microbial Lipids, Vol. 1 (RATLEDGE, WILKINSON, G., Eds.), pp. C., S. 117-201. London: Academic Press. M. A. OSMAN, G., MOSTAFA, A., EL-REFAI, H. H. (1969), Production of lipids and sterols by Aspergillus funigatus. Some culture conditions favouring the formation of lipids and sterols, J. Chem. U.A.R. 12, 185-197. OWUSU-ANSAH, J. (1993), Enzymes in lipid Y. technology and cocoa butter substitutes, in: Technological Advances in Improved and Alternative Sources of Lipids (KAMEL, s., KAKUB. DA, Y., Eds), pp. 360-389. Glasgow: Blackie Academic & Professional. PAN,J. G., RHEE,J. S. (1986), Biomass yields and energetic yields of oleaginous yeasts in batch culture, Biotechnol. Bioeng. 18, 112-114. PARKS, W., RODRIQUEZ, J., MCCAMMON, L. R. M. T. (1984), Sterols of yeast: a model for biotechnology in the production of fats and oils, in: Biotechnology for the Oils and Fats Industry (RATLEDGE c.,DAWSON, RAITRAY,J., Eds.), pp. P., 177-187. Champaign, I L American Oil Chemists Society. PIERINGER, A. (1989), Biosynthesis of non-terR. penoid lipids, in: Microbial Lipids, Vol. 2 (RATLEDGE, c., WILKINSON, G., Eds.), pp. 51-114. S. London: Academic Press. POIRIER, DENNIS, KLOMPARENS, NAWY., D., K., RATH,C., SOMERVILLE, (1992), Perspectives C. on the production of polyalkanoates in plants, FEMS Microbiol. Rev. 103, 237-246. POIRIER, NAWRATH, SCHECHTMAN, A., Y., C., L. SATKOWSKI, M., NODA,I., SOMERVILLE, M. C. (1994), PHB production in plants: some recent advances, Abst. Intern. Symp. on Bacterial PHA, Montreal, Canada, p. 44. RADWAN, S. (1991), Sources of C2,,-polyunsatuS. rated fatty acids for biotechnological use, Appl. Microbiol. Biotechnol. 35, 421430. RADWAN, S., SOLIMAN, H. (1988), ArachiS. A. donic acid from fungi utilizing fatty acids with shorter chains as sole sources of carbon and energy, J. Gen. Microbiol. 134, 387-393. RATLEDGE, (1982), Microbial oils and fats: an C. assessment of their commercial potential, Prog. Ind. Microbiol. 16, 119-206. RATLEDGE, (1986), Lipids, in: Biotechnology, C. 1st Edn., Vol. 4 (REHM,H. J., REED,G., Eds.), pp. 185-213. Weinheim: VCH. RATLEDGE, (1987), Microorganisms as sources C. of phospholipids, in: Lecithin: Technological, Biological and Therapeutic Aspects (HANIN, I.,
194
4 Microbial Lipids
ANSELL, B., Eds.), pp. 17-35. New York PleG. num Press. RATLEDGE, (1988), Biochemistry, stoichiomeC. try, substrates and economics, in: Single Cell Oil (MORETON,R. S., Ed.), pp. 33-70. Harlow: Longman. RATLEDGE, (1989a), Biotechnology of oils and C. fats, in: Microbial Lipids (RATLEDGE, WILC., S. KINSON, G., Eds.), pp. 567-668. London: Academic Press. RATLEDGE, (1989b), Microbiological sources of C. phospholipids, in: Lecithins: Sources, ManufacB. ture and Uses (SZUHAJ, F., Ed.), pp. 72-96. Champaign, I L American Oil Chemists Society. RATLEDGE, (1994), Biodegradation of oils, fats C. and fatty acids, in: Biochemistry of Microbial C., Degradation (RATLEDGE, Ed.), pp. 89-141. Dordrecht: Kluwer Academic Publisher. C., RATLEDGE, EVANS,C. T. (1989), Lipids and their metabolism, in: The Yeasts, 2nd Edn., Vol. 3 (ROSE,A. H., HARRISON, S., Eds.), pp. 367J. 455. London: Academic Press. RATLEDGE, GILBERT, C. (1985), Carnitine C., S. acetyltransferase activity in oleaginous yeasts, FEMS Microbiol. Lett. 21,273-215. RATLEDGE, WILKINSON, G. (1988a), MicroC., S. bial Lipids, Vol. 1. London: Academic Press. S. RATLEDGE, WILKINSON, G. (1989), MicroC., bial Lipids, Vol. 2. London: Academic Press. S. RATLEDGE, WILKINSON, G. (1988b), An C., overview of microbial lipids, in: Microbial LipC., s. ids, Vol. 1 (RATLEDGE, WILKINSON, G., Eds.), pp. 3-22. London: Academic Press. RATLEDGE, BOULTON, A., EVANS,C. T. C., C. (1984), Continuous culture studies of lipid production by oleaginous microorganisms, in: Continuous Culture 8: Biotechnology, Medicine and the Environment (DEAN,A. C. R., ELLWOOD, D. C., EVANS, G. T., Eds.), pp. 272-291. ChiC. Chester: Ellis Honvood. RATTRAY, B. M. (1988), Yeasts, in: Microbial J. C., S. Lipids, Vol. 1 (RATLEDGE, WILKINSON, G., Eds.), pp. 555-697. London: Academic Press. RATI-RAY, B. M. (1991), Plant biotechnology and J. the oils and fats industry, in: Biotechnology of Plant Fats and Oils (RATTRAY, B. M., Ed.), J. pp. 1-35. Champaign, I L American Oil Chemists Society. RATTRAY, B. M. (1994), Biotechnological adJ. vances in improved and alternative sources of lipids, in: Technological Advances in Improved B. and Alternative Sources of Lipids (KAMEL, S., KAKUDA, Y., Eds.), pp. 50-92. London: Blackie. RINGO, JOSTENSEN, P., OLSEN, E. (1992), E., J. R.
Production of eicosapentaenoic acid by freshwater Vibrio, Lipids 27, 564-566. ROESSLER, G. (1990), Environmental control of P. glycerolipid metabolism in microalgae: commercia1 implication and future research directions, J. Phycol. 26,393-399. . Roux, M. P., KOCK, J. L. F., BETHA, A., DU PREEZ,J. C., WELLS, V., BOTES, J. (1994), G. P. Mucor - a source of cocoa butter and gammalinolenic acid, World J. Microbiol. Biotechnol. 10,417-422. SANCHEZ, MARTINEZ, ARACIL, CORMA, N., M., J., A. (1992), Synthesis of oleyl oleate as a jojoba oil analog, J. Am. Oil Chem. SOC. 69, 11501153. S., T., SAWAYAMA, MINOWA, DOTE,Y., YOKOYAMA, S. (1992), Growth of the hydrocarbon-rich microalga Botryococcus braunii in secondarily treated sewage, Appl. Microbiol. Biotechnol. 38, 135-138. SCHAFER, KRAGBALLE, (1991), SupplemenL., K. tation with evening primrose oil in atopic dermatitis: effect of fatty acids in neutrophils and epidermis, Lipids 26,557-560. SCHLEGEL, G., STEINBUCHEL, (Eds.) (1992), H. A. Int. Symp. Bact. Polyalkanoates ISBP 92, FEMS Microbiol. Rev., Vol. 103 (special issue). SCHWEIZER, (1989), Biosynthesis of fatty acids E. and related compounds, in: Microbial Lipids, Vol. 2 (RATLEDGE, WILKINSON, G., Eds.), C., S. pp. 3-50. London: Academic Press. SEKULA, (1992), Wax ester production by yest, B. in: Biotechnology of Plant Fats and Oils (RATJ. TRAY, B. M., Ed.), pp. 162-176. Champaign, I L American Oil Chemists Society. SETO, WANG, L., HESSELTINE, W. (1984), A., H. C. Culture conditions affect eicosapentaenoic acid content of Chlorella minutissima, J. Am. Oil Chem. SOC.61,892-894. SETO,A., KUMASAKA, HOSAKA, KOJIMA, K., M., E., KASHIWAKURA, KATO,T. (1992), ProM., duction of eicosapentaenoic acid by a marine microalgae and its commercial utilization for aquaculture, in: Industrial Applications of Single Cell C., Oils (KYLE, D. J., RATLEDGE, Eds.), pp. 219-234. Champaign, I L American Oil Chemists Society. SHASHI, K., BACHHAWAT, K., JOSEPH, R. A. (1990), ATP:Citrate lyase of Rhodotorula gracilis: purification and properties, Biochim. Biophys. Acta 1033, 23-30. SHAW, (1966), The polyunsaturated fatty acids R. of microorganisms, Adv. Lipid Res. 4,107-174. SHERIDAN, RATLEDGE,C., CHALK, P. A. R., (1989), Pathways to acetyl-CoA formation in Candida albicans, FEMS Microbiol. Lett. 69, 165-170.
8 References
SHIFRIN, S. (1984), Oils from microalgae, in: N. Biotechnology for the Oils and Fats Industry (RATLEDGE, DAWSON, RATTRAY,J., C., P., Eds.), pp. 145-163. Champaign, I L American Oil Chemists Society. S., Y., SHIMIZU, KAWASHIMA,H., SHINMEN, AKIMOTO, K., YAMADA, (1988a), Production of H. eicosapentaenoic acid by Mortierella fungi, J. Am. Oil Chem. SOC.65,1455-1459. SHIMIZU, SHINMEN, KAWASHIMA, AKIS., Y., H., MOTO,K., YAMADA, (1988b),Fungal mycelia H. as a novel source of eicosapentaenoic acid. Activation of enzyme(s) involved in eicosapentaenoic acid production at low temperature, Biochem. Biophys. Res. Commun. 150,335-341. SHIMIZU,S., AKIMOTO,K., KAWASHIMA, H., SHINMEN, YAMADA, (1989a), Production Y., H. of dihomo-y-linolenic acid by Mortierella alpina 1s-4, Am. Oil Chem. SOC. 66,237-241. J. SHIMIZU, KAWASHIMA, AKIMOTO,K., S., H., SHINMEN, YAMADA, (1989b), Microbial Y., H. conversion of an oil containing a-linolenic acid to an oil containing eicosapentaenoic acid, J. Am. Oil Chem. SOC.66,342-437. SHIMIZU, AKIMOTO, SHINMEN, KAWAS., K., Y., SHIMA, H., SUGANO, M., YAMADA, (1991), H. Sesamin is a potent and specific inhibitor of 85desaturase in polyunsaturated fatty acid biosynthesis, Lipids 26, 512-516. SHIMIZU,. KAWASHIMA, WADA, M., YAS, H., MADA, (1992), Occurrence of a novel sterol, H. 24,25-methylenecholest-5-en-3~-ol, in MortierelLipids 21, 481433. la alpina 1s-4, SHINMEN, SHIMIZU, AKIMOTO, KAWAY., S., K., SHIMA, H., YAMADA, (1989), Production of H. arachidonic acid by Mortierella fungi, Appl. Microbiol. Technol. 31,ll-16. SHLOMAI, BEN-AMOTZ, MARGALITH, P., A,, P. (1991),Production of carotene stereoisomers by Phycomyces blakesleeanus, Appl. Microbiol. Biotechnol. 34,458-462. SHUKLA, K. S. (1994), Present and future outV. look of the world fats and oil.supply, in: Technological Awareness in Improved and Alternative B. Y., Sources of Lipids (KAMEL, S., KAKUDA, Eds.), pp. 1-15. London: Blackie. SIMOPOULOS, P. (1989),Summary of the NATO A. Advanced Research workshop on dietary 03 and 06 fatty acids: biological effects and nutritional essentiality,J. Nutr. 119,521-528. SINCLAIR, GIBSON, (Eds.) (1992), Essential A., R. Fatty Acids and Eicosanoids, 3rd Int. Congr. Champaign, I L American Oil Chemists Society. SMIT, H., YKEMA,A., VERBREE, C., VERE. WOERT, I. G. S., KATER,M. M. (1992),ProI. duction of cocoa butter equivalents by yeast mu-
195
tants, in: Industrial Applications of Single Cell C., Oils (KYLE, D. J., RATLEDGE, Eds.), pp. 185-195. Champaign, I L American Oil Chemists Society. E., SMITH, WHITE,K. A., HOLT, D., FENTEM,P. S. A., BRIGHT, W. J. (1994), Expression of polyhydroxybutyrate in oilseed rape, Abstr. Znt. Symp. Bacterial PHA, Montreal, Canada, p. 44. SMITH,P. F. (1988), Arachaebacteria and other specialized bacteria, in: Microbial Lipids, Vol. 1 (RATLEDGE, WILKINSON, G., Eds.), pp. C., S. 489-553. London: Academic Press. SMITH, R. L. (1981),Some aspects of ICIs Single S. Cell Protein Process, in: Microbial Growth on C, Compounds (DALTON,H., Ed.), pp. 343-348. London: Heyden and Sons. STADLER, MOLLION, VERDUS, C., KARAT., J., M. MANOS, Y., MORVAN, CHRITIAEN, (Eds.) H., D. (1988), Algal Biotechnology. Amsterdam: Elsevier. STEINSUCHEL, HUSTEDE, LIEBERGESELL, A., E., U., A., H. M., PIEPER, TIMM, VALENTIN, (1992), Molecular basis for biosynthesis acids in bacteria, FEMS Microbiol. Rev. 103,217-230. SZUHAJ,B. F. (Ed.) (1989), Lecithins: Sources, Manufacture and Uses. Champaign, IL: American Oil Chemists Society. THOMAS, M., HOLUB,B. J. (1994), Nutritional L. f aspects o fats and oils, in: Technological Awareness in Improved and Alternative Sources of Lips., Y., ids (KAMEL, KAKUDA, Eds.), pp. 16-49. London: Blackie. TORNABENE, G., BOURNE, F., RAZIUDDIN, T. T. S., BEN-AMOTZ, (1985), Lipid and lipopolyA. saccharide constituents of cyanobacterium Spirulina platensis, Mar. Ecol. Prog. Ser. 22, 121TOTANI, OBA,K. (1987),The filamentous funN., gus Mortierella alpina, high in arachidonic acid, Lipids 22, 1060-1062. TOTANI, OBA,K. (1988), A simple method for N., production of arachidonic acid by Mortierella alpina, Appl. Microbiol. Biotechnol. 28, 135-137. TOTANI, SOMEYA, OBA, K. (1992), IndusN., K., trial production of arachidonic acid by Mortierella, in: Industrial Applications of Single Cell Oils (KYLE,D. J., RATLEDGE, Eds.), pp. 52-60. C., Champaign, I L American Oil Chemists Society. TYRRELL, (1967),The fatty acid composition of D. 17 Entomophthora isolates, Can. J. Microbiol. TYRRELL, (1968),The fatty acid composition of D. some Entomophthoraceae 11, Lipids 3,368-372. TYRRELL, (1971),The fatty acid composition of D. some Entomophthoraceae 111, Can. J. Microbiol.
125.
13,755-760.
17,1115-1118.
196
4 Microbial Lipids
Eds.), pp. 98-117. Champaign, I L American Oil TYRRELL, WEATHERSTONE,(1976), The fatD., J. Chemists Society. ty acid composition of some Entomophthoraceae IV. The Occurrence of branched-chain fatty WILKINSON, G. (1988), Gram-negative bacteria, S. acids in Conidiobolus species, Can. J. Microbiol. in: Microbial Lipids, Vol. 1 (RATLEDGE,C., WILKINSON, G., Eds.), pp. 299-488. London: 22, 1058-1060. S. Academic Press. VANDYK,M. S., KOCK,J. L. F., COETZEE, J., D. T., W., AUGUSTYN, P. H., NIGAM, S. (1991), Isola- WILLNER, SITZMANN, WEBER,K. (1993), 0. Dry fractionation for cocoa butter replacers, in: tion of a novel arachidonic acid metabolite 3-hyOil and Fats in the Nineties, pp. 162-175. Lysdroxy-5,8,11,14-eicosatetraenoic acid (3-HETE) trup: International Food Science Centre AIS. from the yeast Dipodascopsis uninucleata UOFS-Y128, FEBS Lett. 283, 195-198. WITHOLT, HUISMAN, W., PREUSTING, B., G. H. (1990), Bacterial poly(3-hydroxyalkanoates) in: VAN DYK, M. S., KOCK, J. L. F., BOTHA, A. Novel Biodegradable Microbial Polymers (1994), Hydroxy long chain fatty acids in fungi, (NATO AS1 Series No. 186), (DAWES,E. A., World J. Microbiol. Biotechnol. 10, 495-504. Ed.), pp. 161-173. Dordrecht: Kluwer. VEGA, E. Z., GLATZ,B. A., HAMMOND, G. E. H., M. (1988), Optimization of banana juice fermenta- WI-ITER,B., DEBUCH, STEINER, (1974), Die Lipide von Endomycopsis vernalis bei verschietion for the production of microbial oil, Appl. dener Stickstoff-Ernahrung, Arch. Microbiol. Environ. Microbiol. 54, 748-752. 101,321-335. VENTOSA, NIETO, J. (1995), Biotechnological A., J. WOOD,B. J. B. (1988), Lipids of algae and protoapplications and potentialities of halophilic miC., zoa, in: Microbial Lipids, Vol. 1 (RATLEDGE, croorganisms, World J. Microbiol. Biotechnol. WILKINSON, G., Eds.), pp. 807-867. London: S. 11,85-94. Academic Press. VERT, M., FEIJEN,J., ALBERTOSON, Scorn, A., WOODBINE, (1959), Microbial fat: microorganM. G., CHIELLINI, (Eds.) (1992), Biodegradable E. isms as potential producers, Prog. Ind. MicroPolymers and Plastics. Cambridge: Royal Socibiol. 1, 179-245. ety of Chemistry. V. L. VERWOERT, I. G. S., YKEMA, VALKENBURG, YADWAD, B., WARD,0. P., NORONHA, C. I. A., (1991), Application of lipase to concentrate the J. A. C., VERBREE, C., NIJKAMP, J. J., E. H. docosahexaenoic acid (DHA) fraction of fish oil, SMIT,H. (1989), Modification of the fatty acid Biotechnol. Bioeng. 38,956-959. composition in lipids of the oleaginous yeast H., S., Y. Apiotrichum curvatum by intraspecific sphero- YAMADA, SHIMIZU, SHINMEN, (1987a), Production of arachidonic acid by Mortierella plast formation, Appl. Microbiol. Biotechnol. 32, elongata IS-5, Agric. Biol. Chem. 51, 785-790. 327-333. H. S., Y., VOLKMANN,K. (1989), Fatty acids of microalgae YAMADA, S., SHIMIZU, SHINMEN, KAJ. WASHIMA, H., AKIMOTO, (1987b). ProducK. used as feedstocks in aquaculture, in: Fats for the tion of arachidonic acid and eicosapentaenoic R. Future (CAMBIE, C., Ed.), pp. 263-283. Chiacid by microorganisms in: Proc. World Con$ Chester: Ellis Horwood. Biotechnol. for the Oils and Fats Industry (APVONSHAK, COHEN, RICHMOND, (1985), A., Z., A. PLEWHITE, H., Ed.), pp. 173-177. Champaign, T. The feasibility of mass cultivation of PorphyridI L American Oil Chemists Society. ium, Biomass 8,13-25. YAMADA, SHIMIZU, SHINMEN, AKIMOH., S., Y., WATANABE, YAZAWA, (1992), Cloning of K., K. TO, K., KAWASHIMA, JAREONKITMONGKOL, H., DNA coding for EPA - biosynthesis from a maS. (1992), Production of dihomo-y-linolenic acid, rine bacterium into E. coli, J. Jpn. Oil Chem. arachidonic acid and eicosapentaenoic acid by SOC. 42, 42-49. filamentous fungi, in: Industrial Applications of WATANABE, ISHIKAWA, INOVE,H., CENK., C., Single Cell Oils (KYLE,D. J., RATLEDGE, C., HUA, D., YAZAWA, K., KONDO,K. (1994), InEds.), pp. 118-138. Champaign, I L American corporation of exogenous docosahexaenoic acid Oil Chemists Society. into various bacterial phospholipids, J. Am. Oil YAMAGUCHI, NAKANO, MURAKAMI, K., H., M., Chem. SOC. 71,325-330. S., O., M., KONOSU, NAKAYAMA, KANDA, NAWEETEJ. D. (1980), Lipid Biochemistry of Fungi KANUIRA, IWAMOTO, (1987), Lipid comA., H. and other Organisms. New York: Plenum position of a green alga, Botryococcus braunii, Press. Agric. Biol. Chem. 51, 493-498. H., T., WEETE,J. D., GANDHI, R. (1992), Enhancement YAMAUCHI, MORI,H., KOBAYASHI, SHIMIS. zu, S. (1983), Mass production of lipids by Lipoof C,, polyunsaturated fatty acid production in myces starkeyi in microcomputer aided fed-batch Pythium ultimum, in: Industrial Applications of culture, Ferment. Technol. 61,275-280. Single Cell Oils (KYLE, D. J., RATLEDGE, C.,
8 References
YANO, Y., NAKAYAMA, SAITO,H., ISHIHARA, A., K. (1994), Production of docosahexaenoic acid by marine bacteria isolated from deep sea fish, Lipids 29,527-528. YAZAWA,K., ARAKI, OKAZAKI, WATAK., N., NABE, K., ISHIKAWA, INOVE, A., NUMAO, C., N., KONDO,K. (1988), Production of eicosapentaenoic acid by marine bacteria, J. Biochem. 103, 5-1 1. YAZAWA, K., WATANABE, K., ISHIKAWA, C., KONDO,K., KIMURA, (1992), Production of S. eicosapentaenoic acid from marine bacteria, in: Industrial Applications of Single Cell Oils (KYLE, D. J., RATLEDGE,C., Eds.), pp. 29-51. Champaign, I L American Oil Chemists Society. YKEMA, VERBREE, C., VAN VERSEVELD, A., E. H. W., SMIT,H. (1986), Mathematical modelling of lipid production by oleaginous yeasts in continuous culture, Antonie van Leeuwenhoek 52,491YKEMA,A., VERBREE,A. C., KATER, M. M., SMIT,H. (1988), Optimization of lipid production in the oleaginous yeast Apiotrichum curvatum in whey permeate, Appl. Microbiol. Biotechnol. 29,211-218. YKEMA,A., VERBREE, C., NIJKAMP, J. J., E. H. SMITH, (1989), Isolation and characterization H. of fatty acid auxotrophs from the oleaginous yeast Apiotrichum curvatum, Appl. Microbiol. Biotechnol32,76-84. YKEMA, VERBREE, C., VERWOERT, I. G. A., E. I. S., VAN DER LINDEN,K. H., NIJKAMP, J. J, H.
197
SMIT,H. (1990), Lipid production of revertants of Ufa mutants from the oleaginous yeast Apiotrichum curvatum, Appl. Microbiol. Biotechnol. 33, 176-182. YONGMANITCHAI, WARD, 0. P. (1989), OmeW., ga-3 fatty acids: alternative sources of production, Process Biochem. 24, 117-125. YONGMANITCHAI, W., WARD, 0. P. (1991), Screening of algae for potential alternative sources of eicosapentaenoic acid, Phytochemistry 30,2963-2967. YONGMANITCHAI, W., WARD, 0. P. (1992), Growth and eicosapentaenoic acid production by Phaeodactylum tricornutum in batch and continuous culture systems, Am. Oil Chem. SOC.69, YONGMANITCHAI, WARD, 0. P. (1993), PosiW., tional distribution of fatty acids, and molecular species of polar lipids, in the diatom Phaeodactylum tricornutum, J. Gen. Microbiol. 139, 465472. YOON,S. H., RHEE,J. S. (1983), Quantitative Physiology of Rhodotorula glutinis for microbial lipid production, Process Biochem. 18, 2-4. YOON, S. H., RHIM, W., CHOI, S. Y., RYU, D. D. J. W., RHEE,J. S. (1982), Effect of carbon and nitrogen sources of lipid production of Rhodotorula gracilis, J. Ferment. Technol. 60,243-246. ZVGAGINTSEVA,S., PITRYUK, A., BABEVA, I. I. I. P., RUBAN,E. L, (1975), Fatty acid composition of lipids of soil and epiphytic yeasts, Mikrobioloiya USSR (Int. Edn.) 44, 625-631.
584-590.
506.
Biotechnology
5 Microbial Siderophores
GUNTHER WINKELMANN HARTMUT DRECHSEL

Tiibingen, Federal Republic of Germany
1 Introduction 200 2 General Aspects 200 3 Bacterial Siderophores 201 3.1 Catecholate Siderophores 201 3.2 Hydroxamate Siderophores 209 3.3 Peptide Siderophores 212 3.4 Mycobactins and Related Siderophores 217 3.5 Citrate Hydroxamate Siderophores 220 3.6 Carboxylate Siderophores 223 3.7 Keto Hydroxy Bidentates 225 4 Fungal Siderophores 225 4.1 Ferrichromes 225 4.2 Coprogens 227 4.3 Rhodotorulic Acid 228 4.4 Fusarinines (Fusigens) 228 4.5 Rhizoferrins 229 5 Miscellaneous Compounds 230 6 Transport Mechanisms 233 7 Conclusion and Perspectives 235 8 References 236
200
1 Introduction
Iron nutrition in microbes and higher organisms has become a most fascinating topic in microbiology and biotechnology. Extensive studies in a variety of microorganisms have shown that low-molecular weight, ferric-specific ligands, named siderophores, are essential for growth and survival in natural environments. The biosynthesis of siderophores and their cognate membrane transport systems in bacteria are regulated at the transcriptional level by internal iron sensors named ferric uptake regulation proteins (Fur) which respond to the external iron concentration and function as iron-loaded repressors. The literature on siderophores has increased considerably during the past five years and several comprehensive books offer detailed descriptions of structures and functions of the et various siderophores (WINKELMANN al., 1987; WINKELMANN, 1991a; BARTON and HEMMING, 1993; WINKELMANN WINGE, and 1994). Attention has been paid primarily to the particular functions of siderophores, their occurrence and distribution among bacteria and fungi. Thus, plenty of information on siderophores already exists. From a biotechnological point of view iron is an extremely important element for fermentation processes in order to enhance growth and production yields. Thus, concentration, presence or absence of iron and other ions or medium constituents may greatly affect synthesis and excretion of fermentation products. The present review is intended to update the existing structural diversity of siderophores and to specifically address some biotechnological aspects of siderophores which had already been the aim of the previous review in Vol. 4 of the first edition of this series published about ten years ago (WINKELMA, 1986). Since then, a number of novel siderophore structures have been described which are not only variations of known structures but even represent novel types or classes of siderophores. Overall, the topic of siderophores seems to be a never ending story that continues to inspire the community of microbiologists. The biotechnological value of these results still is not fully appreciated. It
may be predicted that a variety of fermentation processes will be optimized and precisely controlled by the addition or withdrawal of iron and siderophores. Other examples will be briefly addressed at the end of this review.
2 General Aspects
The majority of aerobic and facultative anaerobic microorganisms respond to a decreasing iron content in their environment by the expression of siderophores and their cognate uptake systems. Because of the highly specific ferric iron binding, the ligands have attracted the attention of chemists who developed strategies for the synthesis of a number of natural siderophores (LEE and MILLER, 1983; BERGERON MCMANIS,1991) and and and LIBMAN, biomimetic analogs (SHANZER 1991). The isolation of siderophores from biological fluids is based on various extraction and and detection methods (NEILANDS NAKAMURA, 1991). However, yields are very low sometimes because of the fact that production and excretion of siderophores are negatively regulated by iron. Therefore, mutants defective in regulation of siderophore biosynthesis or in the expression of outer membrane receptors have been used for the production of siderophores (YOUNG and GIBSON,1979; WINKELMANN al., 1994). et The biosynthesis and excretion of many siderophores can be detected in culture filtrates by the appearance of color due to metal-toligand charge transfer bands. Thus, ferric hydroxamate complexes show absorption maxima at 420-440 nm, resulting in a yellowbrown color, while ferric catecholate complexes are red-blue in color due to an absorption maximum at 495-520 nm. Ferric carboxylate siderophores show only a weak absorpet tion at 335 nm (DRECHSEL al., 1992) which results in a pale yellow color. Survival of microorganisms in an aerobic atmosphere where the concentration of soluble ferric ions in solution at neutral pH is smaller than lO-M has been a challenge for about 3.5 billion years and enabled the
3 Bacterial Siderophores
201
evolution of siderophores and their cognate cally stable chromic complexes have been membrane transport systems. Moreover, side- used to study the metabolism of siderophores rophore-mediated iron acquisition has been in various microorganisms. The physicochemcorrelated with the ability of various microor- ical properties of natural and synthetic sideroganisms to establish and maintain infection of phores have been compiled in a recent review (1991). The reader is also rea host, although examples of virulence en- by CRUMBLISS hancement by siderophores are still scarce ferred to a comprehensive description de(BAGG and NEILANDS,1987; HESEMANN, voted solely to the solution and structural et et 1987; ACTISet al., 1988; ENARD al., 1988; chemistry of siderophores (MATZANKE al., 1989). STOJILJKOVIC HANTKE, and 1992). The history of siderophores started with the discovery of growth factors for mycobacteria and fungi (reviewed in: WINKELMANN, 1991a). NEILANDS (1952) was the first to isolate a siderophore in crystalline form and thus opened an era of intensive search for new iron-binding compounds, which were previously named sideramines or sideroch- 3.1 Catecholate Siderophores romes and are now collectively designated as Enterobactin (Fig. l), also referred to as siderophores. Siderophores are designed for the solubilization, transport, and storage of enterochelin, is the prototype catecholate siiron in microorganisms. It is now generally derophore of enteric bacteria. It forms highly agreed that siderophores including their bio- stable octahedral complexes with ferric iron The three catecholamide-binding synthesis and the expression of the corre- (&= sponding membrane-located transport sys- subunits of enterobactin are attached to a tritems are carefully regulated by iron. Iron- L-serine lactone ligand backbone. While the binding compounds that are not regulated ligand is uncharged at neutral pH, the ferric seem to occur but do not fall into the category form is a trianionic complex. Convenient of natural siderophores. Moreover, the term methods for the isolation of enterobactin (ensiderophore should be reserved for the metal- terochelin) have been described by YOUNG free ligand, although in some cases the names had been given to the metal complex earlier (ferrichrome, ferrioxamine, coprogen). The corresponding iron-free compounds then need the prefix desferri- or deferri-; siderophores which represent the iron-free form (enterobactin, aerobactin, staphyloferrin, rhihi OH zoferrin) need the prefix ferri- or ferric when iron is bound to the ligand. The ligands involved in iron(II1) binding OH are either phenolates or catecholates, hydroxamates, oxazolines, a-hydroxy carboxylates (e.g., citrate derivatives), or keto hydroxy bidentates. Ferric siderophores are octahedral com lexes in which the coordinated metal ion is d , high spin, and rapidly exchangeable. 0 After reduction, the ferrous ion shows little HO affinity for the ligand, and this is obviously the mechanism by which iron is removed from siderophores within the cell or at the HO cell surface of microorganisms. Non-reducible . aluminum and gallium complexes and kineti- Fig. 1 Enterobactin.
202
and GIBSON(1979), NEILANDS and NAKAMURA (1991), and BERNER et al. (1991). HPLC separation and isolation from culture fluids have been described which also allow the simultaneous detection of degradation products of linear 2,3-dihydroxybenzoyl serine derivatives (WINKELMANN al., 1994). et The production procedure of YOUNG and GIBSON (1979) employs a mutant strain CfepA) of Escherichia coli which is unable to transport the ferric enterobactin complex into the cell. Whereas wild-type strains produce enterobactin only during severe iron deficiency ( < 0.2 p,M) and terminate production after iron transport into the cell, fepA mutants continue to produce high amounts of enterobactin in a medium not necessarily iron-deficient. Enterobactin can be estimated from aqueous solutions (4 mL) after acidification with conc. H2S04and extracted with an equal volume of ethyl acetate. If required, the extract may be washed with an equal volume of sodium phosphate buffer (0.1 M, pH 7) to remove any of the hydrolysis products of enterobactin which might be present in varying amounts (WINKELMANN al., 1994), and the concenet tration is determined at 316 nm. The concentration of ferri-enterobactin is determined using the molar extinction coefficient (~~~~~,,,=5600 M-'cm-'). The transport of ferric enterobactin into E. coli requires the expression of the FepA outer membrane protein (81 kDa), which has been cloned, sequenced (LUNDRIGAN and KADNER, 1986), and also crystallized (JALAL and VAN DER HELM, 1989). Enterobactin is the most powerful natural iron sequestering agent (reviewed by CRUMBLISS, 1991). Physicochemical properties, analogs as well as mechanistic aspects of transport have been discussed in detail (MATZANKE, 1991; SHANZER and LIBMAN, 1991). The FepA receptor recognizes the ferric enterobactin complex present in a delta-cis configuration as shown by comparison of natural enterobactin and synthetic lambda-cis-enantio-enterobactin(NEILANDS et al., 1981). Purification of FepA by FPLC has been reported from the plasmidharboring E. coli strain UT5600/pBB2 (ZHOU et al., 1995). Successful crystallization of the purified FepA protein had been reported ear-
lier by this group (JALAL and VAN DER HELM,1989) and crystal structures are to be expected in the near future. Five additional gene products are required to complete the transport into the cells, of which only FepB and has been sequenced so far (ELKINS EARHARDT, 1989). The actual mechanism of enterobactin transport is still unsettled. Due to the very low redox potential ( -790 mV vs. NHE at pH 7.4) a direct reduction model could be excluded and an esterolytic degradation prior to reduction has been proposed. Alternative mechanisms involving protonated molecular species of the ferric enterobactin complex during membrane transport have been discussed (CASS et al., 1989; MATZANKE, 1991). An additional internal redox reaction has also been suggested as 55Fe-enterobactin is taken up while 67Ga-enterobactin is not. Recent HPLC data from the authors' laboratory have shown that the esterase seems to attack ferric enterobactin after entrance via FepA, producing several linear dihydroxybenzoyl serine products (WINKELMANN et al., 1994) which then can be easily reduced (-350mV vs. NHE). These results are in favor of a combined esterase-reductase mechanism and thus obviate any protonation or internal redox reaction of the ferric enterobactin complex.
Chrysobactin
Chrysobactin (Fig. 2), a simple catecholtype siderophore containing only one 2,3-dihydroxybenzoyl group connected to a dipepOH
OH
CH20H
Fig. 2. Chrysobactin.
203
azotochelin can be extracted from the culture filtrate using ethyl acetate. Aminochelin is 4-N-2,3-dihydroxybenzoyl-l ,Cdiaminobutane, and azotochelin is N,N'-bis-(2,3-dihydroxybenzoy1)-lysine. It was anticipated that these Aminochelin, Azotochelin, Myxochelin A, compounds may possibly represent sideroand Protochelin phore precursors or degradation products originating from a more complex siderophore Azotobacter vinelandii produces several which indeed has been found recently and catecholate siderophores under iron-limiting was named protochelin (Fig. 5) (TARAZ al., et et conditions (FUKASAWA al., 1972; PAGE 1990). Although recently detected in Azotoand TIGERSTROM, 1988; CORBIN BULEN, bacter, (CORNISH PAGE,1995), protocheand and 1969; DEMANGE al., 1988; BUDZIKIEWICZlin has been previously shown to occur in a et et al., 1992): 2,3-dihydroxybenzoic acid bacterium (DSM 5746) which is able to grow (DHBA), aminochelin (Fig. 3), and azotoche- on methanol as the sole carbon source. The lin (Fig. 4). These three compounds seem to structure of protochelin is a direct combinabe biogenetically related. Aminochelin and tion of aminochelin and azotochelin. As shown by PAGEand HUYER (1984) DHBA is produced constitutively by Azotobacter, whereas azotochelin and azotobactin are proOH duced only under iron limitation. Myxochelin A (Fig. 4) is a new catecholate siderophore isolated from Angiococcus disciOH formis (Myxobacteria). The structure is N,N'bis-(2,3-dihydroxybenzoyl)-lysinol(KUNZEet al., 1989) which is related to aminochelin Fig. 3. Aminochelin. (N,N '-bis-(2,3-dihydroxybenzoyl)-lysine) pro-
tide was isolated from Erwinia chrysanthemi which is a phytopathogenic species in Saintet paulia plants (PERSMARK al., 1989; PERSMARK and NEILANDS, 1992). Structure elucidation revealed that chrysobactin is N-[N2(2,3-dihydroxybenzoyl)-~-lysyl]-~-serine. While for similar natural monocatechol siderophores, like 2,3-dihydroxybenzoyl serine and 2,3-dihydroxybenzoyl lysine, an L-configuration was determined, the lysyl residue in chrysobactin has a D-configuration. However, as shown by transport experiments, the configuration of the amino acids in chrysobactin seems to be of minor importance compared to that of the catechol-iron center (PERSMARK et al., 1992).
Fig. 4. Azotochelin: R = COOH; myxochelin A: R = CH20H.
mo"
Fig. 5. Protochelin.
0
OH OH
204
duced by Azotobacter vinelandii (PAGEand TIGERSTROM, 1988). Myxochelin A exerts weak antibiotic activity on various gram-positive bacteria, e.g., Bacillus brevis, B. cereus, B. subtilis, B. megaterium, Micrococcus luteus, Staphylococcus aureus, Arthrobacter simplex, Brevibacterium linens, Corynebacteriurn faset cians, and Nocardia corallina (KUNZE al., 1989). Agobactin Agrobactin (Fig. 6a) is the characteristic siderophore of Agrobacterium tumefaciens (ONGet al., 1979). In low-iron media A . tumefaciens B6 was found to produce a neutral, ethyl acetate-extractable substance. The UV spectrum revealed absorption maxima at 316 and 252nm. Agrobactin was shown by its blue fluorescence in the ultraviolet range and positive Arnow reaction to belong to the catechol-type siderophores. Acid hydrolysis yielded DHBA, spermidine, and L-threonine. The carboxyl group of DHBA as well as the
amino group and the Phydroxyl group of threonine are linked to an oxazoline ring. The crystal structure of the ligand was published by ENG-WILMOT VAN DER HELM(1980). and Exposure to acid leads to the formation of the oxazoline ion which opens to yield the ester and subsequently under neutral conditions undergoes the N-O-acyl shift to give the amide. This open form was named agrobactin A (Fig. 6b). Whereas enterobactin adopts a delta-cis configuration about the iron center, the configuration of agrobactin is lambda-cis. Agrobactin and several other polyamine catecholamide chelators, such as parabactin and vibriobactin (see below), have been chemically synthesized (reviewed by BERGERON and MCMANIS, 1991). Recently, the structure of serratiochelin from Serratia marcescens has been elucidated, which is a derivative of agrobactin (Fig. 6a) lacking the tetramethylenedihydroxybenzoylamide (EHLERT al., 1994). et
Parabactin
In 1975, TAITisolated a siderophore from low-iron cultures of Micrococcus denitrificans (now Paracoccus denitrificans) which he named compound 111. A reinvestigation (PETERSON and NEILANDS, 1979) Of the StrUCture (Fig. 6a) revealed an analogous structure to agrobactin lacking the OH group of the central catechol in position 3. This compound was named parabactin. Again two forms are conceivable: one possessing a closed oxazoline ring (parabactin) and an open form (parabactin A). Inspection of the molecular models of agrobactin and parabactin revealed that the tertiary N-atom of the oxazoline ring and the o-hydroxyphenyl function are involved in the sixcoordinate ferric complex. Because of the optically active substituent L-threonine, a particular coordination isomer of parabactins and agrobactins can be expected. The CD spectra revealed a lambda-cis configuration as in ferrichrome. In addition, steric constraints ruled out the possible presence of geometrical isomers of the trans variety. The binding strength of parabactins and agrobactins is comparable to that of enterobactin which is approximately lo5*at pH 7.4.
Fig. 6a. Agrobactin: R=OH, parabactin: R=H; b agrobactin A.
205
The study of biological activities of agrobactins and parabactins in Agrobucterium tumefuciens and Purucoccus denitrificuns showed that only the closed but not the open oxazoline forms counteract growth retardation caused by EDTA. Using both labeled metal and labeled ligand complexes, BERGERON et al. (1985) presented evidence that parabactin operates by the iron taxi mechanism. Iron was delivered to the cell, but the ligand was not taken up in higher amounts. The same authors also presented evidence that enunrio-parabactin possessing a delta coordination isomer was unable to transfer higher amounts of labeled metal to Purucoccus denitrificans. Therefore, the parabactin receptor stereoselectively recognizes the natural lambda coordination isomer.
mO
Vibriobactin
From low-iron cultures of Vibrio cholerue the siderophore vibriobactin (Fig. 7a) was isolated by GRIFFITHet al. (1984). V. cholerue (serovar 0 and 0139) is known to have 1 caused several severe pandemic diarrheal diseases, and the latest one (the 7th) is still existing in Asia, Australia, and South America (WACHSMUTH al., 1994). Vibriobactin, like et agrobactin, contains three 2,3-dihydroxybenzoyl residues and two residues of L-threonine per molecule both of which form an oxazoline ring. The polyamine backbone proved to be N-(3-aminopropyl)-l,3-diaminopropane (norspermidine), in contrast to spermidine which is found in agrobactin and parabactin. Cells of V. cholerue Lou15 were used for the original isolation of vibriobactin (GRIFFITH et al., 1984). The yield of the preparation varied between 10-25 mg per 6 L batch. Thin layer chromatography (TLC) on silica gel plates using chloroform-methanol 4: 1 gave one spot detected by its blue fluorescence under UV light or by spraying with ferric chloride or iodine. The UV absorption spectrum is qualitatively the same as that of agrobactin. Biological activity testing revealed that the growth of the producing v. cholerue strain Loul5, which is only weakly pathogenic, was stimulated by vibriobactin and also by agrobactin with nearly equal effi-
Fig. 7a. Vibriobactin: R = OH; vulnibactin: R =H, b fluvibactin.
ciency. Vibriobactin is not the only siderophore utilized by V. cholerue as mutants defective in vibriobactin synthesis can alternatively use ferric citrate (SIGELet al., 1985) or obtain iron from heme (STOEBNERand PAYNE,1988). As shown below, additional carboxylate transport systems for vibrioferrin may exist. The outer membrane receptor for vibriobactin has been identified (STOEBNER et al., 1992) and the gene (viuA) encoding the outer membrane receptor (74 kDa) has recently been cloned and shown to be negatively regulated by the fur gene at the transcriptional level (BUTTERTONet al., 1992). The role of fur in the iron-regulated expression of the outer membrane receptor for vibriobactin (ViuA) and the outer membrane virulence protein (IrgA) has been demonstrated recently (LITWINand CALDERWOOD, 1994). The fact that further proteins were observed which were negatively regulated by iron but independent of Fur suggested that the gene
206
regulation in V. cholerue by Fur and iron is much more complex than previously assumed.
ture elucidation of the amonabactins have shown that four structural varieties of amonabactins exist (TELFORD al., 1994). Amonaet bactins contain 2 mol DHBA, glycine and lysine and 1mol of either tryptophan or phenylalanine. Two of them are glycine-deleted Vulnibactin forms (Fig. 8). A survey on various AeromonThe group of YAMAMOTO has also re- as strains revealed that although most strains ported on the structures of further polyam- produced amonabactins, some produced an ine-containing catecholate siderophores from enterobactin-like or a so far uncharacterized et the human pathogen Vibrio vulnificus (OKU- siderophore (ZYWNO al., 1992). In strains producing amonabactins a biosynthetic gene JO et al., 1994a). The principal compound isolated contained two oxazoline-bound salicyl (umoA) was identified in a Suu3A gene libramoieties and one outer DHBA residue and ry of Aeromonus hydrophilu 495A2 chromosomal DNA (BARGHOUTHI al., 1991). This et was named vulnibactin (Fig. 7a). gene seems to correspond to the first 2,3DHBA biosynthetic enzyme (isochorismate et synthetase) in E. coli (MASSAD al., 1994). Fluvibactin Fluvibactin (Fig. 7b) from Vibrio fluvialis (YAMAMOTO al., 1993) resembles agrobac- Anguibactin et tin with respect to the three DHBA residues While the characteristic moiety of the sideand the central oxazoline ring. However, fluvibactin, like vibriobactin and vulnibactin, rophores described above are oxazoline rings, contains norspermidine as a backbone (iden- the following structures contain a thiazoline tical propane spacer between the two nitro- ring. The first member of this series is anguigens). This is consistent with the finding that bactin (Fig. 9a). Isolated from the fish pathoVibrio species contain norspermidine in its gen Vibrio unguillurum (ACTISet al., 1986), it free form as a major polyamine (YAMAMOTO was one of the first siderophores regarded as an important plasmid-encoded virulence facet al., 1991). tor (CROSA, 1989). Structure elucidation based on single-crystal X-ray diffraction studies revealed that anguibactin consists of 2,3Amonabactins dihydroxybenzoic acid linked to cysteine by a Several amonabactins have been isolated thiazoline ring which in turn is linked by a hyfrom the fish pathogen Aeromonus hydrophi- droxamic acid bound to eN-hydroxy histala (BARGHOUTHI al., 1989). Recent struc- mine (JALALet al., 1989). et
OH
OH
/+/H +=O
Fig. 8. Amonabactins. Phe can be substituted by Trp, Gly-deleted forms occur.
3 Bacterial Siderophores R'

OH
207
ferri-ferrithiocin is produced by Streptomyces antibioticus (TU 1998; Fig. 9b) (NAEGELIand ZAHNER, 1980) which resembles the structure of aeruginic acid published previously (Fig. 9c). Desferri-ferrithiocin contains a hydroxy pyridine residue linked to a thiazoline ring. Although a function in iron transport has never been shown in the producing strain, it has been suggested to be a potent, orally available iron chelator.
PyocheIin
HO
Fig. 9. Anguibactin: R'=OH, R2=A, R3=H, X = CH; desferri-ferrithiocin: R'= H, R2= CH3, R3= COOH, X =N; aeruginic acid: R' = R2= H, R3= COOH, X = CH.
Desferrithiocin
Besides the indicator antibiotics p- and 7rubromycins the iron-binding compound des-
Pyochelin (Fig. 10a) is an iron-chelating compound isolated from low-iron cultures of Pseudomonas aeruginosa strain PAO-1. Pyochelin has also been found in clinical isolates. The structure, 2-(2-o-hydroxyphenyl-2-thiazolin -4-yl) -3-methylthiazolidine -4 -carboxylic acid, is presumed to be biosynthesized from salicylcysteinyl cysteine by cyclization and hydration of the thiazolidine ring. Pyochelin was purified by preparative TLC on silica gel G (Cox et al., 1981) and detected both by fluo-
OH
Fig. 1Oa. Pyochelin; b yersiniabactin.
208
rescence and a red reaction using an iron spray. It forms a stable, red complex with ferric iron. Although the ligand groups involved in iron chelation have not been identified, it may be assumed that iron is bound to the phenyl O H group, the carboxyl group, and one nitrogen in a 2:l ligand-to-iron ratio. The simpler compound 2-o-hydroxyphenyl-2thiazoline-4-carboxylic acid, named aeruginic acid, is known to occur as a fermentation product of P. aeruginosu. Ferrithiocin isolated from Streptomyces untibioticus is related to aeruginic acid, but contains hydroxy pyridine instead of phenyl and a methyl group in the vicinity of the carboxyl group resulting in an additional chiral center. Yersiniabactin From cultures of Yersinia enterocolitica, an iron-complexing and iron-transporting compound named yersiniabactin was isolated (HAAG et al., 1993; CHAMBERS SOKOL, and 1994). The structure of the siderophore was determined by a variety of spectroscopic methods, including 2D NMR experiments on the metal-free ligand as well as its gallium complex (DRECHSEL al., 1995b). The novel et siderophore contains a benzene and a thiazolidine ring as well as two thiazoline rings (Fig. lob) and thus resembles the thiazoline-containing siderophores pyochelin, anguibactin, and desferri-ferrithiocine. The molecule contains five chiral centers, four of them inherent in the connectivity and substituents of the thiazoline rings. The compound forms stable complexes with trivalent cations such as ferric iron and gallium. Iron binding constants have not yet been determined. From pH dependent UV and CD measurements it is clear, however, that the yersiniabactin-iron(II1) complexes are stable at least between pH 411. The complexes with Ga3+ and Fe3+ are monomeric, 1:1 complexes. Yersiniabactin constitutes the principal siderophore of Yersiniu enterocolitica sought for a long time. The same siderophore has been isolated as aluminium complex and has been termed yersiniophore (CHAMBERS al., 1996). et
II
0
Hzo
Fig. 11. Proferrorosamine A.
Ferrorosamine A
Erwinia rhupontici has been shown to produce proferrorosamine A, ~-(2-pyridyl)-lpyrroline-5-carboxylic acid, a Fe(I1)-complexing agent (Fig. 11) which after complexing gives a reddish pigment named ferrorosamine A (FEISTNER al., 1983). While proferroroset amine A in neutral solution exists in the bicyclic form, the pyrroline ring opens under acidic conditions to give the a-amino acid residue which no longer complexes ferrous ions. The proferrorosamine producing Pseudomonas sp. strain G H (LMG 11358), has later been reclassified as E. rhuptontici (VANDE WOESTYNEet al., 1991; DE Vos et al., 1993).
Siderochelin A Siderochelin A (Fig. 12) isolated from fermentation broths of Nocardia sp. SC 11340 was identified as truns-3,4-dihydro-4-hydroxy-
Fig. 12. Siderochelin A.
209
5 - (3- hydroxy -2-pyridinyl) -4-methyl-2H -pyrrole-Zcarboxamide (LIUet al., 1981).
Ferrioxamines
Streptomyces pilosus produces a group of closely related desferrioxamines (Al, AZ, B, 3.2 Hydroxamate Siderophores C, D1, Dz, E, F, G, and H) among which desferrioxamine B or E generally predominate et et Generally, hydroxamate siderophores in (BICKEL al., 1960 KELLER-SCHIERLEIN bacteria can be divided into two main groups: al., 1964). Although commonly isolated in (1) exclusively containing hydroxamate their ferric forms as ferrioxamines, the stucgroups as iron-binding bidentates, (2) con- tures shown in Fig. 13 represent the iron-free taining additional iron-binding ligands, e.g., forms. Desferrioxamine B is produced indus. a-hydroxycarboxylate residues, like citrate or trially by large-scale fermentation of S pilocatecholate groups as found in the pyover- sus strain A 21748 as the methane sulfonate dins. Members of the first group comprise the salt (DesferaP). Desferale still is the only ferrioxamines originally isolated from the drug approved for treatment of iron overload gram-positive streptomycetes and are now diseases although several other useful iron known to occur also in the gram-negative chelators are currently being developed et genus Pseudomonas and in Enterobacteria, (DIONIS al., 1991). such as P. stutzeri (MEYERand ABDALLAH, Desferrioxamine B is a linear trihydrox1980) and Erwinia, Enterobacter, Hafnia amate. Acid hydrolysis yields acetic acid, suc(BERNERand WINKELMANN, 1990; REISS- cinic acid, and 1-amino-5-hydroxylaminopentane in a ratio of 1:2:3. After addition offerBRODT et al., 1990).
NH
R '
N-C
N-C
N-C
Hb
Hb
I,
HO
\c=o
b
NH
I
N-C N-C
HO
N-C
a Ferrioxamine B: R' =H, R2= CH3, n =5; ferrioxamine D1:R' = COCH3, R2 = CH3, n =5; ferrioxamine G1: R = H, R2= CH2CH2COOH,n =5; ferrioxamine G2:R' =H, R2 = CH2CH2COOH,n =4; b ferriox-
Fig. 13. Ferrioxamines (iron-free).
HO
amine E; ferrimycin A,: R 1=A, R2 = CH3.
'
210
ric iron, a 1:1complex is formed with a stabil- ing number and size of the consituent diity constant of Kf=1030.6 showing a red- amines, cyclic structure, and C-terminal acetyl 1995). brown color due to the metal-to-ligand charge residues (FEISTNER, Ferrioxamine B is the most intensely studtransfer band at h,,,=428 nm ( ~ = 2 8 0 0 M cm -I). Since ferrioxamine B possesses a ied ferrioxamine. Today it is used in the mefree amino group (pK=9.74), it migrates as a sylate form (DesferaP) for the treatment of cation during electrophoresis in a weak acid iron storage diseases in man. The biological buffer. Ferrioxamine D, (N-acetyl-ferriox- activity of ferrioxamine B has been studied amine B) was isolated as a minor constituent using growth promotion tests with iron-auxoof the ferrioxamine fraction and can be pre- trophic bacteria, such as Microbacterium lactipared synthetically by acetylation of ferriox- cum ATCC 8181, Arthrobacter flavescens JG9 amine B with acetic acid anhydride. The li- ATCC 29091, and A. terregens as reviewed by (1991a). Transport experigand of ferrioxamine E is identical with the WINKELMANN antibiotic nocardamine isolated from No- ments with iron-labeled ferrioxamines in cardia. The crystal structure of ferrioxamine Streptomyces pilosus (ATCC 19797) de(1984) E was determined (VAN DER HELMand POL- scribed by MULLER and RAYMOND ING, 1976), and the coordination about the and MULLERet al. (1984) showed that feriron center is cis. The ligand is a cyclic struc- rioxamine B has the highest ratio of iron upture consisting of 3mol succinic acid and take ( K , =0.2 pM). Chromic and gallium 3 mol l-amino-5-hydroxylamino pentane. complexes of ferrioxamine B were transFormal hydrolysis of one peptide bond in fer- ported at similar rates indicating that the inrioxamine E leads to ferrioxamine G which in tact complexes were taken up. Differences in turn can be converted to ferrioxamine E by cis and trans isomers were not observed. In treatment with dicyclohexyl carbodiimide. contrast, ferrioxamine transport studies using Some of the isolated ferrioxamines contained in vivo Mossbauer spectroscopy and radioaca mixture of l-amino-5-hydroxylamino pen- tive labeling in Enterobacteria, such as Ertane and l-amino-4-hydroxylamino butane winia herbicola (now called Pantoea herbico(molar ratio 2:l) and were designated as fer- la), revealed that the ligand is not accumurioxamine A (related to ferrioxamine B), and lated inside the cells (MATZANKEet al., ferrioxamine A*, (related to ferrioxamine D1) 1991). and ferrioxamine DZ(related to ferrioxamine E). Recent investigations in the authors laboratory have shown that ferrioxamines are not confined to the genus Streptomyces but are Bisucaberin and Alcaligin also characteristic siderophores of several en-, Bisucaberin (Fig. 14a), a dihydroxamate siterobacterial genera, e.g., Erwinia, Pantoea, Enterobacter, Ewingella, and Hafnia (BER- derophore isolated from low-iron cultures of NER et al., 1988; BERNER and WINKELMANN,the marine bacterium Alteromonas haloplanc1990; REISSBRODT al., 1990). Earlier re- tis is a cyclic dimer of succinyl-(N-hydroxy)et ports have shown that Pseudomonas stutzeri cadaverine. It thus resembles ferrioxamine E is able to produce desferri-ferrioxamine E, which is a cyclic molecule composed of succipreviously named nocardamine (MEYERand nyl diaminopentane residues. From the heterotrophic marine bacterium ABDALLAH, 1980). Although the name proferrioxamine has been suggested for the li- Alcaligenes denitrificans a similar cyclic dihygand instead of desferrioxamine, deferriox- droxamate siderophore named alcaligin (Fig. et amine, or deferoxamine (FEISTNERet al., 14b) was isolated (NISHIO al., 1988). Alcal1993), the new terminology of pFOs has not igin is a cyclic molecule consisting of two been accepted in the biological literature so N1-hydroxy-3(0H)-putrescine residues linked far but seems to be of great value in distin- by two succinic acid residues. Alcaligin has guishing the vast number of structurally dif- recently been found in the obligate pathogens fering ligand derivatives by a simple subscript Bordetella pertussis and B. bronchiseptica et (e.g., desferrioxamine B = PFOsss~c)indicat- (MOOREet al., 1995; BRICKMAN al., 1996).
211
Ferrimycins and Albomycins

The ferrimycins (Fig. 13f) represent ferrioxamine-type siderophores possessing antibiotically active residues. Ferrimycins (A,, A2, B) and the compound A-22765 are ferrioxamine-type antibiotics isolated from Streptomyces griseoflavus and other species. The ferrimycins inhibit growth of gram-positive bacteria, such as Staphylococcus aureus and Bacillus subtilis. A revised structure of ferrimycin A was published by KELLER-SCHIERLEIN et al. (1984). Agar diffusion tests revealed that the antibiotic activity of the ferrimycins is antagonized by the presence of ferrioxamines. This has been explained by the existence of a transport antagonism in which the ferrioxamines and ferrimycins compete for the same target in the transport system Fig. 14a. Bisucaberin; b alcaligin, R =OH; putre- (ZAHNERet al., 1977). It remains an open bactin, R =H. question whether the antibiotically active group is split off from the transport moiety or whether ferrimycin acts as a whole after enA related cyclic dihydroxamate named putre- tering the cell. Albomycins (al, &, E) (Fig. 15) have also bactin (Fig. 14c), containing putrescine instead of hydroxy-putrescine, was isolated been isolated from Streptomyces strains, alfrom Shewanella putrefaciens (ALISONBUT- though the triornithine peptide is a characteristic feature of the fungal siderophores, e.g., LER, personal communication). ferrichromes. The originally published albomycin structure has been revised (BENZet al.,
Fig. 15. Albomycins. Albomycin al: X = 0; albomycin &: X =NCONH2; albomycin
E:
X =NH.
212
1982) and was shown to consist of a linear peptidonucleoside (Oml-Om2-Om3-Ser-nucleoside). Albomycins are highly active against gram-negative and gram-positive bacteria. Ferrimycins and albomycins use the iron transport systems of ferrioxamine and ferrichrome, respectively. Albomycin is taken up in E. coli via the FhuA receptor protein and the tonB gene product, as shown with F u A and tonB mutants (HARTMANN al., et 0 L o \ NH I 1979). Moreover, using 3H- and 35S-labeled albomycin it was confirmed that, contrary to the iron atom, the ligand is not accumulated inside the cell. From the small portion of labeled sulfur detected within the cells it was OH concluded that the nucleoside residue was split off to exert its antibiotic activity. Al- Fig. 1 . Alterobactin. 6 though inhibition of protein biosynthesis was demonstrated, it was also suggested that the inhibition of oxidative phosphorylation is the bacterium, Alteromonus luteovioluceu (REID initial inhibitory effect of albomycins. Growth et al., 1993). It has been shown to possess an of E. coli is inhibited at very low concentra- exceptionally high formation constant tions (lo-' M). Because of the high frequen- (K=1049-1053). Alterobactin A is a cyclic sicy of resistant mutants, ferrimycins and albo- derophore containing a lactone ester bond, mycins have not found any application in the while alterobactin B is the corresponding open-chain form. So far, only two siderotherapy of bacterial infections so far. phores from Alteromonus have been isolated - alterobactin and bisucaberin from Alfero3.3 Peptide Siderophores monus huloplunctis (TAKAHASHIet al., 1987). Alterobactin
A 0
OYNH.f
Alterobacin, a peptide siderophore containing the sequence cyc1.-(DHBA-Ser-GlyArg-/?-OH-Asp-POH-Asp) (Fig. 16), was isolated from a marine Pseudomonds-like
Ferrocins
Ferrocins (Fig. 17) represent a new group of iron-containing peptide antibiotics pro-
Fig. 1 . Ferrocins. 7 Ferrocin A: R'=H, R2=R3=R4=H,ferrocin B: R'=OH, R2=R3=R4=H,ferrocin C and D: R'=H, RZ,R3, R4=2-H, 1* CH3.
213
duced by Pseudomonas fluorescens YK-310 (KATAYAMA al., 1993). These siderophore et antibiotics showed antibacterial activity against E. coli and P. aeruginosa and strong therapeutic effects on P. aeruginosa. The structure elucidation revealed that the ferrocins are cyclic decapeptides containing three hydroxamate residues .(TSUBOTANI al., et 1993). Four different ferrocins (A, B, C, D) have been isolated. Ferricrocin A is a cyclic peptide containing a fatty acid residue and a lactone ring between the C-terminal Gly and the N-terminal Ser: (Z)-3-Decenoicacid-cyc1.Ser-X-Gly-Val-Ser-X-Ala-Gly-X-Gly. X may represent N-Acetyl-N-OH-Orn or N-propionyl-N-OH-Om. Because of the three orni-
thine residues and the peptidic nature of the ferrocins they resemble the albomycins, although the latter contain a nucleoside residue.
Pseudobactins and Pyoverdms

Strains of the Pseudomonas fluorescens group (P. fluorescens, P. putida, P. aeruginosa, P. syringae, and related species) produce yellow-green fluorescent siderophores. Typical structures are the linear pseudobactin (Fig. 18) (TEINTZE al., 1981) and pyoveret din Pfl2, containing a cyclic substructure (Fig. 19) (BUDZIKIEWICZ, 1993). Pseudobactins and pyoverdins contain a common structural
Fig. 18. Pseudobactin.
Fig. 19. Pyoverdin Pfl12.
214
element, the chromophore which is a derivative of 2,3-diamino-6,7-dihydroxyquinoline, possessing an additional 1-carboxyl-pyrimidino ring (S configuration) and an amino group acylated with different dicarboxylic acid residues: succinic acid, succinamide, 2-0x0-glutaric acid, glutamic acid, malic acid (amide). The chromophore is connected via the C-1 carboxyl group predominantly (for exceptions, see BUDZIKIEWICZ, to the N-ter1993) minus of a peptide chain of varying chain length (6-12 amino acids), half of which possess D configuration. In addition to the catecholate bidentate of the quinoline residue two further amino acids within the peptide chain function as iron bidentates among which N5acyl-N5-hydroxy ornithine is present in all compounds studied so far. The N5-acyl group of ornithine may be formyl, acetyl, or p-hydroxybutyryl or may even be connected to its own carboxyl group (cyclic OH-ornithine). The second iron-binding amino acid is generally threo-p-hydroxy aspartic acid. However, some of the pyoverdins lack p-hydroxy aspartic acid and possess two NS-acyl-N5-hydroxy ornithine residues instead. In addition to the cyclic NS-OH-Orn the peptide chain may possess larger internal cyclic substructures where &-aminogroups (Lys) may be connected via amide bonds to distant C-terminal carboxyl groups or where Ser and Thr form additional tetrahydropyrimidine rings. Although internal rings have been detected in a variety of pyoverdins, pseudobactin is a linear molecule as confirmed by X-ray analysis (TEINTZE et al., 1981). A list of the currently known pyoverdin structures is shown in Fig. 20. After chelation with ferric iron the complex adopts a red-brown color. At pH 7 the Fe complexes show UV-VIS absorption maxima at 400 (log &=4.2), 320 and 280nm and additional charge transfer bands at 470 and 550 nm. Formation constants have been reported to be in the range of 1024-1026.
KIEWICZ et al. (1992) and LINGET et al. (1992) have generally confirmed the composition of ferribactin. Both compounds, ferribactin and pyoverdin, isolated from the same strain have been shown to differ in the chromophore part but are identical in their peptide sequence (BUDZIKIEWICZ al., 1992). et Thus, the composition of ferribactin isolated from Pseudomonus uptutu was similar to pyoverdin Pap (P. uptufa)but lacked the fluorescent chromophore and contained Glu and Tyr/Dbu instead. Ferribactin is a much weaker iron-binding compound than the pyoverdins as the catechol group provided by the chromophore is absent. However, the phenol group of Tyr and the carboxylate group of Glu in ferribactin may possibly substitute the chromophore ligand binding sites. Because of the diversity of the pyoverdin structures produced by the various Pseudornonus species a pronounced specificity during uptake of ferric pyoverdins can be expected and has indeed been observed by HOHNADEL MEYER(1988) and MEYER and (1992). The structural gene for the ferric pseudobactin receptor (PupA) in Pseudomonus putidu WCS358 has been cloned and sequenced (MARUGG al., 1989; BITTER et al., 1991). et An inducible ferric pseudobactin receptor (PupB) of P. putidu WCS358 has also been identified and characterized (KOSTERet al., 1993). Different types of pyoverdin-defective (pvd) mutants have been isolated and characterized (VISCAet al., 1992). The pvd-l mutant is an ~-N-hydroxy ornithine auxotroph unable to oxygenate L-ornithine and requiring ~-N-hydroxy ornithine for pyoverdin production. Other types of mutants appear to be blocked in further steps of the biosynthetic pathway leading to pyoverdin, the acylation of ~-N~-hydroxy ornithine (pvd-2)and chromophore biosynthesis (pvd-3).The oxygenase gene pvd-A (previously named pvd-2) from P. ueruginosu has been cloned (VISCAet al., Fembactin 1994). From a gene bank using the broad-host range cosmid pLAFR3 mobilized in a pvdADuring the isolation of iron-binding pig- defective mutant a trans-complementing cosments from P. fluorescens a non-fluorescent mid pPV4 was obtained which restored pyoprecursor named ferribactin was isolated verdine synthesis and oxygenase activity in (MAURER al., 1968). The studies of BUDZI- the pvdA mutant. et
Bacteria
Pseudomoms putidu ATCC 12633 Pseudomonas putidu CFBP 246 1 Pseudomonas chlororaphis ATCC 9446 PseudomonasjluorescensATCC I3525 PseudomonasjluorescensSB 83 PseudomonasjluorescenrATCC I 7400 (Demange et al., 1990a) PseudomonasjluorescensCCM 2798 (Demange et al., 1990b) Pseudomonasaeruginosa ATCC I 5692 (Briskot et al., 1989) Pseudomoms rolnasii NCPPB 2 192 (Demange et at., 1990b)
215
Structures of pyoverdins and related compounds
Chr-L-Asp-L-Lys-D-OHsp-(D,L)Ser-L-Thr-D-Ala-D-Glu-(D,L)Ser-L-cOHOrn Chr-L-Asp-(D,L)-Lys-D-OHAsp-D-Ser-L-Thr-D-AIa-(D,L)-Lys-L-Thr-L-cOHOrn
Chr-D-Ser-L-Lys--Gly-(D,L)-FoOHOrn-c~-Lys-~,L)-FoOHOrn-L-Ser] Chr-D-Ser-L-Lys-Gly-(D,L)-FoOHOrn-c~-Lys-(D,L~FoOHOrn-L-Ser] Chr-D-Ala-L-Lys-L-Thr-D-Scr-L-AcOHOrn-L-cOHOrn
Chr-D-Ala-D-Lys-Gly-Gly-L-OHAsp-D-GlnCTHPMD-L-Ser-D-Ala-L-cOHOrn
Chr-SerCTHPMD-Gly-L-Ser-D-OHAsp-L-Ala-Gly-D-Ala~ly-L~OHOrn
Chr-D-Ser-L-Arg-DSer-L-FoOHOrn-c~-Lys-L-FoOHOrn-L-~r-L-Thr]
Chr-D-Ser-L-Lys-L-Ser-DSer-L-Thr-D-Ser-L-AcOHOrn-L-Thr-D-Ser-D-cOHOrn
Azorobacter vinelandii CCM 289 Chr A-L-Asp-DSer-L-Hse-Gly-D-OHAspL-Ser-P (azotobactin D) (Demange et al., 1988) PseudomonasjluorescensATCC I 3525 L-Glu-D-TyrCTHPMD-DSer-L-Lys-Gly~D,L)-FoOHOrn~[L-Lys-(D,L)-F~HOrn-L-Ser] (desferribactin)(Linget et al., 1992) Azomonas macrocyrogenes ATCC I2334 Chr-L-Hse-D-AcOHOrn-D-Ser-L-AcOHOrn-D-Hse-L-CTHPMD (azovcrdin) (Bcrnardini et al., 1996) For additional structures see:Abdallah 1991 and Budzikiewicz 1993.
Chr = Chromophore, ChrA = chromophoreof azotobactin, OHOm and OHAsp = N'-hydroxyornithine and Bhydroxyaspartic acid, c = cyclic, cOHOrn = cyclo-N*-hydroxyornithine,Fo and Ac = formyl and acetyl present at the N'-nitrogen atom of hydroxyornithines, SelCTHPMD, TyrCTHPMD and GlnCTHPMD represent the amino acids derived from tetrahydropyrimidine resulting from ring formation of L-2,4-diaminobutyric acid with Ser, Gln or Tyr:
R = succinic acid (amide)
malic acid (amide) 2-ketoglutaric acid (amide) glutamic acid Chromophore
OH
-NH
Fig. 20. Structures of pseudobactins and pyoverdins.
k30
SerCTHPMD
TyrCTHPMD
GlnCTHPMD
216
Azotobactin The peptide siderophore azotobactin (Fig. 21) seems to be the most prominent siderophore of the Azotobacteriaceae showing a striking similarity to the pyoverdins. The main structural difference between the azotobactins and pyoverdins is that azotobactins form an extra imidazolone ring connecting the amino group at C-5 to the chromophore N-4 by an urea unit, whereas the pyoverdins possess an amidically bound dicarboxylic acid at that position (CORBIN al., 1970). As this et is a minor structural difference, some authors have included the azotobactins in the pyoverdin group (MENHARDT al., 1991). et
OH
OH
0
nithine acylated with 3-hydroxy butyric acid (BUDZIKIEWICZal., 1992). et Using 55Fe-labeled compounds it was shown that azotochelin and azotobactin do function as siderophores in A. vinelandii (KNOSP al., 1984). However, siderophoreet mediated uptake of iron was not observed until substantial non-specific binding of iron to the cells was eliminated. Transport of azotobactin and azotochelin was also decreased by high concentrations of cations, such as Na +, K + , Li +, Mg2+, due to some interference or with the siderophore transport system. VISWANATHA his group (MENHARDT and et al., 1991) have recently discussed the heterogeneity of the pyoverdins and azotobactins observed during growth under iron limitation. From their pulse labeling experiments it was inferred that an amino acid deficiency or a lack of stringency may be responsible for the observed heterogeneity of the produced azotobactins and pyoverdins. Ornibactins Ornibactins (Fig. 22) represent modified tetrapeptide siderophores containing N-terminal3-hydroxy-acyl residues of different chain lengths (C4, C6, (3). The peptide contains two t-Orn (N5-OH, N5-acyl), one Asp (pOH), one L-Ser and a C-terminal lP-diamino butane residue (STEPHAN al., 1993a, b). et Thus the ornibactins resemble the pyoverdins but they possess an acyl residue instead of a chromophore and are much smaller peptides. They have been isolated from Pseudomonas cepacia-like strains. The cepacia group
i
Fig. 2 . Chromophore of azotobactin. 1
FUKASAWA al. (1972) described the first et structure of the yellow-green fluorescent peptide, azotobactin 0, produced by Azotobacter vinelandii (strain 0) which contained only two bidentate chelating groups. Two further structures have been elucidated by now: azotobactin D (DEMANGE al., 1988) and azotobacet tin from A. vinelandii DSM 87 (ATCC 12837),the latter containing an N-hydroxy or-
0
I
Fig. 22. Ornibactins. Ornibactin C4 (R =methyl); ornibactin C6 (R=propyl); ornibactin C8 (R = pentyl).
217
(rRNA homology group 11) has been recently transferred to the genus Burkholderia (YABUUCHI et al., 1992). Ornibactin production has been found in clinical isolates as well as in nitrogen-fixing bacterial isolates from the rhizosphere of rice plants. Moreover, ornibactinmediated iron uptake was observed in all Burkholderia strains but was absent in Pseudomonas aeruginosa, P. fluorescens, and P. stutzeri (MEYERet al., 1995).
in ethanol was added to convert the mycobactins completely to the ferric form. An equal volume of chloroform was added, followed by water until two layers were formed. The chloroform layer was washed three times with water to remove excess iron; it was then dried with MgS04, evaporated to dryness, and the residue was dissolved in methanol. As pointed out by SNOW(1970), the characteristic UV absorption spectra of the iron-free mycobactins arise from two parts of the molecule, the 2-(o-hydroxyphenyl)-oxazoline structure and the acyl hydroxamic acid. De3.4 Mycobactins and Related pending on the substitution of the benzene Siderophores ring the following A,, values in methanol are given: (no methyl group) 243, 249, and Mycobacteria produce a series of lipid-sol- 304 nm with inflection at 258 nm; (with meuble, iron-binding compounds termed myco- thyl group at position 6) 250 and 311 nm with bactins (Fig. 23a) which have been described a shoulder at 254nm and an inflection at in previous reviews by SNOW (1970) and 265 nm. Growth of Mycobacterium paratuberculosis RATLEDGE (1987). Among the different mycobactins (A, F, H, M, N, P, R, S, T), myco- (formerly named M. johnei) was stimulated in bactin P, isolated from Mycobacterium phlei, the presence of 5-15 ng mycobactin P (FRANand mycobactin T, isolated from M. tubercu- CIS et al., 1953). However, other mycobactins losis, were analyzed in more detail (SNOW, also gave significant effects so that there 1965 ). After splitting the ester link, two prod- seems to be little specificity. M. paratubercuucts known as mycobactic acid and cobactin losis and some Mycobacteria related to Mycowere obtained. Regarding the P type mycobacterium avium have lost the ability to synbactin, chemical analysis of the mycobactic thesize mycobactins and are, therefore, stimuacid unit yields (1) an aromatic acid, either lated by the addition of exogenous mycobacsalicylic acid or 6-methyl salicylic acid, the lat- tins. Nocobactins have been isolated from ter undergoing further degradation to m-cre- Nocardia species as compounds equivalent in sol and carbon dioxide; (2) a P-hydroxy ami- function to the mycobactins and similar in and no acid, either serine or threonine; (3) a mix- structure (RATLEDGE PATEL,1976). Mycobactins are extremely lipophilic comture of homologous long-chain fatty acids, usually with a double bond adjacent to the pounds and are generally not excreted into carboxyl group; (4) N6-hydroxy lysine. The the medium. Therefore, RATLEDGE(1987) cobactin unit yields a hydroxy acid, either 3- suggested that mycobactins function as an hydroxy butyric acid or 3-hydroxy-2-methyl iron shuttle within the lipid boundary of the pentanoic acid, and a second molecule of N6- cell surface. Although salicylic acid is excreted into the medium, its contribution to hydroxyl lysine. The mycobactins are associated with the iron solubilization seems to be insignificant. mycobacterial cell walls and have to be ex- Instead, water-soluble exochelins are thought tracted by organic solvents. In order to avoid to function as iron scavengers and to transextraction of undesired fatty material, the port iron to the mycobactins (STEPHENSON most convenient method is to suspend the and RATLEDGE,1979). The mechanism procells in cold ethanol. HALL and RATLEDGE posed for iron uptake into Mycobacterium (1982) have reported a simple method for the smegmatis involves the mediation of mycoproduction and isolation of mycobactins: My- bactin in shuttling iron across the boundary and MARSHALL, cobacteria were grown on solid media and layers of the cell (RATLEGE scraped from the agar after growth. The bac- 1972). The loading of iron into mycobactin is teria were then held for 24 h in ethanol. FeC13 presumed to occur via exochelins which are
218
a
HO
C
a
m q N , 0
OH
OH
Fig. 23a-c. General structure of a mycobactins (for side chains and nomenclature see SNOW,1970); b exochelin MS: c exochelin MN.
able to solubilize iron from the environment (MACHAM et al., 1975; STEPHENSONand RATLEDGE, 1979). Structure elucidation of exochelins from M.smegmatis has shown the presence of a modified linear pentapeptide
containing three N'-hydroxy-ornithines (Fig. Z3b) (SHARMAN al., 1995a). Another novel et water-soluble iron-binding linear hexapeptide named exochelin MN has recently been isolated from M. neoaururn (SHARMAN al., et
219
1995b) containing P-hydroxyhistidine, 2 @ alanines, N"-methyl-'N-hydroxyornithine, ornithine, and cyclic 'N-hydroxyornithine (Fig. 23c). In addition, exochelins showing structural similarities with mycobactins have been isolated from M. tuberculosis (GOBINet al., 1995, LANEet al., 1995; RATLEDGE Ewand ING, 1996). These exochelins contain dicarboxylic acid methyl esters instead of the long chain fatty acids occurring in the mycobactins.
Acinetobactin
has previously been reported to produce a novel citrate-based siderophore, acinetoferrin (OKUJO al., 1994b). et
Frankobactin
From low-iron cultures of Acinefobacter baumannii ATCC19606 another compound with both catecholate and hydroxamate functional groups was isolated containing 2,3-dihydroxy benzoic acid and threonine combined to an oxazoline ring and o-N-hydroxy histamine (YAMAMOTO al., 1994a). The et structure was elucidated by chemical degradation, fast atom bombardment mass spectrometry, and 'H and 13C NMR spectroscopy. Acinetobactin (Fig. 24) was identified in 4 of
0
The siderophore of Frankia (strain 52065), named francobactin, has recently been isolated and shown to be a novel hydroxamatecontaining peptide siderophore (Fig. 25). The following components have been tentatively identified (MW 731): a phenyl oxazoline ring, serine, glycine, ornithine, and P-HO-aspartate (BOYERand ARONSON, 1994; ARONSON and BOYER,1994). The actinomycete Frankia is a gram-positive, aerobic, filamentous, sporulating, and nitrogen-fixing bacterium that exists as a free-living saprophyte or as a symbiont in the roots of the actinorhizal plants including alder trees (e.g., Afnus glufinosa), Ceanofhus americanus, Myrica gale, and several species of Casuarina. Unlike the Rhizobium strains, Frankia strains are also able to fix nitrogen in aerobic culture which makes Frankia an interesting organism to study siderophore production under symbiotic and non-symbiotic conditions.
Maduraferrin
7
NH
Fig. 24. Acinetobactin.
12 clinical isolates of A . baumannii. Acinetobactin is structurally related to anguibactin, a siderophore isolated earlier from Vibrio anguiffarumwhich has a thiazoline ring instead of an oxazoline ring (JALALet al., 1989). Another Acinefobacter species, A . huemolyticus,
An Acfinomadura madurae strain (Actinomycetes) produces the novel siderophore maduraferrin (KELLER-SCHIERLEIN al., et 1988). Maduraferrin (Fig. 26) is the iron complex of an oligopeptide siderophore composed of salicylic acid, p-alanine, glycine, Lserine, N5-hydroxy-N2-methyl-~-ornithine, and ~-hexahydropyridazine-3-carboxylic acid. Although most Acfinomadura strains produced various ferrioxamines, some strains, like strain DSM 43067, were found to pro-
Serine
Glycine
ornithine
HYhXYaSpertate
Fig. 25. Frankobactin.
220
R'
OH
N H y ( C H d m-N
Fig. 26. Maduraferrin.
duce desferri-maduraferrin. Because of the low growth in chemically defined media, maduraferrin production was performed in a complex medium (soy, yeast extract) with the addition of AlC13 which mimics iron deficiency. Transport of iron mediated by maduraferrin was demonstrated in the producing strains but was absent in E. coli or Staphylococcus aureus, suggesting a special uptake route in Actinomadura strains.
+
H H
i -C n -R'
0
3.5 Citrate Hydroxamate Siderophores

Schizokinen
Schizokinen (Fig. 27a), originally described as a factor that reduced the division lag of Bacillus megaterium, B. cereus, and B. subtilis, was isolated from a culture of B. megaterium ATCC 19213 (BYERS et al., 1967). The growth factor activity was confirmed with a schizokinen-auxotroph of B. megaterium. The structure of schizokinen was elucidated by MULLISet al. (1971). Hydrolysis of schizokinen yielded l-amin0-3-(N-hydroxyamino) propane, and oxidation with periodate yielded acetate as the acyl moiety of the hydroxamic acid bonds. The central backbone is composed of a citric acid residue. Titration and CPK models confirmed that both the central hydroxyl and carboxyl group can bind to the metal ion to give an octahedral six-oxygen coordination sphere. At neutral pH the ferric complex is an anion indicating the presence of ionized citrate hydroxyl. The absorption
C
Fig. 27. Citrate hydroxamates.
Schizokinen: R'=R2=H, R3=R4=CH3,n=2; rhizobactin 1021: R' = R2=H, R3= A, R4= CH3, n=2; acinetoferrin: R'=R2=H, R3=R4=B,n=2; arthrobactin: R' = R2= H, R3= R4= CH3, n =4; aerobactin: R ' =R2= COOH, R3=R4= CH3, n =4; nannochelin A: R ' =R2= COOCH3, R3= R4= C, n =4; nannochelin B: R' = COOCH3, R2= COOH, R3= R4=C, n=4; nannochelin c R1=R2=COOH,R3=R4=C,n=4.
(Amax =390 nm) was pH independent between pH 5 and 9.5, but shifted to red below pH 5. The cultures were grown in a chemically defined low-iron medium at 35 "C with strong aeration for one week (3% inoculum). After removing the cells, the supernatant was concentrated to 1/10 of the original volume, acidified with conc. HCI to pH 2, extracted with chloroform-phenol (1 :l), and transferred
221
with ether to the aqueous phase. This was then concentrated by evaporation and applied to a column of the acetate form of Dowex AG-2-XlO. The column was washed with distilled water and schizokinen was eluted with 0.2M ammonium chloride and further purified on Bio-Gel P-2. TLC on silica gel with methanol (Rf=0.60) using tetrazolium chloride or FeC13 in 0.05 N HCl as sprays was performed to confirm purity. Schizokinen was also found in Anabaena strains (LAMMERS and SANDERS-LOHR, 1982) and may be a siderophore in other cyanobacteria. A derivative of schizokinen, schizokinen A (Fig. 28), has recently been isolated from alcalophilic Bacillus strains containing a cyclic side chain due to the condensation of the carboxyl at the quarternary C atom with one of the amide NH groups (STEPHAN al., unet published results). Schizokinen A has been synthesized by LEE and MILLER(1983). A similar compound was isolated earlier from B. megaterium. In the original publication, however, this product was incorrectly described to have a six-membered ring (MULLIS et al., 1971). It is still unclear whether or not cyclization to schizokinen A is an artifact arising during the isolation procedure of schizokinen or if it is a real natural product. The occurrence of imido forms in citrate containing siderophores has also been reported for staphyloferrin A (KONETSCHNY-RAPP al., et 1990) and rhizoferrin (DRECHSELet al., 1992). Rhizoferrin, e.g., tends to give both, imidorhizoferrin and bis-imidorhizoferrin (DRECHSEL al., 1992). et BYERSand co-workers were the first to investigate the transport of iron in B. megaterium, using double-labeled 59Fe-3H-schizokin-
en. The results indicated an initial temperature-independent binding of ferric schizokinen, followed by a temperature-dependent (active) transport of the chelate into the cell, and an enzyme-catalyzed separation of iron from the chelate (ARCENEAUX al., 1973). Moreet over, a mutant of B. megaterium S K ll unable to synthesize schizokinen showed recognition capacity for the chemical structure of schizokinen. Ferrioxamine B (Desferalm) was taken up at lower rates and aerobactin did not supet port uptake of iron (HAYDON al., 1973). A recent investigation has shown that B. subtilis utilizes three types of hydroxamate siderophores: ferrichromes, ferrioxamines, and schizokinen (SCHNEIDER and HANTKE,1993). Moreover, the transport system showed significant homology to the binding protein-dependent components of E. coli. As a periplasm is missing in gram-positive bacteria, the FhuD-corresponding protein is anchored as a lipoprotein in the cytoplasmic membrane.
Rhizobactin 1021
Rhizobactin 1021 (Fig. 27b) is a novel asymmetric citrate-based dihydroxamate isolated from Rhizobium meliloti strain 1021. While acinetoferrin contains two identical hydroxamic acid acyl groups as (E)-Zoctenoic acids, rhizobactin 1021 contains two different acyl residues, (E)-2-octenoic acid and (E)-2decenoic acid (PERSMARK al., 1993). Ferric et rhizobactin 1021 predominantly exists in solution in the lambda configuration in an apparent equilibrium of a monomeric and a dimeric species. Acinetoferrin Acinetoferrin (Fig. 27c) has been isolated from Acinetobacter haemolyticus and shown to belong to the citrate-based dihydroxamates (OKUJOet al., 1994b). Structure elucidation revealed that acinetoferrin is a structural analog of schizokinen as it contains a citric acid backbone linked amidically to two l-amino3-(N-hydroxy propane) residues which in turn are condensed by hydroxamic acid bonds to two (E)-2-octenoic acid moieties. Fast atom
c-0 I
Fig. 28. Schizokinen A.
222
bombardment mass spectroscopy revealed a quasi molecular ion peak (MH +) at mlz =585 corresponding to a molecular mass of 584 amu. Maximal production of acinetoferrin occurred when A . huemolyticus ATCC 17906 was grown in a chemically defined medium containing approximately 0.1 pM iron. Absorption spectra of acinetoferrin showed an absorption maximum at 210 nm (2.9.104 M- cm-) with a shoulder at 250 nm, indicative of unsaturated bonds. After addition of iron a red-brown color develops that is characteristic for a ligand-to-metal charge transfer 1. band (h,,,=486 nm; ~ =9 .8 -1 0 * M- cm-
1969). Aerobactin consists of a conjugate of N6-acetyl-N6-hydroxy lysine and citric acid. After complexing with ferric iron the internal citrate carboxyl and the a-carboxyls of the amino acid residues remain free resulting in a ferric complex with three negative charges at neutral pH. The synthesis of aerobactin was described by MAURERand MILLER(1982). Coordination chemistry, stability constants, ligand protonation, and metal ligand equilibria were determined by HARRISet al. (1979). The ferric aerobactin complex shows an overall formation constant of log Kf =22.93 and an absorption of A,,,=398 nm (e=2170 M- cm-). Based on the pM value (stability under specified concentration and pH) aerobactin can effectively compete with transferrin. Arthrobactin The comparatively low redox potential Arthrobactin (Fig. 27d), previously known (-0.336 V) allows reduction via physiological as the terregens factor, is produced by reductants and reuse of the ligand, which strains of Arthrobacter and supported the seems to be an advantage for the producing growth of the iron-auxotrophs Arthrobacter organism. The CD spectrum revealed that terregens and A . jlavescens JG9 ATCC 29091. ferric aerobactin predominantly exists as the The structure of arthrobactin was elucidated lambda optical isomer in aqueous solution. Aerobactin can be isolated from strains of by LINKE al. (1972). It contained two l-amiet no-5-(N-acetyl-N-hydroxy)aminopentane re- Enterobacter aerogenes using chemically desidues linked symmetrically to citric acid by fined media without added iron as described peptide bonds. Arthrobactin was isolated by GIBSON and MAGRATH(1969). After from low-iron cultures of Arthrobacter pm- inoculation and incubation for 20 h at 37C cens (ATCC 13346) in an asparagine-salts with stirring and aeration, the cells are sepamedium. After inoculation with a spore sus- rated and the supernatant is passed through a pension the culture was incubated for 24 h at Dowex-1 (Cl -) column. All the material giv27C. Then FeC13 was added, the culture fil- ing a red color with FeC13 is held on the cotrate was saturated with ammonium sulfate, lumn. The column is then eluted with an amand extracted with benzyl alcohol. Diethyl- monium chloride gradient (0.4-1 M). 2,3-Diether was added to transfer the ferric arthro- hydroxy benzoate derivatives are eluted first, bactin complex into the aqueous phase. The followed by aerobactin. The fraction containcrude material was purified by countercurrent ing aerobactin is then evaporated to dryness, extraction and by chromatography and gel fil- dissolved in water, and further purified on a tration. Arthrobactin and schizokinen were Dowex 50W-X8 ( H + ) (200-400 mesh) colsynthesized chemically (LEE and MILLER, umn with water. FeC13-positive fractions are collected, freeze-dried, and dried over P205 1983). for several days. Aerobactin is also produced by Enterobacter cloacae and Shigella flexneri (PAYNE, Aerobactin 1980), Erwinia carotovora (ISHIMARU and Examination of supernatants from cultures LOPER,1992), and by Escherichia coli strains 1981). of Aerobacter aerogenes 62-1 (now called En- containing the plasmid ColV (BRAUN, terobacter aerogenes) and related strains It was shown that aerobactin uptake requires showed the presence of a hydroxamic acid for the expression of an outer membrane recepwhich the trivial name, aerobactin (Fig. 27e), tor (Iut A, 74 kDa) which serves as a common was suggested (GIBSON and MAGRATH, binding site for aerobactin and the bacterio-
223
cin cloacin (VAN TIEL-MENKVELD al., et 1982; BINDEREIF al., 1982). The genes codet ing for aerobactin synthesis were analyzed by GROSSet al. (1985) and DELORENZO al. et (1986). Aerobactin production also seems to be widespread in other bacterial genera. Besides in enterobacteria, it has been detected OH in cultures of a halophilic pseudomonad (BUYERet al., 1991) and in a culture of a COOH u gram-positive DSM 4640 strain (WINKELMANN, unpublished data). Fig. 29. Rhizobactin DM4. Nannochelins Citrate hydroxamate siderophores named nannochelins have been isolated from lowiron cultures of Nannocystis excedens (Myxobacteria) (KUNZE al., 1992) (Fig. 27f). Nanet nochelins resemble aerobactin, but contain two N-E-cinnamoyl residues instead. While nannochelin C contains two non-esterified lysyl carboxyl groups, nannochelin B has one and nannochelin A two methyl-esterified carboxyl groups.
NH
3.6 Carboxylate Siderophores

Because of their high stability constants and their pronounced colored ferric complexes, catecholate and hydroxamate siderophores were the first microbial iron chelates detected. However, with novel chemical and biological assay methods being available, e.g., the chrome azurole S test (CAS test), additional carboxylate-type siderophores have been detected (SCHWYNand NEILANDS, 1987). Rhizobactin DM4 Rhizobactin DM4 (Fig. 29) isolated from Rhizobium meliloti DM4 (SMITH al., 1985) et should not be confused with rhizobactin 1021 et isolated from R. meliloti 1021 (PERSMARK al:, 1993), although both have been isolated from isolates of the same species. Rhizobactin DM4 contains a characteristic ethylene di-
amine bridge between a (non-amidically linked) alanine and lysine residue, the latter being N6-acylated by malic acid. Thus, rhizobactin DM4 is regarded as a complexone type or carboxylate type siderophore, while rhizobactin 1021 is a citrate-based dihydroxamate. It was anticipated that the iron complex of rhizobactin DM4 is hexacoordinated involving the two ethylene diamine nitrogens, the two amino acid carboxylates, and the a-hydroxy acid function of malic acid (NEILANDS, 1993). There is structural similarity of rhizobactin DM4 and the mugineic acids isolated as phytosiderophores from grasses having a diamine backbone with (a-hydroxy)carboxylate side functions. Mugineic acids may be also regarded as complexone-type siderophores (NOMOTO al., 1987). et Staphyloferrin A Staphyloferrin A (Fig. 30) isolated from staphylococci is a carboxylate type siderophore consisting of two citric acid residues linked amidically to D-ornithine (KONETSCHNY-RAPPet al., 1990; MEIWESet al., 1990). Staphylococcus hyicus DSM 20459 was used to produce staphyloferrin A in larger amounts. In addition, three dehydration products were obtained, two mono forms and one diimido form, as a result of intramolecular condensation of the central carboxyl groups with the amide NH groups. Interestingly feeding with D- and L-ornithine led to incorporation and synthesis of D-ornithinecontaining staphyloferrin A. Transport experiments with 55Fe-labeled compounds
224
a
COOH
I
taric acid residue can easily form a five-membered ring by nucleophilic addition of the amide nitrogen to the carbonyl carbon. NMR experiments showed NOE contacts confirming the existence of a cyclization product as shown in the structural formula of staphyloferrin B. Whether this shift of equilibrium from the open to the cyclized form is caused by sample preparation or measurement conditions is still unclear. Vibrioferrin
Fig. 3Oa. Staphyloferrin A, b staphyloferrin B; c vibrioferrin.
A structurally novel siderophore was isolated by successive column and TLC from Vibrio parahaemolyticus (YAMAMOTO al., et 1992). This organism is known to occur in marine and estuarine environments and causes diarrhea associated with seafood consumption. Vibrioferrin was negative in the chemiand (for descripcal assays of ARNOW CSAKY tion, see NEILANDS NAKAMURA, and 1991) excluding the presence of catecholates and hydroxamates. The constituents of the siderophore found after hydrolysis were L-alanine, ethanolamine, citric acid, and 2-oxoglutaric acid. Fast atom bombardment mass spectrometry revealed a mass of 446 amu. A complete structure has recently been reported (YAMAMOTO al., 1994b). 2-oxoglutaric et acid is amidically linked to L-alanine and, like in staphyloferrin B, exists in equilibrium with a five-membered hydroxylactam (Fig. 3012).
Cepabactin Cepabactin (Fig. 31) is a natural bacterial pyridinone siderophore (l-hydroxy-5-methoxy-6-methyl-2(1H)-pyridinone) isolated from Pseudomonas cepacia ATCC 25416 (MEYERet al., 1989). Cepabactin can be re-
showed that ferric staphyloferrin A acts as a siderophore in staphylococci. Staphyloferrin B Staphyloferrin B has also been isolated from Staphylococci, however, the structural components differ completely from staphyloferrin A (DRECHSEL al., 1993a) in that citet ric acid is linked by a 2,3-diamino propionic acid residue at one side and by a diamino ethane and a 2-0x0 glutaric acid residue at the other side (Fig. 30b). The molecular mass was determined by pneumatically assisted electro spray from aqueous solution and indicates hydration of the 2-0x0 function. The 2-oxoglu-
Fig. 3 . Cepabactin. 1
OH
4 Fungal Siderophores
225
garded as a cyclic hydroxamate or as a keto hydroxy bidentate.
3.7 Keto Hydroxy Bidentates

There seems to be a wide range of compounds in nature possessing keto hydroxy bidentate ligands, most of which have not been shown to be involved in iron transport. However, some selected keto hydroxy bidentate ligands proved to be very effective in iron transport, as demonstrated recently (THIEKEN and WINKELMANN, 1993) using bacteria of the group Proteeae (Proteus, Providencia, Morganella) all of which are able to produce a-keto acids as a result of oxidative deaminaet tion of amino acids (DRECHSEL al., 1993b). Deamination to 2-0x0 acids is a characteristic trait of the Proteeae and is only of minor importance in other enterobacterial genera. Because of the high production rates in Proteeae and their activity in iron nutrition, a-keto acids have been assigned a siderophore function in Proteeae and in other genera (DRECHSEL et al., 1993b; REISSBRODT al., 1994). et Interestingly, strains of Proteus mirubilis were found to be unable to transport iron via the carboxylate type siderophore rhizoferrin but still were able to transport ferric complexes of keto hydroxy bidentate ligands (THIEKEN and WINKELMANN, 1993), suggesting that the transport systems for ferric rhizoferrin and ferric complexes of keto hydroxy bidentate ligands are different. Although rhizoferrin is a
R
0
fungal siderophore (see Sect. 4.3, uptake kinetics and characterization of the genes involved in Morganella morganii have recently been described (CARRANO al., 1996 KUHN et et al., 1996).
4.1 Ferrichromes
Ferrichrome (Fig. 32), first isolated in 1952 from Ustilago sphaerogena (NEILANDS, 1952), was found later in other Ustilago species (DEML,1985) as U. maydis (BUDDEand LEONG, 1989), in Neovossiu indica (Tilletiales) (DEML et al., 1984), Lipomycetaceae (VAN WALTet al., 1990) and the dermader tophyte Trichophyton mentugrophytes (MOR et al., 1992). Besides ferrichrome, Ustilago species also produce ferrichrome A, the structure of which is closely related to ferrichrome. Crystal structures of ferrichrome A and ferrichrome were determined by ZALKIN al. et (1966) and VAN DER HELM et al. (1980) showing a cyclic, modified Om-0rn-0rn-GlyGly-Gly hexapeptide. Ferrichrome may be regarded as a prototype siderophore from which a variety of other ferrichromes are derived, e.g., ferricrocin, ferrichrysin, ferrichrome c , ferrichrome A, ferrirubin, ferrirhodin, malonichrome. An updated detailed
(
O0
Fig. 3 . Ferrichromes. 2 Ferrichrome: R =R2=H, R3=CH3; ferrichome A: R=R2=H, R 3 = A ferricrocin: R=H, R2= CH20H, R3=CH3; ferrichrysin: R = R2= CH20H, R3= CH3; ferrirubin: R = R2= CH20H, R3=B; ferrirhodin: R = R2= CH20H, R3= C; asperchromes: R = R2= CH,OH, R3=mixture of C and CH3.
226
description is given by VAN DER HELMand moved on a Sephadex LH 20 column and a WINKELMANN (1994). With the exception of final separation on a silica gel-silica gur (1:l) tetraglycyl ferrichrome, a heptapeptide column with dichloro methane-methanol (DEMLet al., 1984), all other members of the (2:l) as an eluting solvent (DEML et al., ferrichrome family possess a cyclic hexapep- 1984). HPLC separation using acetonitrileet tide backbone. Ferrichrome type sidero- water mixtures (KONETSCHNY-RAPP al., phores contain a characteristic tripeptide se- 1988) or TLC on silica gel plates using dichloquence of NS-acyl-NS-hydroxy-~-ornithinero methane-methanol-water (65:25:4) as a and a variable tripeptide sequence of short solvent system are recommended for identifiamino acids (Gly, Ser, Ala). Variations in the cation and isolation. A detailed survey on the latter tripeptide sequence and in the nature of methods of isolation and identification of funthe N-acyl substituents of the ornithine resi- gal siderophores is given by JALALand VAN dues account for the number of distinct fer- DER HELM(1991). The biological properties of siderophores richrome-type siderophores mentioned above. X-ray diffraction studies and CD can be studied by growth promotion tests or measurements in solution have revealed that by transport experiments with radio-labeled the configuration of all ferrichromes is lamb- siderophores as discussed in previous reviews da-cis (VAN DER HELMet al., 1980; LEONG (WINKELMANN, 1991a, 1993). Originally ferand RAYMOND, 1974). The stability of desfer- richrome and ferrichrome A were characterri-ferrichrome with ferric ions is in the range ized by their growth-stimulating effect on Pilobolus and Arfhrobacter species. A. flavesc= of K ~ 1029-1031. ens (now reclassified as Aureobacterium flavVAN DER HELMand co-workers have isolated the so-called asperchromes as minor escens) JG9 (ATCC 25091) is a siderophoreconstituents in low-iron culture filtrates of auxotroph and requires trace amounts of ferAspergillus ochraceus, containing a common richrome or other hydroxamate siderophores cyclic Om-0rn-0rn-Ser-Ser-Gly hexapeptide but does not respond to catechols (RABSCH sequence as in ferrichrysin, ferrirhodin, and and WINKELMANN, 1991; SHANZERet al., (1971) was the first to show ferrirubin but different ornithyl-N-acyl 1988). EMERY groups (JALALet al., 1984). Interestingly, a that ferrichrome indeed functions as a sideroferrichrome-type siderophore with an incom- phore in Ustilago sphaerogena. Ferrichrome is plete, open-chain peptide was detected and actively taken up. The iron-free ligand is exnamed des(diserylglycy1)ferrirhodin (JALAL creted again and can be used for another et al., 1985). The latter has recently also been round of transport. Contrary to that, ferrifound in the edible mushroom Agaricus bi- chrome A does not enter the cells but donates iron to the external face of the cell membrane sporus (ENG-WILMOT al., 1992). et Ferrichrome was originally isolated from where it is removed by a reduction process Ustilago sphaerogena grown in iron-contain- (ECKERet al., 1982a). A mutant of Neurospoing medium. A key observation was made ra crassa, unable to synthesize siderophores when the production of ferrichrome could be in the absence of ornithine, has been used to enhanced by growing Ustilago in an iron-defi- study uptake of ferrichrome-type sideroand cient medium. The composition of the me- phores and coprogen (WINKELMANN dium for the production of ferrichrome and ZAHNER,1973). Uptake of ferrichromes in N. ferrichrome A is a modified Grimm-Allen crassa and Penicillium parvum has been medium. To isolate ferrichromes, iron is ad- shown to be highly stereospecific (WINKELded under stirring to the cell-free supernatant MA, 1979; WINKELMANN and BRAUN, of a low-iron culture after a week of growth 1981). Thus enanfio-ferrichromes (NAEGELI 1978) with strong aeration. The ferrichromes can and KELLER-SCHIERLEIN, were not reeasily be adsorbed on a column of XAD-2 recognized by the fungal siderophore transport sin and, after washing with distilled water to systems (HUSCHKA al., 1985, 1986). et remove salts and other polar constituents, the ferrichromes can be desorbed with acetonewater (1:l) or methanol. Lipids can be re-
227
and VAN DER ia sp. (for review, see JALAL HELM,1991). Coprogen (Fig. 33) has been isolated from Analysis of coprogen transport in wild type Penicillium species, from Neurospora crassa, and mutant strains of Neurospora crassa using Curvularia, and a variety of other fungi (JAL- single-labeled 55Fe-coprogen, 14C-coprogen, AL and VAN DER HELM,1991). A method for or double-labeled 55Fe-'4C-coprogen (WINet coprogen production and isolation from N. KELMANN and ZAHNER, 1973; HUSCHKA crassa was described by WONGet al. (1983). al., 1985) indicated that the intact coprogen The structure of coprogen was elucidated by molecule can enter the cell. Furthermore, KELLER-SCHIERLEIN DIEKMANN and (1970) Mossbauer spectroscopic studies (MAT1981) confirmed and revealed a linear trihydroxamate sidero- ZANKE and WINKELMANN, phore composed of 3 mol of N5-acyl-N5-hy- that after uptake a major portion of intraceldroxy-L-ornithine, 3 mol of anhydro meva- lular coprogen remains in the iron complex lonic acid and 1 mol of acetic acid. Desacetyl form and iron is removed by reduction only coprogen, named coprogen B, was isolated slowly. Moreover, an additional ferrichromefrom Fusarium dimerum (DIEKMANN, 1970). type siderophore, ferricrocin, was found to be A further coprogen, triornicin, was found in synthesized only as an intracellular siderolow-iron culture filtrates of Epicoccum pur- phore functioning as an iron storage compurescens (FREDERICK al., 1981). Curvu- pound from which iron is gradually removed et laria lunata produces neocoprogen I and neo- by reduction. Uptake and competition expericoprogen 11, in which the two outer anhy- ments have shown that N. crassa possesses dromevalonic acid residues are substituted by two different siderophore recognition sites, one or two acetyl groups, respectively. The one for ferrichrome-type siderophores one CD spectrum of coprogen showed a delta confor coprogen, both of which donate their sidefiguration in solution (WONG et al., 1983), rophores to a common transport system loand the crystal structure of neocoprogen I re- cated in the cytoplasmic membrane (HUSCHvealed a delta-trans configuration (HOSSAIN KA et al., 1985; CHUNG et al., 1986). Details et al., 1987). Further coprogen derivatives, of the actual transport process are still unresuch as hydroxy coprogen, N"-dimethyl co- solved. However, there is evidence that the progen as well as their corresponding neo- transport of siderophores in fungi is functionderivatives have been isolated from Alternar- ally connected to the membrane potential
4.2 Coprogens
Fig. 33. Coprogens. a Coprogen: R' =H, R2=COCH3, R3= R4 = A. Coprogen B: R ' = R 2 = H , R3=R4 =A. Neocoprogen I: R' =H, R2= COCH3, R3= CH,, R4= A. Neocoprogen 11: R'= H, R~ =COCH,, R3= R4= CH3. Nu-dimethyl coprogen: R' = R =CH,, ~ R3=R4=A. b Desferricoprogen: R3=R4 =A.
H"^ O(
228
generated by the plasma membrane ATPase (HUSCHKA al., 1983). et
4.3 Rhodotorulic Acid

A compound with the properties of a secondary hydroxamic acid strongly binding iron(II1) was isolated from supernatants of iron-deficient cultures of Rhodotorula pilimanae (ATKINand NEILANDS, 1968). This compound, named rhodotorulic acid (RA) (Fig. 34a), was characterized as the diketo piperazine of N-acetyl-L-N-hydroxy ornithine and has subsequently been found in species of Leucosporidium, Rhodosporidium, Sporidiobolus, and Sporobolomyces (ATKIN et al., 1970). The diketopiperazine moiety is also present in coprogen as dimerum acid (Fig. 34b). Thus, rhodotorulic acid (RA) is structurally related to the coprogen family. The iron complex of rhodotorulic acid at neutral pH has been found to be dimeric with the formula Fe2(RA)3,where both iron atoms have and the delta-cis configuration (CARRANO RAYMOND, 1978). Below pH 3 this complex dissociates to give the monomer, FeRA+, in which each iron is bound to two hydroxamate ligands. Ferric RA has an absorption maxiM-' cm-') at pH 7 mum at 425 nm ( ~ = 2 7 0 0 and becomes red at pH 2 with an absorption maximum at 480 nm ( ~ = 1 7 5 0 M-' cm-'). Ferric RA functions as an iron transport agent in Rhodotorula pilimanae and probably in other RA-producing heterobasidiomycetous yeasts. RA mediates iron transport to the cell but does not actually transport iron into the cell (CARRANOand RAYMOND, 1978). Citrate was found to be as effective in iron transport as RA in this fungus.
NH
Fig. 35. Fursarinines. a Fusarinine A. n =2, fusarinine B: n = 3; b fusarinine C (Fusigen), R = H ; triacetyl fusarinine C (triacetyl fusigen), R = COCH3.
4.4 Fusarinines (Fusigens)

Fusarinines (Fig. 35) had originally been isolated from Fusarium cubense (DIEKMANN, 1967) and F. roseum (SAYER and EMERY, 1968) and were later shown to occur in other genera, e.g., Giberella, Aspergillus, Penicillium, and a variety of other fungi (DIEKMANN, 1968; JALALet al., 1986, 1987). Recently, fusarinine C was described as an endogenous siderophore in Agaricus bisporus (ENG-WILMOT al., 1992). Alkaline degradet ation yielded 3 mol of the monohydroxamic acid cis-fusarinine (Fig. 35a) in which N5-hydroxy-L-ornithine is N-acylated by cis-5-hydroxy-3-methyl-2-pentenoic acid. The three fusarinine molecules are esterified head-totail to build the cyclic triester fusigen (Fig. 35b). The free amino groups show a pK of 7.1
Fig. 3 . Rhodotorulic acid: R = CH3; dimerum acid 4 R=A.
A:
&
OH
229
and the ferric iron complex is stable up to pH 1. In addition to the cyclic triester, the linear trimer (fusarinine B), the linear dimer (fusarinine A), and the monomer (fusarinine) can be found in the culture filtrate. Another member of the fusigen family, N,N',N"-triacetyl fusarinine C (TAFC) or triacetyl fusigen (Fig. 35c), has been isolated from Aspergillus and KREZfumigatus strains (DIEKMANN DORN, 1975) and from Penicillium sp. 1976). The ferric tri(MOOREand EMERY, acetyl fusarinine C is a relatively flat molecule with a total thickness of about 45 pm as determined from the crystal structure (HOSSAIN et al., 1980). Two different coordination isomers of the triacetyl fusigen molecule have been observed. Crystals from ethanolbenzene adopted a lambda-cis absolute configuration. However, in solution and in the morphologically different crystals from chloroform, the molecule predominantly existed as a delta-cis isomer, as determined by circular dichroism. Fusigen can be isolated from low-iron cultures of Fusarium cubense in an asparaginesalts medium (DIEKMANN and ZAHNER, 1967). The medium is inoculated with a suspension of spores and incubated at 27C on a rotary shaker or in a fermenter with aeration. After 3-5 d FeC& solution is added, and the siderophores are adsorbed on Servachrome XAD-2, washed with three volumes of distilled water, and desorbed with one volume of methanol. The crude siderophore solution is evaporated to dryness and dissolved in 0.01 M ammonium acetate buffer (pH 5 ) and passed through a CM-Sephadex column equilibrated and eluted with the same buffer. Bound fusigen is then eluted with 0.1 M ammonium acetate buffer, and again adsorbed on XAD-2, washed, and desorbed as described before. A further purification can be achieved on Sephadex LH-20 in methanol. Triacetyl fusigen can be isolated from Penicillium and Aspergillus strains using the same medium as described for the production of fusigen (DIEKMANN and KREZDORN,1975). Transport experiments with 55Fe-labeled fusigen in Aspergillus fumigatus revealed high uptake rates, suggesting rapid utilization of the chelated iron by the producing strain (WIEBE and WINKELMANN, 1975). EMERY (1976) reported Fusarium roseum to contain
an enzyme which hydrolyzes the ornithine ester bonds of fusigen (fusarinine C) but not the ferric chelate of fusigen. Penicillium sp. was found to hydrolyze triacetyl fusigen and was fully active on the ferric trihydroxamate chelate. The iron-free form of triacetyl fusigen was also reported to act as an antibiotic on a variety of bacteria grown on minimal media (ANKE, 1977). Bacillus brevis, Clostridium pasteurianum, Pseudomonas fluorescens, and Streptomyces viridochromogenes were found to be sensitive even on complex media.
4.5 Rhizoferrins
Rhizoferrin (Fig. 36) is a novel carboxylatetype siderophore first isolated from low-iron cultures of Rhizopus microsporus var. rhizopodiformis (DRECHSELet al., 1991, 1992; WINKELMANN, 1992). Further studies have shown that rhizoferrin seems to be the characteristic siderophore in the order of Mucorales (Zygomycetes) as a variety of other strains from different genera (Mucor, Phycomyces, Chaetostylum, Absidia, Cokeromyces, Cunninghamella, Mycotypha, and Mortierella) produce rhizoferrin as their only siderophore (THIEKEN and WINKELMANN, 1992). Hydroxamates could not be found. However, recent results from the authors' laboratory point to the existence of hydroxamate siderophores in strains of Basidiobolus and Coniduniobolus (THIEKENand WINKELMANN, published results) which are regarded as a separate genus of Zygomycetes (ZYCHAand SIEPMANN, 1969). The stability constant of rhizoferrin with ferric iron has been determined to be Kf= loB (ALBRECHT-GARY, personal communication). The existence of rhizoferrins as a novel carboxylate siderophore in Mucorales
Fig. 3 . Rhizoferrin. 6
'COOH
HOOC'
230
is especially interesting, since the Zygomycetes are the only fungal class where ferritins as iron storage proteins have been found yet. Ferritins have not been detected so far in Ascomycetes and Basidiomycetes (MATZANKE, 1994). Therefore, rhizoferrin iron transport and ferritin iron storage might coexist without metabolic disturbance. The siderophore activity has been investigated using Fe-labeled rhizoferrin (DRECHSEL et al., 1991; CARRANO al., 1996). Furet thermore, a comparison with ferrioxamines revealed that iron uptake rates mediated by rhizoferrin and ferrioxamines in R. microsporus were similarly effective, suggesting that fungi of the Mucorales are also able to use ferrioxamines as an iron source (DRECHSEL et al., 1991). This has important consequences for their role as pathogens, since some members of this group have been repeatedly isolated during fatal cases of mucormycosis in dialysis patients (BOELAERT al., 1993, 1994). et Thus, a substantial number of patients treated with desferrioxamine (DesferaP) for either aluminium or iron overload have developed mucormycosis caused by Rhizopus (BOELAERT et al., 1991). A simple growth promotion assay using bacteria of the Proteus group has been developed to detect bioactive siderophores containing keto hydroxy bidentate ligands including rhizoferrin and simple keto and hydroxy 1993). acids (THIEKEN and WINKELMANN, This bioassay not only allows the detection of further fungal carboxylate siderophores but also discriminates between carboxylate and hydroxamate producing fungi. Several derivatives of rhizoferrin have been prepared by directed fermentation (DRECHSEL et al., 1995a). Thus the addition of analogous diamino acids or diamines to the culture medium yielded rhizoferrin analogs with either shorter (norrhizoferrin) or longer diamino residues (homorhizoferrin). Even substituted amines were incorporated leading to 2-ketorhizoferrin and oxarhizoferrin. Variation of the citryl residues of rhizoferrin by addition of tricarballylic acid to the culture medium yielded the corresponding mono- and didesoxy rhizoferrin derivatives. As the latter alterations affect the iron-liganding properties of the rhizoferrins, the chelate stability with iron should be
reduced. Uptake of iron rhizoferrin and its metal analogs (Cr, Rh, Ga) was studied in Absidiu spinosu (Mucorales) and in Morgunellu morgunii (Proteeae) (CARRANO al., et 1996). While uptake kinetics in the fungus could be explained by an active transport system via a shuttle mechanism, growth promotion assays with M. morgunii could not easily be explained and were dependent on a number of factors including the nature of the chelating agents used to induce iron deficiency. Two genes, rumA and rumB (rhizoferrin uptake into Morgunellu), encoding an outer membrane protein and a periplasmic protein, respectively, have been identified in M. morgunii (KUHNet al., 1996).
5 Miscellaneous Compounds
Several miscellaneous compounds have been isolated from fungi containing ironbinding catecholate, phenolate, or keto hydroxy bidentate ligands. Most of these compounds originate from the shikimate-chorismate pathway and represent condensation products of tyrosine. They have been characterized earlier as pigments from fungi, mainly of the Boletales, e.g., Boletus, Suillus, Paxillus, Hydnellum, Polyporus, and Xerocomus (GILL and STEGLICH, 1987). The blueing of members of the Boletaceae, and the red and yellow stains of several Aguricus and Cortinurius species have long been used as taxonomic characteristics by fungal taxonomists. In many cases quinones react by the combined action of alkali, oxygen, or ferric choride. Thus compounds like the terphenyl quinones (polysporic acid, atromentin, leucomelone, and variegatin) (Fig. 37) possess both keto hydroxy and phenolic iron binding functionalities. Internal cyclization leads to cycloleucomelone, cyclovariegatin, and thelephoric acid (Fig. 38) having only phenolic groups. Another series of compounds are the pulvinic acids (vulpinic acid, atromentic acid, xerocomic acid, and variegatic acid) (Fig. 39) which, depending on the kind and number of
5 Miscellaneous Compounds
231
b
COZH
OH
I
Fig. 37s. Polysporic acid; b atromentin; c leucomelone; d variegatin.
functional groups, possess good iron-binding properties. Because of the quinoid and polyphenolic structure these compounds can be regarded as iron-binding pigments, but their involvement in iron transport has still to be proven. A novel pyridinone-type iron-binding compound has recently been isolated from strains of the fungus Tolypocladium named tolypo-
cin (HL) (Fig. 40) and terricolin (FeL,), respectively (JEGOROV al., 1993). Members et of the entomopathogenic genus Tolypocladium are known for their ability to produce cyclosporin A and a variety of pigments. Although the iron complex (terricolin) has been characterized by X-ray structure determination, iron transport properties have not been reported. The absolute configuration of terri-
232
6H
Fig. 38a. Cycloleucomelone: R =H,cyclovariegatin: R = OH, b thelephoric acid.
OH
Colin has been shown to be lambda-cis both in solution and in the solid state. Pulcherriminic acid (Fig. 41a) and its iron complex pulcherrimin have been isolated from Candida pulcherrima (KLUYVER et al., 1953). Pulcherrimin, obtained by extraction of the cells with methanolic KOH as an amor- d phous red powder represents a polymer of dihydroxy-pyrazine dioxide which is insoluble in water and may be regarded as an iron storage compound rather than a siderophore. Several related compounds, like the antibiotic mycelianamide (Fig. 41b) from Penicil\ lium griseofulvum (OXFORD RAISTRICK, and OH 1948) or aspergillic acid (Fig. 41c), hydroxy aspergillic acid, muta-aspergillic acid, and Fig. 39a. Vulpinic acid b atromentic acid; c xeroneo-aspergillic acid from culture fluids of As- comic acid; d variegatic acid. pergillus flavus and A. oryzae (DUTCHER, 1958) were isolated. The latter compounds OH I are water-soluble and showed growth promotion activity with A. flavescens JG. However, siderophore activity in the producing strains has not been demonstrated.
Fig. 4 . Tolypocin. 0
6 Transport Mechanisms
233
OH
(1) Diffusion of siderophores across the membranes along a diffusion gradient. (2) Active transport of the entire iron chelate into the cell and intracellular reductive removal of iron with degradation or expulsion of the free ligand. (3) Transport of siderophores only to the membrane surface, with or without reduction of iron, and donation of iron to the cell interior without penetration of the complex or ligand into the cell (taxi model). Indeed, all three modes of siderophore uptake have been shown to occur in certain microorganisms under certain experimental conditions (WINKELMANN, 1991a).
Siderophore iron transport in fungi has recently been reviewed with emphasis on the conditions and precautions of kinetic measurements (WINKELMANN, 1993). Generally, there are several approaches to unravel the Fig. 41a. Pulcherrimic acid b mycelianamide; c as- mechanisms of fungal iron transport: radioacpergillic acid. tive labeling, photoaffinity labeling, and specand 14C-labeledsiderotroscopy. Using 55Fephores four types of transport have been defined (VAN DER HELMand WINKELMANN, 1994): (1) a shuttle mechanism (e.g., ferrichromes in Neurospora); (2) a taxi cab mechTo determine whether or not an iron-bind- anism (e.g., Fe-rhodotorulate in Rhodotoruing compound can be regarded as a sidero- la); (3) a hydrolytic mechanism (e.g., fusarinphore, its physiological function and genetic ines in Mycelia sterilia); and (4) a reductive regulation must be studied. An increasing mechanism (siderophores and ferric ions in number of microbial iron transport systems Saccharomyces and other non-siderophore are currently being identified and character- containing yeasts). Using a photoaffinity label ized at the molecular level. Siderophore (p-azidobenzoyl coprogen) 50% irreversible transport in enterobacteria and pseudomon- inhibition of coprogen and ferrichrome upads has been reviewed by BRAWN and take was measured in Neurospora crassa et HANTKE(1991) and MARWGGand WEIS- (BAILEY al., 1986) - an argument in favor of a common transport system for these sideBEEK (1991). Fungal iron transport, although mainly characterized in terms of structure- rophores. These results correspond to earlier function relationship (VAN DER HELMand transport kinetics showing a common transport system but different siderophore recepWINKELMANN, 1994) is coming to the same et molecular level of understanding that has tors (HUSCHKA al., 1985). Although a varibeen achieved in bacterial systems (MEI and ety of siderophores may be recognized by LEONG, 1994; LESWISSE LABBE, 1994). fungal siderophore receptors, recognition is and Siderophore-mediated iron transport has highly stereoselective as determined by combeen studied in bacteria and fungi using ra- parative studies with ferrichrome and enandio-labeled siderophores in wild type strains tio-ferrichrome (WINKELMANN, 1979; WINand mutants. In principle, three mechanisms KELMANN and BRAUN, 1981). These results of uptake into microorganisms are conceiv- confirm that fungal siderophore transport is not a diffusion-controlled process but reable:
6 Transport Mechanisms
234
quires specific interactions with components of the transport system, although proteins of fungal siderophore transport systems have not been identified so far. There are several reports on the application of EPR and Mossbauer spectroscopic measurements to fungal siderophore iron uptake (MATZANKE WINKELMANN, and 1981; ECKERet al., 1982; MATZANKE al., 1988) et supporting the view that iron undergoes subsequent reductive removal and that under certain conditions siderophores can also serve as iron storage compounds (see also the review of MATZANKE, 1994). Interestingly, siderophores are also found in conidiospores and seem to have an important function during sporulation and germination (HOROWITZ et al., 1976; MATZANKE al., 1987; WINKELet MANN, 1991b). Up to now only few reports on the transport of carboxylate type siderophores, like rhizoferrin, in fungi are available (DRECHSEL, al., 1991; THIEKEN et and WINKELMANN, 1992). The mechanism of uptake in fungi of the Mucorales (Zygomycetes) is still under investigation. The mechanism of hydroxamate iron transport in enterobacteria was examined in early studies with E. coli K-12 and Salmonella typhimurium LT-2 mutants defective in enterobactin synthesis using 55Fe-and 3H-labeled ferrichrome and their chromic analogs (LEONG NEILANDS, and 1976). These investigations already showed that ferrichrome is taken up as an intact chelate in both E. coli and Salmonella. Later studies revealed that the ferrichrome ligand is modified by acylation of the hydroxamate N-OH groups after iron delivery (HARTMANN BRAUN, and 1980) thereby reducing the stability of ferrichrome with iron within the cell. Our current knowledge of ferrichrome uptake in E. coli has increased considerably by a detailed genetic analysis of the various membrane components involved (BRAUN and HANTKE, 1991). In gram-negative bacteria, the transport of iron bound to hydroxamate siderophores generally depends on the expression of outer membrane receptor proteins. Ferrichrome, e.g., is taken up via the FhuA receptor protein (78kDa) localized in the outer membrane of gram-negative bacteria. FhuA also binds the phages T1, T5, @80, to colicin M,
and the antibiotic albomycin (HANTKEand BRAUN, 1975). Although ferrichrome is first bound to the FhuA receptor, it has to be processed further by additional gene products (FhuB, C, D) to permit completion of siderophore transport into the cell (KOSTER, 1991). This mechanism of nutrient translocation is known as a periplasmic binding-dependent transport which earlier was described for other transport processes (AMES,1986). Different ferric hydroxamate uptake (Fhu) outer membrane receptors of E. coli have been shown to mediate the transport of various fungal siderophores, e.g., ferrichromes, coprogen, and rhodotorulic acid (HANTKE, 1983). Surprisingly, the fungal siderophores coprogen and Fe rhodotorulate enter the cells of E. coli by the same outer membrane receptor protein (FhuE) as the bacterial ferrioxamines, whereas members of enterobacterial Erwina herbicola (Pantoea herbicola) use a separate outer membrane receptor FoxA to transport ferrioxamines (BERNERand WINKELMANN, 1990). While FhuD is located in the periplasmic space, FhuB and FhuC are integrated in the cytoplasmic membrane. All fhu genes are organized in an operon; they have been sequenced, and the deduced molecular weights of the polypeptides are in accordance with those found by SDS-PAGE. Although the hydroxamate siderophores enter the cells by different outer membrane receptors, e.g., ferrichromes (FhuA), coprogens (FhuE) and Fe rhodotorulate (FhuE), aerobactin (Iut), and ferrioxamines (FoxA), their transport across the cytoplasmic membrane is accomplished by the same transport proteins (FhuD, B, C). In addition, a TonB protein is essential for an energized state of the outer membrane receptors as shown by tonB mutations which prevent correct interaction of the TonB protein with outer membrane receptor proteins (BRAUN al., 1991). Thus TonB is regarded et as a coupling device between the outer and inner membrane. The interaction of TonB with the receptor proteins requires the presence of a common binding domain which is characterized on the genetic level by a consensus sequence, the so-called TonB box, present in all TonB-dependent receptor proteins (BRAUN and HANTKE, 1991).
7 Conclusion and Perspectives
235
ERNST al. (1978) were the first to ob- ments of receptor and binding proteins inet serve a constitutive production of sidero- volved. There is a structure-function relationphores in a S. typhimurium mutant and ship between iron-containing siderophores coined the term fur (ferric uptake regulation) and the surface of transport proteins. Conforfor the gene responsible for the regulation of mation of siderophores seems to be the most siderophore biosynthesis and expression of important factor. In gram-negative bacteria the cognate receptor proteins. The fur gene two membranes and a periplasmic space have later was also found in E. coli and was subse- to be penetrated before iron can be utilized quently cloned and sequenced (HANTKE, by the cell metabolism. Although molecular 1984). The corresponding protein was charac- genetics has become the most powerful tool terized by WEE et al. (1988). It turned out in analyzing the transport of siderophores, that all flu genes and other siderophore re- many open questions remain, e.g.: Which part ceptor genes (enterobactin, aerobactin, ferri- of the siderophore molecule is recognized and oxamines) are regulated by iron via the f i r what kind of interaction is to be expected? gene product. This also applies to the biosyn- Should the outer membrane receptor protein be a channel protein through which the sidethetic genes of siderophores. Uptake of ferrichrome in fungi is stereo- rophore can pass with specific interaction, as specific, as proven with synthetic enantio-fer- has been inferred from recent results (KILLO richrome which has the opposite chirality MANN et al., 1993; RUTZet al., 1993)? H W is 1981). However, the outer membrane receptor gated or ener(WINKELMANN BRAUN, and uptake of enantio-ferrichrome (delta configu- gized by the TonB protein? What kind of inration) in E. coli was still 50% as compared to teractions prevail at the periplasmic binding uptake of the natural ferrichrome (lambda proteins (FhuD) and, finally, how does the configuration), indicating either an incom- translocation through the inner cytoplasmic plete recognition or additional routes for iron membrane proceed? In fungi similar structure-function relationuptake. Enantioselectivity seems to be more pronouced with the cloned Yersiniu ferri- ships have been documented and allocated to chrome receptor protein (FcuA) accepting putative cytoplasmic membrane proteins. ferrichrome but excluding enantio-ferri- However, since siderophore transport muchrome completely (BAUMLER HANTKE, tants are not available, membrane proteins and 1992). A recent analysis of the siderophore involved in siderophore binding and transspecificity of different Fhu receptors in E. coli port have never been identified. However, has revealed that certain ferrichromes like the principal mechanisms of interaction with ferrichrysin and ferrirubin may not only enter specific membrane proteins also seem to be via FhuA but may also use the FhuE receptor valid in fungal siderophore transport. The basic question as to whether the entire chelate 1992). (KILLMANN BRAUN, and molecule or only the iron atom enters the cell has been solved with double-labeled siderophores in various fungi. According to our present knowledge only some siderophores, 7 mainly ferrichromes, can enter the fungal cell and remain intact. Several other siderophores need to interact with membrane receptors to allow exchange of iron. In every case, howevThe knowledge of siderophores and their er, specific interaction is observed. Thus the cognate iron transport systems in microorgan- essential features of structure-function relaisms is crucial for an understanding of basic tionship among bacterial and fungal sideroevents, such as growth, metabolic activity, phore systems seem to be identical. Future applications of siderophores are host invasion, and virulence. In all cases iron nutrition is a prerequisite. The diversity of si- manifold. The lag phase of slowly growing miderophores is remarkable and can only be croorganisms, e.g., might be shortened to a discussed in relation to the structural require- certain extent by simply adding very small
Conclusion and Perspectives
236
amounts of genus-specific siderophores. Moreover, addition of siderophores to biological fluids or food may overcome iron restriction due to the iron-withholding properties of et iron-binding proteins (SOUKKA al., 1992) and may thus allow outgrowth and detection of otherwise dormant microorganisms. The use of genetically engineered organisms that possess improved iron transport systems recognizing specific siderophores may be an alternative approach to enhance growth and metabolite production. Also the use of nonutilizable or antibiotic-linked siderophores (NIKAIDO and ROSENBERG, 1990) designed for reducing the bioavailablity of iron by contaminating bacteria might be an interesting future biotechnological application. The best examples of siderophore application can be found in nature where biocontrol and host defense mechanisms based on nutritional iron deficiency are active in a finely tuned manner without negative effects on the surrounding ecosystems. The biotechnological applications of siderophores to medicine, agriculture, the environment, and the food industry are numerous. For example, the affinity of siderophores for toxic metals, such as chromium and plutonium, suggests that siderophores may be useful agents for remediation of polluted environments. Likewise, the development of orally effective siderophores suitable for iron chelation therapy through directed fermentation is a virtually unexplored area of investigation. A possible role of bacterial siderophores in reducing inflammation (COFFMANN et al., 1990) as well as protecting cells from free radical damage has also been suggested. However, addressing of possible applications of siderophores in more detail is beyond the scope of this chapter. The authors are convinced that the future will bring exciting advances in this area. The present review aims at understanding the wealth of the structural diversity of siderophores, their production, and mode of action and will hopefully provide a stimulus for future research in microbial physiology, ecology, and fields of applied biotechnology.
8 References
ABDALLAH, M. A. (1991), Pyoverdins and pseudobactins, in: Handbook of Microbial Iron Chelates (WINKELMANN, Ed.), pp. 139-153, Boca RaG., ton, FL: CRC Press. ACTIS, A., FISH, CROSA, H., KELLERMAN, L. W., J. K., ELLENBERGER, HAUSER, M., SANS. R., F. J. DERS-LOHR, (1986), Characterization of anguibactin, a novel siderophore from Vibrio anguillarum 775 (pJMl), J. Bacteriol. 167, 57-65. ACTIS, A., TOMALSKY, E., FARREL, H., L. M. D. CROSA,J. H. (1988), Genetic and molecular characterization of essential components of the Vibrio anguillarum plasmid-mediatediron transport system, J. Biol. Chem. 263,2853-2860. AMES, . F. L. (1986), Bacterial periplasmic transG port systems: structure, mechanism, and evolutio, Ann. Rev. Biochem. 55, 397-425. ANKE, (1977), Metabolic products of microorH. ganisms: desferritriacetylfusigen, an antibiotic from Aspergillus deflectus, J. Antibiot. 30, 125128. ARCENEAUX,E. L., DAVIS, B., DOWNER, J. W. D. N., HAYDON, H., BYERS, R. (1973), Fate A. B. of labeled hydroxamates during iron transport from hydroxamate-iron chelates, J. Bacteriol. 115,919-927. ARONSON, B., BOYER, . L. (1994), Growth D. G and siderophore formation in six iron-limited strains of Frankia, Soil Biol. Biochem. 265, 561567. ATKIN, L., NEILANDS, B. (1968), RhodotorulC. J. ic acid, a diketopiperazine dihydroxamic acid with growth factor activity: isolation and characterization, Biochemistry 7, 3734-3739. ATKIN, L., NEILANDS, B., PHAFF, J. (1970), C. J. H. Rhodotorulic acid from species of Leucosporidium, Rhodosporidium, Rhodotorula, Sporidiobolus, and Sporobolornyces and a new alanine containing ferrichrome from Cryptococcus melibiosum, J. Bacteriol. 103, 722-733. BAGG, A., NEILANDS, B. (1987), Molecular J. mechanism of regulation of siderophore-mediated iron assimilation, Microbiol. Rev. 51,509518. BAIIEY, T., KIME-HUNT, M., CARRANO, C. E. C. J., HUSCHKA, WINKELMANN, . (1986), A H., G photoaffinity label for the siderophore-mediated iron transport system in Neurospora crassa, Biochim. Biophys. Acta 883, 299-305. BARGHOUTHI, YOUNG, ARCENEAUX,E. S., R., J. L., BYERS, R. (1989), Physiological control of B. arnonabactin biosynthesis in Aeromonas hydrophila, BioMetals 2, 155-160. BARGHOUTHI, PAYNE, M., ARCENEAUX, S., S. J. E. L., BYERS, R. (1991), Cloning, mutageneB.
8 References
sis, and nucleotide sequence of a siderophore biosynthetic gene (amoA) from Aeromonas hydrophila, J. Bacteriol. 173,5121-5128. BARTON,L. L., HEMMING, C. (Eds.) (1993), B. Iron Chelation in Plants and Soil Microorganisms. San Diego, CA: Academic Press. BAUMLER, J., HANTKE, (1992), FerrioxamA. K. ine uptake in Yersinia enterocolitica: characterization of the receptor protein FoxA, Mol. Microbiol. 6, 1309-1321. BENZ,G., SCHRODER, KURZ,J., WUNSCHE, T., C. K. W., STEFFENS, PFITZNER, SCHMIDT, G., J., D. (1982), Konstitution der Desferriform der Albomycine a,, &, E, Angew. Chemie 94, 552-553. BERGERON, J., MCMANIS, S. (1991), SyntheR. J. sis of catecholamide and hydroxamate siderophores, in: Handbook of Microbial Iron Chelates (WINKELMANN, Ed.), Boca Raton, F1: CRC G., Press. BERGERON, J., DIONIS, B., ELLIOT,G. T., R. J. KLINE,S. J. (1985), Mechanism and stereospecificity of the parabactin-mediated iron transport system in Paracoccus denitrificans, J. Biol. Chem. 260,7936-7944. BERNER, WINKELMANN, (1990), FerrioxamI., G. ine transport mutants and the identification of the ferrioxamine receptor protein (FoxA) in Erwinia herbicola (Enterobacter agglomerans), BioMetals 2, 197-202. BERNER,I., KONETSCHNY-RAPP, JUNG, G., S., WINKELMANN, (1988), Characterization of G. ferrioxamine E as the principal siderophore of Envinia herbicola (Enterobacter agglomerans), BioMetals 1, 51-56. I., M., J., G., BERNER, GREINER, METZGER, JUNG, WINKELMANN,(1991), Identification of enteG. robactin and linear dihydroxybenzoylserine compounds by HPLC and ion spray mass spectrometry (LC/MS and MSIMS), BioMetals 4, 113-118. BERNHARDINI, LINGET-MORICE, HOH, F., J., C., COLLINSON, K., KYSLIK,P., PAGE, W. J., S. DELL,A., ABDALLAH, A. (in press), BacteriM. al siderophores: Structure elucidation and H, 13C, I5N two-dimensional NMR assignments of azotoverdin and related siderophores synthesized by Azomonas macrocytogenes ATCC 12334, BioMetals 9, 107-120. BICKEL, H., BOSSHARDT,R., GAUMANN, E., REUSSER, VISCHER, VOSER,w., WE=P., E., STEIN, A., ZAHNER, (1960), StoffwechselproH. dukte von Actinomyceten. Uber die Isolierung und Charakterisierung der Ferrioxamine A-F, neuer Wuchsstoffe der Sideramin-Gruppe, Helv. Chim. Acta 43,2118-2128. BINDEREIF, BRAUN,V., HANTKE,K. (1982), A., The cloacin receptor of ColV-bearing Esch-
237
erichia coli is part of the Fe-aerobactin transport system, J. Bacteriol. 150, 1472-1475. BITTER, W., MARUGG, D., DE WEGER,L. A., J. TOMMASSEN, WEISBEEK, J. (1991), The J., P. ferric-pseudobactin receptor PupA of Pseudomonas putida WCS 358: homology to TonB-dependent Escherichia coli receptors and specificity of the protein, Mol. Microbiol. 5, 647-655. BOELAERT, R., FENVES, Z., COBURN, W. J. A. J. (1991), Deferoxamine therapy and mucormycosis in dialysis patients: report of an international registry, Am. J. Kidney Dis. 18, 660-667. BOELAERT, R., DELOCHT, VAN CUTSEM, J. M., J., KERBELS, CANTINIEAUX, VERDONCK, V., B., A., VAN LANDUYT, SCHNEIDER, Y.-J. (1993), Mucormycosis during desferoxamine therapy is a siderophore-mediated infection. In vitro and in vivo animal studies, J. Clin. Invest. 91, 19791986. BOELAERT, R., VANCUTSEM, DELOCHT, J. J., M., SCHNEIDER, Y.-J., CRICHTON, R. (1994), DeR. feroxamine augments growth and pathogenicity of Rhizopus, while hydroxypyridinone chelators have no effect, Kidney Int. 45, 667-671. BOYER,G. L., ARONSON, B. (1994), Iron upD. take and siderophore formation in the actinorhizal symbiont Frankia, in: Biochemistry of the Metal Micronutrients in the Rhizosphere (MANTHEY,J. A., CROWLEY, E., LUSTER, G., D. D. Eds.), pp. 41-54. Boca Raton, FL: CRC Press. BRAUN, (1981), Escherichia coli cells, containing V. the plasmid COW produce the iron ionophore aerobactin, FEMS Microbiol. Lett. 11, 225-228. BRAUN, HANTKE, (1991), Genetics of bacteV., K. rial iron transport, in: Handbook of Microbial G., Iron Chelates (WINKELMANN, Ed.), pp. 107138. Boca Raton, FL: CRC Press. BRAUN,V., GUNTER, C., HANTKE,K. (1991), Transport of iron across the outer membrane, BioMetals 4, 14-22. BRICKMAN, J., HANSEL, T. J.-G., MILLER, J., M. S. ARMSTRONG, K. (1996), Purification spectroscopic analysis, and biological activity of the macrocyclic dihydroxamate siderophore alcaligin produced by Bordetella pertussis and Bordetella bronchiseptica, BioMetals 9, 191-203. BRISKOT, G., TARAZ, K., BUDZIKIEWICZ, H. (1989), Pyoverdin-type siderophores from Pseudomonas aeruginosa, Liebigs Ann. Chem. 1989, 375-384. BUDDE,A. D., LEONG,S. A. (1989), Characterization of siderophores from Ustilago maydis, Mycopathologia 108, 125-133. BUDZIKIEWICZ, (1993), Secondary metabolites H. from fluorescent pseudomonads, FEMS Microbiol. Rev. 104, 209-228.
238
BUDZIKIEWICZ, SCHAFFNER, M., TARAZ, H., E. K. (1992), A novel azotobactin from Azotobacter vinelandii, Nat. Prod. Lett. 1, 9-14. BUTTERTON, R., STOEBNER, A., PAYNE, J. J. S. M., CALDERWOOD, B. (1992), Cloning, seS. quencing, and transcriptional regulation of vizA, the gene encoding the ferric vibriobactin receptor of Vibrio cholerae, J. Bacteriol. 174, 37293738. BUYER,J. S., DELORENZO, NEILANDS, B. V., J. (1991), Production of the siderophore aerobactin by a halophilic Pseudomonad, Appl. Environ. Microbiol. 57, 2246-2250. BYERS, R., POWELL, V., LANKFORD, E. B. M. C. (1967), Iron-chelating hydroxamic acid (schizokinen) active in initiation of cell division in Bacillus megaterium, Bacteriol. 93, 286-294. CARRANO, J., RAYMOND, N. (1978), CoordiC. K. nation chemistry of microbial iron transport compounds: Rhodotorulic acid and iron uptake in Rhodotorula pilimanae, J. Bacteriol. 136, 6974. CARRANO, THIEKEN, WINKELMANN, C., A., G. (1996), Specificity and mechanism of rhizoferrin mediated metal ion uptake, BioMetals 9, 185189. CASS,M. E., GARRETT, M., RAYMOND, N. T. K. (1989), The salicylate mode of bonding in protonated ferric enterobactin analogs, J. Am. Chem. SOC. 111, 1677-1682. CHAMBERS, E., MCINTYRE, D., MOUCK, C. D. M., SOKOL, (1996), Physical and structural characP. terization of yersiniophore, a siderophore produced by clinical isolates of Yersinia enterocolitica, BioMetals 9, 157-168. CHAMBERS, E., SOKOL, A. (1994), CompariC. P. son of siderophore production and utilization in pathogenic and environmental isolates of Yersinia enterocolitica, J. Clin. Microbiol. 32, 32-39. CHUNG, D. Y., MATZANKE, F. M., WINKELT. B. G., K. MANN, RAYMOND, N. (1986), The inhibitory effect of the partially resolved coordination isomers of chromic desferricoprogen on coprogen uptake in Neurospora crassa, J. Bacteriol. 165,283-287. COFFMAN, Cox, C. D., EDECKER, L., BRITIT., B. GAN, B. E. (1990), Possible role of bacterial siderophores in inflammation, J. Clin. Invest. 86, 1030-1 037. CORBIN, L., BULEN, A. (1969), The isolation J. W. and identification of 2,3-dihydroxybenzoic acid and 2-N, 6-N-di (2,3-dihydroxybenzoyI)-~-lysine formed by iron-deficient Azotobacter vinelandii, Biochemistry 8,757-762. CORBIN, L., KARLE,I. L., KARLE,J. (1970), J. Crystal structure of the chromophore from the fluorescent peptide produced by iron-deficient
Azotobacter vinelandii, J. Chem. SOC., Chem. Commun. D 186-187. CORNISH, S., PAGE,W. J. (1995), Production of A. the tricatecholate siderophore protochelin by Azotobacter vinelandii, BioMetals 8, 332-338. Cox, C. D., RINEHART, L., MOORE, M. L., K. J. COOK, C. (1981), Pyochelin: novel structure of an iron-chelating growth promotor for Pseudomonas aeruginosa, Proc. Nut. Acad. Sci. USA 78, 42564260. CROSA, H. (1989), Genetics and molecular bioloJ. gy of siderophore-mediated iron transport in bacteria, Microbiol. Rev. 53, 517-530. CRUMBLISS, L. (1991), Aqueous solution equiliA. brium and kinetic studies of iron siderophore and model siderophore complexes, in: Handbook of Microbial Iron Chelates (WINKELMANN, G., Ed.), pp. 177-233. Boca Raton, FL: CRC Press. DELORENZO, BINDEREIF, PAW, B. H., NEIV., A., LANDS, J. B. (1986), Aerobactin biosynthesis and transport genes of plasmid ColV-K30 in Escherichia coli K-12, J. Bacteriol. 165, 570-578. DEMANGE, BATEMAN, DELL,A., ABDALP., A., LAH,M. A. (1988). Structure of azotobactin D, a siderophore of Azotobacter vinelandii strain D (CCM 289), Biochemistry 27,2745-2752. DEMANGE, BATEMAN, MACLEOD,J. K., P., A., DELL,A., ABDALLAH, A. (1990a), Bacterial M. siderophores: Unusual 3,4,5,6-Tetrahydropyrimidine-based amino acids in pyoverdins from Pseudomonas fluorescens, Tetrahedron Lett. 31, 7611-7614. DEMANGE, BATEMAN, MERZ,C., DELL,A,, P., A., Y., M. PIBMONT, ABDALLAH, (1990b), Bacterial siderophores: Structures of pyoverdines Pt, siderophores of Pseudomonas tolaassii NCPPB 2192, and pyoverdins Pf, siderophores of Pseudomonas fluorescens CCM 2789. Identification of an unusual natural amino acid, Biochemistry 29, 11041-11 051. DEML,G. (1985), Studies in heterobasidiomycetes, Part 34. A survey on siderophore formation in low-iron cultured smuts from the floral parts of Polygonaceae, Syst. Appl. Microbiol. 6, 23-24. DEML,G., VOGES,K., JUNG,G., WINKELMANN, G. (1984), Tetraglycylferrichrome- the first heptapeptide ferrichrome, FEBS Lett. 173, 53-57. DEVOS, P., VANDEWOESTYNE, M., VANCANNEYT, M., VERSTRAETE, w., KERSTERS, K. (1993), Identification of proferrorosamine producing Pseudomonas sp. strain GH (LMG 11358) as Erwinia rhapontici, System. Appl. Microbiol. 16, 252-255. DIEKMANN, (1967), Fusigen - ein neues SiderH. amin aus Pilzen, Arch. Microbiol. 58, 1-5.
8 References
239
DIEKMANN, (1968), Stoffwechselprodukte von ECKER,D. J., LANCASTER, R., EMERY,T. H. J. Mikroorganismen. 68. Die Isolierung von trans(1982), Siderophore iron transport followed by 5-Hydroxy-3-methylpenten-(2)-saure, Arch. Mielectron paramagnetic resonance spectroscopy, crobiol. 62, 322-327. J. Biol. Chem. 257,8623-8626. DIEKMANN, (1970), Stoffwechselprodukte von EHLERT, G., TARAZ, K., BUDZIKIEWICZ, H. H. Mikroorganismen, 81. Mitteilung. Vorkommen (1994), Serratiochelin, a new catechoiate sideround Strukturen von Coprogen B und Dimerumphore from Serratia marcescens, 2. Naturforsch. saure, Arch. Microbiol. 73,65-76. 4 9 ~ 11-17. , DIEKMANN, KREZDORN, (1975), Stoffwech- ELKINS, F., EARHARDT, F. (1989), NucleoH., E. M. C. selprodukte von Mikroorganismen. 150. Ferritide sequence and regulation of the Escherichia crocin, Triacetylfusigen und andere Sideramine coli gene for ferrienterobactin transport protein aus Pilzen der Gattung Aspergillus, Gruppe fuFepB, J. Bacteriol. 171, 5443-5451. migatus, Arch. Microbiol. 106, 191-194. EMERY, (1971), Role of ferrichrome as a ferric T. ionophore in Ustilago sphaerogena, BiochemisDIEKMANN, ZAHNER, (1967), Konstitution H., H. try 10, 1483-1488. von Fusigen und dessen Abbau zu delta-2-Anhydromevalonsaurelacton, Eur. J. Biochem. 3,213EMERY, (1976), Fungal ornithine esterases: relaT. 218. tionship to transport, Biochemistry 15, 27232728. DIONIS, B., JENNY, PETER, H. H (1991), J. H., Therapeutically useful iron chelators, in: Hand- ENARD, DIOLEZ,A., EXPERT, (1988), SysC., D. book of Microbial Iron Chelates (WINKELMANN, temic virulence of Erwinia chrysanthemi on G., Ed.), pp. 339-356. Boca Raton, F L CRC Saintpaulia chrysanthemi 3937 requires a funcPress. tional iron assimilation system, J. Bacteriol. 170, 2419-2426. DRECHSEL, METZGER, FREUND, JUNG, H., J., S., L. G., BOELAERT, WINKELMANN, (1991), ENG-WILMOT, D., VAN DER HELM,D. (1980), J., G. Rhizoferrin - a novel siderophore from the funMolecular and crystal structure of the linear trigus Rhizopus microsporus var. rhizopodiformis, catechol siderophore, agrobactin, J. Am. Chem. SOC. 102,7719-7725. BioMetals 4, 238-243. L. J.P., VAN DER DRECHSEL,H., JUNG, G., WINKELMANN, G. ENG-WILMOT, D., ADJIMANI, HELM, D. (1992), Siderophore mediated iron(1992), Stereochemical characterization of rhizo(111) transport in the mycelia of the cultivated ferrin and identification of its dehydration prodfungus, Agaricus bisporus, J. Inorg. Biochem. 48, ucts, BioMetals 5, 141-148. 183-1 95. DRECHSEL,H., FREUND,S., NICHOLSON, G., L. HAAG, H., JUNG,O., ZAHNER,H., JUNG,G. ERNST,J. F., BENNET,R. L., ROTHFIELD, I. (1978), Constitutive expression of the iron-ente(1993a). Purification and chemical characterizarochelin and ferrichrome uptake systems in a tion of staphyloferrin B, a hydrophilic sideromutant strain of Salmonella typhimurium, J. phore from staphylococci, BioMetals 6, 185Bacteriol. 135, 928-934. 192. G. DRECHSEL, THIEKEN, REISSBRODT, H., A., R., FEISTNER, F. (1995), Suggestion for a new, semirational nomenclature for the free chelators of JUNG,G., WINKELMANN, (1993b), a-Keto G. ferrioxamines, BioMetals 8, 193-196. acids are novel siderophores in the genera Proteus, Providencia, and Morganella and are pro- FEISTNER, KORTH, KO, H., PULVERER, G., H., G., duced by amino acid deaminases, J. Bacteriol. BUDZIKIEWICZ, (1983), Ferrorosamine A H. 175,2727-2733. from Erwinia rhapontici, Curr. Microbiol. 8, 239-243. DRECHSEL, TSCHIERSKE, THIEKEN, H., M., A., G. A. ZAHNER, H., JUNG, G., WINKELMANN, FEISTNER, J., STAHL,D. C., GABRIK, H. G. (1993), Proferrioxamine siderophores of Erwinia (1995), The carboxylate-type siderophore rhizoamylovora. A capillary liquid chromatographic/ ferrin and its analogs produced by directed fermentation, J. Ind. Microbiol. 14, 105-112. electrospray tandem mass spectrometric study, Org. Mass Spectrom. 28, 163-175. DRECHSEL, STEPHAN, LOTZ,R., HAAG, H., H., H., J., H. J., ZAHNER,H., HANTKE, JUNG,G. (1995), FRANCIS, MACTURK, M., MADINAVEITIA, K., SNOW, A. (1953), Mycobactin, a growth facG. Structure elucidation of Yersiniabactin, a siderotor for Mycobacterium johnei. I. Isolation from phore from highly virulent Yersinia strains, LieMycobacterium phlei, Biochem. J. 55, 596-607. bigs Ann. Chem., 1727-1733. C. M. DUTCHER, D. (1958), Aspergillic acid: An anti- FREDERICK, B., BENTLEY, D., SHIVE,W. J. (1981), The structure of triomicin, a new siderobiotic substance produced by Aspergillus flavus. phore, Biochemistry 20, 2436-2438. 111. The structure of hydroxyaspergillic acid, J. Biol, Chem. 232,785. FUKASAWA, GOTO,M., SASAKI, HIRATA, K., K.,
240
albomycin by Escherichia coli K-12, Eur. J. Biochem. 99,517-524. HAYDON, H., DAVIS, B., ARCENEAUX, A. W. J. E. L., BYERS, R. (1973), Hydroxamate recogniB. tion during iron transport from hydroxamateiron chelates, J. Bacteriol. 115, 912-918. HESEMANN,(1987), Chromosomal-encoded sideJ. rophores are required for mouse virulence of enteropathogenic Yersinia species, FEMS Microbiol. Lett. 18, 229-233. HOHNADEL, MEYER, D., J.-M. (1988), Specificity of pyoverdin-mediated iron uptake among fluorescent Pseudomonas strains, J. Bacteriol. 170, 48654873. HOROWITZ, H., CHARLANG, HORN,G., N. G., N. WILLIAMS, P. (1976), Isolation and identification of the conidial germination factor of Neurospora crassa, J. Bacteriol. 127, 135-140. HOSSAIN, B., ENG-WILMOT, L., LOGHRY, M. D. R. A., VAN DER HELM, D. (1980), Circular dichroism, crystal structure, and absolute configuration of the siderophore ferric N, N, N-triacetylfusarinine, J. Am. Chem. SOC. 102, 57665773. HOSSAIN, B., JALAL, A. F., BENSON, A., M. M. B. BARNES, L., VAN DER HELM,D. (1987), c. Structure and conformation of two coprogentype siderophores: Neocoprogen I and Neocoprogen 11, J. Am. Chem. SOC. 109,49484954. HUSCHKA, MULLER,G., WINKELMANN, H., G. (1983), The membrane potential is the driving force for siderophore iron transport in fungi, FEMS Microbiol. Lett. 20, 125-129. HUSCHKA, NAEGELI, H., LEUENBERGER-RYF, H., H., KELLER-SCHIERLEIN, WINKELMANN, W., G. (1985), Evidence for a commQn siderophore transport system but different siderophore receptors in Neurospora crassa, J. Bacteriol. 162, 715-721. HUSCKA, JALAL, A. F., VAN DER HELM, H., M. D., WINKELMANN,(1986), Molecular recognition G. of siderophores in fungi: Role of iron-surrounding N-acyl residues and peptide backbone during membrane transport in Neurospora crassa, J. Bacteriol. 167, 1020-1024. ISHIMARU, A., LOPER, E. (1992), High-affiniC. J. ty iron uptake systems present in Erwinia carotovora subsp. carotovora include the hydroxamate siderophore aerobactin, J. Bacteriol. 174, 2993-3003. JALAL, A. F., VAN DER HELM, (1989), PuriM. D. fication and crystallization of ferric enterobactin receptor protein, FepA, from outer membranes of Escherichia coli UT5600IpBB2, FEBS Lett. 243,366-370. JALAL, A. F., VAN DER HELM, (1991), IsolaM. D. tion and spectroscopic identification of fungal
Y., SATO,S. (1972), Structure of the yellowgreen fluorescent peptide produced by iron-deficient Azotobacter vinelandii strain 0, Tetrahedron 28,5359-5365. GIBSON, MAGRATH, I. (1969), The isolation F., D. and characterization of a hydroxamic acid (aerobactin) formed by Aerobacter aerogenes 62-1, Biochim. Biophys. Acta 192, 175-184. GILL,M., STEGLICH, (1987), Pigments of fungi W. (Macromycetes), in: Progr. Chem. of Organic Natural Products, Vol. 51 (HERZ,W., GRISEBACH, KIRBY, W., TAMM, Eds.). HeiH., G. C., delberg: Springer-Verlag. GOBIN, MOORE, H., REEVE, R. WONG, J., C. J. D. M. K., GIBSON, W., HORWITZ, A. (1995), B. Iron acquisition by Mycobacterium tuberculosis: isolation and characterization of a family of ironbinding exochelins, Proc. Natl. Acad. Sci. USA 92,5189-5193. GRIFFITH, L., SIGEL, P., PAYNE, M., NEIG. S. S. LANDS, B. (1984), Vibriobactin, a siderophore J. from Vibrio cholerae, J. Biol. Chem. 259, 383385. GROSS,R., ENGELBRECHT, BRAUN, (1985), V. Identification of the genes and their polypeptide products responsible for aerobactin synthesis by ColV plasmids, Mol. Gen. Genet. 201, 204-212. HAAG, HANTKE, DRECHSEL, STROJILH., K., H., KOVIC, I., JUNG, G., ZAHNER, (1993), PurifiH. cation of yersiniabactin: a siderophore and possible virulence factor of Yersiniaenterocolitica,J. Gen. Microbiol. 139,2159-2165. HALL,R. M., RATLEDGE, (1982), Mycobactins C. as chemotaxonomic characters for some rapidly growing mycobacteria, J. Gen. Microbiol. 130, 1883-1892. HANTKE, (1983), Identification of an iron upK. take system specific for coprogen and rhodotorulic acid in Escherichia coli K12, Mol. Gen. Genet. 191, 301-306. HANTKE, (1984), Cloning of the repressor proK. tein gene of iron regulated systems in E. coli K12, Mol. Gen. Genet. 197, 337-341. HANTKE, BRAUN, (1975), A function comK., V. mon to iron-enterochelin transport and action of colicins B, I, V in Escherichia coli, FEBS Lett. 59,277-281. HARRIS, R., CARRANO, J., RAYMOND, N. W. C. K. (1979), Coordination chemistry of microbial iron transport compounds. 16. Isolation, characterization, and formation constants of ferric aerobactin, J. Am. Chem. SOC. 101,2722-2727. HARTMANN, BRAUN, (1980), Iron transport A., V. in E. coli: uptake and modification of ferrichrome, J. Bacteriol. 143, 246-255. HARTMANN, FIEDLER,H.-P., BRAUN,V. A., (1979), Uptake and conversion of the antibiotic
8 References
241
siderophores, in: Handbook of Microbial Iron KELLER-SCHIERLEIN, HAGMANN, ZAHW., L., G., Chelates (WINKELMANN, Ed.), pp. 235-269. NER, H., HUHN, W. (1988), StoffwechselproBoca Raton, F L CRC Press. dukte von Mikroorganismen. 167. Maduraferrin, ein neuartiger Siderophor aus Actinomadura JALAL, A. F., MOCHARLA, BARNES, L., M. R., C. M. D. HOSSAIN, B., POWELL, R., ENG-WILMOT, madurae, Helv. Chim. Acta 71, 1528-1540. H., V. S. B. D. L., GRAYSON, L., BENSON, A., VAN KILLMANN, BRAUN, (1992), An aspartate deletion mutation defines a binding site of the DER HELM, D. (1984), Extracellular sideromultifunctional FhuA outer membrane receptor phores from Aspergillus ochraceus, J. Bacteriol. of Escherichia coli, J. Bacteriol. 174, 3479-3486. 158,683-688. H., V. JALAL, A. F., GALLES, L., VAN DER HELM, KILLMANN, BENZ,R., BRAUN, (1993), ConM. J. version of the FhuA transport protein into a difD. (1985), Structure of des(diserylg1ycyl)ferrifusion channel through the outer membrane of rhodin, DDF, a novel siderophore from AsperEscherichia coli, EMBO J. 12,3007-3016. gillus ochraceus, J. Org. Chem. 50,5642-5645. KLUYVER, J. J., VAN DER WALT,J. P., VAN A. JALAL, A. F., LOVE,S . K., VAN DER HELM, M. D. TRIET,A. J. (1953), Pulcherrimin, the pigment (1986,) Siderophore mediated iron(II1) uptake of Candida pulcherrima, Proc. Natl. Acad. Sci. in Gliocladium virens. 1. Properties of cis-fusarU.S.A. 39, 583. inine, trans-fusarinine, dimerum acid, and their KNOSP, TIGERSTROM, PAGE,W. J. (1984), O., M., ferric complexes, J. Inorg. Biochem. 28, 417Siderophore-mediated uptake of iron in Azoto430. bacter vinelandii, J. Bacteriol. 159, 341-347. JALAL, A. F., LOVE,S. K., VAN DER HELM, M. D. S., H., (1987), Siderophore mediated iron(II1) uptake KONETSCHNY-RAPP, HUSCHKA, WINKELG. MA, G., JUNG, (1988), High-performance in Gliocladium virens. 2. Role of ferric monoliquid chromatography of siderophores from and dihydroxamates as iron transport agents, J. fungi, BioMetals 1,9-17. Inorg. Biochem. 29,259-267. KONETSCHNY-RAPP, JUNG,G., MEIWES,J., S., JALAL, A. F., HOSSAIN, B., VAN DER HELM, M. M. ZAHNER, (1990), Staphyloferrin A a strucH. J. D., SANDERS-LOHR, ACTIS,L. A., CROSA, J., turally new siderophore from staphylococci, Eur. H. (1989), Structure of anguibactin, a unique J. Biochem. 191, 65-74. plasmid-related bacterial siderophore from the J., J., M., fish pathogen Vibrio anguillarum, J. Am. Chem. KOSTER, VAN DE VOSSENBERG, LEONG, WEISBEEK, J. (1993), Identification and charP. SOC. 111,292-296. acterization of the pupB gene encoding an induJEGOROV, MATHA, HUSAK, KRATOCHA., V., M., cible ferric-pseudobactin receptor of PseudoP., VIL, STUCHLIK, SEDMERA, HAVLICEK, monas putida WCS358, Mol. Microbiol. 8, 591B., J., V. (1993), Iron uptake system of some members 601. of the genus Tolypocladiurn: Crystal structure of KOSTER,W. (1991), Iron(II1) hydroxamate transthe ligand and its iron(II1) complex, J. Chem. port across the cytoplasmic membrane of SOC. Dalton Trans. 1993, 1287-1293. Escherichia coli, BioMetals 4, 23-32. KATAYAMA, NOZAKI, OKONOGI, HA- KUNZE, B., BEDORF, KOHL,W., HOFLE,G., N., Y., K., N., RADA, S, ONO, H. (1993), Ferrocins, new ironREICHENBACH,(1989), Myxochelin A, a new H. containing peptide antibiotics produced by baciron-chelating compound from Angiococcus disteria. Taxonomy, fermentation and biological acciformis (Myxobacteriales), J. Antibiot. 42, 14tivity, J. Antibiot. 46, 65-70. 17. KELLER-SCHIERLEIN, DIEKMANN, (1970), KUNZE, B., TROWITZSCH, HOFLE, G., REIW., H. W., Zur Konstitution des Coprogens, Helv. Chim. CHENBACH, H. (1992), Nannochelins A, B and Acta 53, 2035-2044. C, new iron-chelating compounds from NannoKELLER-SCHIERLEIN,PRELOG, ZAHNER, W., V., cystis excedens (Myxobacteria), J. Antibiot. 45, H. (1964), Siderochrome, in: Forrschritte der 147-1 50. Chemie organischer Naturstoffe (ZECHMEISTER, KUHN,S., BRAUN, KOSTER,W. (1996), Ferric V., L., Ed.), pp. 279-322. Heidelberg: Springer-Verrhizoferrin uptake into Morganella morganii: lag. characterization of genes involved in the uptake of a polycarboxylate siderophore, J. Bacteriol. KELLER-SCHIERLEIN,HUBER, KAWAGUW., P., 178,496-504. CHI, H. (1984), Chemistry of danomycin, an P. J. iron-containing antibiotic, in: Natural Products LAMMERS, J., SANDERS-LOHR,(1982), Active and Drug Development (KROGSGAARD-LAR- transport of ferric schizokinen in Anabaena sp., J. Bacteriol. 151, 28&294. SEN, P., BROGGER CHRISTENSEN, s., KOFOD, H., Eds.), pp. 213-227. Copenhagen: Munks- LANE, J., MARSHALL, S., UFTON, R. J., RATS. P. gaard. LEDGE, c., EWING, (1995), Novel extracelluM.
242
lar mycobactins, the carboxymycobactins from Mycobacterium avium, Tetrahedron Lett. 36, 41294132. LEE, B. H., MILLER,M. J. (1983), Natural ferric ionophores: Total synthesis of schizokinen, schizokinen A, and arthrobactin, J. Org. Chem. 48, 24-31. J., J. LEONG, NEILANDS, B. (1976), Mechanisms of siderophore iron transport in enteric bacteria, J. Bacteriol. 126, 823-830. LEONG, RAYMOND, N. (1974), Coordination J., K. isomers of biological iron transport compounds. 11. The optical isomers of chromic desferriferrichrome and desferriferrichrysin, J. Am. Chem. SOC.96,66284630. LESUISSE, LABBE, (1994), Reductive iron asE., P. similation in Saccharomyces cerevisiae, in: Metal G., Ions in Fungi (WINKELMANN, WINCE,D. R., Eds.), pp. 149-178. New York: Marcel Dekker. LINGET, STYLIANOU, G., DELL,A., WOLFF, C., D. M. R. E., PIEMONT,Y., ABDALLAH, (1992), Bacterial siderophores: the structure of a desferribactin produced by a Pseudomonas fluorescens ATCC 13525, Tetrahedron Lett. 33,3851-3854. LINKE, W. D., CRUEGER,A., DIEKMANN, H. (1972), Stoffwechselprodukte von Mikroorganismen. 106. Zur Konstitution des Terregens-Faktors, Arch. Microbiol. 85, 44-50. LITWIN, M., CALDERWOOD,B. (1994), AnalyC. S. sis of the complexity of gene regulation by Fur in Vibrio cholerae, J. Bacteriol. 176, 240-248. LIU, W., FISHER,S., WELLS,J. S., RICCA,C. S., PRINCIPE, A., TREJO,W. H., BONNER, P., P. D. GOUGOUTOS, Z., POEPLITZ, SYKES, B. J. B., R. (1981), Siderochelin, a new ferrous-ion chelating agent produced by Nocardia, J. Antibiot. 34, 791-799. LUNDRIGAN, D., KADNER,R. J. (1986), NuM. cleotide sequence of the gene for the ferrienterochelin receptor FepA in Escherichia coli, J. Biol. Chem. 261, 10797-10801. MACHAM, P., RATLEDGE, NOCTON,J. C. L. C., (1975), Extracellular iron acquisition by mycobacteria: Role of the exochelins and evidence against the participation of mycobactin, Infect. 1242-1251. Immun. U , MARUGG, D., WEISBEEK, J. (1991), Molecular J. P. genetics of siderophore biosynthesis in fluorescent pseudomonads, in: Handbook of Microbial G., Iron Chelates (WINKELMANN, Ed.), pp. 155175. Boca Raton, F L CRC Press. MARUGG, D., DE WEGER,L. A., NIELANDER, J. H. B., OORTHUIZEN, RECOURT,K., LUGM., TENBERG, B., VAN DER HOFSTAD, A. J. M., G. WEISBEEK, J. (1989), Cloning and characteriP. zation of a gene encoding an outer membrane protein required for siderophore-mediated up-
take of Fe3+ in Pseudomonas putida WCS358, J. Bacteriol. 171, 2819-2826. MASSAD, ARCENEAUX, E. L., BYERS,B. R. G., J. (1994), Diversity of siderophore genes encoding biosynthesis of 2,3-dihydroxybenzoic acid in Aeromonas spp., BioMetals 7, 227-236. B. MATZANKE, F. (1991), Structures, coordination chemistry and functions of microbial iron chelates, in: Handbook of Microbial Iron Chelates (WINKELMANN, Ed.) pp. 15-64. Boca Raton, G., F L CRC Press. MATZANKE, F. (1994), Iron storage in fungi, in: B. Metal Ions in Fungi (WINKELMANN, WINCE, G., D. R., Eds.), pp. 179-214. New York: Marcel Dekker. MATZANKE, F., WINKELMANN, (1981), SideB. G. rophore transport followed by Mossbauer spectroscopy, FEBS Lett. 130, 50-53. A. MATZANKE, F., BILL, E., TRAUTWEIN, X., B. WINKELMANN, (1987), Role of siderophores G. in iron storage in spores of Neurospora crassa and Aspergillus ochraceus, J. Bacteriol. 169, 5873-5876. MATZANKE, F., BILL,E., TRAUTWEIN, X., B. A. WINKELMANN, (1988), Ferricrocin functions G. as the main intracellular iron-storage compound in mycelia of Neurospora crassa, BioMetals 1, 18-25. G., K. MATZANKE, F., MULLER, RAYMOND, N. B. (1989), Siderophore mediated iron transport, in: Iron Carriers and Iron Proteins (LOEHR, T., Ed.), pp. 1-121. Weinheim: VCH. MATZANKE, F., BERNER, BILL,E., TRAUTB. I., WEIN,A. X., WINKELMANN, (1991), TransG. port and utilization of ferrioxamine-E-bound iron in E. herbicola (Pantoea herbicola), BioMetals 4,181-185. M. MAURER, J., MILLER, J. (1982), Microbial P. iron chelators: Total synthesis of aerobactin and its constituent amino acid N6-acetyl-N6-hydroxylysine, J. Am. Chem. SOC. 104,3096-3101. MAURER,B., MULLER,A., KELLER-SCHIERLEIN, W., ZAHNER, (1968), Ferribactin, ein SideroH. phor aus Pseudomonas fluorescens Migula, Arch. Microbiol. 60, 326-339. MEI, B.,LEONG,S. A. (1994), Molecular biology of iron transport, in: Metal Ions in Fungi (WINKELG., WINGE,D. R., Eds.), pp. 117-148. M" A, New York: Marcel Dekker. MEIWES, FIEDLER, HAAG,H., ZAHNER, J., H., H., KONETSCHNY-RAPP, JUNG,G. (1990), IsolaS, . tion and characterization of staphyloferrin A, a compound with siderophore activity from Staphylococcus hyicus DSM 20459, FEMS Microbiol. Lett. 67, 201-206. T. MENHARDT, THARIATH, VISWANATHA, N., A., (1991), Characterization of the pyoverdins of
8 References
Azotobacter vinelandii ATCC 12837 with regard to heterogeneity, BioMetals 4, 223-232. MEYER,J. M. (1992), Exogenous siderophore-mediated iron uptake in Pseudomonas aeruginosa: possible involvement of porin OprF in iron translocation, J. Gen. Microbiol. 138, 951-958. MEYER,J. M., ABDALLAH, A. (1980), The sideM. rochromes of non-fluorescent pseudomonads: production of nocardamine by Pseudomonas stutzeri, J. Gen. Microbiol. 118, 125-129. MEYER,J. M., HOHNADEL, HALLB,F. (1989), D., Cepabactin from Pseudomonas cepacia, a new type of siderophore, J. Gen. Microbiol. 135, 1479-1 487. MEYER, J.-M., TRAN VAN, V., STINTZI, A., G. BERGE,O., WINKELMANN, (1995), Ornibactin production and transport properties in strains of Burkholderia vietnamiensis and Burkholderia cepacia (formerly Pseudomonas cepacia), BioMetals 8, 309-317. MOORE,R. E., EMERY, (1976), N'-AcetylfusarT. inines: isolation, characterization, and properties, Biochemistry 15, 2719-2722. MOORE,C. H., FOSTER, L.-A., GERBIGJR., D. G., DYER,D. W., GIBSON,B. (1995), Identification of alcaligin as the siderophore produced by Bordetella pertussis and B. bronchiseptica, J. Bacteri01. 7, 1116-1118. MOR, H., KASHMAN, WINKELMANN, BAY., G., RASH, I. (1992), Characterization of siderophores produced by different species of the dermatophytic fungi Microsporum and Trichophyton, BioMetals 5, 213-216. MULLER,G., RAYMOND, N. (1984), Specificity K. and mechanism of ferrioxamine-mediated iron transport in Streptomyces pilosus, J. Bacteriol. 160,304-312. MULLER,G., MATZANKE, F., RAYMOND, N. B. K. (1984), Iron transport in Streptomyces pilosus mediated by ferrichrome siderophores, rhodotorulic acid, and enantio-rhodotorulic acid, J. Bacteriol. 160, 313-318. MULLIS,K. B., POLLACK, R., NEILANDS, B. J. J. (1971), Structure of schizokinen, an iron transport compound from Bacillus megaterium, Biochemistry 10, 4894-4898. NAEGELI, H., KELLER-SCHIERLEIN, (1978), W. Stoffwechselprodukte von Mikroorganismen. Eine neue Synthese des Ferrichroms; enantioFerrichrom, Helv. Chim. Acta 61, 2088-2095. NAEGELI,H., ZAHNER,H. (1980), Stoffwechselprodukte von Mikroorganismen. Ferrithiocin, Helv. Chim. Acta 63, 1400-1406. NEILANDS, B. (1952), A crystalline organo-iron J. pigment from a rust fungus (Ustilago sphaerogena), J. Am. Chem. SOC. 74,4846-4847. NEILANDS, B. (1993), Overview of bacterial iron J.
243
transport and siderophore systems in Rhizobia, in: Iron Chelation in Plants and Soil MicroorganB. isms (BARTON,L. L., HEMMING, C., Eds.), pp. 179-195. San Diego, CA: Academic Press. NEILANDS, B., NAKAMURA, (1991), DetecJ. K. tion, determination, isolation, characterization and regulation of microbial iron chelates, in: Handbook of Microbial Iron Chelates (WINKELG., MANN, Ed.), pp. 1-14. Boca Raton, FL: CRC Press. NEILANDS, B., ERICSON, J., RASTETTER, J. T. W. H. (1981), Stereospecificity of the ferric enterobactin receptor of Escherichia coli K-12, J. Biol. Chem. 256,3831-3832. NIKAIDO, ROSENBERG,E. Y. (1990). Cir and H., Fiu proteins in the outer membrane of Escherichia coli catalyze transport of monomeric catechols: Study with p-lactam antibiotics containing catechol and analogous groups, J. Bacteriol. 172, 1361-1367. NISHIO,T., TANAKA, HIRATAKE, N., JuN., KATSUBE, Y., ISHIDA, Y., ODA, J. (1988), Isolation and structure of the novel dihydroxamate siderophore alcalignin, J. Am. Chem. SOC. 110,87338734. NOMOTO, K., SUGIURA, TAKAGI, (1987), Y., S. Mugineic acids, studies on phytosiderophores, in: Iron Transport in Microbes, Plants and Animals (WINKELMANN, G., VAN DER HELM, D., NEILANDS, B., Eds.), pp. 401-425, Weinheim: J. VCH. S., OKUJO,N., SAITO,M., YAMAMOTO, YOSHIDA, T., MIYOSHI, SHINODA, (1994a), Structure S., S. of vulnibactin, a new polyamine-containing siderophore from Vibrio vulnificus, BioMetals 7, 109-116. OKUJO,N., SAKAKIBARA, YOSHIDA,T., YAY., MAMOTO, S. (1994b), Structure of acinetofferin, a new citrate-based dihydroxamate siderophore from Acinetobacter haemolyticus, BioMetals 7 , 170-176. ONG,S. A., PETERSON, NEILANDS, B. (1979), T., J. Agrobactin, a siderophore from Agrobacterium tumefaciens, J. Biol. Chem. 254, 1860-1865. OXFORD, E., RAISTRICK, (1948), Studies on A. H. the biochemistry of microorganisms. 76. Mycelianamide, a metabolic product of Penicillium griseofulvum. Part I. Preparation, properties and breakdown products, Biochem. J. 42,323. PAGE, W. J., HUYER,M. (1984), Derepression of the Azotobacter vinelandii siderophore system using iron-containing minerals to limit iron repletion, J. Bacteriol. 158, 496-502. PAGE,W. J., TIGERSTROM, (1988), AminocheM. lin, a catecholamine siderophore produced by Azotobacter vinelandii, J. Gen. Microbiol. 134, 453-460.
244
Pasteurella haemolytica, J. Basic Microbiol. 34, PAYNE, M. (1980), Synthesis and utilization of S. 61-63. siderophores by Shigella flexneri, J. Bacteriol. 143, 1420-1424. RUTZ, J. M., LIU, J., LYONS,J., GORANSON, J., PERSMARK, NEILANDS,J. B. (1992), Iron(II1) M., S. M. J. ARMSTRONG, K., MCINTOSH, A., FEIX, B., KLEBBA, P. E. (1993), Formation of a gated complexes of chrysobactin, a siderophore of Erchannel by a ligand-specific transport protein in winia chrysanthemi, BioMetals 5, 29-36. the bacterial outer membrane, Science 258,471PERSMARK, M., EXPERT, D., NEILANDS, B. J. 475. (1989), Isolation, characterization, and synthesis of chrysobactin, a compound with siderophore SAYER, M., EMERY, (1968), Structures of the J. T. activity from Erwinia chrysanthemi, J. Biol. naturally occurring hydroxamic acids, fusarinChem. 264,3187-3193. ines A and B, Biochemistry 7, 184190. PERSMARK, EXPERT,D., NEILANDS, B. SCHNEIDER, HANTKE, (1993), Iron-hydroxaM., J. R., K. (1992), Ferric iron uptake in Erwinia chrysanmate uptake systems in Bacillus subtilis: identifithemi mediated by chrysobactin and related cacation of a lipoprotein as part of a binding protechol-type compounds, J. Bacteriol. 174, 4783tein-dependent transport system, Mol. Micro4789. biol. 8, 111-121. PERSMARK, M., PITTMAN,P., BUYER,J. S., SCHWYN, NEILANDS, B. (1987), Universal B., J. SCHWYN, GILL, P. R., NEILANDS, B. B., J. chemical assay for the detection and determina(1993), Isolation and structure of rhizobactin tion of siderophores, Anal. Biochem. 160, 471021, a siderophore from the alfalfa symbiont 56. Rhizobium meliloti 1021, J. Am. Chem. SOC.115, SHANZER, LIBMAN, (1991), Biomimetic sideA., J. 3950-3956. rophores, in: Handbook of Microbial Iron ChePETERSON, NEILANDS, B. (1979), Revised T., J. lates (WINKELMANN, Ed.), pp. 309-338. G., structure of a catecholamide spermidine sideroBoca Raton, FL: CRC Press. phore from Paracoccus denitrificans, Tetrahe- SHANZER, LIBMAN, LAZAR,R., TOR, Y., A., J., dron Lett. 50,48054808. EMERY,T. (1988), Synthetic ferrichrome anaRABSCH, WINKELMANN,(1991), The specifW., G. logues with growth promotion activity for Aricity of bacterial siderophore receptors probed throbacter flavescens, Biochem. Biophys. Res. by bioassays, BioMetals 4, 244-250. Commun. 157,389-394. RATLEDGE, (1987), Iron metabolism in mycoC. SHARMAN, WILLIAMS, H., EWING, F., G., D. D. bacteria, in: Iron Transport in Microbes, Plants RATLEDGE, (1995a), Isolation, purification C. and Animals (WINKELMANN, G., VAN DER and structure of exochelin MS, the extracellular HELM, NEILANDS, B., Eds.), pp. 207-233, D., J. siderophore from Mycobacterium smegmatis, Weinheim: VCH. Biochem. J. 305, 187-196. RATLEDGE, EWING, (1996), The occurrence C., M. SHARMAN, WILLIAMS, H., EWING, F., G., D. D. of carboxymycobactin, the siderophore of paRATLEDGE, (1995b), Determination of the C. thogenic mycobacteria, as a second extracellular structure of exochelin MN the extracellular sidesiderophore in Mycobacterium smegmatis, Mirophore from Mycobacterium neoaurum, Chem. crobiology 142, 2207-2212. Biol. 2, 553-561. RATLEDGE, MARSHALL, J. (1972), Iron C., B. SIGEL,S. P., STOEBNER, A., PAYNE,S. M. J. transport in Mycobacterium smegmatis: the role (1985), Iron-vibriobactin transport system is not of mycobactin, Biochim. Biophys. Acta 279, 58required for virulence of Vibrio cholerae, Infect. 74. Immun. 47,360-362. RATLEDGE, PATEL, V. (1976), The isolation, C., P. SMITH,M. J., SHOOLERY, N., SCHWYN, J. B., properties and taxonomic relevance of lipid-solHOLDEN, NEILANDS,J. B. (1985), RhizobacI., uble iron-binding compounds (the nocobactins) tin, a structurally novel siderophore from Rhizofrom Nocardia, J. Gen. Microbiol. 93, 141-152. bium meliloti, J. Am. Chem. SOC. 107, 1739REID,R. T., LIVE,D. H., FAULKNER, J., BUTD. 1743. LER,A. (1993), A siderophore from a marine G. bacterium with an exceptional ferric ion affinity SNOW, A. (1965), Isolation and structure of mycobactin T, a growth factor from Mycobacterium constant, Nature 366,455-458. REISSBRODT, RABSCH, CHAPEAUROUGE, tuberculosis, Biochem. J. 97, 166-175. R., W., A., JUNG,G., WINKELMANN, (1990), Isola- SNOW,G. A. (1970), Mycobactins: iron chelating G. growth factors from Mycobacteria, Microbiol. tion and identification of ferrioxamine G and E Rev. 34, 99-125. in Hafnia alvei, BioMetals 3, 54-60. J., REISSBRODT, ERLER,W., WINKELMANN, SOUKKA, T., TENOVUO, LENANDER-LUMIKARI, R., G. M. (1992), Fungicidal effect of human lactoferrin (1994), Iron supply of Pasteurella multocida and
8 References
against Candida albicans, FEMS Microbiol. Lett. 90,223-228. STEPHAN, FREUND, BECK,W., JUNG,G., H., S., WINKELMANN,(1993a), Ornibactins - a new G. family of siderophores from Pseudomonas, BioMetals 6, 93-100. STEPHAN, FREUND, MEYER,J.-M., WINH., S., KELMANN, G., JUNG,G. (1993b), Structure elucidation of the gallium-ornibactin complex by 2D-NMR spectroscopy, Liebigs Ann. Chem. 1993,4348. STEPHENSON, C., RATLEDGE, (1979), Iron M. C. transport in Mycobacterium smegmatis: Uptake of iron from ferriexochelin, J. Gen. Microbiol. 110,193-202. STOEBNER, A., PAYNE, M. (1988), Iron-regulJ. S. ated hemolysin production and utilization of heme and hemoglobin by Vibrio cholerae, Infect. Immun. 56,2891-2895. STOEBNER, A., BUTTERTON, R., CALDERJ. J. S. S. WOOD, B., PAYNE, M. (1992), Identification of the vibriobactin receptor of Vibrio cholerae, J. Bacteriol. 174, 3270-3274. STOJILJKOVIC, HANTKE, (1992), Hemin upI., K. take system of Yersinia enterocolitica:similarities with other TonB-dependent systems in gram-negative bacteria, EMBO J. 11, 4359-4367. TAKAHASHI, NAKAMURA, KAMEYAMA, A., H., T., KURASAWA, NAGANAWA, OKAMI, S., H., Y., TAKEUCHI, UMEZAWA, (1987), BisucaberS, H. in, a new siderophore, sensitizing tumor cells to macrophage-mediated cytolysis. 11. Physicochemical properties and structure determination, J. Antibiot. 40,1671-1676. TARAZ,K., EHLERT,G., GEISEN, BUDZIKIEK., G. WICZ,H., KORTH, H., PULVERER, (1990), Protochelin, ein Catecholat-Siderophor aus einem Bakterium (DSM 5746), 2. Naturforsch. 45b,1327-1332. TEINTZE,M., HOSSAIN, B., BARNES, L., M. C. D. LEONG, VAN DER HELM, (1981), Structure J., of ferric pseudobactin, a siderophore from a plant-deleterious Pseudomonas, Biochemistry 20,6446-6457. TELFORD, R., LEARY, A., TUNSTAD, M. G., J. J. L. BYERS, R., RAYMOND, N. (1994), AmonaB. K. bactin: Characterization of a new series of siderophores from Aeromonas hydrophila, J. Am. Chem. SOC. 116,44994500. THIEKEN, WINKELMANN,(1992), RhizoferA., G. rin: A complexone type siderophore of the Mucorales and Entomophthorales, FEMS Microbiol. Lett. 94, 37-42. THIEKEN, WINKELMANN, (1993), A novel A., G. bioassay for the detection of siderophores containing keto-hydroxy bidentate ligands, FEMS Microbiol. Lett. 111, 281-286.
245
TSUBOTANI, KATAYAMA, FUNABASHI, S., N., Y., ONO, H., HARADA, (1993), Ferrocins, new S. iron-containing peptide antibiotics produced by bacteria. Isolation, characterization and structure elucidation, J. Antibiot. 46, 287-293. VAN DER HELM, D., POLING, M. (1976), The crystal structure of ferrioxamine E, J. Am. Chem. SOC. 98, 82-86. VAN DER HELM, D., BAKER, R., ENG-WILMOT, J. D. L., HOSSAIN, B., LOGHRY, A. (1980), M. R. Crystal structure of ferrichrome and comparison with the structure of ferrichrome A, J. Am. Chem. SOC. 102, 42244231. VAN DER HELM, D., WINKELMANN,(1994), HyG. droxamates and polycarboxylates as iron transport agents (siderophores) in fungi, in: Metal Ions in Fungi (WINKELMANN, WINGE, R., G., D. Eds.), pp. 39-98. New York: Marcel Dekker. VAN DER WALT, J. P., BOTHA,A., EICKER, A. (1990), Ferrichrome production by Lipomycetaceae, Syst. Appl. Microbiol. 13, 131-135. VAN TIEL-MENKVELD, G.J., MENTJOX-VERVUURT,M., OUDEGA,B., DEGRAAF, F. K. (1982), Siderophore production by Enterobacter cloacae and a common receptor protein for the uptake of aerobactin and cloacin DF13, J. Bacteriol. 150, 490-497. VANDEWOESTYNE, M., BRUYNEEL, MERB., GEAY, VERSTRAETE, (1991), The Fez+ M., W. chelator proferrorosamine A is essential for the siderophore-mediated uptake of iron by Pseudomonas roseus fluorescens, Appl. Environ. Microbiol. 57, 949-954. VISCA,P., SERINO, ORSI, N. (1992), Isolation L., and characterization of Pseudomonas aeruginosa mutants blocked in the biosynthesis of pyoverdin, J. Bacteriol. 174, 5727-5731. VISCA,P., CIERVO, ORSI,N. (1994), Cloning A., and nucleotide sequence of the pvdA gene encoding the pyoverdin biosynthetic enzyme L-ornithine N5-oxygenasein Pseudomonas aeruginosa, J. Bacteriol. 176, 1128-1140. WACHSMUTH, K., BLAKE,P. A., OLSVIK,0. I. (Eds.) (1994), Vibrio cholerae and Cholera Molecular to global perspectives. Washington, DC: Am. SOC. Microbiol. J. M. WEE, S., NEILANDS, B., BITTNER, L., HEMB. R. MING, c.,HAYMORE, L., SEETHARAM, B. (1988), Expression, isolation and properties of Fur (ferric uptake regulation) protein of Escherichia coli K12, BioMetals 1, 62-68. WIEBE, WINKELMANN,(1975), Kinetic studC., G. ies on the specificity of chelate iron uptake in Aspergillus, J. Bacteriol. 123, 837-842. WINKELMANN, (1979), Evidence for stereospeG. cific uptake of iron 14 chelates in fungi, FEES Lett. 97, 43-46.
246
WINKELMANN, (1986), Iron complex products G. (siderophores), in: Biotechnology, 1st Edn., Vol. 4, (REHM,H.-J., REED,G., Eds), pp. 215-243. Weinheim: VCH. WINKELMANN, (Ed.) (1991a), Handbook of MiG. crobial Iron Chelates. Boca Raton, F L CRC Press. WINKELMANN, (1991b), Importance of sideroG. phores in fungal growth, sporulation and spore germination, in: Frontiers in Mycology (HAWKSWORTH, L., Ed.), pp. 49-65. Wallingford D. CAB International. WINKELMANN,(1991c), Specificity of iron transG. port in bacteria and fungi, in: Handbook of MiG., crobial Iron Chelates (WINKELMANN, Ed.), pp. 65-105. Boca Raton, F L CRC Press. G. WINKELMANN, (1992), Structures and functions of fungal siderophores containing hydroxamate and complexone type iron binding ligands, Mycol. Res. %, 529-534. WINKELMANN, (1993), Kinetics, energetics, and G. mechanisms of siderophore iron transport in fungi, in: Iron Chelation in Plants and Soil MiL. B. croorganisms (BARTON, L., HEMMING, C., Eds.), pp. 219-239. Boca Raton, F L CRC Press. WINKELMANN, BRAUN, (1981), StereoselecG., V. tive recognition of ferrichrome by fungi and bacteria, FEMS Microbiol. Lett. 11, 237-241. WINKELMANN, WINCE, D. R. (Eds.) (1994), G., Metal Ions in Fungi. New York: Marcel Dekker. WINKELMANN, ZAHNER, (1973), StoffwechG., H. selprodukte von Mikroorganismen. 115. Eisenaufnahme bei Neurospora crassa. Zur Spezifitat des Eisentransportes, Arch. Microbiol. 88, 4960. WINKELMANN, VAN DER HELM, D., NEIG., LANDS, B. (Eds.) (1987), Iron Transport in MiJ. crobes, Plants and Animals. Weinheim: VCH. WINKELMANN, CANSIER, BECK,W., JUNG, G., A., G. (1994), HPLC separation of enterobactin and linear 2,3-dihydroxybenzoylserinederivatives: a study on mutants of Escherichia coli defective in regulation (fur), esterase (fes) and transport (FepA),BioMetals 7 , 149-154. WONG,G. B., KAPPEL,M. J., RAYMOND, N., K. G. MATZANKE,B., WINKELMANN, (1983), Coordination chemistry of microbial iron transport compounds. 24. Characterization of coprogen and ferricrocin, two ferric hydroxamate siderophore, J. Am. Chem. SOC.105,810-815. YABUUCHI, KOSAKO, OYAIZU, YANO, E., Y., H., I., HOTTA,H., HASHIMOTO, EAZAKI,T., Y., ARAKAWA, (1992), Proposal of Burkholderia M.
gen. nov. and transfer of seven species of the genus Pseudomonas homology group I1 to the new genus, with the type species Burkholderia and cepacia (PALLERONI HOLMS1981) comb. nov., Microbiol. Immunol. 36, 1251-1275. YAMAMOTO, CHOWDHURY, A. R., KUROS., M. DA, M., et al. (1991), Further study on polyamine compositions in Vibrionaceae, Can. J. Microbiol. 37, 148-153. YAMAMOTO, FUJITA, OKUJO, MINAMI, S., Y., N., S., S. C., MATSUURA, SHINODA, (1992), Isolation and partial characterization of a compound with siderophore activity from V. parahaemolyticus, FEMS Microbiol. Lett. 94, 181-186. YAMAMOTO, OKUIO, FUJITA, SAITO, S., N., Y., M., YOSHIDA, SHINODA, (1993), Structures of T., S. two polyamine-containing catecholate siderophores from Vibrio fluvialis, J. Biochem. 113, 538-544. YAMAMOTO, OKUJO, N., SAKAKIBARA, S., Y. (1994a), Isolation and structure elucidation of acinetobactin, a novel siderophore from Acinetobacter baumannii, Arch. Microbiol. 249-254. YAMAMOTO, OKUJO,N., YOSHIDA, MATS., T., SUURA, SHINODA, (1994b), Structure and S., S. iron transport activity of Vibrioferrin, a new siderophore from Vibrio parahaemolyticus, J. Biochem. 115, 868-874. I. F. YOUNG, G., GIBSON, (1979), Isolation of enterobactin from Escherichia coli, Methods Enzymol. 56,394-398. ZAHNER,H., DIDDENS, KELLER-SCHIERLEIN, H., W., NAEGELI, (1977), Some experiments with H. semisynthetic sideromycin, J. Antibiot. 30 (Suppl.), s-201-206. ZALKIN, FORRESTER, D., TEMPLETON, H. A., J. D. (1966), Ferrichrome A tetrahydrate. Determination of crystal and molecular structure, J. Am. Chem. SOC. 88, 1810-1814. ZHOU, x. H., VAN DER HELM, VENKATRAD., MANI, (1995), Binding characterization of the L. iron transport receptor from the outer membrane of Escherichia coli (FepA): differentiation between FepA und FecA, BioMetals 8, 129136. ZYCHA,H., SIEPMANN, (Eds.) (1969), MucoR. rales - Eine Beschreibung aller Gattungen und Arten dieser Pilzgruppe. Mit einem Beitrag zur Gattung Mortierella von G. Linnemann, Lehre: Verlag von Cramer. M., ZYWNO, R., ARCENEAUX,E. L., ALTWEG, S. J. BYERS,B. R. (1992), Siderophore production and DNA hybridization groups of Aeromonas spp., J. Clin. Microbiol. 30, 619-622.
Biotechnology
6 Advances in the Molecular Genetics of fl=LactamAntibiotic Biosynthesis
PAULL. SKATRUD
Indianapolis, IN, USA
TORSTEN SCHWECKE
Cambridge, UK
HENK VAN LIEMPT

Bonn, Germany
MATTHEWB. TOBIN
Indianapolis, IN, USA
1 Introduction 249 2 Organisms which Produce p-Lactam Antibiotics 250 3 Biosynthesis of @Lactam Antibiotics 251 4 Development of Genetic Transformation Systems 253 5 Enzymes Involved in Penicillin Biosynthesis 253 5.1 &(L-a-Aminoadipy1-L-Cysteinyl-D-Valine Synthetase (ACVS) 254 5.2 Isopenicillin N Synthase (IPNS) 256 5.3 Acyl-Coenzyme A:Isopenicillin N Acyltransferase (AT) 257 6 Cloning of the Genes Involved in the Biosynthesis of Penicillin G, Cephalosporin C, and Cephamycin C 259 7 Clustering of /I-Lactam Biosynthetic Genes 259
248
6 Advances in the Molecular Genetics of P-Lactam Antibiotic Biosynthesis
8 Compartmentalization of P-Lactam Biosynthetic Enzymes 261 9 Evolution of the &Lactam Biosynthetic Pathway: The Horizontal Transfer Hypothesis 262 10 p-Lactam Antibiotics and Recombinant DNA Technology: Practical Applications 264 11 Regulation of &Lactam Biosynthetic Gene Expression 265 12 Summary 267 13 References 267
I Introduction
249
1 Introduction
120 years ago scientists observed growth inhibition when certain microbes were cultured together and they recognized the potential therapeutic value of this antagonism (PASTEUR and JOUBERT,1877). Half a century later a similar observation in the British laboratory of SIR ALEXANDER FLEMING piqued interest in this phenomenon (FLEMING, 1929). Over a decade later, these discoveries and other observations led SIR EDWARD ABRAHAM and his colleagues to the first amazingly successful clinical use of penicillin in 1941 (ABRAHAM al., 1941). In the ensuet ing years, the clinical use of plactam antibiotics (i.e., penicillins, cephalosporins, and cephamycins) became the most successful foray of mankind in his on-going battle against infectious diseases. Indeed, it has been suggested that in the United States the use of antibiotics has added approximately ten years to life expectancy (MCDERMOTT ROGERS,1982). and Due to their intrinsically high therapeutic index and clinical success, plactam antibiotics have been and continue to be the most intensely used class of antibacterial compound. The mode of action of p-lactam antibiotics on bacteria has been extensively studied since the identification of penicillin (reviewed by WAXMANand STROMINGER, 1982). WISE and PARK(1965) and TIPPER STROMINand GER (1965) demonstrated that penicillin inhibits the transpeptidation reaction required for peptidoglycan crosslinking, and hypothesized that penicillin acts as a structural analog of the acyl-D-ala-D-ala terminus of the stem peptide (Park nucleotide) found in peptidoglycan. Other studies have demonstrated that penicillin also binds at the same active-site serine in two bacterial D-alanine carboxypeptidases (YOCUM al., 1979). Regeneration of et these enzymes requires cleavage of the acylenzyme complex by hydrolysis (of carboxypeptidases) or transaminolysis (of transpeptidases). The scissile bond of the p-lactam ring replaces the natural D-ala-D-ala substrate at the enzyme active site. Acylation of the catalytically active amino acid would occur with the opening of the plactam ring, forming an
inactive penicilloyl enzyme resistant to hydrolysis or aminolysis. Why the inhibition of cell wall biosynthesis should ultimately lead to cell death is a complex question (reviewed by SHOCKMAN al., et 1982). Peptidoglycan assembly is performed by several specific enzymes forming glycosidic, amide, and peptide bonds, and as such requires some kind of coordinate regulation of these enzymes. Early studies conducted by LEDERBERG (1957) assumed that inhibition of cell wall biosynthesis led directly to a shortage of building blocks, so that as growing cells increased in size, the limited or weakened cell wall was unable to contain the contents and lysed. However, more recent observations indicate that not all bacterial species undergo lysis after treatment with cell wall inhibitors such as penicillins. Indeed, the historically accepted dogma concerning that bacteriolysis is the direct result of penicillin treatment of susceptible bacteria has recently been challenged. GIESBRECHT al. (1993) et were able to distinguish between the onset of killing and the onset of bacteriolysis after treatment of S. aureus with p-lactam antibiotics, suggesting that the penicillin-induced bacteriolysis of sensitive bacteria is not the cause of death, but rather its consequence. The unparalleled clinical success of p-lactam antibiotics stimulated intense interest and research into these compounds as well as the organisms that produce them. For decades, industrial research was focused on improving overall productivity and on understanding the biosynthesis of penicillin G in Penicillium chrysogenum and of cephalosporin C in Cephalosporium acremonium (syn. Acremonium chrysogenum). Strain improvement studies utilized natural selection and extensive mutagenesis programs coupled with biochemical selections. These strain improvement studies resulted in strains vastly improved in their capacity to produce p-lactam antibiotics. By 1985, the central reaction pathways for the biosynthesis of p-lactams had been described (O'SULLIVAN SYKES, and 1983). Most of the intermediates had been isolated and the order of reactions could be assigned (QUEENER and NEUSS,1982) (see Fig. 1). The stage was set for the complete elucidation of biosynthetic pathway mechanisms, including the cat-
250
alytic entities and the corresponding genes. Specifically the production of modified P-lactam antibiotics seemed feasible and this spurred research in both industrial groups and academic centers. Initial recombinant DNA studies on plactam producing fungi were done in the industrial setting (QUEENER et al., 1985; PENALVA al., 1985) and as colet laborative studies between industry (Eli Lilly and Company, U.S.) and academia (Oxford University, U.K.) (SAMSONet al., 1985). Since that time both communities have contributed heavily to the expansion of our knowledge in this arena. In the years since 1983, two lines of research have come together to prepare the way for a rationally designed exploitation of plactam antibiotic producing organisms. The first major step was the identification and cloning of the plactam biosynthetic genes which engendered a means to influence the catalytic properties of the corresponding enzymes. And second, the development of transformation protocols for achieving stable expression of foreign DNA in P-lactam producing strains led to an encompassing array of methods for drug design and production. Several reviews have been written in the past few years which address progress made
in understanding the molecular biology of p lactam biosynthesis as well as the genetic manipulation and characterization of organisms producing these compounds. The reader is referred to several of these reviews for more extensive coverage of topics discussed in this chapter (MILLER and INGOLIA, 1989; INGOLIA and QUEENER,1989; SKATRUD, 1991, 1992; MARTfN et al., 1991; AHARONOWITZ et al., 1992; VICHITSOONTHONKUL 1994; et al., MART~N GUTIERREZ, and 1995). On the following pages, the developments in these lines of research are highlighted and examples of the use of these advances in basic knowledge for the improvement of antibiotic production are given.
2 Organisms which Produce P-Lactam Antibiotics

A variety of prokaryotic and eukaryotic microorganisms are able to synthesize bicyclic and monocyclic p-lactams (Tab. 1). Of these
Tab. 1 Representative P-Lactam-Producing Organisms (modified from TURNER, . 1992)
@-Lactam Structure
Fungi
Bacteria Actinomycetes Eubacteria
Penam
Cephem
Aspergillus nidulans Penicillium chrysogenum Epidermophyton Trichophyton Polypaecilum Malbranchea Microsporum Cephalosporium acremonium Paecilomyces Scopulariopsis Diheterospora Spiroidium
Streptomyces Nocardia
Flavobacterium Lysobacter Xanthomonas
Clavam Carbapenem
Streptomyces Streptomy ces
Erwinia Serratia
3 Biosynthesis of p-Lactam Antibiotics
251
compounds, the bicyclic p-lactams possess the most potent antimicrobial activity. Thus far, only P. chrysogenum and C. acremonium have been used extensively at the industrial scale for production of bicyclic p-lactam compounds. Despite their relatively potent antimicrobial activity, only a minority of these compounds are directly used in the clinical setting. However, chemical modification of bicyclic p-lactam fermentation products has produced a wide array of clinically useful semi-synthetic p-lactams. For the purposes of this chapter we will focus on the biosynthesis, molecular genetics, and enzymology of bicyclic p-lactams.
ase (IPNS) enzymes, respectively. In Nocardia lactamdurans and S. clavuligerus, a-aminoadipic acid is formed from L-lysine by lysine-E-aminotransferase, encoded by the lat gene. In the fungi, a-aminoadipate is recruited from L-lysine biosynthesis. Subsequent to the formation of isopenicillin N (IPN), the penicillin and cephalosporin/cephamycin biosynthetic pathways diverge. In P. chrysogenum and A. nidulans, hydrophobic penicillins (e.g., penicillin G) are formed by a transacylation reaction catalyzed by acylcoenzyme A:isopenicillin N acyltransferase (AT), encoded by the penDE gene. In cephalosporin/cephamycin producing organisms (e.g., C. acremonium, N. lactamdurans, and several strains of Streptomycetes), IPN is converted to penicillin N by an epimerase, encoded by the cefD gene. The thiazolidine ring of the penicillin is then ring-expanded to form deacetoxycephalosporin C (DAOC), containing the dihydrothiazine ring characteristic of the cephalosporins. DAOC is subsequently A clear understanding of p-lactam biosyn- hydroxylated at the C-3 position, forming thesis (QUEENER NEUSS,1982) enabled deacetylcephalosporin C (DAC). In the filand investigators to efficiently dissect and modify amentous fungus C. acremonium, the cefEF this pathway through the use of molecular gene product (DAOC/DAC synthase, expanbiology. All p-lactams contain the four-mem- dase/hydroxylase, REX/H) catalyzes both of bered azetidinone ring system in which anti- these reactions. In the prokaryotes S. clavulimicrobial activity resides (Fig. 1). Bicyclic @- gerus and N. lactamdurans, these two reaclactam antibiotics fall into two natural classes: tions are catalyzed by separate enzymes, penicillins with the second ring system being DAOC synthase (REX, expandase) and a five-membered thiazoladine ring and cepha- DAC synthase (hydroxylase), encoded by two losporins/cephamycins in which the second separate genes which bear striking homology ring system is a six-membered dihydrothia- to one another (cefE and cefc respectively). zine ring. The biosynthetic pathways leading At this stage, the biosynthetic pathways of ceto the production of penicillin G in P. chryso- phalosporins and cephamycins diverge. In C. genum, cephalosporin C in C. acremonium, acremonium, DAC is converted to cephalosand cephamycin C in Streptomyces clavulige- porin C, the end product of cephalosporin rus are illustrated in Fig. 1 as a branched biosynthesis, by an acetyltransferase reaction pathway. Designations for the genes encoding at the C-3 position. The enzyme catalyzing enzymes of this pathway were established by this reaction is encoded by the cefG gene. The INGOLIAand QUEENER(1989) and SKA- formation of cephamycin C in S. clavuligerus involves further modification of DAC. A 3'TRUD (1991); pcb refers to genes common to all three end products, pen defines genes hydroxymethylceph-3-em-0-carbamoyltransfound only in penicillin biosynthesis, cef ref- ferase, encoded by the cmcH gene, catalyzes ers to genes found only in cephalosporin C the carbamoylation at the C-3 position of biosynthesis, and cmc denotes genes specific DAC to form 0-carbamoyldeacetylcephalosfor cephamycin C biosynthesis. The so-called porin C. The last two steps in cephamycin C early genes (pcbAB and pcbC) encode the 6 biosynthesis involve 7 '-a-hydroxylase and (L-a-aminoadipoy1)-L-cysteinyl-D-valine syn- methyltransferase reactions, encoded by the thetase (ACVS) and isopenicillin N synth- cmcl and cmcJ genes, respectively.
3 Biosynthesis of P-Lactam Antibiotics
252
6 Advances in the Molecular Genetics of p-Lactam Antibiotic Biosynthesis

L-alpha-aminoadipic acid
L-cysieine
L-valine
I
LLDACV
ACV SYNTHETASE W A B )
4-membered
CqH
0
I S O F ' E N I C I L ~ ~ ~ f l T H E T A(IPNS) SE
6 A P A ACYLCoA
ACYL'IRANSlTRASE
p h c n y a c e ty l C o A
(a)
CqH
PENICILLIN N
F'ENICILLING
uacl?roXYCQWALCSFCfUNC OAW
0
DAOCS (EXF'ANDASE) W F in C.
DE4cliTYL-
( a E F in C. acremonium) ( U F in S.
C OAO
3-HYDROXYMETHYLCEPH-3-EM
CEPHALOSPORIN C SYNIRETASE
II
(3-3
CEPHALOSPORWC
C ~ H
CqH
Fig. 1. Biosynthesis of p-lactam antibiotics.
5 Enzymes Involved in Penicillin Biosynthesis
253
4 Development of Genetic Transformation Systems

The application of recombinant DNA technology was based on the development of genetic transformation systems for p-lactam producing organisms and cloning of biosynthetic genes. The ability to transform industrially relevant organisms (specifically the filamentous fungi P. chrysogenum and C. acremonium) would allow investigators to manipulate the biosynthetic pathway in a precise manner and provide an avenue for practical applications (e.g., gene dosage studies for rate-limiting steps, gene disruptions to alter end products, or the addition of novel genes to modify the end product). Experimental conditions utilized in the development of the first fungal transformation system (i.e., Saccharomyces cerevisiae, HINNEN al., 1978) et were relied upon heavily in establishing conditions for transformation of plactam producing fungi. The first fungal plactam producing organism to be transformed was Aset pergillus nidulans (YELTON al., 1984), however, this organism is not used on the industrial scale for production of p-lactam antibiotics. C. acremonium was transformed the following year (QUEENER al., 1985; PENALVA et et al., 1985). While the transformation of A . nidulans relied on auxotrophic complementation, transformation of C. acremonium made use of a dominant selectable marker, hygromycin B phosphotransferase. Mutagenesis of a highly developed production strain frequently resulted in a decline in overall productivity, thus the use of a dominant selectable marker which permitted transformation of a broad range of C. acremonium strains without mutagenesis was a significant advance. Several laboratories also developed transformation systems for P. chrysogenum shortly thereafter (BERIand TURNER, 1987; CANTORAL et al., 1987; SANCHEZ al., 1987; SKAet TRUD et al., 1987a). Typical for filamentous fungi, the transformation frequencies for these organisms are low (i.e., 1-10 transformants per Fg of transforming DNA). Subsequently, utilization of a homologous promoter (pcbC promoter) attached to the hygromy-
cin B phosphotransferase (HPT) gene significantly increased the rate of transformation for C. acremonium (SKATRUD al., 1987b). et Transformation occurs by integration rather than replication - in contrast to S. cerevisiae where both modes of transformation are readily available. In C. acremonium and P. chrysogenum, most transformation events occurred by ectopic integration into genomic DNA; only a small percentage of the transformants harbored homologous integration events. Despite these limitations, transformation systems such as these opened the door for the application of recombinant DNA technology to these organisms. Transformation systems for prokaryotes which produce p-lactam antibiotics have also been developed. Recently, an efficient genetic transformation system was developed for the prokaryote Nocardia lactamdurans, a cephamycin C-producing organism, using an endogenous plasmid found in a species of Amycolatopsis (KUMAR al., 1994). This developet ment made possible the genetic manipulation of Nocardia species and the related genus Amycolatopsis, some of which are used on an industrial scale for the production of antibiotics (e.g., efrotomycin, vancomycin, and rifamycin).

While the basic biosynthetic routes reading to the different p-lactam antibiotics (see Fig. 1) have been known for several decades, the enzymes catalyzing the reactions have only recently been isolated and studied in detail. Knowledge about primary and tertiary structures and catalytic properties of these rather unique enzymes provided a rationale for planning new biosynthetic routes and for the design of novel biocatalysts. Here, we will focus on the three enzymes leading to the formation of penicillin G. As stated above, isopenicillin N is a precursor for both penicillin G and cephalosporins, thus the first two en-
254
6 Advances in the Molecular Genetics of b-Lactam Antibiotic Biosynthesis
zymes are common to all organisms producing penicillin and cephalosporin-type compounds.
WITZ et al., 1992). So far, the multienzymes LIEMFTet al., 1989) S. from A. niduluns (VAN cluvuligerus (BALDWIN al., 1990; JENSEN et et et al., 1990; SCHWECKE al., 1992a; ZHANG et al., 1992), and C. ucremonium (BALDWIN et 5.1 S(L-a-Aminoadipy1)-Land al., 1990 ZHANG DEMAIN,1990a) have been purified to homogeneity and enzymatiCysteinyl-D-Valine Synthetase cally characterized (ZHANG,1991; ZHANG (ACVS) and DEMAIN,1992). The S. cluvuligerus ACVS was a monomeric protein (JENSEN et et 8- (L - aAminoadipyl) - L- cysteiny1- D - valine al., 1990; SCHWECKE al., 1992b; ZHANG (ACV), the first common intermediate in the and DEMAIN,1990b) whereas the C. ucreformation of all penicillins and cephalospo- monium ACV synthetase appeared to be a rins by eukaryotic and prokaryotic microor- homodimer with weak physical interactions and DEganisms, was initially thought to be synthe- between the two subunits (ZHANG et sized by two different enzymes in analogy to MAIN,1990a, b, 1992; ZHANG al., 1992; 1991). In analogy to many known the structurally similar tripeptide glutathione. ZHANG, This type of process would predict that a dipeptide synthetases, ACVS activates the peptide existed as a free intermediate. How- three substrate amino acids at the expense of ever, BANKOet al. (1986, 1987) showed that the a-p-phosphate bond of ATP (SCHWECKE an extract of C. ucremonium catalyzed the et al., 1992a; VAN LIEMPTet al., 1989). Forsynthesis of ACV at a substantially higher mation of aminoacyl adenylates with all three rate from the individual amino acids than amino acids was demonstrated by measuring from G(L-a-Aminoadipy1)-L-cysteine (AC) formation of labeled ATP from [32P]-PPiin and L-valine, and also faster than AC was the presence of the single amino acids. One formed from first component amino acids. mole of ATP was consumed per peptide bond The authors concluded that ACV synthesis in formed (KALLOW al., 1994). Substrate speet C. ucremonium involved the action of a single cificity studies have demonstrated that ACVS multifunctional enzyme. JENSEN al. (1988) accepts a broad range of amino acid analogs et reported on a cell-free extract of S. cluvulige- (SHIAUet al., 1995; ZHANG and DEMAIN, rus capable of synthesizing ACV from the ap- 1992). The a-aminoadipate dependent ATP/ propriate component amino acids. The actual PPi exchange reaction catalyzed by the S. cludemonstration of catalysis by a multienzyme vuligerus enzyme was inhibited by Tris buffer was achieved by VAN LIEMPTet al. (1989) (SCHWECKE al., 1992a) as well as tripepet et with the isolation of a large synthetase from tide formation (ZHANG al., 1992). ZHANG A. niduluns which catalyzed the formation of and DEMAIN (1992) compared several P-lacACV from the L-amino acid precursors. The tam biosynthetic enzymes in crude cell-free corresponding gene was identified and found extracts of C. ucremonium and S. cluvuligerus to direct the biosynthesis of a protein of and found that ACVS possessed only 1-10% 422kDa (MACCABEet al., 1991). To date of the specific activity of isopenicillin synseveral other genes encoding ACV synthetase thase, isopenicillin epimerase, and deacetoxyfrom prokaryotes as well as from eukaryotes cephalosporin C-synthetase. After being actihave been cloned and characterized. The de- vated as aminoacyl adenylates, the substrate duced sizes of the encoded polypeptides were amino acids were covalently bound to the cysbetween 404 and 424 kDa (AHARONOWITZ teamine moieties of 4g-phosphopantetheine et al., 1993). A common feature found in all (ROLAND al., 1975) which were attached to et these enzymes was the presence of three re- each activation domain of the enzyme by a peated modules, each comprised of more posttranslational modification (BALDWIN et than 500 amino acids. These modules exhi- al., 1991; SCHLUMBOHM al., 1991). In the et bited extensive sequence homology with non- final product formed by ACVS, valine ocribosomal peptide synthetases such as firefly curred in the D-configuration, however, only luciferase and acyl-CoA ligases (AHARONO- the L-configuration of covalently bound val-
255
ine could be released from the enzyme indi- ing blocks in the final product, as has been cating that epimerization took place upon or convincingly shown for other peptide synthetet after peptide bond formation (SHIAUet a]., ases (HORIet al., 1991; KRAUSE al., 1985; and MARAHIEL, 1988; KUROTSO et 1995; SCHWECKE VON DOHREN,unpub- KRAUSE and lished data). The putative thioesterase func- al., 1991; PIEPERet al., 1995; SAITOet al., et et tion located in the C-terminal region of 1995; STACHELHAUS al., 1995; TURGAY et ACVS has been proposed to specifically real., 1992; VAN SINDREN al., 1993; WEBER lease the LLD-tripeptide (KLEINKAUF and et al., 1994). Heterologous expression of individual VON DOHREN, 1996). The order of the partial reactions involved modules in E. coli has so far been hampered in the biosynthesis of the complete tripeptide by the fact that the fragments expressed were et is still not completely resolved. A particulate insoluble within the bacterium (SCHWECKE Masters Thesis, preparation of lysed protoplasts from a Ce- al., 1992b; V. UHLMANN, phalosporium sp. incorporated DL-[14C]-val- Technical University Berlin, cited in KLEINine into penicillin when incubated with the di- KAUF and VON DOHREN,1996). After renapeptide AC and ATP (LODER and ABRA- turation of protein from unfolded inclusion HAM, 1971b). BANKO and colleagues ob- bodies containing the N-terminal module of et tained ACV from AC and valine (BANKO et the S. clavuligerus ACVS, SCHWECKE al. al., 1986,1987). However, SHIAU al. (1995) (1992b) demonstrated adenylation of all three et recently reported on the formation of cyste- constituent amino acids with valine showing inyl-valine (CV) as an intermediate in ACV highest activity in the ATP/[32]PPi exchange biosynthesis implicating a different mecha- reaction followed by cysteine and a-aminonism, which implies the initial formation of adipate. This somewhat surprising result the second peptide bond. Nevertheless, the showed that this approach to determine kiyield was low and no biosynthesis of ACV oc- netic constants is of limited use in intact curred from a-aminoadipate and the dipep- ACVS. However, upon incubation with all constituent 14C-labeled amino acids only atide CV (SHIAU al., 1995). et Such results do not resolve the issue of aminoadipate gave rise to radioactivity covalwhether the order of peptide bond formation ently bound to the refolded fragment, supis ACV or VCA. A critical requirement for porting the idea of a modular arrangement of analyzing and manipulating ACVS is, there- activation units on the multienzyme that refore, the ability to characterize each individu- flects the order of amino acids in the product. al module or domain thereof. TAVANLAR No investigation into the nature of the covaland co-workers used limited proteolysis to ent bond was conducted. Therefore, it cannot isolate and analyze ACVS fragments from C. be ruled out that the observed enzyme-bound radioactivity was at least partially due to an acremonium (TAVANLAR, SCHWECKE, VAN LIEMPT, and VON DOHREN, unpublished unidentified contaminant known to be preset data, cited in KLEINKAUF VON DOHREN, ent in the 14C-a-aminoadipate(SCHWECKE and 1996). Three major bands with a Mr of ap- al., 1992a). While cysteine and valine are available in proximately 116 kDa were observed and purified after digestion of a homogenous ACVS all cells, a-aminoadipate has to be synthesample using either subtilisin or proteinase K. sized exclusively for p-lactam formation from The digestion products were investigated by lysine in actinomycetes (KERN et al., 1980 et et sequence analysis and biochemical assays. KIRKPATRICK al., 1973; MADDURI al., The activation of cysteine as aminoacyl ade- 1989). Cephalosporin and cephamycin pronylate could undoubtedly be attributed to the ducing strains possess an enzyme, L-lysine &middle fragment by means of ATP/PP, ex- aminotransferase (LAT), which mediates the change assay. Surprisingly, the N-terminal formation of a-aminoadipate by removal of et fragment activated valine to a higher degree the &-aminogroup from lysine (MADDURI et than a-aminoadipate. These results were in al., 1989, 1991; VINING al., 1990). The recontrast to the assumption that the order of sulting 1-piperidine-6-carboxylic acid was modules are colinear with the order of build- postulated to be oxidized by a yet unknown
256
dehydrogenase in analogy to a pipecolic acid pathway found in Pseudomonas sp. (CALVERT and RODWELL, 1966). The gene encoding LAT has been mapped to the pcblcef gene cluster in plactam antibiotic producing strains of Strepromyces (COQUEet al., 1991b; MADDURI al., 1991) and is absent in nonet producers (MADDURI al., 1989). MALMet BERG and colleagues found 4-fold elevated levels of LAT and 2-5 times increased cephamycin production after introduction of an additional copy of the gene into the chromoet some of S. clavuligerus (MALMBERG al., 1993, 1995). Fungal p-lactam producing strains make a-aminoadipate as an intermediate in the synthesis of lysine and do not seem to require such an enzyme.
two histidine residues, rendering the otherwise highly reactive iron less damaging to the cell. The IPNS enzyme catalyzed both cyclizations found in the structure of isopenicillin N. Since the initial purification of IPNS, the enzyme has been isolated from many different sources, including fungi, gram-positive and gram-negative bacteria. In each case, IPNS was characterized as a protein with a molecular mass very close to that of IPNS from C. acremonium. The IPNS reaction mechanism has been addressed in elaborate studies by BALDWIN and coworkers (BALDWIN ABRAHAM, and 1988; BALDWIN BRADLEY, and 1990). By analyzing kinetic isotope effects, they were able to show that the reaction proceeds in a two-step mechanism, the intermediate being a monocyclic plactam in which the sulphur is bound to iron dioxygen of the enzyme. Based on 5.2 Isopenicillin N Synthase cell-free catalytic systems using purified re(IPNS) and colleagues combinant IPNS, HUFFMAN demonstrated that IPNS could recognize and A key step in the biosynthesis of p-lactam utilize substrates similar to ACV. Such studantibiotics is the conversion of the tripeptide ies led to the formation of many novel plactam compounds, some of which possessed anACV into the bicyclic fused plactam-thiazolet idine ring structure. This particular step has tibiotic activity (HUFFMAN al., 1992). The three-dimensional structure of IPNS been the subject of many studies over several decades. The structure of the plactam ring has recently been elucidated by BALDWIN was correctly established through the work of and coworkers (ROACH et al., 1995). In a ABRAHAM colleagues over 50 years ago. complex with manganese, the recombinant and The structure of the direct precursor (ACV) IPNS protein from A. nidulans was crystalwas determined in 1971 (LODERand ABRA- lized, and the structure was determined at a f HAM,1971a). The enzymatic conversion o resolution of 2.5 A. Structural analysis of the tripeptide ACV into a p-lactam com- crystalized IPNS revealed the presence of 10 pound was finally established in 1979 ( O W L - helices and 16 P-strands, eight of which fold into a jelly-roll motif. The active site was buLIVAN et al., 1979). ABRAHAM al. (1981) et noted that IPNS required Fez+ and molecu- ried within the p-barrel. A glutamine residue lar oxygen for activity. The catalytic activity which is conserved in all IPNS primary strucof IPNS was greatly stimulated by the pres- tures available to date interacted with the meence of ascorbate and thiol compounds such tal ion (manganese in the case of the crystalas D m and by catalase (ABRAHAM al., line form of IPNS) as well as two histidines et 1981). The first purification of IPNS was re- and one aspartate residue. Furthermore, two ported from C. acremonium in 1984 (PANGet water molecules were coordinated to the al., 1984). Analysis of purified IPNS enzyme manganese. Based on these findings, a reacpredicted a protein of 38 kDa, which required tion mechanism was proposed in which ACV ferrous iron and ascorbate and used dioxygen and dioxygen bind to the coordination sites as a co-substrate during catalysis. The nature occupied by the water molecules and glutaof the predicted iron-binding site within the mate. IPNS is now the most well understood enactive site cleft was addressed by CHENand colleagues (KRIAUCIUNASal., 1991). They zyme in the p-lactam biosynthetic pathway. et suggested that the metal ion was bound to The elucidation of the reaction mechanism
257
cursing fermentations with phenylacetic acid or phenoxyacetic acid, respectively. The composition of the active form of AT has been the subject of some controversy. Initial purifications of the A T enzyme suggested that it was a monomeric protein of approximately 30 kDa (ALVAREZ al., 1987; ALONet so et al., 1988). However, subsequent protein 5.3 Acyl-Coenzyme A: purifications (e.g., WHITEMAN al., 1990) et have indicated that A T is a heterodimer comIsopenicillin N Acyltransferase posed of two dissimilar subunits of 11 kDa (a subunit) and 29 kDa ( p subunit). (AT) The eventual cloning of the gene encoding The final step in the biosynthesis of hydro- A T (penDE) from both P. chrysogenum phobic penicillins in P. chrysogenum and A . (BARREDO al., 1989c; TOBIN al., 1990) et et niduluns involves the removal of the L-a-ami- and A. niduluns (MONTENEGRO al., 1990; et noadipoyl side chain from isopenicillin N and TOBIN al., 1990), demonstrated that the et its exchange with one of many coenzyme A- penDE gene is composed of four exons enderived monosubstituted acetic acids (e.g., coding 357 amino acids, translated as a phenylacetyl, forming benzylpenicillin) 40 kDa proenzyme, and posttranslationally et (BRUNNER al., 1968; FAWCE-IT al., 1975; processed to generate the two subunits et ALVAREZet al., 1987, 1993; WHITEMAN et (Fig. 2). Based on NH2-terminal amino acid al., 1990). Side chain exchange either occurs sequences, proenzyme cleavage occurs beet directly or as a two-step process, forming 6- tween Glylo2 and Cyslo3 (BARREDO al., WHITEMAN al., 1990). However, the et aminopenicillanic acid (6-APA) as an inter- 1989~; mediate (QUEENER and NEUSS, 1982) (see presence of the -11 kDa subunit in active A T has been questioned by some investigaFig. 3). et A single multifunctional enzyme, acyl- tors (e.g., BARREDO al., 1989~). Recent electrospray mass spectrometric coenzyme A:isopenicillin N acyltransferase (AT), catalyzes all of the possible steps in- analysis of both recombinant and native P. volved in this transacylation, including 6- chrysogenum AT has verified the location of APA formation via IPN side chain removal the proenzyme cleavage site between Gly'O' (IPN amidohydrolase, IAH), 6-APA acyla- and CySlo3(APLINet al., 1993a, b). These retion (acyl-CoA:6-APA acyltransferase, ports also indicated that both the 11 kDa (a AAT), IPN transacylation (acyl-CoA:IPN subunit) and the 29 kDa ( p subunit) proteins acyltransferase, IAT), and acyl-CoA hydroly- are present in purified AT forming an a,@ sis (ALVAREZ al., 1993). The kinetic pa- heterodimer, in agreement with the report of et et rameters reported for each of these reactions WHITEMAN al. (1990). To further investisuggests that removal of the a-aminoadipoyl gate the requirement for both of these subside chain from IPN is the rate-limiting step units, TOBIN al. (1993) separated the reet in the acyltransferase reaction (ALVAREZ et gions of penDE encoding each subunit. Indeal., 1993). pendent production of either subunit in E. The ability of the A T enzyme to accept a coli did not yield active AT, and post-expresvariety of acyl-CoA derivatives (LUENGOet sion complementation by the mixing of cell al., 1986; ALONSO al., 1988), together with sonicates containing separately produced subet the capacity of P. chrysogenum to form a units did not regenerate activity. However, number of acyl-CoA thioesters, has allowed reconstitution of A T activity was accomthe in vivo production of more than 100 dif- plished by coproduction of the two subunits ferent penicillins in P. chrysogenum fermen- produced from separate plasmids in the same tations containing the corresponding precur- cell. Further, in vitro refolding of separately sor acid (COLE, 1966). The production of produced 11 kDa and 29 kDa subunits only penicillin G or V, e.g., is accomplished by pre- resulted in active A T when mixed together aided by knowledge of the three dimensional crystal structure combined with the known amino acid sequences of IPNS molecules from many different sources may provide an avenue for rational design of IPNS proteins with new desired activities.
258
pcbAB
1
I 1
penDE
I I
'"
I1
I1
II
transcription
-1.4-kb mRNA
AAA translation
AT 40-kDa proenzyme
I co;
I
I
cleavage
I 11-kDa a-subunitI I
- 1
29-kDa Esubunit
Mature AT (a, heterodimer?)
Fig. 2. Transcription,translation, and posttranslational modification of AT in P. chrysogenum.
prior to refolding (TOBIN,1994). These results indicate that the 11 kDa and the 29 kDa proteins interact at an intermediate step in the folding pathway to form AT enzyme. Studies of the posttranslational cleavage of the proenzyme have been hampered by the inability to repeatably isolate the -40 kDa AT proenzyme from P. chrysogenum. Using a heterologous penDE expression system for E. coli APLINet al. (1993a) observed the soluble production of heterodimeric AT, suggesting that other specific P. chrysogenum factors were not required for proenzyme cleavage. Using insolubly produced, resolubilized, and purified 40 kDa recombinant protein, TOBIN (1994) further demonstrated cleavage of the AT proenzyme by in vitro refolding. Cleavage occurred between Gly'O2 and Cys103, as observed for native and solubly produced recombinant AT, suggesting that this event is autolytic. Mutational analysis and recombinant expression of the penDE gene has identified three amino acid residues essential for A T activity. Substitution of several residues in the proenzyme cleavage site region demonstrated that Cyslo3 was absolutely required for AT proenzyme cleavage and AT activity. The presence of Ser or Ala at this position yielded uncleaved and inactive recombinant AT (ToBIN et al., 1995). Ser227,residing in an S-Q-N motif in AT and other plactam biosynthetic enzymes was also examined for a potential role in AT activity (TOBINet al., 1994). Sub-
stitution of this residue with Cys did not destroy AT activity, however, Ala at this position did not result in either proenzyme cleavage or activity. Further, a thioesterase-like domain (G-X-S309-X-G) has been identified in AT (ALVAREZ al., 1993). TOBIN al. et et (1994) observed that Ser3@+Cys retained activity, however, Ser3@+Ala was not active. AT enzyme containing either of these mutations was posttranslationally cleaved, indicating that this residue is involved in the AT enzyme mechanism itself. Together with the sensitivity of AT activity to chloromercuribenzoate and iodoacetamide and the requirement for reducing agents (e.g., DTT) for A T activity (ALVAREZ al., 1993), these results et imply that a thioesterase activity may be involved in the AT enzyme mechanism. Although there are five conserved Cys residues in P. chrysogenum and A. nidulans AT, the requirement for Cys lo3 tempts speculation that this residue may bind the CoA-derived acyl group, subsequently transferred to 6APA or IPN by a thioesterase activity. The apparent absolute requirement for proenzyme cleavage to produce active AT additionally suggests that proenzyme cleavage may be an enzyme activation mechanism. Substrate specificity studies have demonstrated that, in addition to IPN and 6-APA, AT will hydrolyze and acylate penicillin N forming 6-APA and penicillin G, respectively (ALVAREZet al., 1987; TOBIN,1994). Further, it has recently been shown that the ceph-
7 Clustering of PLactam Biosynthetic Genes
259
The isolation of antibiotic biosynthetic genes from Streptomyces species revealed that these genes tend to be organized in clusters, and that they are often associated with antibiotic resistance genes (MALARTIDA and HOPWOOD,1984; SENO and BALTZ, 1989). This clustering was also to be true for p-lactam biosynthetic genes when BAILEY al. et (1984) observed the production of clavulanic acid by introducing a cosmid clone containing a segment of S. clavuligerus DNA into a nonproducing mutant. BURNHAM al. (1987) et provided evidence that the cefE, cefF, and cmc genes were clustered in S. clavuligerus. KOVACEVIC al. (1989) later isolated a coset . mid clone from S clavuligerus containing both the pcbC and cefE genes involved in cephalosporin Ckephamycin C biosynthesis. This clone was subsequently used to isolate the cefD (KOVACEVIC al., 1990), cefF et (KOVACEVIC and MILLER, 1991), lat, and pcbAB genes (TOBIN al., 1991). et Both DiEz et al. (1989) and BARREDO et al. (1989b) demonstrated that the pcbC and penDE genes were adjacent to one another in A combination of approaches has been the filamentous fungus P. chrysogenum. A used to identify, clone, and characterize genes study employing genetic complementation involved in plactam biosynthesis including and heterologous cloning techniques later ingenetic complementation, gene disruption, dicated that all of the genes involved in the genetic linkage, and reverse genetics (MIL- biosynthesis of penicillins resided in close LER and INGOLIA, 1989; SKATRUD, 1991). proximity to one another in this organism Over the past decade, a substantial array of (SMITHet al., 1990b). In addition to demonplactam biosynthetic genes have been cloned strations of linkage in s. clavuligerus and P. and sequenced from several different organ- chrysogenum, clustering of plactam biosynisms (Tab. 2). A striking similarity exists be- thetic genes has been demonstrated in Aspertween corresponding genes and enzymes iso- gillus nidulans (SMITH al., 1990a) as well as et lated from various organisms, despite their in the bacteria Flavobacterium and N. lactamphylogenetic diversity. This characteristic has durans (COQUEet al., 1991a, b). In addition, been exploited to permit the isolation of a p-lactamase gene has been identified within genes from several different organisms by the Flavobacterium (Lysobacter) and N. laccross-hybridization techniques. tamdurans clusters (KIMURAet al., 1990; COQUEet al., 1993). Fig. 3 illustrates the arrangement of p-lactam biosynthetic genes in various representative organisms. Recent results have demonstrated that the clavulanic acid biosynthetic gene cluster in S. clavuligerus is adjacent to the gene cluster for p-lac-
em nucleus 7-aminodeacetoxycephalosporank acid (7-ADCA) is poorly acylated by recombinant AT, forming deacetoxycephalosporin G (TOBIN,1994). Transacylation of cephalosporins containing aliphatic side chains (e.g., deacetoxycephalosporin C, deacetylcephalosporin C, and cephalosporin C) has not been observed. However, the acceptance of 7-ADCA and penicillin N (the latter containing the D-configured a-aminoadipoyl side chain) as a substrate would suggest that AT could be used to produce hydrophobic cephalosporins for industrial application to semi-synthetic cephalosporin production, pending mutagenesis, and screening for the optimization of these activities.
7 Clustering of P-Lactam Biosynthetic Genes
6 Cloning of the Genes Involved in the Biosynthesis of Penicillin G, Cephalosporin C, and Cephamycin C
260
6 Advances in the Molecular Genetics of @-LactamAntibiotic Biosynthesis
Tab. 2. DNA Sequences Reported for @-LactarnBiosynthetic Genes
Gene (Enzyme Endocded) lat (lysine-E-arninotransferase) pcbAB (ACV synthetase)
Organism S. clavuligerus N. lactamdurans P. chrysogenum A. nidulans C. acremonium

S. clavuligerus (partial)
Reference TOBIN al. (1991) et COQUEet al. (1991a) SMITH al. (199Oc), D ~ E et al. (1990) et Z MACCABE al. (1991) et GUTIBRREZ al. (1991) et HOSKINS al. (1990) et TOBIN al. (1991) et SAMSON al. (1985) et CARR et al. (1986), BARREDOet al. (1989a) RAMdN et a]. (1987), WEIGEL et al. (1988) WEIGEL al. (1988) et LESKIW al. (1988) et SHIFFMAN al. (1988) et GARCIA-DOMINGUEZ (1991) et al. SHIFFMAN al. (1990) et COQUEet al. (1991b) BARREDO al. (1989~) et MONTENEGRO al. (1990), TOBIN al. et et (1990) KOVACEVIC al. (1990) et COQUE al. (1993) et SAMSON al. (1987) et KOVACEVIC al. (1989) et COQUEet al. (1993) KOVACEVIC MILLER and (1991) GUTII~RREZ (1992) et al. COQUE al. (1995a) et COQUEet al. (1995b) COQUEet al. (1995b)
pcbC (IPN synthase)
C. acremonium P. chrysogenum A. nidulans S. lipmanii S. clavuligerus S. jumonjinensi S. griseus Flavobacterium sp. 12154 N. lactamdurans P. chrysogenum A . nidulans S. clavuligerus N. lactamdurans C. acremonium S. clavuligerus N. lactamdurans S. clavuligerus C. acremonium N. lactamdurans N. lactamdurans N. lactamdurans
penDE (IPN acyltransferase) cefD (IPN epirnerase) cefEF (REZH) cefE (expandase) cefF (hydroxylase) cefG (acyltransferase) cmcH (carbarnoyltransferase) cmcl (7 -a-hydroxylase) cmcl (rnethyltransferase)
tam antibiotic biosynthesis (WARD and and QUEENER (1989) reported that the early HODGSON, 1993). genes were located on chromosome VI and An anomaly regarding linkage of biosyn- the late genes on chromosome I1 in an industhetic genes occurs in the filamentous fungus trial strain of C. acremonium (strain 394-4). C. acremonium. The early genes (pcbAB, However, a later report (FIERRO al., 1994) et pcbC) are linked (HOSKINS al., 1990), how- demonstrated that in C. acremonium ATCC et ever, they reside on a different chromosome 28901 the early and late genes resided in from the late (cefEF, cefG) genes. SKATRUD chromosomes VII and I, respectively. In both
8 Compartmentalization of PLactam Biosynthetic Enzymes
I
I 4
P. chrysogenumand A. nidulans
pcbAS
pcbC penD
-I
261
I I
C. acremonium *I
pcbAB pcbC
4-
ceff c e f ~
w - a
ORF OAF 10 8, 9
S. clavuligerus
ORFpCbC
1
c - w -1
/at cefD
ORF 2-7
pcbAS
cef cefF
N. lactamdurans
Fig. 3. Clustering of p-lactam biosynthetic genes in various organisms.
bla ORF J I
10
I
1-
H pcbc cmc
4+-----q
/at cefDcefE CmcTORFpbp
12
pcbAB
Lysobacter (Flavobacterium)
bla cefD ceff cefE pcbC pcbAS
studies the early and late genes were in separate clusters. It is conceivable that these apparently conflicting reports are due to differences between the strains. The location of the cefD gene in C. acremonium has not been determined. In A. niduluns ATCC 28901 the penicillin biosynthetic pathway is located on chromosome VI (3.0-Mb) (MONTENEGRO al., et 1992). In P. chrysogenum AS-P-78 and P2 the cluster is located on chromosome I (FIERRO et al., 1994). In P. norarum, the organism originally used for penicillin production, the biosynthetic genes reside on chromosome I1 (FIERROet al., 1994). The orientation of transcription of the plactam biosynthetic genes relative to one another is not entirely conserved. In the prokaryotes (Flavobacrerium,N. lactamdurans, and S. clavuligerus) the early genes tend to be transcribed in the same direction. PETRICH and JENSEN (1994) reported that the sequential arrangement of the lat, pcbAB, and pcbC genes affords coordinate regulation at the level of transcription, involving transcription units of various lengths. In addition to a short transcript for pcbC alone, a polycistronic message encoding LAT, ACVS, and IPNS has been suggested by these investigators. Whether or not the lut gene may be transcribed by itself is not clear (S. JENSEN,University of Alberta, personal communication). In addi-
tion, KOVACEVIC al. (1989) reported that et the cefD and cefE genes are cotranslated. Thus, portions of the biosynthetic pathway constitute operons. In the fungi, however, the early genes are transcribed in opposite directions. While clustering of biosynthetic genes in fungi is common separate genes have their own promoter, so relative orientation is not important in a regulatory sense (TURNER, 1992).
8 Compartmentalization of P-Lactam Biosynthetic Enzymes

The role of subcellular compartmentalization in the biosynthesis of penicillin in P. chrysogenum has been elucidated in some detail. Early studies indicated that ACVS activity was associated with a particular fraction of the crude cell homogenate (FAWCETT and ABRAHAM, 1976). IPNS is apparently a soluble enzyme, although its activity in cell-free extracts appears to be stimulated by Triton X100 or sonication (SAWADAet al., 1980). More recently, KURYLOWICZ al. (1987) et suggested that penicillin G is synthesized in
262
the Golgi system. Pursuant to this finding, MULLERet al. (1991) determined that the last step in penicillin biosynthesis (acyl transfer) is accomplished in organelles of 200-800 nm in diameter. A subsequent report by this group determined that a short signal peptide (AlaArg-Leu) located at the COOH-terminus of the AT enzyme is required for its localization in organelles (microbodies) and for benzylpenicillin biosynthesis (MULLERet al., 1992). This signal sequence is similar to that found at the COOH-terminus of several proteins known to be transported to peroxisomes (GOULDet al., 1989). In addition, evidence for the compartmentalization of the amino acid precursors required for ACV formation has been obtained (AFFENZELLER KUBIand CEK,1991; LEDENFELD al., 1993) and conet firms the suspicions of earlier studies regarding the limitation of ACVS activity in vivo by one of its substrates (HONLINGER al., 1988; et HONLINGERand KUBICEK,1989). LEDENFELD et al. (1993) ultimately concluded that the a-aminoadipate, L-cysteine, and L-valine precursors for ACV formation are derived from the vacuole, and ACVS is located either within or bound to the vacuolar membrane. A current model (as suggested by LEDENFELD et al., 1993) regarding the compartmentalization of events involved in penicillin biosynthesis in P. chrysogenum is illustrated in Fig. 4.
In light of the compartmentalization of the penicillin biosynthetic pathway it seems likely that the alleviation of substrate limitations, particularly with regard to the formation of ACV, would be an effective route to strain improvement.
9 Evolution of the P-Lactam Biosynthetic Pathway: The Horizontal Transfer Hypothesis

The improbable identity demonstrated between particular genes and enzymes involved in p-lactam biosynthesis from a wide diversity of microorganisms has generated a great deal of speculation concerning the horizontal transfer of the p-lactam biosynthetic pathway from a prokaryote to a eukaryote. The first in theory - as put forward by T. D. INGOLIA CARR al. (1986) and later expanded (WEIet GEL et al., 1988; SKATRUD, 1991) - was designed to explain the identity of the pcbC genes of P. chrysogenum and C. acremonium and the presence of IPNS activity in gram-positive bacteria of the Sfreptomyces genus. As
PSC
A
Pen
Extracellular
PSC
Microbodies Microbodies
PSC-CoA
IPNS ACVS IF 6ACV
a-AAA
P 6 ! C
-1
Intracellular
9
Vacuole
a-AAA
Val C s a-AAA y
Fig. 4. Compartmentalizationof the penicillin biosynthetic pathway in P. chrysogenum (adet apted from LENDENFELD al., 1993). Open circles represent possible transport steps, filled circles represent enzymatic steps; a-AAA: a-aminoadipic acid; PSC: precursor side chain; PSC-CoA: CoA-activated PSC, Pen: penicillin; 6-OPC 6-0x0piperidine-2-carboxylic acid; enzymes are indicated in bold letters.
9 Evolution of the p-Lactam Biosynthetic Pathway: The Horizontal Transfer Hypothesis
263
originally stated, the so-called horizontal and Staphylococcus aureus from unknown et transfer hypothesis argued for the transfer of sources have been proposed (DOWSON al., the pcbC genes from Streptomyces to a euka- 1989; MATSUHASHI al., 1986). Thus, in a et ryotic organism. Since then, the DNA se- mechanistic sense, a horizontal transfer of the quences of several more pcbC and other /?- p-lactam biosynthetic genes from a prokalactam biosynthetic genes have been eluci- ryote to a eukaryote appears plausible. dated, and the theory remains consistent with While a great deal of attention has been the new information. Supportive evidence paid to the acquisition of the biosynthetic from other genes, particularly the pcbAB, pathway by various organisms, little notice penDE, cefD, cefE, cefF, and cefEF genes, has has been given to the explanation of how a lent credence to the theory of horizontal number of genes recognizing similar sub1991). The close linkage strates appeared and how genetic linkage transfer (SKATRUD, of the genes in a variety of organisms makes a arose. There are two general scenarios: first, single transfer event plausible. However, the it is conceivable that several of the genes in DNA sequence of the pcbC gene from a the pathway arose via convergent evolution gram-negative bacterium, Flavobacterium, of substrate specificity to perform the various has produced an extension/modification of reactions. The second possibility is that some the suggested transfer theory. Based on this of the genes may have arisen by divergent DNA sequence AHARONOWITZal. (1992) evolution, requiring (1) a gene duplication, et suggested that more than one horizontal and (2) divergent evolution of function. transfer event may have occurred during the There does not appear to be a great deal of DNA or amino acid homology between most evolution of this biosynthetic pathway. It is virtually impossible to accurately pre- genedenzymes for different steps in the pathdict or suggest how this horizontal transfer way, with the exception of a conserved motif might have occurred. However, there are sev- (S ...Q ...N) identified by KOVACEVICand eral known mechanisms of DNA transfer, in- MILLER(1991) and TOBIN al. (1994). This et cluding conjugative plasmids, DNA transfor- motif is surrounded by a block of similar amimation, protoplast fusion, and bacteriophage no acids in all the enzymes examined. Howevtransduction. In one study plasmids in plac- er, the case of the expandase and hydroxylase tam producing streptomyces were detected functions provides an elegant testimonial for et recently (KINASHI al., 1995) The plasmids the second possibility. In C. acremonium, one ranged in size from 12 kb to 450 kb - an ade- gene (cefEF) encodes both activities, howevquate size to contain the entire P-lactam bio- er, in S. clavuligerus, these two activities are synthetic pathway in at least some cases. encoded by separate genes (cefE and cefF). In However, hybridization analysis did not pro- the latter organism, each enzyme retains a revide evidence for the genes on these plasmids sidual level of the other activity, however, the (KINASHI al., 1995). The ability to conju- DNA and amino acid sequences are strikingly et gate E. coli to both Saccharomyces cerevisiae similar (KOVACEVIC MILLER, and 1991). The (HEINEMANN SPRAGUE, and 1989) and Schi- level of identity indicates that the separate et zosaccharomyces pombe (SIKORSKI al., cefE and cefF genes arose by gene duplication 1990) in the laboratory is a direct demonstra- in S. clavuligerus (KOVACEVIC MILLER, and tion of DNA transfer between a prokaryote 1991). An extended account of the horizontal and a eukaryote. BORK and DOOLITTLE transfer hypothesis has been given in SKA(1992) have proposed that the fibronectin TRUD (1991). type I11 domain of animal proteins has been acquired by bacteria through horizontal transfer from mammalian cells. In addition, bacterial acquisition of antibiotic resistance by plasmids, transposable elements, and gene transfers is well known (DAVIES, 1994) and recent examples of horizontal DNA transfers of PBP genes to Streptococcus pneumoniae
264
6 Advances in the Molecular Genetics of fi-Lactam Antibiotic Biosynthesis
These investigators returned multiple copies of this gene via genetic transformation to a different strain of C. acremonium. They were able to show a direct correlation between cefG copy number, message level, and cephalosporin C titer suggesting a different ratelimitation in their strain. It is interesting to note that in the study by SKATRUDet al. (1989) the fragment containing the cefEF The use of gene dosage of genes involved gene was large enough to include the cefG in penicillin biosynthesis for industrial strain gene as well. Thus cefG encoded activity may improvement of P. chrysogenum was unsuc- have been augmented as well, depending on cessful (SKATRUD et al., 1987a). Shortly the site of integration within the transforming thereafter, evidence emerged which explained DNA, and may have exerted some influence such results. Analysis of industrial strains of on the results in that study. P. chrysogenum revealed that the cluster of Several steps in the biosynthesis of cephagenes responsible for penicillin production losporin C are oxygen dependent. DEMOwas reiterated several times (BARREDO al., DENA et al. (1993) reasoned that increasing et 1989b; SMITH et al., 1989; TOBIN, unpub- the intracellular availability of oxygen might lished data). Thus gene dosage was already positively affect overall antibiotic productiviutilized by these strains as a means of increas- ty. To accomplish this goal they expressed a ing overall productivity. To further increase bacterial hemoglobin gene in C. acremonium yields in such strains, it may be more fruitful (strain C10). Fermentation results demonto study and alter regulation of gene expres- strated increased cephalosporin C production sion within the penicillin biosynthetic clus- in some transformants carrying the bacterial hemoglobin gene. It should be noted, howevter. In contrast, gene dosage studies in C. acre- er, that the strain transformed was not a highmonium did meet with success. In the first ex- ly developed strain in terms of cephalosporin ample, characterization of an industrial strain C production. Also, fermentation analyses of C. acremonium suggested that the expand- were performed in shake flasks (baffled and ase/hydroxylase step was rate-limiting in the unbaffled) in which oxygen limitation is an inbiosynthesis of cephalosporin C due to the ac- herent problem. It is not clear that the same cumulation of isopenicillin N (see Fig. 1). To type of effect would be observed in largeovercome this rate limitation, this strain was scale fermentations. transformed with a plasmid containing the Another exciting area for application of recefEF gene encoding expandase/hydroxylase. combinant DNA technology to plactam proIsolate LU4-79-6 contained one extra copy of ducing organisms was the production of intercefEF ectopically integrated into its genome. mediate compounds useful in the manufacIn fermentations utilizing strain LU4-79-6, ture of semi-synthetic cephalosporins (SKAisopenicillin N no longer accumulated, sug- TRUD, 1992). Medically important cephalogesting that this rate-limitation was over- sporins such as cephalexin, cephradine, and come. Further analysis revealed that this cefadroxil are made by the addition of differstrain possessed twice the amount of expand- ent side-chains to the 7-amino group of 7-amiase/hydroxylase activity as compared to the nodeacetoxycephalosporin C (7-ADCA) et recipient and produced approximately 15% (BUNNELL al., 1986). This procedure normore cephalosporin C when scaled up to in- mally includes several chemical modification dustrial level (SKATRUD al., 1989). In an- steps with production of organic solvent et other example, the cefG gene which is direct- waste products. A biosynthetic/enzymatic ly adjacent to cefEF was cloned from C. acre- process to produce 7-ADCA was proposed by monium (MATHISON al., 1993). This gene CANTWELL al. (1990) which would elimiet et encodes the enzyme which carries out the last nate three chemical modifications and reduce step in cephalosporin C production (Fig. 1). pollution. The proposed process involved in
10 P-Lactam Antibiotics and Recombinant DNA Technology: Practical Applications
11
Regulation of P-Lactam Biosynthetic Gene Expression
265
vivo ring expansion of penicillin V in P. chrysogenum via a genetically engineered cefE gene to form cephalosporin V and subsequent enzymatic deacylation resulting in the desired end product, 7-ADCA. The initial step in establishing feasibility of such a process was accomplished by these investigators. . The cefE gene from S clavuligerus was modified by fusing it in frame with the pcbC promoter of P. chrysogenum. The modified gene was transformed into an industrial strain of P. chrysogenum. Transformants were recovered which maintained their capacity to produce penicillin and possessed expandase activity (CANTWELL al., 1990). In order for this et work to achieve maximal impact, the amount of cefE gene product produced in P. chrysogenum transformants would have to be increased and the substrate specificity of cefE would have to be modified to efficiently act on penicillin V. CRAWFORD al. (1995) took a somewhat et different approach to achieve the production of 7-ADCA in P. chrysogenum. Their process . involved the expression of the S clavuligerus cefE or the C. acremonium cefEF gene in genetic transformants of P. chrysogenum, and subsequent fermentation in the presence of adipic acid. Transformants produced adipoyl6-APA (presumably by acyltransferase-catalyzed transacylation of IPN and adipoylCoA), which is subsequently ring-expanded, producing high titers of adipoyl-7-aminodeacetoxycephalosporanic acid (ad-7-ADCA) (cefE-catalyzed), or adipoyl-7-aminodeacetylcephalosporanic acid (ad-7-ACA) (cefEF-catalyzed). The adipic acid side chain from ad7-ADCA is removed by the Proteus SY77-1 cephalosporin acylase, producing 7-ADCA in high yields. Current work is directed at attaching this acylase to a resin to enable the removal of the adipic acid from the acylase reaction. One apparent drawback to this approach may involve isolation of ad-7-ADCA or ad-7-ACA from the fermentation broth, however, these products may prove to be extractable by an organic solvent. It is interesting to note that the observed i vivo ring exn pansion of adipoyld-APA could not be duplicated in vitro with cell-free extracts of S. cluvuligerus (MAEDA et al., 1995). YEH et al. (1994) similarly reported no in vitro ring ex-
pansion of adipoyld-APA with expandase or expandase/hydroxylase produced by recombiet nant E. coli cells. In contrast, BALDWIN al. (1987), working with cell-free extracts from C. ucremonium, were able to detect ring expansion of this substrate.
11 Regulation of P-Lactam Biosynt hetic Gene Expression

Investigations of fungal strains used in the production of plactam antibiotics have revealed two general strategies to enhance productivity. In general terms, increasing the amount of biosynthetic enzymes present in an organism should increase productivity as long as primary metabolites or other factors utilized in biosynthesis are not depleted. Two routes for increasing the concentration of biosynthetic enzymes are readily apparent. An increase in the rate of transcription (i.e., upregulation) or an increase in the number of genes (i.e., gene dosage) or a combination of both, should achieve enhanced production of plactams. As detailed earlier, genes involved in the biosynthesis of p-lactam antibiotics are closely linked in several organisms. Such close physical proximity of the biosynthetic genes entails obvious benefits regarding the inheritance of these genes and may simplify the possibility of coordinate transcriptional regulation. In P. chrysogenum, physical linkage of the penicillin biosynthetic genes proved to be an advantage in terms of improving overall penicillin productivity by gene dosage. Two groups independently demonstrated that the penicillin biosynthetic cluster in P. chrysogenum was amplified several-fold in strains highly developed for penicillin production (SMITH al., 1990a; D ~ E Z al., 1990). The et et three genes formally recognized as a part of the biosynthetic pathway are contained within an approximately 15-kb region. The total amplified region was about 35 kb in size. In a et more recent study, FIERRO al. (1995) found
266
6 Advances i the Molecular Genetics of P-Lactam Antibiotic Biosynthesis n
a 106.5-kb region repeated several-fold in penicillin biosynthesis is sensitive to environtandem on chromosome I in a different strain mental factors such as nitrogen source, carof P. chrysogenum which was highly devel- bon source, and growth stage in P. chrysogeoped for the production of penicillin. An- num suggesting the presence of several reguand (1992) other strain they investigated contained a latory factors. ESPESO PENALVA 58-kb repeat containing the penicillin bio- described a temporal pattern of expression of synthetic pathway. These investigators also penicillin biosynthetic genes in A. nidulans noted a hexanucleotide repeat (TTTACA) which was dependent upon carbon source. which bounded the amplified units in all An enhancement of that knowledge came et strains examined, regardless of copy number when FENG al. (1994) verified that expreset or size of amplified unit. FIERRO al. (1995) sion of the penicillin biosynthetic genes in P. suggested that amplification occurred by mu- chrysogenum was regulated by these factors tation-induced site-specific recombination at at the level of transcription. Using a reporter the conserved hexanucleotide sequence. gene analysis, their study revealed that proA salient point regarding the amplified unit moters found within the intergenic region beof genomic DNA in P. chrysogenum is that tween pcbAB and pcbC were sensitive to nithe length of the pcbAB, pcbC, and penDE trogen repression, glucose repression, and genes is only a fraction of the amplified regrowth stage. The gene (NRE) encoding the gion. It is possible that other genes involved major nitrogen regulatory protein has recentin penicillin biosynthesis and the regulation ly been cloned from P. chrysogenwn (HAAS thereof are encoded by this region. Indeed, a et al., 1995). This gene resembles other major recent report suggests that the pcbAB and nitrogen regulatory genes (e.g., 60% identical pcbC genes are subject to truns-acting regula- to ureA from A. niduluns and 30% identical tion (CHU et al., 1995). A transcriptional reg- to nit2 from Neurosporu crussu). HAASand ulatory protein which binds to the upstream MARZLUF (1995) carried these studies further regions of both pcbAB (-32 to -200 and by demonstrating that the NRE encoded pro-188 to -344) andpcbC( -10 to -131) was tein binds specifically to the intergenic proidentified by electromobility shift assay. Bind- moter regions of nitrate assimilation and pening of this regulatory protein to these regions icillin biosynthetic gene clusters of P. chrysoresults in a repression of transcription of both genum. All three of the above mentioned magenes. A similar situation has been observed jor nitrogen regulatory proteins are members et in A . niduluns (PEREZ-ESTEBAN al., 1993). of the GATA protein family - as such they Whether these elements effect penDE tran- are DNA binding proteins characterized by scription, either directly or indirectly by re- the presence of a zinc finger motif (Cys-X2pression of pcbAB and pcbC transcription, Cys-X17-Cys-X2-Cys) and they recognize the et has not been reported. FENG al. (1995) re- sequence GATA which is present in the inported several nuclear proteins which bind to tergenic region of pcbABlpcbC. ESPESO al. et the intergenic region between pcbAB and (1993) and TILBURN al. (1995) demonet pcbC. One particularly abundant nuclear pro- strated that the pcbC gene was regulated by tein, nuclear factor A (NF-A), recognizes and PACC, a regulatory protein which mediates binds to a specific 8 base pair sequence pH response of several other A. niduluns (GCCAAGCC) found in the intergenic re- genes as well. THEN BERGH et al. (1996) gion. BRAKHAGE and VANDENBRULLEidentified a major cis-acting DNA element, (1995) utilized a reporter gene system in A . located between pcbAB and pcbC in A . niduniduluns to identify recessive truns-acting mu- luns which regulates expression of both genes. tations residing in two complementation In this organism 872 base pairs separate these groups (prgAl and prgB1) which regulate p two divergently transcribed genes. Deletion lactam biosynthesis. Mutations in these comof nucleotides -353 to -432 upstream of plementation groups resulted in a depression pcbAB lead to an altered expression of both of penicillin production. genes. Further, analysis revealed the seAlthough not clearly understood at the mo- quence CCAAT (designated PENR1) found lecular level, it has been known for years that within this region, was bound specifically by a
I3 References
267
protein or protein complex. Evidence was also provided which suggested that the corresponding genes in C. aremonium and P. chrysogenum contained a similar regulatory circuit. The regulation of genes involved in cephalosporin C production in C. acremonium or cephamycin C production in the streptomycetes has not been studied as extensively as those involved in penicillin production. It is interesting to note that the first two genes of the biosynthetic pathway (pcbAB and pcbC) in C. acremonium are physically arranged in the same manner as in A. nidulans and P. chrysogenum (see Fig. 2). Despite a similar physical arrangement, gene amplification for the cephalosporin C biosynthetic pathway has not been observed. This observation suggests that alteration in gene regulation produced enhanced cephalosporin C production in industrial strains of C. acremonium. The relative strength of the pcbAB and pcbC promoters was recently studied with a reporter gene system (MENNEet al., 1994). Results of that study suggested that the pcbC promoter was approximately five times stronger than the pcbAB promoter. Such a contrast in the relative strength of promoters in the same biosynthetic pathway might suggest differences in protein stability, protein localization (i.e., compartmentalization vs. cytoplasmic), or specific activity of the enzymes involved in these steps. Regulation of the rate of transcription for the cephalosporin C biosynthetic pathway can be positively influenced by the addition of exogenous methionine to the medium (VELASCO al., 1994). et
provement studies). Currently, strain improvement may be approached not only from the standpoint of enhanced productivity but also biosynthetic pathway engineering resulting in production of valuable new end products useful in the manufacture of semisynthetic p-lactam compounds (SKATRUD, 1992). In the same time frame, the clinical effectiveness of p-lactam antibiotics was once again threatened by wide-spread emergence of serious resistance problems (e.g., methicillin-resistant Staphylococcus aureus [MRSA], penicillin-resistant Streptococcus pneumoniae). The primary mode of resistance in these organisms was not due to the elaboration of p-lactamases which has been a major clinical problem in the past. Rather, the targeted penicillin-binding proteins (PBPS) were either modified by mutation (PBP2x of S. pneumoniae) or replaced with a novel PBP (PBP2a in MRSA) resulting in a target with low affinity for p-lactam antibiotics (SPRATT, 1984). Once 1989; HARTMAN and TOMASZ, again recombinant DNA technology may have a dramatic impact on the use of p-lactam antibiotics. However, in this case the application of this technology will also be on the organisms which are killed by these antibiotics, not on the producing organisms. One novel approach to development of new antibiotics targeted at the modified PBPs may involve structure-based drug design in which the three-dimensional structure of a low-affinity PBP will be determined (Wu et al., 1992). Based on that information, modifications to existing compounds may lead to new more effective plactam antibiotics and novel approaches for rapid diagnostic detection of problematic organisms (UNALet al., 1992).
12 Summary
The last decade has produced numerous significant advances in the p-lactam antibiotic field. Most noteworthy perhaps has been the application of recombinant DNA technology to. the organisms which produce these lifesaving compounds. This application of technology has opened new doors to many avenues of research which were in some instances becoming less fruitful (e.g., strain im-
13 References
ABRAHAM, P., CHAIN,E, FLETCHER, M., E. C. GARDNER, D., HEATLEY, G., JENNINGS, A. N. M. A., FLOREY, W. (1941), Further observaH. tions on penicillin, Lancet 241 (2), 177-188. ABRAHAM, P., HUDDLESTON, A., JAYATIE. J. LAKE, G . S., O'SULLIVAN, WHITE, R. L. J., (1981), in: Recent Advances in the Chemistry of
268
J. E. PLactam Antibiotics, R. SOC. Chem., Special BALDWIN E., ABRAHAM, P. (1988), The biosynthesis of penicillin and cephalosporins, Nat. Publication No. 38, pp. 125-134. Prod. Rep. 5, 129-145. AFFENZELLER, KUBICEK, P. (1991), EviK., C. J. M. dence for a compartmentation of penicillin bio- BALDWIN, E., BRADLEY, (1990), Isopenicillin N synthase mechanistic studies, Chem. Rev. 90, synthesis in a high- and a low-producing strain of 1079-1088. Penicillium chrysogenum, J. Gen. Microbiol. 131, BALDWIN, E., ADDLINGTON, M., COATES, J. R. J. 1653-1660. B., CRABBE, J., CROUCH, P., KEEPING, M. N. J. AHARONOWITZ, COHEN,G., MARTIN, J. F. Y., W., KNIGHT, C., SCHOFIELD, J., TING, B. C. H.(1992), Penicillin and cephalosporin biosynthetic H., VALLEJO,C. A,, THORNILEY, ABRAM., genes: structure, organization, regulation and HAM, P. (1987), Purification and initial charE. evolution, Annu. Rev. Microbiol. 46, 461495. acterization of an enzyme with deacetoxycephaAHARONOWITZ, BERGMEYER, CANTORAL, Y., H., losporin C synthase and hydroxylase activities, J. M., COHEN,G., DEMAIN, L., FINK,U., A. Biochem. J. 245, 831-841. KINGHORN, LANCE,M., MCCABE, PALJ., A., J. J. ISA, H., PFEIFER,E., SCHWECKE, VAN BALDWIN, E., BIRD, W., FIELD,R. A., OCALT., LAGHAN, M., SCHOFIELD, J. (1990), IsolaN. C. LIEMPT, H., VON DOHREN,H., ZHANG,J. (1993), S(L-cu-Aminoadipyl)-L-cysteinyl-D-val- tion and partial characterization of ACV synthetase from Cephalosporium acremonium and ine synthetase, the multienzyme integrating the Streptomyces clavuligerus, J. Antibiot. 43, 1055four primary reactions in plactam biosynthesis, 1057. as a model peptide synthetase, BiolTechnology 11,807-810. BALDWIN, E., BIRD, W., FIELD,R. A., OCALJ. J. LAGHAN, M., SCHOFIELD, J., WILLIS, N. C. A. ALONSO, J., BERMEJO, REGLERO, FERM. F., A., C. (1991), Isolation and partial characterization DE NANDEZ-CAN6N, J. M., GONZALEZ BUIof ACV synthetase from Cephalosporium acreTRAGO, LUENGO, M. (1988), Enzymatic G., J. monium and Streptomyces clavuligerus:evidence synthesis of penicillins, J. Antibiot. 41, 1074for the presence of phosphopantothenate in 1084. ACV synthetase, J. Antibiot. 44,241-248. ALVAREZ, CANTORAL, M., BARREDO, L., E., J. J. A. D~Ez, MART~N, F. (1987), Purification to BANKO,G., WOLFE, S., DEMAIN, L. (1986), B., J. Cell-free synthesis of S(L-a-aminoadipy1)-L-cyshomogeneity and characterization of acyl-coenteine, the first intermediate of penicillin and zyme A: 6-aminopenicillanic acid acyltransferase cephalosporin biosynthesis, Biochem. Biophys. of Penicillium chrysogenum, Antimicrob. Agents Res. Commun. 131,528-535. Chemother. 31, 1675-1682. G., A. ALVAREZ, MEESSCHAERT, MONTENEGRO, BANKO, DEMAIN, L., WOLFE,S. (1987), 6 E., B., (L-a-Aminoadipyl)-L-cysteinyl-D-vaIine synthetE., GUTIERREZ, D~Ez, BARREDO, L., S., B., J. ase (ACV synthetase): a multifunctional enzyme J. MART~N, F. (1993), The isopenicillin-N acylwith broad substrate specificity for the synthesis transferase of Penicillium chrysogenum has isoof penicillin and cephalosporin precursors, J. penicillin-N amidohydrolase, 6-aminopenicilAm. Chem. SOC. 109,2858-2860. lanic acid acyltransferase and penicillin amidase activities, all of which are encoded by the single BARREDO, L., CANTORAL, M., ALVAREZ, J. J. E., penDE gene, Eur. J. Biochem. 215,323-332. D~Ez,B., MART~N, F. (1989a), Cloning, seJ. quence analysis and transcriptional study of the APLIN, T., BALDWIN, E., COLE,S. C. J., SuR. J. isopenicillin N synthase of Penicillium chrysogeM. THERLAND, J. D., TOBIN, B. (1993a), On the num AS-P-78, Mol. Gen. Genet. 216,91-98. production of a,P-heterodimeric acyl-coenzyme A:isopenicillin N acyltransferase of Penicillium BARREDO, L., DIEz, B., ALVAREZ, MART~N, J. E., chrysogenum: studies using a recombinant J. F. (1989b), Large amplification of a 35-kb source, FEBS Lett. 319, 166-170. DNA fragment carrying two penicillin biosynthetic genes in high penicillin producing strains J. P. APLIN, T., BALDWIN, E., ROACH, L., RoR. of Penicillium chrysogenum, Curr. Genet. 1 , 6 BINSON, c. V., SCHOFIELD, c. J. (1993b), Inves453459. tigations into the post-translational modification and mechanism of acyl-coenzymeAisopenicillin BARREDO, L., VANSOLINGEN, D~Ez, ALJ. P., B., VAREZ, E., CANTORAL, M., KAITEVILDER, J. N acyltransferase using electrospray mass spectrometry, Biochem. J. 294, 357-363. A., SMAAL, B., GROENEN, A. M., VEENE. M. STRA, A. E., MARTfN, J. F. (1989~), Cloning and BAILEY, R., BUTLER,M. J., NORMANSELL, C. I. characterization of the acyl-coenzyme A:6-amiD., ROWLANDS, T., WINSTANLEY, J. R. D. nopenicillanic acid acyltransferase gene of Peni(1984), Cloning a Streptomyces clavuligerus gecillium chrysogenum, Gene 83,291-300. netic locus involved in clavulanic acid biosynthesis, BiolTechnology 2, 808-811. G. BERI,R. K., TURNER, (1987), Transformation
13 References
of P. chrysogenum using the A . nidulans amdS gene as a dominant selective marker, Curr. Genet. 11, 639-641. BORK,P., DOOLITTLE, F. (1992), Proposed acR. quisition of an animal protein domain by bacteria, Proc. Natl. Acad. Sci. USA 89, 8990-8994. BRAKHAGE, A., VANDENBRULLE,(1995), A. J. Use of reporter genes to identify recessive transacting mutations specifically involved in the regulation of Aspergillus nidulans penicillin biosynthesis genes, J. Bacteriol. 177(lo), 2781-2788. BRUNNER, ROHR, M., ZINNER, (1968), Zur R., M. Biosynthese des Penicillins, Hoppe-Seylers 2. Physiol. Chem. 349.95-103. BUNNEL, A., LUKE,W. D., PERRY, M., JR. C. F. (1986), in: PLactam Antibiotics for Clinical Use (QUEENER,. F., WEBER,J. A., QUEENER, S S. W., Eds.), pp. 255-284. New York: Marcel Dekker. BURNHAM, K. R., HODGSON, E., NORMANM. J. SELL, I. D. (1987), Isolation and expression of genes involved in the biosynthesis of p-lactams, Eur. Patent Application No. 0233715. CALVERT, F., RODWELL, W. (1966), MetabA. V. olism of pipecolic acid in a Pseudomonas sp., J. Biol. Chem. 241,409-414. CANTORAL, M., DIEZ, B., BARREDO, L., ALJ. J. VAREZ, E., MART~N,E. (1987), High-frequenJ. cy transformation of P. chrysogenum, BiolTechnology 5,494-497. CANTWELL, A., BECKMANN, J., DOTZLAF, C. R. J. E., FISHER, L., SKATRUD, L., YEH,W.-K., D. P. QUEENER, W. (1990), Cloning and expression S. of a hybrid S. clavuligerus cefE gene in P. chrysogenum, Curr. Genet. 17,213-221. CARR,L. G., SKATRUD, L., SCHEETZ, E., P. M. S. T. QUEENER, W., INGOLIA, D. (1986), Cloning and expression of the isopenicillin N synthetase gene from Penicillium chrysogenum, Gene 48,257-266. CHU, Y. W., RENNO,D., SAUNDERS, (1995), G. Detection of a protein which binds specifically to the upstream region of the pcbAB gene in Penicillium chrysogenum, Curr. Genet. 28(2), 184189. COLE,M. (1966), Microbial synthesis of penicillins: Part 1, Process Biochem. 1, 334-338. COQUE, J. R., J. LIRAS, LAIZ,L., MARTIN, J. F. P., (1991a), A gene encoding lysine 6-aminotransferase, which forms the p-lactam precursor aaminoadipic acid, is located in the cluster of cephamycin biosynthetic genes in Nocardia lactamdurans, J. Bacteriol. 113, 6258-6264. COQUE, J. R., MARTIN, F., CALZADA, G., J. J. J. LIRAS, P. (1991b), The cephamycin biosynthetic genes pcbA B, encoding a large multidomain peptide synthetase, and pcbC of Nocardia lac-
269
tamdurans are clustered together in an organization different from the same genes in Acremonium chrysogenum and Penicillium chrysogenum, Mol. Microbiol. 5, 1125-1133. COQUE,J. J. R., LIRAS, MARTIN, F. (1993), P., J. Genes for a plactamase, a penicillin-binding protein and a transmembrane protein are clustered with the cephamycin biosynthetic genes in Nocardia lactamdurans, EMBO J. 12,631- 639. COQUE, J. R., LLARENA, J. P., ENGUITA, J., J. F. F. DE LA FUENTE, L., MART~N, F., LIRAS, J. J. P. (1995a), Characterization of the cmcH genes of Nocardia lactamdurans and Streptomyces clavuligerus encoding a functional 3 -hydroxymethylcephem 0-carbamoyltransferase for cephamycin biosynthesis, Gene 162,21-27. COQUE, J. R., ENGUITA, J., MART~N,F., LIJ. F. J. RAS, P. (1995b), A two-protein component 7acephem-methoxylase encoded by two genes of the cephamycin C cluster converts cephalosporin C to 7-methoxycephalosporin C, J. Bacteriol. 177 (8), 2230-2235. CRAWFORD, STEPAN, M., MCADA,P. C., L., A. J. M. V. RAMBOSEK, A., CONDER, J., VINCI, A., C. REEVES, D. (1995), Production of cephalosporin intermediates by feeding adipic acid to recombinant Penicillium chrysogenum strains expressing ring expansion activity, BiolTechnology 13, 58-62. DAVIES, (1994), Inactivation of antibiotics and J. the dissemination of resistance genes, Science 264,375-382. DEMODENA, A., GUTIERREZ, VELASCO, J. S., J., FERNANDEZ, J., FACHINI, A., GALAZZO, F. R. J. L., HUGHES, E., MART~N,F. (1993), The D. J. production of cephalosporin C by Acremonium chrysogenum is improved by the intracellular expression of a bacterial hemoglobin, BiolTechnology 11, 926-929. D~Ez, BARREDO, L., ALVAREZ, CANTOB., J. E., RAL, J. M., VAN SOLINGEN, GROENEN, P., M. A. J. A. M., VEENSTRA, E., MARTIN, F. (1989), Two genes involved in penicillin biosynthesis are linked in a 5.1 kb Sal I fragment in the genome of Penicillium chrysogenum, Mol. Gen. Genet. 218, 572-576. D~Ez,B., GUTIERREZ, BARREDO, L., VAN S., J. SOLINGEN, VAN DER VOORT, L. H. M., P., J. MART~N, F. (1990), The cluster of penicillin biosynthetic genes: Identification and characterization of the pcbAB gene encoding the a-aminoadipyl-cysteinyl-valinesynthetase and linkage to the pcbC and penDE genes, J. Biol. Chem. 265, 16358-16365. DOWSON, G., HUTCHISON, BRANNIGAN, C. A., J. A., GEORGE, C., HANSMAN, LINARES, R. D., J., TOMASZ, MAYNARD A., SMITH, SPRATT, J., B.
270
induced death, in: 50 Years of Penicillin ApplicaG. (1989), Horizontal transfer of penicillin bindH., tion: History and Trends (KLEINKAUF, VON ing protein genes in penicillin-resistant clinical DOHREN, Eds.), Technische Universitat BerH., isolates of Streptococcus pneumoniae, Proc. Natl. lin, Germany (printed by Public Ltd., Czech ReAcad. Sci. USA 86,8842-8846. public), pp. 353-363. ESPESO, A., PENALVA, A. (1992), Carbon caE. M. G.-A., HOSKEN, WILN., tabolite repression can account for the temporal GOULD,S. J., KELLER, KINSON, SUBRAMANI,(1989), A conserved J., S. pattern of expression of a penicillin biosynthetic tripeptide sorts proteins to peroxisomes, J. Cell gene in Aspergillus nidulans, Mol. Microbiol. 6 Biol. 108, 1655-1664. ( l l ) , 1457-1465. E., ESPESO, A., TILBURN, ARST,H. N., JR., PEN- GUTIBRREZ,S., D~Ez, B., MONTENEGRO, E. J., J. MART~N, F. (1991), Characterization of the ALVA,M. A. (1993), pH regulation is a major determinant in expression of a fungal biosynCephalosporium acremonium pcbAB gene encoding a-aminoadipyl-L-cysteinykbvahe synthetic gene, EMBO J. 12, 3947-3956. FAWCETT, A., ABRAHAM, P. (1976), & ( ~ - a - thetase, a large multidomain synthetase: linkage P. E. to the pcbC gene as a cluster of early cephaloAminoadipy1)cysteinylvaline synthetase, Methsporin biosynthetic genes and evidence of multiods Enzymol. 43,471-473. ple functional domain, J. Bacteriol. 173, 2354FAWCETT, A,, USHER,J. J., ABRAHAM, P. P. E. 2365. (1975), Behavior of tritium-labelled isopenicillin S., J., F. N and 6-aminopenicillanic acid acyltransferase GUTIERREZ, VELASCO, FERNANDEZ, J., MARTfN, J. F. (1992), The cefF gene of Cephafrom Penicillium chrysogenum, Biochem. J. 151, losporium acremonium is linked to the cefEF 741-746. gene and encodes a deacetylcephalosporin C FENG,B., FRIEDLIN, MARZLUF, A. (1994), E., G. acetyltransferase closely related to homoserine A reporter gene analysis of penicillin biosynthe0-acetyltransferase, J. Bacteriol. 174, 3056sis gene expression in Penicillium chrysogenum 3064. and its regulation by nitrogen and glucose cataG. bolite repression, App. Environ. Mirobiol. 60 HAAS,H., MARZLUF, A. (1995), NRE, the major nitrogen regulatory protein of Penicillium (12), 4 4 3 2 4 3 9 . chrysogenum, binds specifically to elements in FENG,B., FRIEDLIN, MARZLUF, A. (1995), E., G. the intergenic promoter regions of nitrate assimNuclear DNA-binding proteins which recognize ilation and penicillin biosynthetic gene clusters, the intergenic control region of penicillin biosynCurr. Genet. 28, 177-183. thetic genes, Curr. Genet. 27, 351-358. G., FIERRO, MONTENEGRO, MARCOS,A. T., HAAS,H., BAUER,B., REDL, B., STOFFLER, F., E., MARZLUF, A. (1995), Molecular cloning and G. MART~N, F. (1994), Chromosomal location of J. anlysis of nre, the major nitrogen regulatory the plactam biosynthetic genes in fungal overgene of P. chrysogenum, Curr. Genet. 27, 150producing strain, presented at 7th Int. Symp. Ge158. net. Ind. Microorg., Montreal, Quebec, Canada, HARTMAN, J., TOMASZ, (1984), Low-affinity B. A. June 26-July 1, 1994. penicillin-binding protein associated with P-lacFIERRO, BARREDO, L., D~Ez,B., GUTIBRF., J. tam resistance in Staphylococcus aureus, J. BacREZ, S., FERNANDEZ, J., MART~N, F. F. J. teriol. 158, 513-516. (1995), The penicillin cluster is amplified in tandem repeats linked by conserved hexanucleotide HEINEMANN, SPRAGUE, F. (1989), BacteJ. A., G. sequences, Proc. Natl. Acad. Sci. USA 92, 6200rial conjugative plasmids mobilize DNA transfer 6204. 4, between bacteria and yeast, Nature 3 0 205209. FLEMING, (1929), On the antibacterial action of A. cultures of a Penicillium, with special reference HINNEN, HICKS, B., FINK, R. (1978), TransA., J. J. to their use in the isolation of B. injluenzae, Br. formation of yeast, Proc. Natl. Acad. Sci. USA J. Exp. Pathol. 10, 226-236. 75,1929-1933. GARCIA-DOMINGUEZ, LIRAS, MARTIN, M., P., J. HONLINGER, KUBICEK, P. (1989), RegulaC., C. F. (1991), Cloning and characterization of the tion of &(L-a-aminoadipy1)-L-cysteinyl-D-valine isopenicillin N synthase gene of Streptomyces and isopenicillin N biosynthesis in Penicillium griseus NRRL 3851 and studies of expression chrysogenum by the a-aminoadipate pool size, and complementation of the cephamycin pathFEMS Microbiol. Lett. 65, 71-76. way in Streptomyces clavuligerus, Antimicrob. HONLINGER, HAMPEL, A., ROHR,M., KuC., W. Agents Chemother. 35,4452. BICEK, P. (1988), Differential effects of generC. GIESBRECHT, FRANZ, KRUGER, LABISP., M., D., al amino acid control of lysine biosynthesis on CHINSKI, WECKE, (1993), Bacteriolysis of H., J. penicillin formation in strains of Penicillium Staphylococci is only a side-effect of penicillinchrysogenum, J. Antibiot. 41, 255-257.
I3 References
HORI,K., YAMAMOTO, TOKITA, SAITO, Y., K., F., KUROTSU, KANDA, OKAMURA, FuT., M., K., RUYAMA, J., SAITO, (1991), The nucleotideY. sequence for a proline-activating domain of gramicidin-S-synthetase-2 gene from Bacillus brevis, J. Biochem. 110, 111-119. J., N., S. HOSKINS, ~'CALLAGHAN,QUEENER, W., CANTWELL, A., WOOD, J. S., CHEN,V. J., C. P. SKATRUD, L. (1990), Gene disruption of the pcbAB gene encoding ACV synthetase in Cephalosporium acremonium, Curr. Genet. 18, 523-530. HUFFMAN, W., GESELLSCHEN, D., TURNER, G. P. R. H. J. R., ROTHENBERGER,B., OSBORNE, E., MILLER, D., CHAPMAN, L., QUEENER, F. J. S. W. (1992), Substrate specificity of isopenicillin N synthetase, J. Med. Chem. 35(10), 1897-1914. INGOLIA, D., QUEENER, W. (1989), PLactam T. S. biosynthetic genes, Med. Res. Rev. 9 (2), 245264. JENSEN, E. (1985), Biosynthesis of cephalospoS. rins, Crit. Rev. Biotechnol. 3, 277-301. JENSEN, E., WESTLAKE, W. S., WOLFE,S. S. D. (1988), Production of the penicillin precursor 6 (L-a-aminoadipyl)-L-cysteinyl-D-valine(ACV) by cell-free extracts from Streptomyces clavuligerus, FEMS Microbiol. Lett. 49, 213-218. JENSEN, E., WONG,A., ROLLINS, J., WESTS. M. LAKE, W. S. (1990), Purification and partial D. characterization of S(L-a-aminoadipy1)-L-cysteinyl-D-Valine synthetase from Streptomyces clavuligerus, J. Bacteriol. 172, 7269-7271. KALLOW, W., VON DOHREN,H., KENNEDY, J., TURNER, (1994), Structure and functional G. study of ACV synthetase, 7th Int. Congr. Bacterial, Appl. Microbiol. Mycol. Prague, Czech Republic, p. 426. B. D., E. KERN, A., HENDLIN, INAMINE, (1980), LLysine-E-aminotransferaseinvolved in cephamycin biosynthesis, in Streptomyces lactamdurans, Antimicrob. Agents Chemother. 17,679-685. KIMURA, H., MIYANOSHITA, SUMINO, H., Y. (1990), DNA and applications thereof, Jpn. Patent Application No. 2-3762. KINASHIH., DOI, M., NIMI,0. (1995), Isolation of large linear plasmids from p-lactam producing actinomycete strains, Biotechnol. Lett. 17(3), 243-246. KIRKPATRICK, R., DOOLIN, E., GODFREY, J. L. L. W. (1973), Lysine biosynthesis in Streptomyces lipmannii: implications in antibiotic biosynthesis, Antimicrob. Agents Chemother. 4, 542-550. KLEINKAUF, VON DOHREN, (1996), A nonH., H. ribosomal system of peptide biosynthesis, Eur. J. Biochem. 236,335-351. KOVACEVIC, MILLER, R. (1991), Cloning and S., J. sequencing of the Plactam hydroxylase gene
271
(cefEF)from Streptomyces clavuligerus: gene duplication may have led to separate hydroxylase and expandase activities in the actinomycetes, J. Bacteriol. 173, 398-400, S., B. M. KOVACEVIC, WEIGEL, J., TOBIN, B., INGOLIA, T. D., MILLER,J. R. (1989), Cloning, characterization, and expression in Escherichia coli of the Streptomyces clavuligerus gene encoding deacetoxycephalosporin C synthetase, J. Bacteriol. 171, 754-760. KOVACEVIC, TOBIN,M. B., MILLER,J. R. S., (1990), The p-lactam biosynthesis genes for isopenicillin N epimerase and deacetoxycephalosporin C synthetase are expressed from a single transcript in Streptomyces clavuligerus, J. Bacteriol. 172, 3952-3958. KRAUSE, MARAHIEL, A. (1988), OrganizaM., M. tion of the biosynthesis genes for the peptide antibiotic gramicidin S, J. Bacteriol. 170, 46694674. KRAUSE, MARAHIEL, A., VON DOHREN, M., M. H. H., KLEINKAUF, (1985), Molecular cloning of an ornithine-activating fragment of the gramicidin-S-synthetase-2-genefrom Bacillus brevis and its expression in Escherichia coli, J. Bacteriol. 162, 1120-1125. KRIAUCIUNAS, FROLIK, A., HASSELL, C., A., C. T. SKATRUD, L., JOHNSON, G., HOLBROOK, P. M. V. N. L., CHEN, J. (1991), The functional role of cysteines in isopenicillin N synthase, J. Biol. Chem. 266, 11779-11 788. KUMAR,C. V., COQUE,J.-J. R., MART~N, F. J. (1994), Efficient transformation of the cephamycin C producer Nocardia lactamdurans and development of shuttle and promoter-probe cloning vectors, Appl. Environm. Microbiol. 60(11), 4068-4093. KUROTSO, HORI,K., SAITO, (1991), CharacT., Y. terization and location of the L-proline activating fragment from the multifunctional gramicid i n 4 synthetase-2, J. Biochem. 109, 763-769. KURYLOWICZ, KURZATKOWSKI, KURZATW., W., KOWSKI, (1987), Biosynthesis of benzylpenicilJ. lin by Penicillium chrysogenum and its Golgi apparatus, Arch. Immunol. Ther. Exp. 35, 699724. LEDERBERG, (1957), Mechanism of action of J. penicillin, J. Bacteriol. 73, 144. LENDENFELD, GHALI, WOLSCHEK, KuT., D., M., BICEK-PRANZ, M., KUBICEK, P. (1993), E. C. Subcellular compartmentation o penicillin biof synthesis in Penicillium chrysogenum: the amino acid precursors are derived from the vacuole, J. Biol. Chem. 268,665471. B. Y., LESKIW, K., AHARONOWITZ, MEVARECH, L. D. M., WOLFE,S., VINING, C., WESTLAKE, W. S., JENSEN, E. (1988), Cloning and nucleoS.
272
6 Advances in the Molecular Genetics of p-Lactam Antibiotic Biosynthesis cephalosporin antibiotic biosynthesis, in: Molecular Industrial Mycology: Systems and ApplicaS tions for Filamentous Fungi (LEONG, . A., BERKA, R. M., Eds.), pp. 149-196. New York Marcel Dekker. MATHISON, SOLIDAY, STEPAN,T., ALL., C., DRICH, RAMBOSEK, (1993), Cloning, charT., J. acterization, and use in strain improvement of the C. acremonium gene cefG encoding acetyl transferase, Curr. Genet. 23, 33-41. MATSUHASHI, SONG,M. D., ISHINO, WAM., F., CHI, M., Dot, M., INOUE,M., UBUKATA, K., YAMASHITA, KONNO, (1986), Molecular N., M. cloning of the gene of a penicillin-binding protein supposed to cause high resistance to p-lactam antibiotics in S. aureus, J. Bacteriol. 167, 975-980. MCDERMOTI-, ROGERS,D. E. (1982), Social W., ramifications of control of microbial disease, Johns Hopkins Med. J. 151, 302-312. S., MENNE, WALZ,M., KUCK, U. (1994), Expression studies with the bidirectional pcbAB-pcbC promoter region from Acremonium chrysogenum using reporter gene fusions, Appl. Microbiol. Biotechnol. 42, 57-66. MILLER, R., INGOLIA, D. (1989), Cloning pJ. T. lactam genes from Streptomyces spp. and fungi, in: Genetics and Molecular Biology of Industrial Microorganisms (HERSHBERGER, C . L., S. G., QUEENER, W., HEGEMAN, Eds.), pp. 246255. Washington, D C American Society for Microbiology. J. MONTENEGRO, BARREDO, L., G U T I ~ R R E Z , E., S., DLEz, B., ALVAREZ,E., MARTIN,J. F. (1990), Cloning, characterization of the acylCoA:6-aminopenicillianic acid acyltransferase gene of Aspergillus nidulans and linkage to the isopenicillin N synthase gene, Mol. Gen. Genet. 221,322-330. MONTENEGRO, FIERRO, FERNANDEZ, J., E., F., F. GUTIERREZ, MARTIN, J. F. (1992), Resolus., tion of chromosomes 111 and VI of Aspergillus nidulans by pulse-field gel electrophoresis shows that the penicillin biosynthetic pathway genes pcbAB, pcbC, andpenDE are clustered on chromosome VI (3.0 Megabases), J. Bacteriol. 174, 7063-7067. MULLER, H., VAN DER KRIFT,T. P., KROUWW. ER, A. J. J., WOSTEN, H. A. B., VAN DER VOORT,L. H. M., SMAAL, B., VERKLEIJ, E. A. J. (1991), Localization of the pathway of the penicillin biosynthesis in Penicillium chrysogenum, EMBO J. la, 489-495. MULLER, H., BOVENBERG, A. L., GROOTW. R. HUIS,M. H., KATTEVILDER, SMAAL, B., F., E. VAN DER VOORT, L. H. M., VERKLEIJ, A. J. (1992), Involvement of microbodies in penicillin
tide sequence of the IPNS gene from Streptomyces clavuligerus, Gene 62, 87-196. LODER,P. B., ABRAHAM, P. (1971a), Isolation E. and nature of intracellular peptides from a cephalosporin C-producing Cephalosporium sp., Biochem. J. 123,471-476. LODER,P. B., ABRAHAM, P. (1971b), BiosynE. thesis of peptides containing a-aminoadipic acid and cysteine in extracts of a Cephalosporium sp., Biochem. J. 123,477-482. LUENGO, M., IRISO,J. L., L6PEZ-NIET0, M. J. J. (1986), Direct enzymatic synthesis of natural penicillins using phenylacetyl-CoA:6-APA phenylacetyl transferase of Penicillium chrysogenum: minimal and maximal side chain requirements, J. Antibiot. 39, 1754-1759. MACCABE, P., VAN LIEMPT, PALISSA, A. H., H., S. M. UNKLES, E., RIACH, B. R., PFEIFER,E., VON DOHREN, H., KINGHORN, R. (1991), S J. (L-a-Aminoadipyl)-L-cysteinyl-D-valine synthetase from Aspergillus nidulans, J. Biol. Chem. 266,12646-12654. MADDURI, K., STUTTARD, C., VINING,L. C. (1989). Lysine catabolism in Streptomyces spp. is primarily through cadverine: p-lactam producers also make a-aminoadipate, J. Bacteriol. 171, 299-302. MADDURI, K., STUTTARD, C., VINING,L. C. (1991), Cloning and location of a gene governing lysine E-aminotransferase, an enzyme initiating p-lactam biosynthesis in Streptomyces sp., J. Bacteriol. 173, 985-988. MAEDA, K., LUENGO,J. M., FERRERO,O., WOLFE,S., LEBEDEV, Y., FANG,A., DEM. MAIN, A. L. (1995), The substrate specificity of deacetoxycephalosporin C synthase (expandase) of Streptomyces clavuligerus is extremely narrow, Enzyme Microb. Technol. 17, 231-234. L. D. MALMBERG, H., Hu, W.-S., SHERMAN, H. (1993), Precursor flux control through targeted chromosomal insertion of the lysine &-aminotransferase (LAT) gene in cephamycin C biosynthesis, J. Bacteriol. 175, 6916-6924. D. MALMBERG, H., Hu, W. S., SHERMAN, H. L. (1995), Effects of enhanced lysine E-aminotransferase activity on cephamycin biosynthesis in Streptomyces clavuligerus, Appl. Microbiol. Biotechnol. 44,198-205. MALPARDITA, HOPWOOD, A. (1984), MolecF., D. ular cloning of the whole biosynthetic pathway of a Streptomyces antibiotic and its expression in a heterologous host, Nature 309,462-464. MARTIN, J. F., GUTI~RREZ, S. (1995), Genes for plactam antibiotic biosynthesis, Antonie van Leeuwenhoek 67, 181-200. MARTIN, J. F., INGOLIA, D., QUEENER, W. T. s. (1991), Molecular genetics of penicillin and
I3 References
213
biosynthesis, Biochim. Biophys. Acta 1116, 210ROACH,P. L., CLIFTON, J., FULOP, V., HARLOS, I. 213. K., BARTON,G. J., HAJDU,J., ANDERSSON, I., O'SULLIVAN, SYKES, B. (1986), PLactam anJ., R. SCHOFIELD, J., BALDWIN, E. (1995), CrysC. J. tal structure of isopenicillin N synthase is the tibiotics, in: Biotechnology 1st. Edn., Vol. 4 first from a new structural family of enzymes, (REHM, H.-J., REED, G., Eds.), pp. 247-273. Nature 375, 700-704. Weinheim: VCH. O'SULLIVAN, BLEANEY,R. C., HUDDLESTON, ROLAND,I., FROYSHOV, LALAND, (1975), J., O., G. J. A., ABRAHAM, P. (1979), Incorporation of E. On the presence of pantothentic acid in the 3H from &(~-(~-amino(4,5-3H)adipyl)-~-cystei- three complementary enzymes of bacitracin synthetase, FEBS Lett. 60,305-308. nyl-D-(4,4-3H)valine into isopenicillin N, Biochem. J. 184,421-426. SAITO,F., HORI, K., KANDA,M., KUROTSU,T., PANG,C. P., CHAKRAVARTI, ADLINGTON, B., R. SAITO, (1995), Entire nucleotide sequence for Y. M., TING, H., WHITE,R. L., JAYATILAKE, H. G. Bacillus brevis Nagano Grs2 gene encoding gramicidin S synthetase 2: a multifunctional pepS., BALDWIN, E., ABRAHAM, P. (1984), PuJ. E. rification of isopenicillin N synthetase, Biochem. tide synthetase, J. Biochem. 116, 357-367. J. 222, 789-795. SANCHEZ, LOZANO,M., RUBIO,V., PENALVA, F., PASTEUR, JOUBERT, (1877), Charbon et sepL., J. M. A. (1987), Transformation of P. chrysogeticemie, C. R. Acad. Sci. Paris 85, 101-115. num, Gene 51,97-102. PENALVA, A., TOURINO, PATINO,C., SAN- SAMSON, . E., BELAGAJE, BLANKENSHIP, M. A., S R., D. CHEZ, F., FERNANDEZ SOUSA, J. M., RUBIO,V. T., CHAPMAN, L., PERRY,D., SKATRUD, J. P. (1985), Studies of transformation of CephalospoL., VAN FRANK, M., ABRAHAM, E. P., R. rium acremonium, in: Molecular Genetics of FilBALDWIN, E., QUEENER, W., INGOLIA, J. S. T. amentous Fungi (TIMBERKLAKE, E., Ed.), W. D. (1985), Isolation, sequence determination pp. 59-68. New York: Alan Liss. and expression in E. coli of the isopenicillin N synthetase gene from Cephalosporium acremonP~REZ-ESTEBAN, OREJAS, B., M., G6MEZ-PARium, Nature (London) 318, 191-194. DO, E., PENALVA, (1993), Molecular characA. terization of a fungal secondary metabolism pro- SAMSON,S. M., DOTZLAF,J. E., SLISZ, M. L., G. R. BECKER, W., VAN FRANK, M., VEAL,L. moter: transcription of the Aspergillus nidulans isopenicillin N synthetase gene is modulated by E., YEH, W. K., MILLER, R., QUEENER, W., J. S. INGOLIA, D. (1987), Cloning and expression T. upstream negative elements, Mol. Microbiol. 9 of the fungal expandaselhydroxylase gene in(4), 881-895. volved in cephalosporin biosynthesis, BiolTechPETRICH, K., JENSEN,S. E. (1994), TranscripA. nology 5,1207-1214. tional regulation of the genes involved in the Y., J. P. early steps of the cephamycin biosynthetic path- SAWADA, BALDWIN, E., SINGH, D., SOLOA. MON,N. A., DEMAIN, L. (1980), Cell-free cyway in Streptomyces clavuligerus, presented at 7th Int. Symp. Genet. Ind. Microorg. Montreal, clization of &(L-cY-aminoadipyl)-L-cysteinyl-Dvaline to isopenicillin N, Antimicrob. Agents Canada, June 26-July 1, 1994. Chemother. 18,456-470. PIEPER,R., HAESE, SCHRIDER, A., W., ZOCHER, R. (1995), Arrangement of catalytic sites in the SCHLUMBOHM, STEIN,T., ULLRICH, VAW., C., multifunctional enzyme enniatin synthetase, TER, J., KRAUSE, M., MARAHIEL,M. A., Eur. J. Biochem. 230, 119-126. KRUFI-,V., WI~MANN-LIEBOLD, B. (1991), An active serine is involved in covalent substrate QUEENER, W., NEUSS,N. (1982), The biosyntheS. amino acid binding at each reaction center of sis of p-lactam antibiotics, in: The Chemistry and gramicidin S synthetase, J. Biol. Chem. 266, Biology of p-Lactam Antibiotics (MORIN,R. B., 23 135-23 141. MORGAN,M., Eds.), pp. 1-81. London: AcaSCHWECKE, AHARONOWITZ, PALISSA, T., Y., H., demic Press. VON DOHREN, H., KLEINKAUF, H., VAN QUEENER, W., INGOLIA, D., SKATRUD, L., S. T. P. LIEMFT,H. (1992a), Enzymatic characterization J. CHAPMAN, L., KASTER,K. R. (1985), A sysof the multifunctional enzyme G(L-a-aminoaditem for genetic transformation of Cephalospopy1)-L-cysteinyl-D-valine synthetase from Streprium acremonium, in: Microbiology I985 tomyces clavuligerus, Eur. J. Biochem. 205, 687(LIEVE, L., Ed.), pp. 468-472. Washington, DC: 694. American Society for Microbiology. T., M., S., RAM6N, D., CARRAMOLINO, L., PATINO, c., SAN- SCHWECKE, TOBIN, KOVACEVIC, MILLER, J. R., SKATRUD,P. L., JENSEN, S. E. CHEZ,F., PENALVA, A. (1987), Cloning and M. (1992b), The A module of ACV synthetase: excharacterization of the isopenicillin N synthetase pression, renaturation and functional characterigene mediating the formation of the plactam zation, in: Industrial Microorganisms: Basic and ring in Aspergillus nidulans, Gene 57, 71-181.
274
6 Advances in the Molecular Genetics of
p-.Lactam Antibiotic Biosynthesis
P. S. Applied Molecular Genetics (BALTZ, R. H., SKATRUD, L., QUEENER, W., CARR,L. G., FISHER, D. L. (1987b), Efficient integrative G. P. HEGEMAN, E., SKATRUD, L., Eds.), p. 291. transformation of C. acremonium, Curr. Genet. Washington D C American Society for Micro12,337-348. biology. SKATRUD, L., TIETZ, A. J., INGOLIS, D., P. T. SENO, T., BALTZ, H. (1989), Structural organE. R. CANTWELL, A., FISHER, L., CHAPMAN, C. D. J. ization and regulation of antibiotic synthesis and L., QUEENER, W. (1989), Use of recombinant S. resistance genes in Actinomycetes, in: RegulaDNA to improve production of cephalosporin C tion of Secondary Metabolism in Actinomycetes in C. acremonium, Bio/Technology 7, 411-485. (SHAPIRO, Ed.), pp. 1-48. Boca Raton, FL: S., SMITH, J., BULL,J. H., EDWARDS, TURNER, D. J., CRC Press. G. (1989), Amplification of the isopenicillin N SHIAU, Y., BALDWIN, E., BYFORD, F., SoC. J. M. synthetase gene in a strain of Penicillium chrysoBEY, J., SCHOFIELD, J. (1995), &(L-cr-Amw. C. genum producing high levels of penicillin, Mol. inoadipyl)-L-cysteinyl-D-vahe synthetase: the Gen. Genet. 216,492-497. order of peptide bond formation and timing of SMITH, J., BURNHAM, K. R., BULL,J. H., D. M. the epimerisation reaction, FEBS Lett. 358, 97P., HODGSON, E., WARD, J. M., BROWNE, J. 100. BROWN, BARTON, EARL, J., TURNER, J., B., A. SHIFFMAN, MEVARECH, JENSEN,S. E., D., M., G. (1990a), PLactam antibiotic biosynthetic Y. COHEN, AHARONOWITZ, (1988), Cloning G., genes have been conserved in prokaryotes and and comparative sequence analysis of the gene eukaryotes, EMBO J. 9,741-747. coding for isopenicillin N synthase in StreptoSMITH, J., BURNHAM, K. R., EDWARDS, D. M. J., myces, Mol. Gen. Genet. 214, 562-569. EARL,A. J., TURNER, (1990b), Cloning and G. SHIFFMAN, COHEN,G., AHARONOWITZ, D., Y., H., PALISSA, VON DOHREN,H., KLEINKAUF, expression of the penicillin biosynthetic gene cluster from Penicillium chrysogenum, BiolTechH., MEVARECH, (1990), Nucleotide sequence M. nology 8,39-41. of the isopenicillin N synthase gene (pcbC) of SMITH, J., EARL, A. J., TURNER, (1990c), D. G. the gram-negative Flavobacterium sp. SC 12,154, The multifunctional peptide synthetase performNucleic Acids Res. 18, 660. ing the first step in penicillin biosynthesis in PenSHOCKMAN, D., DANEO-MOORE, McDowG. L., icillium chrysogenum is a 421 037 Dalton protein ELL,T. D., WONG,W. (1982), The relationship similar to Bacillus brevis peptide antibiotic synbetween inhibition of cell wall synthesis and bacthetases, EMBO J. 9, 2743-2750. terial lethality, in: The Chemistry and Biology of B. PLactam Antibiotics, Vol. 3 (MORIN,R. B., SPRATT, G. (1989), Resistance to p-lactam antibiotics mediated by alterations of penicillinGORMAN, Eds.), pp. 303-338. London: AcaM., binding proteins, in: Handbook of Experimental demic Press. L. Pharmacology, Vol. 91 (BRYAN, E., Ed.), pp. W., LEVIN,H. L., SIKORSKI, S., MICHAUD, R. 77-100. Berlin: Springer-Verlag. BOEKE,J. D., HIETER,P. (1990), Trans-king- STACHELHAUS, SCHNEIDER, MARAHIEL, T., A., dom promiscuity, Nature 345,581-582. M. A. (1995), Rational design of peptide antiSKATRUD, L. (1991), Molecular biology of the P. biotics by targeted replacement of bacterial and plactam producing fungi, in: More Gene Manifungal domains, Science 269, 69-72. pulations in Fungi (BENNET, W., LASURE, J. L. THENBERGH, LITZKA, BRAKHAGE, A. K., O., A. L., Eds.), pp. 364-395. New York: Academic (1996), Identification of a major cis-acting DNA Press. element controlling the bidirectionally tranSKATRUD, L. (1992), Genetic engineering of p P. scribed penicillin biosynthesis genes acvA lactam antibiotic biosynthetic pathways in fil(pcbAB) and ipnA (pcbC) of Aspergillus niduamentous fungi, Trends BioTechnol. 10(9), 324lans, J. Bacteriol. 178(13), 3908-3916. 329. TILBURN, SARKAR, WIDDICK, A., ESPEJ., S., D. SKATRUD, L., QUEENER, W. (1989), An elecP. S. SA, E. A., OREJAS, M., MUNGROO, PENALJ., trophoretic molecular karyotype for an indusVA,M. A., ARST,H. N., JR. (1995), The Aspertrial strain of Cephalosporium acremonium, gillus PacC zinc finger transcription factor meGene 78,331-338. diates regulation of both acidic- and alkaline-expressed genes by ambient pH, EMBO J. 14,779SKATRUD, L., FISHER,D. L., CHAPMAN, L., P. J. 790. S. CANTWELL, A., QUEENER, W. (1987a), C. J. Strain improvement studies in P. chrysogenum TIPPER,D. J., STROMINGER,L. (1965), Mechanism of action of penicillins: a proposal based on using the cloned P. chrysogenum isopenicillin N their structural similarity to acyl-D-alanyl-D-alasynthetase gene and the amdS gene of Aspergilnine, Proc. Natl. Acad. Sci. USA 54, 1133-1141. lus nidulans, SIM News 37(4), 77.
13 References
275
valine synthetase from Aspergillus nidulans, J. TOBIN, B. (1994), Genetic engineering of the M. Biol. Chem. 264,3680-3684. acyl-coenzyme A:isopenicillin N acyltransferase from Penicillium chrysogenum, Thesis, Univer- VAN SINDREN, GALLI,G., COSMINA, DED., F., sity of Oxford. FERRA, WITTHOFF, VENEMA, GRANS., S. G., DI,G. (1993), Characterization of the srfA locus TOBIN, B., FLEMING, D., SKATRUD, L., M. M. P. J. of Bacillus subtilis: only the valine-activating doMILLER, R. (1990), Molecular characterizamain of srfA is involved in the establishment of tion of the acyl-coenzyme A:isopenicillin N acylgenetic competence, Mol. Microbiol. 8, 833transferase gene (penDE)from Penicillium chry841. sogenum and Aspergillus nidulans and activity of recombinant enzyme in Escherichia coli, J. Bac- VELASCO, GUTIERREZ, FERNANDEZ, J., J., S., F. teriol. 172, 5908-5914. MARCOS,A. T., ARENOS,C., MART~N, F. J. (1994), Exogenous methionine increases levels TOBIN, B., KOVACEVIC, MADDURI, HosM. S., K., of mRNAs transcribed from pcbAB, pcbC, and KINS, J., SKATRUD, L., VINING, C., STUDP. L. cefEF genes, encoding enzymes of the cephaloDARD, MILLER, R. (1991), Localization of C., J. the lysine E-aminotransferase (lat) and S(L-(Y- sporin biosynthetic pathway, in Acremonium chrysogenum, J. Bacteriol. 176(4), 985-991. aminoadipyl)-L-cysteinyl-D-valine synthetase (pcbAB) genes from Streptomyces clavuligerus VICHITSOONTHONKUL, T., CHU, Y. W., SODHI,H. and production of lysine e-aminotransferase acG. S., SAUNDERS, (1994), p-Lactam antibiotics tivity in E. coli, J. Bacteriol. 173, 6223-6229. produced by genetically engineered filamentous fungi, in: Recombinant Microbes for Industrial J. TOBIN,M. B., BALDWIN, E., COLE, S. C. J., Y., MILLER, R., SKATRUD, L., SUTHERLAND, and Agricultural Applications (MUROOKA, J. P. IMANAKA, Eds.), pp. 119-135. New York: E., J. D. (1993), The requirement for subunit interMarcel Dekker. action in the production of Penicillium chrysogenum acyl-coenzyme A:isopenicillin N acyltrans- VINING, C., SHAPIRO, MADDURI, STUTL. S., K., ferase in Escherichia coli, Gene 132, 199-206. TARD, (1990), Biosynthesis and control of pC. lactam antibiotics: the early steps in the classiTOBIN,M. B., COLE, S. C. J., KOVACEVIC, S., cal tripeptide pathway, Biotechnol. Adv. 8,159MILLER, R., BALDWIN, E., SUTHERLAND, J. J. J. 183. D. (1994), Acyl-coenzyme A:isopenicillin N acyltransferase from Penicillium chrysogenum: WARD,J. M., HODGSON, E. (1993), The biosynJ. effect of amino acid substitutions at Ser, thetic genes for clavulanic acid and cephamycin Ser and Ser3won proenzyme cleavage and acproduction occur as a super-cluster in three tivity, FEMS Microbiol. Lett. 121, 9-46. Streptomyces, FEMS Microbiol. Lett. 110, 239242. TOBIN, B., MILLER, R., BALDWIN, E., SuM. J. J. THERLAND, J. D. (1995), Amino acid substituWAXMAN, J., STROMINGER, L. (1982), p-LacD. J. tions in the cleavage site of acyl-coenzymeA:isotam antibiotics: biochemical modes of action, in: penicillin N acyltransferase from Penicillium The Chemistry and Biology of p-Lactam Antichrysogenum: effect on proenzyme cleavage and biotics, Vol. 3 (MORIN,R. B., GORMAN, M., activity, Gene 162, 29-35. Eds.), pp. 209-285. London: Academic Press. TURGAY,K., KRAUSE,M., MARAHIEL, A. WEBER,G., SCH~RGENDORFER,SCHNEIDERM. K., (1992), Four homologous domains in the primaSCHERZER, LEITNER, (1994), The peptide E., E. ry structure of GrsB are related to domains in a synthetase catalyzing cyclosporine production in superfamily of adenylate forming enzymes, Mol. Tolypocladium niveum is encoded by a giant Microbiol. 6, 529-546. 45.8-kilobase open reading frame, Curr. Genet. 26, 120-125. TURNER, (1992), Genes for the biosynthesis of G. plactam compounds in microorganisms, in: Sec- WEIGEL, J., BURGETT, G., CHEN, J., SKAB. S. V. ondary Metabolites: Their Function and EvoluTRUD,P. L., FROLIK, A., QUEENER, W., C. S. tion (CHADWICK, J., WHELAN, Eds.), pp. D. J., T. INGOLIA, D. (1988), Cloning and expression 113-128. Chichester: Wiley and Sons (Ciba in Escherichia coli of isopenicillin N synthetase Foundation Symposium 171). genes from Streptomyces lipmanii and Aspergillus nidulans, J. Bacteriol. 170, 3817-3826. J., J. UNAL,S., HOSKINS, FLOKOWITSCH, E., Wu, P. C. Y. E., PRESTON,D. A., SKATRUD, L. WHITEMAN, A., ABRAHAM, P., BALDWIN, P. E. J. (1992), Detection of methicillin-resistant staphyE., FLEMING, D., SCHOFIELD, J., SUTHERM. C. lococci by using the polymerase chain reaction, LAND, J. D., WILLIS, A. C. (1990), Acyl coenJ. Clin. Microbiol. 30,1685-1691. zyme A:6-aminopenicillanic acid acyltransferase VAN LIEMPT, H., VON DOHREN, KLEINKAUF, from Penicillium chrysogenum and Aspergillus H., nidulans, FEBS Lett. 262, 342-344. H. (1989), G(L-a-Aminoadipyl)-L-cysteinyl-D-
276
WISE,E. M., JR., PARK,J. T. (1965), Penicillin: its basic site of action as an inhibitor of a peptide cross-linking reaction in cell wall mucopeptide synthesis, Proc. Natl. Acad. Sci. USA 54, 75-81. Wu, C. Y. E., HOSKINS, BLASZCZAK, C., J., L. PRESTON, A., SKATRUD, L. (1992), ConD. P. struction of a water-soluble form of penicillinbinding protein 2a from a methicillin-resistant Staphylococcus aureus isolate, A ntimicrob. Agents Chemother. 36,533-539. YEH, W. K., DOTZLAF, E., HUFFMAN, W. J. G. (1994), Biochemical characterization and evolutionary implication of plactam expandaselhydroxylase, expandase and hydroxylase, in: 50 Years of Penicillin Application; History and Trends (KLEINKAUF, VON DOHREN, H., H., Eds.), pp. 208-223. Prague: Public. YELTON,M. M., HAMER,J. E., TIMBERLAKE, W. E. (1984), Transformation of Aspergillus nidufans using a trpC plasmid, Proc. Natl. Acad. Sci. USA 81,1470-1474. YOCUM,R. R., WAXMAN, J., RASMUSSEN, R., D. J.
STROMINGER, L. (1979), Mechanism of peniJ. cillin action: penicillin and substrate bind covalently to the same active site serine in two bacterial D-alanine carboxypeptidases, Proc. Natl. Acad. Sci. USA 76,2730-2734. ZHANG,J. (1991), ACV synthetase in cephalsporin biosynthesis, Thesis, Massachussets Institute of Technology, Cambridge, MA. ZHANG,J., DEMAIN, L. (1990a), Purification A. from Cephalosporium acremonium of the initial enzyme unique to the biosynthesis of penicillins and cephalosporins, Biochem. Biophys. Res. Commun. 137,528-535. ZHANG, DEMAIN, L. (1990b), Purification of J., A. ACV synthetase from Streptomyces clavuligerus, Biotechnol. Lett. 12, 649454. ZHANG,J., DEMAIN, L. (1992), ACV synthetA. ase, Crit. Rev. Biotechnol. 12, 245-260. ZHANG, WOLFE,S., DEMAIN, L. (1992), BioJ., A. chemical studies of ACV synthetase activity from Streptomyces clavuligerus, Biochem. J. 383, 691698.
Biotechnology
7 Peptide Antibiotics
HORST KLEINKAUF HANS VON DOHREN

Berlin, Germany List of Abbreviations 278 1 Introduction 279 2 Structures 280 2.1 Ribosomal Origin 280 2.1.1 Structural Classifications 280 2.1.2 Similar Structures of Ribosomal and Nonribosomal Peptides - Monocyclic Structures 280 2.1.3 D-Amino Acids in Ribosomally Produced Peptides 282 2.1.4 Posttranslational Modifications of Side Chains 282 2.1.5 Multicyclic Peptides 283 2.1.6 Linear Peptides 284 2.1.7 Production of Ribosomally Formed Peptides 284 2.1.8 Similar Open Questions in Gene-Encoded and Enzymatically Formed Peptides 284 2.2 Nonribosomal Origin 285 3 Biotechnology of Peptides 285 3.1 Biosynthesis 285 3.1.1 Biosynthetic Gene Clusters 287 3.1.2 Biosynthetic Modules 287 3.1.3 Biosynthetic Enzyme Systems 293 3.2 Production 293 3.2.1 Fermentation Procedures 293 3.2.2 Structural Alterations 295 3.2.2.1 Family Exploitations 295 3.2.2.2 Biosynthetic Manipulations 297 4 Future Prospects 298 4.1 Peptides of Ribosomal Origin 298 4.2 Peptides of Nonribosomal Origin 299 5 Compilation of Compounds 299 6 References 316
278
List of Abbreviations
Note: If no D- or D-prefix is given, all amino acids are in the L-configuration. l-C1-D-vinylGly 1-chloro-D-vinylglycine 2-amino-3-hydroxy-42a3h4buOA butyl-octanoic acid 2-amino-4-methyl-hex-42a4m4HEA enoic acid 2-amino-3-hydroxy-42a3h4mOA methyl-octanoic acid 2a3h4,8m2NA 2-amino-3-hydroxy-4,8dimethyl-nonanoic acid 2-chloro-~-alanine 2-Cl-~Ala 2-fluoro-~-alanine 2-F-~Ala 2-hydrox y-asparagine 2hAsn 2-hydroxy-butyric acid 2hBu 2-hydroxy-isocaproic acid 2hiCap 2-h ydroxy-phen ylalanine 2hPhe 2-hydroxy-valine 2hVal 2-hydroxy-3-methyl-valeric 2h3mVa acid 2,3-hydroxy-3-methyl2,3h3mP pentanoic acid 2,4-hydroxy-3-methyl2,4h3mP pentanoic acid 2,5-hydroxy-3-methyl2,5h3mP pentanoic acid 3-hydroxy-asparagine 3hAsn 3-hydroxy-cyclohexyl3hCHA alanine 3-h ydroxy-tetradecanoic acid 3-hydroxy-leucine 3hLeu 3-thioproline 3S-Pro 3,4-dehydro-proline 3,4APro 4-hydroxy-proline 4hPro 4-thioproline 4S-Pro 4,5-hydroxy-isoleucine 4,5hIle 6 'hTrp 6 '-hydroxy-tryptophan a-Aad a-aminoadipic acid pAla palanine dehydro-alanine AAla dehydro-aminobutyric acid AAbu aminobutyric acid Abu 2-amino-S-oxo-9,lO-epoxiAeo decanoic acid aIle ah-isoleucine allylGly allyl-glycine amino-octanoic acid AOC
aThr Aze Bmt
allo-threonine azetidine-2-carboxylic acid (4R)-4-[(4E)-Zbutenyl]-4methyl-L-threonine C cysteine linked as thioether Cit citrulline CPG cyclopropyl-glycine cyclodihydroBmt cyclodehydro-Bmt (see Bmt) cyclopropylGly cyclopropyl-glycine deoxyBmt deoxy-Bmt (see Bmt) diaminobutyric acid Dbu formyl-isoleucine fIle formyl-valine Nal f-ornithine fOrn 2-hydroxy-4-methyl-valeric Himv acid hydroxy-isovaleric acid Hiv hydroxy-methyl-pentanoic HmP acid 4-hydroxy-proline hPro 4-h ydroxy-proline HYP lactic acid Lac lysergic acid LYSA meta-fluoro-phenylalanine mFPhe 4-methyl-hydroxy-anthranMha ilic acid methyloxazole (formed by Mtz cyclization of threonyl side chain) N,Zmethyl-aspartic acid N2MeAsp norleucine Nle norvaline Nva ortho-fluoro-pheny lalanine oFPhe O-methyl-serine OMeSer oxazole (formed by cyclioxz zation of seryl side chain) pyro-glutamic acid pGlu phenyl-serine PhSer phosphinothricine PPT quinaldic acid Q quinaldinic acid Qaa quinazol-4-one-3-acetic Qoa acid quinoxaline-2-carboxylic Qxa acid sarcosine Sar modified serine contained Ser* in tetrahydropyridine moiety of thiopeptides thioproline Spro
I Introduction
279
Main applications of peptides still are pharmacological formulations applied in both human and animal health care. Besides antibacterial uses, increasing attention is paid to the introduction of antifungal compounds while immunomodulators have been firmly established. A considerable number of proteinase inhibitors and other enzyme inhibitors are continuously emerging and find various applications. In Sect. 5 a compilation of recently described peptides can be consulted for current screening targets and their results. As will be shown below, antibiotics evolved in the ribosomal or enzymatic nonribosomal system may have similar structures and properties. Similar genetic backgrounds for the manufacturing of complex metabolites can be followed from prokaryotes to lower and highPeptides as a chemically defined group of er eukaryotes, and boundaries between fields metabolites have attracted considerable at- of microbial, plant or animal origin may distention for a variety of reasons. The first anti- appear. This becomes obvious in the screenbiotics discovered - penicillins, tyrocidines, ing of marine organisms where in some inand gramicidins - are peptides, and new pep- stances metabolites isolated from sponges tides emerge from screenings at a steady rate. could be traced to associated microorganisms. However, in addition to the enzymatic path- Such possibly symbiotic relationships may inways of biosynthesis which, e.g., produce po- deed trigger the production of useful metablyketides or terpenoids, there is the additional olites. Likewise, experimental simulations of route of ribosomal peptide synthesis. The microbial invasions lead to the production of number of publications on peptide structures antimicrobials, and this principle has been apreflects that increasing attention is paid to this plied successfully, e.g., in the detection of riroute. About a half of the 100 references on bosomally produced antibiotics of animal oripeptide antibiotics traced from December gin. 1995 until December 1996 deal with peptides The aim of these new approaches is to unof ribosomal origin. Those detected via non- derstand the ecological significance of such antibiotic properties like hormone functions metabolites and to try to trace the concepts of are not even part of this cluster. In recent the evolution of structures. These concepts treatises on commercially important micro- will be beneficial in the design of selection bial products peptides have an increasing screens in biosynthetic approaches utilizing share (VINING STUTTARD1995; STROHL, the natural combinatorial potential. In addiand 1996). In this volume, further chapters deal tion, boundaries between compound strucwith highly significant fields: the established tures and their biosynthetic origins show the plactams (SKATRUD al., Chapter 6), the tendency of opening up due to the unifying et promising antibacterials called dalbaheptides concept of genetic structures. This is particu(LANCINI and CAVALLERI, Chapter 9), and larly evident from the similar genetic backthe ribosomally formed lantibiotics (JACKet ground of various polyketide structures and al., Chapter 8). The currently predominant their combination with peptide forming sysimmunosuppressor cyclosporin is treated in tems, e.g., in the prominent immunomodulaChapter 12 by FLIRIet al. Other compounds tors FK506 and rapamycin (SCHWECKE al., et of peptidic origin are contained in the chapt- 1995). ers on siderophores (WINKELMANN and DRECHSEL,Chapter 5) and on antitumor agents (GRAFEet al., Chapter 14).
dehydrated threonyl side chain linked as thioether tbuAla t-butyl-alanine tbuGly t-butyl-glycine Thz thizole (formed by side chain cyclization of cysteine) vinylGly vinyl-glycine D-prefix for D-configuration, used withoutMe-prefix used for N-methylNme-prefix used for N-methyl-
1 Introduction
280
2 Structures
2.1 Ribosomal Origin
Ribosomally encoded peptide antibiotics in microorganisms have been a topic throughout the history of antibiotic research. The promising field of lanthionine-containing peptides for which the term lantibiotics has been coined is treated in Chapter 8 of this volume by JACK et al. These peptides are used in the food industry. As animal antibiotics of ribosomal origin they have led to very promising fields of research. Defenses against microbial infections - besides the highly specific system of adaptive immunity in vertebrates - were detected in a variety of targets. Thus compounds with antimicrobial activity, mainly peptides, were found in animals, plants, and bacteria. Animal antibiotics are located at surfaces like skin or mucosal surfaces, e.g., epithelial cells, and are components of body fluids like neutrophils, macrophages, and killer cells in mammals and haemocytes in insects. These antibacterials function as a primary defense system which acts faster and simpler and saves energy as well as information compared to the slow and highly specific memory-based 1995). system of adaptive immunity (BOMAN, Fundamental problems are the growth rates of bacteria compared to lymphocytes which differ 50- to 100fold. Focusing on peptides, molecular genetics have opened up research to the complicated network of innate immunity. Current work describes the structure of gene clusters, regulation of the expression of peptide genes, their posttranslational processing and modification, and their fate in the respective environment. Obviously, these approaches are not limited to antimicrobial peptides and apply as well to, e.g., hormones or toxins. It has been pointed out that peptides signaling defense reactions in plants exhibit analogies to defense signalet ing in animals (BERGEY al., 1996). Thus the release of the 18-residue peptide systemin from a 200-amino acid precursor is induced by mechanical wounding of tomato plants. Systemin induces a lipid-based signaling cascade leading to the expression of wound re-
sponse proteins. Signaling functions have now been reported in a variety of cases including the involvement of peptide factors like the 45residue ComS in competence in Bacillus subet et tilis ( D S o u z ~ al., 1994; HAMOEN al., et 1995; MAGNUSONet al., 1994; SOLOMON al., 1995). In this chapter several aspects of mainly antimicrobial peptides are presented. As the et field has been reviewed recently (LEHRER et 1995), al., 1993; BOMAN al., 1994; BOMAN, we focus on biotechnological aspects.
2.1.1 Structural Classifications

There have been several general approaches to group or classify peptides of ribosomal origin. Two groups of linear peptides with or without disulfide linkages is a common scheme, which is further refined into respective subgroups of dominating residues, antibacterial actions, or processed from precursors, or two, three, or four disulfides (BoMAN, 1995). Recently, HANCOCK al. introet duced the group of cationic bactericidal pepet tides (HANCOCK al., 1995) which are defined as polypeptides with less than 100 amino acid residues carrying a net charge of at least +2. It is obvious that there will be various overlaps within such restricted systems as many compounds show several properties, including linear portions, disulfides and antibacterial properties. Their linear parts may be dominated by certain residues or they may include cationic regions. Schemes of this kind will be useful in some respects, but some compounds will be part of several subgroups. As will be shown below, there is a similar situation in the classification of nonribosomally generated peptides.
2.1.2 Similar Structures of Ribosomal and Nonribosomal Peptides - Monocyclic Structures

Many cyclic and branched cyclic peptides of microbial origin containing nonprotein constituents are well known antimicrobials. Cyclic peptides of ribosomal origin were char-
2 Structures
281
acterized from bovine neutrophils (bactenecins (I); ROMEOet al., 1988), and from the skin of bullfrogs (brevinin 1 (11), brevinin 2 (111), Rana brevipode; esculentin (IV), Rana esculenta; MORIKAWA al., 1992; SIMMACO et et al., 1993; ranalexin (V), Rana catesbeiana; CLARK al., 1994). et R4L-C I
\
three identical 198-base pair open reading frames which encode identical 66-amino acid peptides, but differ in their presumably regulatory untranslated 5 '-regions. The amino terminal prepropeptide region is composed of a signal sequence separated by a prohormone processing site from the acidic region preced-
+R+I+V+V
R+C+C+R+I FLPVLAGIAAKVVPALF+ C+ K+ I +T I1
\
1
C+K+K
+
I11
GIMDTLKNLAKTAGKGALQSLLNKAS +C A +V+T \ 1 C+K+K

GIFSKLGRKKIKNLLISGLKNVGKEVGMDVVRTGIDIAG C + K-' I K
+ +
IV FLGGLIKIVPAMI V
+
1
C + E +G
C+ A +V+T
\
1
C+K+K ing the antibiotic peptide. Similar sequences are known from other propeptides (BEVINS and ZASLOFF, 1990). These signal sequences resemble sequences found in amphibian opioid peptides including dermorphins and deltorphins (VIII and IX). These opioid receptor-binding heptapeptides were first isolated from the skin of South American frogs (Phyllomedusa sp.) and have been shown to
+
The cycloheptapeptide moiety appears to be a common structural element of the Rana peptides, carrying positively charged side chains as found in polymyxin (VI), a nonribosomal antibacterial from Bacillus polymyxa. Cyclononapeptide structures as in bactenecins are found, e.g., in nonapeptidolactones like lysobactidkatanosin (VII, Lysobacter sp. and Cytophaga sp.). VI VII
acyl+ Dbu +Thr +DDbu+ D bu+ Dbu DLeu+Thr 1 ThrcDbutDbu DLeu +Leu+PhSer+3hLeu +Leu+ Arg+ Val
I
1
Sert3hAsn +Gly+aThr Variable side chains point to additional functional features, possibly even linked to the processing of precursor structures. CLARKet al. (1994) isolated the gene for ranalexin using ' the 'cDNA approach and detected an acidic amino acid-rich propeptide sequence at the amino terminal end. The ranalexin structutral gene has been found to be contained in contain D-amino acids (ERSPAMER al., et 1989). VIII IX
Tyr-DAla-Phe-Gly-Tyr-Pro-SerNH2 . .
Tyr-DAla-Phe-AsP/G1u-val-valGlyNH2
282 X
A different route of conversion is found in lantibiotics where Ser or Thr residues are iniXI pGlu-Leu-Tyr-Glu-Asn-Lys-Pro- tially reduced to the dehydro derivatives (SKAUGEN al., 1994). These may further et Arg-Arg-Pro-Tyr-Ile-Leu react with Cys side chains to thioethers, but Both types of bioactive peptides are produced isolated dehydro residues are known, e.g., in in granular glands of the skin, compounds all lantibiotics (JACK et al., Chapter 8, this like dermorphin, caeruleins (X), and neuro- volume). These reactive side chains may be tensin (XI) being noxious to predators. The involved in covalent binding of peptides to common processing/transport signals indicate target proteins, as has been shown for the cyanobacterial cyclopeptide microcystin and their common secretory pathway. et protein phosphatase 1 (MACKINTOSH al., 1995). A desdehydro derivative of nisin A 2.1.3 D-Amino Acids in shows reduced antibacterial activity (ROLLERibosomally Produced Peptides MA et al., 1996). pGlu-Asp-Tyr( HS03)-Thr/Leu/MetGly-Trp-Met-Asp-PheNH2 Epimerizations of amino acid residues in gene-encoded peptides have been verified for dermorphin (VIII) and dermenkephalin (XII). Dermorphin binds exclusively to the popioid receptor and is 1000 times more potent than morphine upon intracerebroventricular injection. The D-residue is of crucial importance for this high affinity. Isolation of the respective gene revealed the Ala codon as precursor of D-Ala (RICHTERet al., 1987). The predicted biosynthetic precursor contains 4 dermorphin sequences together with one for dermenkephalin flanked by putative cleavage sites. In addition, D-amino acids have been found in the neuroexcitatory snail peptides from Achatia fulica achatin I (XIII) and fulicin (XIV; FUJITA al., 1995), and the 48 amiet no acid-containing Ca2+ antagonist WAgaIVC from the venom of the Agelopsis aperra spider (XV). This venom has been the source of a cofactor-independent isomerase acting on peptides with the common sequence Leu-Xua-Phe-Ala. This enzyme has been shown to isomerize Ser, Cys, OMeSer, and Ala residues in a reversible reaction (HECKet al., 1995). At least in the case of dermorphins the conversion to D-Ala was shown to proceed at the prepropeptide stage (MOR et al., 1991).
2.1.4 Posttranslational Modifications of Side Chains
Non-disulfide cyclic peptides include lantibiotics (JACKet al., Chapter 8, this volume; JACK al., 1995) with stable thioether links et forming multiple 3- to 6-membered cyclic peptide structures. These unique antibacterials, which are approved as food components, have been found exclusively in bacteria so far. Examples given are epidermin (XVI) and subtilin (XVII). Several respective gene clusters have been characterized and, besides the structural genes of the propeptides, they contain modifying enzymes and proteins involved in export. Structural modifications to direct specificity and improved stability are under investigation (for details, see Chapter 8). In contrast to disulfide bonds, enzymes opening thioether bonds have not yet been identified. Side chain modifications of cysteine and serine residues to oxazoles and thiazoles, established for nonribosomal products for a long time, have recently been detected in ribosomal peptides as well, including microcins (MORENO al., 1995) and the rhizobial pepet tide trifolitoxin (BREILet al., 1996; TRIPLETT, XI1 Tyr-DMet-Phe-His-Leu-Met-AspNH2 personal communication). A respective transforming enzyme complex has been identified XI11 Gly-DPhe-Ala-Asp by the analysis of the microcin B17 (XVIII) VIV Phe-DAsn-Glu-Phe-ValNH2 XV EDN CIAEDIGK CTWGGTK CC RGRP CR CSMIGTN C E CTPRLIMEGL-D-Ser-SFA
I I
I
(XVI)
IASKFICTPGCAKTGSFNSYCC
(XVII) WKSETLCTPGCVTGALQTCFLQTLTCNCKISK -rV T gene cluster and it contains three proteins (LI et al., 1996). Microcin B17 is the first known gyrase inhibitor of peptidic nature. Its 69-residue precursor peptide is modified at Gly-Cys and Gly-Ser segments, respectively, or at GlySer-Cys or Gly-Cys-Ser segments, with the formation of bicyclic thiazol-oxazole derivatives. Replacement of oxazoles by thiazoles leads to an inactive product (BAYERet al., 1995). All respective examples reported so far seem again to be restricted to bacterial sources, including thiopeptides like thiostrepton (STROHL FLOSS,1995), the antitumor and peptide bleomycin, bacitracin, and various marine peptides like ulithiacylamide (XIX). These cyclic peptides isolated from sponge extracts are likely to originate from associated microorganisms, although this has not been generally established (BEWLEYet al., 1996). XVIII VGIGGGGGGGGGG-Oxz-ThzGGQGGG-Thz-GG-Thz-SNG-ThzOxz-GGNGG-Thz-GG-Thz-GSHI Leu-Thz-Cys-Mtz Mtz- Cys-Thz-Leu bosomal origin ..as now been verified by the detection of the structural gene during the Bacillus subtilis genome sequencing program. (XXI) 4,5hIle-6hTrp-Gly 4hPro
--
2 Structures
283
Asn-Cy s-Gly
I I
SO Ile
1 I
(XXII) Qxa Qaa Qoa-DSer-Ala-Cys-NMeVal
NMeVal-Cys-Ala-D Aer -Qoa Qaa Qxa

L
Qxa Qaa Qoa-DSer -Ala-Cys- NMeVal
XIX
YS
NMeVal-Cys-Ala-dSer-Qoa Qaa Qxa
(XX) [Q]-Ile-Ala-AAla-Ala- [Ser*]-Thz-Thr-AAbu-DCys-3,4hIle-Thz-Thr-Thz-[Ser*] -Thz-AAlaI I I I AAla(NH2)
2.1.5 Multicyclic Peptides

Prominent examples of multicyclic peptides from microorganisms include thiostrepton (XX), amanitins (XXI), and quinomycin (XXII). For the latter in vivo transformation of triostins (XXIII) to quinomycins was demonstrated (WILLIAMSON al., 1982). More et recently, the structural elucidation of the Bacillus subtilis antibacterial subtilosin (XXIV) introduced a new level of complexity. Its ri-
The principle of multiple rings in gene-encoded peptides is well established for numerous structures containing up to 4 disulfide links (see compilation in Sect. 5 ) . Common cyclization patterns emerge and point to a similar conserved enzymology. Properties of the peptides identified so far are mainly antimicrobial, cytotoxic, or like trifoil peptides involved in tissue repair and cell proliferation (CHINERY COFFEY,1996). The rat intesand tinal trifoil factor rITF has even been shown to form covalent dimers (CHINERY COFand
284
(XXIV) GLGLWGNKGCATCSIGAACLVDGPIPDG*IAGAX
LS-S]
IXI
FEY, 1996). Recent excitement has been provoked by the discovery of microbial multicyclics with promising antiviral properties like aborycin (RP71955, XXV) (POTTERATet al., et 1994; FRECHET al., 1994).
(XXV)
r CLGIGSCNDFAGCGXAVVCFW ,
- -I
s-s s-s
2.1.6 Linear Peptides

A number of peptides without any side chain modifications or disulfide bonds with diverse biological activities are known. For antimicrobial peptides at least several groups have been described of (1) amphipathic a-helical peptides (e.g., magainin) and (2) two ahelices linked by a hinge region (e.g., cecropins). Three other groups show high contents of (1) glycine (e.g., diptericin), (2) proline and arginine (e.g., apedaecin), or (3) tryptophan (e.g., indolicidin) (LEHRERet al., 1993; BoMAN et al., 1994; BOMAN, 1995). The tryptophan-rich indolicidin (XXVI) shows antibacterial and hemolytic activity. Replacement of tryptophan by phenylalanine abolishes only hemolytic activity (SUBBALAKSHMI al., et 1996) while antibacterial activities remain, even when proline residues are substituted by alanine. (XXVI) ILPWKWPWWPWRRNHZ A compilation of structures can be found in Sect. 5.
er, regioselective formation of several disulfide bonds still remains a challenge for chemists and overall yields are in the range of a few percent (KELLENBERGER al., 1995). et Microbial peptides produced in their hosts do not present a special problem. The production of animal or plant antibacterial peptides, in bacterial species, however, poses the problem of a suicide situation. This could be overcome by simultaneously introducing resistance or export genes. So far, either the expression of fusion constructs has been practiced (PIERSet al., 1993) or insensitive yeast cells have been employed (REICHHART al., et 1992).
2.1.8 Similar Open Questions in Gene-Encoded and Enzymatically Formed Peptides

The findings on multicyclic peptides illustrate well that the formation of constrained disulfide linkages to stabilize and functionalize peptide chains is not restricted to higher eukaryotes. The production of various multicyclic peptides may thus be well achieved by microbial cultures. However, a number of questions might be relevant, brought up recently in connection with defensins (SELSTED and OULE-I'TE, 1995): How can the spontaneous lysis of granules producing cytolytic peptides be prevented? The same question is unresolved for the many microbial compounds with lytic properties; there might be natural antagonists or storage molecules. Are there extracellular roles for neutrophil defensins, or are there intracellular roles for enteric defensins? This question applies to most metabolites with, e.g., antimicrobial properties and it has not yet been answered for any secondary metabolite convincingly. Do defensin isoforms cooperate to form heteromer assemblies in target cell mem-
2.1.7 Production of Ribosomally Formed Peptides

Besides solid-phase synthesis production by fermentation is in an early phase. Howev-
3 Biotechnology of Peptides
285
branes? The question of assembly is of considerable importance since dimerization or multimerization properties cannot be predicted yet. The human defensin HNP-3 forms dimers, and dimer formation is essential for the function of many microbial peptides like the linear gramicidin pore formation or vancomycin binding to peptidoglycan precursors.
In the biotechnology of peptides fermentation of pure microbial cultures is used traditionally. An increasing number of processes deals with cell culture techniques that are still dominated by larger peptides and proteins as products. Obviously, the manufacturing of most bioactive peptides, especially antibiotics, needs self-protective systems. Such systems of self-immunity are inherent in producer cells but may limit the levels of production. Problems arise when non-self products are to be generated, e.g., the expression of animal-derived antibacterials in bacterial cultures. Solutions of such problems might be the use of in vitro production systems or the introduction of resistance genes into the expression organism.
2.2 Nonribosomal Origin

In contrast to those of ribosomal origin, more unusual structures and compounds are found in nonribosomally formed peptides. Systematic approaches to peptide structures are difficult, since in reports which are basically oriented towards structural chemistry the compounds may be found as peptide alka1996), as indole alkaloids (IHAloids (LEWIS, RA and FUKUMOTO, 1996), or they may be contained in source-oriented reviews (e.g., 1996, and marine metabolites; FAULKNER, references therein). Peptide structures linked with polyketides and terpenoids may raise additional problems of compound classification. In a biosynthetic approach simply the types of enzyme structures involved are used. Since all peptides originate from linear precursors followed by various processing reactions, several approaches to systematics are possible (VON DOHREN, 1990). However, in many cases there are several structural features involved and convenient approaches should then permit the search for a structural profile.
3.1 Biosynthesis
The essential questions to be addressed refer to the biosynthetic origin of the compound and to the availability of precursors. It has to be decided whether a structural gene or a protein template is involved. This decision can be made upon inspection of the structural features. A collection of modified amino acids found in peptides of ribosomal origin is given in Tab. 1. A respective collection of amino acids encountered in nonribos-
Tab. 1. Unusual Features of Ribosomally Derived Peptides
Modified Amino Acid Glu Ala Met ASP Ser Thr CyslSer Cys/Thr SeriThr CYS
Modification pyroGlu D-Ala D-Met D-As~ D-Ala D-Abu lanthionine methyl-lanthionine oxazole-derivatives thiazole-derivatives
Process terminal modification in-chain epimerization enzymatic dehydration enzymatic dehydration and thioether linkage enzymatic cyclization
286
SYSTEM Nucleic Acid Dependent

1 2 3
PRODUCT
Ribosomal System
linear polypeptidesof any sequence and Imgihfmm 20 proteinogenicamino acids activationas sminoacyl adenyla1.s aminoacyl-tRNA-.ynUIetases high specificitjthmugh proof reading mechanisms high energy requiring processingof peptides by poomanslational processes
TYPE 1
choica of Program Asssmbiy Line
Complex Mechaniam Up t 180 Compomntr Participating o
TYPE 2
Fixed Program Assembly Line
& %bP& a Multionryme

All Intermediates Remain Enzyme-Bound Sequence Is Determined on the MuitlfunctionaiEnzyme-Chain 4-Phosphopantetheine is Cofactor MultlenzymesIdentified so far only in Bacteria and Fungi
Nucleic Acid Independent
Enzymatic Systems
--
PolyenzyrneSystem
iinoar, branchedand cyclic pptidea and depsipeptides. e.g. Gramicidin S activetimas amlmcyl adenyiates proteinand nonprotein amino acids (Dconfiguratlon. omithine. hydroxy acids) lowspecificity Incorporationof analmodifkatbn, 9.9. Nmethylation length of peptides up to about 20 amino acids
Single Step Enzymes

linear peptides, 9.9. OlUblhlolr, bacterialcoil wall activationas phosphates or aminoacyl adenylates synthesis of short peptides (generally less than 5 amino acids) also non-protein amino acids are components
separate enzymes Intarmediatesare soluble
only one component is activated
Fig. 1 Mechanisms of peptide biosynthesis. Scheme comparing ribosomal (type 1)and nonribosomal systems (type . 2 - multistep systems, type 3 - single-step systems).
omally formed peptides is presented in Sect. 5. The principal differences of peptide forming systems are shown in Fig. 1. Nonribosoma1 peptides are formed either by single-step enzymes with free intermediates or by multistep systems. Well-known examples of peptides made by single-step enzymes are glutathione, pantetheine, or muropeptides. Thiol peptides termed phytochelatins which upon exposure to Cd2+, Cu2+, or Zn2+ are induced in the fission yeast Schizosaccharomyces pombe (KONDO al., 1984) or in culet tured plant cells (GRILLet al., 1989) are composed of Glu, Cys, and Gly of the composi-
tion (*Flu-Cys),Gly, with n=2-11 (RAUSER, 1990 STEFFENS, 1990). Besides Gly also p Ala or Ser may function as C-terminal residues (GEKELER al., 1989; KLAPHECK al., et et 1994). In maize seedlings two additional peptide families of the type (fllu-cys),, and (+lu-Cys),Glu have been observed upon exposure to Cd2+ (MEUWLY al., 1995). Bioet synthetic systems for these peptides are supposed to resemble glutathione forming enzymes. The modification of peptide structures by alteration of their genetic background is an emerging approach (see Sect. 3.2.2). In gener-
287
al, the structural genes for peptides or peptide synthetases are clustered together with those for modifying enzymes, regulatory signals, and - if secreted - for export proteins.
3.1.1 Biosynthetic Gene Clusters

Clusters identified for peptide biosynthesis are listed in Tab. 2. So far, only microbial clusters have been characterized in detail. Genome sequencing has revealed that there are no nonribosomal peptide biosynthetic activities in yeast, Haemophilus influenzae, Mycoplasma genitalium, Synechocystis, and Methanococcus jannaschii. However, evidence for multiple peptide forming clusters has been obtained in Bacillus subtilis, Microcystis aeruginosa, Amycolatopsis mediterranei, and Bacillus brevis. Gene structures generally permit the identification of the involved biosynthetic reactions. Structural genes for ribosomally encoded peptides are a direct proof of their origin (e.g., subtilosin). A respective nonribosoma1 system is readily identified by the presence of very large genes (up to 45 kb) with modular structures. The most prominent structural element is the amino acid activating module or adenylate domain. Domains like this can be identified by a number of highly conserved motifs (KLEINKAUF and VON DBHREN, 1996). The signature sequence SGTTGxPKG is most prominent and is also contained in acyl-CoA synthetases. Peptide forming systems are recognized by the repeated occurrence of these modules directly related to the number of amino acids contained in the respective peptide.
tems, or reductions and dehydration in the polyketide systems. A module is thus comprised of submodules. Boundaries of these submodules have been determined at the protein level by limited proteolysis studies and sequence alignments. Therefore, the respective DNA modules have corresponding functional protein modules, which contain domains and subdomains of defined structures and functions. All modules can be readily identified by the presence of highly conserved core sequences (PFEIFERet al., 1995; STEIN and VATER,1996) (Fig. 2); a selection is given in Tab. 3. Arrangement of modules Modules and their corresponding functional domains show a linear arrangement corresponding to the sequence of catalytic events. As these sequential reactions are catalyzed by protein surfaces this strict linearity is not obvious. Domains which are not directly linked need to interact frequently, e.g., in condensation reactions (see below). Domains may be reused for repeated sequences as in the formation of the cyclodepsipeptide enniatin (XXVII). The respective synthetase contains only two, not six activation modules. NMeVal-Hiv-NMEVal I I (XXVII) Hiv-NMeVal-Hiv Adenylate forming domains
3.1.2 Biosynthetic Modules

The term module has been introduced for a gene-encoded functional unit performing an elongation step in either polyketide or peptide biosynthesis (DONADIO al., 1991; et KLEINKAUF VON DBHREN,1996). Durand ing the process the particular residue entering the elongation cycle may be structurally altered, which implies epimerization, methylation, or hydroxylation in peptide forming sys-
In contrast to polyketide forming systems peptide synthetases activate their building blocks, i.e., amino acids, imino acids, and hydroxy acids, as adenylates on an integrated activation domain. These activation domains show structural similarities to the carboxyl-activating acyl-CoA synthetases of the polyketide systems. The positioning of activating domains corresponds in sequence to the amino acid sequence of the product in either wholly or partially integrated interacting enzyme systems (Fig. 3). The reaction sequence includes binding of carboxyl substrate and MATPZ(with M=Mg2+, Mn2+) and formation of the
288
Tab. 2. Peptide Biosynthetic Clusters Identified

Compound Type" Organism Reference
Ribosomally Formed Peptides: Cytolysins P-?-M Epidermin P-21-M Lactocin P-37-M Lactococcin P-27-M Microcin B17 P-43-M Nisin P-34-M Pep-5 P-34-M Subtilin P-32-M Trifolitoxin P-11-M Nonribosomally Formed Peptides: A54145 R-P-13-L-10 Actinomycin R-(L-5)2
Anguibactin Ardacin Bacilysin Bacitracin Bialaphos CDA Clavulanic acid Cephalosporin Cephamycin Coronatin Cyclosporin Daptomycin Destruxin Enniatin Enterochelin Fengymycin Ferrichrome Gramicidin S HC-toxin Immunomycin Iturin Lysobactin Microcystin Nikkomycins Nosiheptide Penicillin R-P-2-M P-7-M P-2-M R-P-12-C-7 P-3 R-P-11-L-10 modified amino acid P-3-M P-3-M acyl amino acid c-11 R-P-13-L-10 L-6 D-6 P-C-E-3 R-P-lo-? C-6 C-(P-5)* c-4 modified polyketide C-8 P-11-L-9 c-7 modified peptides R-P-13-C-10-M P-3-M
Enterococcus faecalis Staphylococcus epidermidis Lactobacillus sake Lactococcus lactis Escherichia coli Lactococcus lactis Staphylococcus epidermidis Bacillus subtilis Rhizobium leguminosarum
SAHLet al., 1995 SCHNELL al., 1992 et SKAUGEN al., 1994 et RINCE al., 1994 et LI et al., 1996 JACKet al., 1995 K A L E ~et al., 1989 A HANSEN, 1993 BREIL al.. 1996 et
Streptomyces fradiae Streptomyces chrysomallus Vibrio anguillarum Kibdelosporangium aridum Bacillius subtilis Bacillus licheniformis Streptomyces Streptomyces Streptomyces Acremonium viridochromogenes coelicolor clavuligerus chrysogenum
Lysobacter lactamgenus Nocardia lactamdurans Pseudomonas syringae Tolypocladium niveum Streptomyces roseosporus Metarhizium anisopliae Fusarium scirpi Escherichia coli Bacillus subtilis Bacillus subtilis Ustilago maydis Bacillus brevis Helminthosporium carbonum Streptomyces sp. Bacillus subtilis Lysobacter sp. Microcystis aeruginosa Streptomyces tendae Streptomyces actuosus Aspergillus nidulans Penicillium chrysogenum
BALTZ,1996 SCHAUWECKERal., et 1996 CHENet al., 1996 PIECQet al., 1994 SAKAJOH al., 1987; et HILTON al., 1988 et HERZOG-VELIKONJA et al., 1994 SCHWARTZ al., 1996 et CHONG al., 1996 et HODGSON al., 1995 et MART~N GUTIERand REZ,1995 KIMURA al., 1996 et COQUEet al., 1995a,b BENDER al., 1996 et WEBERet al., 1994 BALTZ,1996 BAILEY al., 1996 et HAESEet al., 1993 REICHERT al., 1992 et LIU et al., 1996 MEI et al., 1993 TURGAY MARAand HIEL, 1995 PITKIN al., 1996 et MOTAMEDI, 1996 HUANG al., 1993 et BERNHARD al., 1996 et MEISSNER al., 1996 et BORMANN al., 1996 et STROHL FLOSS,1995 and SMITH al., 1990; MACet CABEet al., 1990 DIEZ et al., 1990
Tab. 2. Continued
289
Compound Phaseolotoxin Pristinamycin A Pristinamycin M Pyoverdin Rapamycin Saframycin SDZ214-103 Surfactin Syringomycin Syringostatin Thiostrepton Tolaasin Tyrocidine Yersiniabactin
a
Type" P-4-M R-C-6 polyketide/peptide R-P-8-M modified polyketide P-4-M L-11 L-8 R-L-9 R-P-17-C-10-M R-P-18-L-5 c-10 R-P-3-M
Organism Pseudomonas syringae Streptomyces pristinaespiralis Pseudomonas fluorescens Streptomyces hygroscopicus Myxococcus xanthus Cylindrotrichum oligospermum Bacillus subtilis Pseudomonas syringae Streptomyces laurentii Pseudomonas tolaasi Bacillus brevis Yersinia enterocolitica
Reference TURGAY and MARAHIEL, 1995 BLANC al., 1997 et STINTZI al., 1996 et SCHMECKE al., 1995; et MOLNAR al., 1996; et APARICIO al., 1996 et POSPIECH al., 1996 et BERNHARD al., 1996 et COSMINA al., 1993 et Mo et al., 1995 STROHL FLOSS, and 1995 RAINEY al., 1993 et MITTENHUBER al., et 1989 GUILVOUT al., 1993 et
The abbreviations used are: P peptide, C cyclopeptide, L lactone, D depsipeptide, E ester, R acyl, M modified. The structuraltypes are defined by the number of amino-, imino- or hydroxy acids in the precursor chain. The ring sizes of cyclic structures are indicated by the number following C, L, D, or E, defining the type of ring closure.
Fig. 2. Domain construction of peptide synthetases.Peptide synthetase sequencescan be identified by a spacing of motifs, which have been assigned to adenylate formation (A1 and A2), aminoacylation of pantetheine (S), N-methylation of aminoacylresidues (M), condensation (peptide bond formation) (C), epimerization (E), and The respective consensus sequences are given in Tab. 3. thioesterase (TE).
acyladenylate which, as a highly reactive mixed anhydride, is stabilized against hydrolysis by the two subdomains. The released MPF- may reverse the activation reaction leading to the two substrates. This reaction employing [P-321-PPi usually is used as a sen-
sitive assay for peptide synthetases and their activation domains. In the presence of high concentrations of ATP the adenylate may react to form AP4A, but it is not known if this reaction has any physiological significance.
290
Tab. 3. Consensus Sequences of Peptide Synthetase Motifs Adenylate Domains A: LTxxELxxxAxxLxR B: AVxxAxAxWxIDxxYPxER C. YSTGTTGxPKG D: IIxxYGxT E: GELxIxGxxVAR F: RLYRTGDL G: IEYLGRxDxQVKIRxxRIELGEIE H: LxxYMVP I: LTxxGKLxRKAL
Acyl Carrier Domains J: LGGxSIxAI

Elongation (1) Epimerization" (2) Domains K (Bl): YPSVxxQxRMYIL, L (Bl): LIxRHExL, M (Bl): DMHHIIxDGxSxxI N (Bl): LSKxGQxDIIxGTPxAGR 0 (Bl): IxGMFVNTxLALR,
(B2): LxPIQxWF (B2): LxxxHD (B2) HHxxVDxVSWxIL (B2): VxxEGHGRE (B2): TVGWFTxxxPxxL P: PxxGxGy
Q:
N-Methyl-Transferasesb M1: GxDFxxWTSMYDG; M 2 LEIGTGTGMVLFNLxxxxGL; M3: VxNSVAQYFP M4: ExxEDExLxxPAW, M5: HVExxPKxMxxxNELSxYRYxAV N6: GxxVExSxARQxGxLD.
a
Derived from Bacillus synthetases only. Derived from fungal domains.
Aminoacylation or carrier domains

Adenylates formed on acyl-CoA synthetases are transferred to CoA, and these thioesters are the stable transport form of aliphatic carboxylic acids. Amino acids are transferred to acyl carrier modules integrated within the multienzyme structures. These carrier modules have a structure similar to acyl carrier proteins involved in various polyketide forming systems including fatty acid biosynthesis (STEINand VATER, 1996; STACHELHAUS et al., 1996a). Their essential functional group, 4'-phosphopantetheine, has to be introduced by posttranslational modification from CoA. The respective transferases have recently been identified with the aid of the holo-acyl
carrier protein synthase structure from Escherichia coli (LAMBALOT al., 1996). Reet spective genes have been detected in several biosynthetic gene clusters, and for each cluster a specific transferase to activate the pathway has been proposed. The finding that apoenzymes form unusually tight complexes with the transferases in the absence of CoA suggests an additional control mechanism (LAMBALOT and WALSH,1995; PFEIFER al., unet published data). As in the case of the acyl transfer to CoA the adenylate is cleaved by the phosphopantetheine thiol to release AMP and an enzymebound intermediate thioester is produced (PFEIFERet al., 1995).
291
Fig. 3. Domains in peptide forming multienzyme systems. ACVS: ACV synthetase, ES: enniatin synthetase, Ent: enterobactin biosynthetic enzymes EntE and EntF, GS: gramicidin S synthetases 1 and 2, T Y tyrocidine synthetases 1 and 2, Srf: surfactin synthetases 1, 2 and 3, CY: cyclosporin synthetase, H C HCtoxin synthetase. Amino or hydroxy acid specificity of the domains is indicated. Note that two different types of condensation domains are found, slightly deviating in their consensus sequences, and one very closely related epimerization domain. The E-C domain in the HC-system catalyzes both epimerization and condensation.
(KLEINKAUFand VON DOHREN,1997) (Fig. 4). Adjacent to the carrier domains condensaIn several systems D-amino acid residues tion or epimerization domains with similar are formed during a transfer reaction bestructures are found. These are structurally tween two multienzymes. In these cases an stable regions of about 50 kDa (DIECKMANN epimerization domain terminates the first enet al., unpublished results). Epimerization do- zyme while the second starts with a condensamains are readily identified by two additional tion domain (DE CRECY-LAGARD al., et motifs (see Tab. 3). Epimerizations are cata- 1995). If the transfer occurs within an enzyme lyzed at the aminoacyl or peptidyl stage re- system and is thus an intraenzymatic transfer, sulting in a thioester-bound mixture of both a third type of domain with a size of about isomers (KLEINKAUF and VON DOHREN, 100 kDa is found resembling a fusion of both 1997). In the following step the selection of types of domains. Such a domain was dethe stereospecific intermediate is controlled. tected first in HC toxin synthetase, a multienThe current model of peptide bond formation zyme forming the plant pathogenic cyclotetraassumes that aminoacyl and peptidyl interme- peptide HC toxin (SCOTT-CRAIG et al., diates of two adjacent modules are directed 1992). towards peptide bond formation at the enclosed condensation domain. In order to ex- N-methylation domains plain the direction of bond formation aminoMethylation of amino groups is catalyzed acyl and peptidyl binding sites have been postulated in analogy to the ribosomal system by a transferase domain at the aminoacyl-
Condensation and epimerization domains
292
4-Phosphopantetheine
/
ondensation domain 1
'2
Activation domain
domain
Thiogsterasedomain
Exit site
Fig. 4. Schematic view of peptide biosynthesis using a tripeptide model system, ACV synthetase. The three amino acids are activated at the activation domains 1-3. Each activation domain is composed of two subdomains. Adjacent to each activation domain is a carrier domain, which contains the cofactor 4 '-phosphopantetheine. This factor may swing into three positions: binding the activated amino acid (activation domain), transport of the activated amino acid to the condensation site (A-site), and transport of the starter unit or peptidyl intermediate to the condensation P-site. In the terminating reaction epimerization occurs and the peptide is stereospecifically released at the thioesterase domain.
thioester stage (BILLICH and ZOCHER,1987, 1990). This domain is inserted within the adenylate domain and has a size of about 50 kDa (HAESEet al., 1993; WEBERet al., 1994). The highly conserved core motifs (see Tab. 3) are not found in other known methyl transferases which all share the substrate S-adenosyl-methionine (BURMESTER al., 1995). The reacet tion has been investigated in some detail and methods to detect methylation domains at the protein or gene level have been developed (ZOCHER al., unpublished data). et
change of a highly conserved serine residue contained in the motif GXSXG in the thioesterase domain, G(L-a-aminoadipy1)-L-cysteinyl-L-valine is released as the main product instead of exclusively S(L-a-aminoadipy1)-Lcysteinyl-D-valine (KALLOW et al., 1996). This result indicates that the thioesterase catalyzes the stereospecific release of the product. Similarily condensation domains adjacent to epimerization domains are thought to catalyze the stereospecific condensation of only one of the peptidyl or aminoacyl intermediates.
Thioesterase domains Enzymes still to be identified

Domains with similarities to thioesterases have been detected in peptide synthetases and polyketide synthases as well. Their positioning in the C-terminal region of terminating multienzymes indicates their functions in terminating a catalytic cycle. Such functions have been demonstrated in several polyketide forming enzyme systems. Experimental work on peptide synthetases is only available so far on ACV synthetase producing the tripeptide precursor of plactam antibiotics (for details, see Chapter 6, this volume). Since a linear peptide is released the thioesterase function seems to be obvious in this case. Upon A number of reactions known to be involved in enzymatic peptide biosynthesis have not been studied yet. It could not be established so far, if these proceed within the context of integrated multienzymes or if separate single-step enzymes interact with the respective systems. Such reactions include the modification of amino acids (like hydroxylation or O-methylation), side chain modifications (like oxazole and thiazole ring formation of serine, threonine and cysteine side chains; cf. Sect. 2.1.4), and ether linkages or direct linkages of aromatic side chains.
293
of Bacillus subtilis genome sequencing has revealed a peptide biosynthetic cluster with a yet unidentified product (TOGNONI al., et 1995). The conditions of expression of this Multienzyme systems for most types of cluster are unknown. Tentatively, from correpeptides have been characterized (KLEIN- lating activation modules with known structures, as has been discussed by COSMINA et KAUF and VON DOHREN,1996) (Tab. 4). It seems evident that systems of eukaryotic ori- al. (1993), the already identified peptide fengin are fully integrated while prokaryotic sys- gycin could be the product. tems contain up to three interacting multienzymes. The only system deviating from this rule is ACV synthetase forming the p-lactam 3.2 Production precursor tripeptide. This enzyme is recovThe best studied system for peptide proered in a fully integrated state from both prokaryotic and eukaryotic sources. Isopenicillin duction is the penicillin fermentation where N synthase forming the bicyclic penam pre- relevant parameters have been investigated in cursor from the tripeptide is located adjacent considerable detail. These include precursor to the synthetase in the respective gene clus- concentrations and transport, intracellular ters and has been implicated in horizontal compartmentation, levels and activity of biogene transfer (AHARONOWITZ al., 1992; synthetic enzymes, their regulation, expreset BUADES and MOYA,1996). Transfer, howev- sion, and stability in relation to the gene copy er, is thought to proceed from prokaryotic to number, their posttranslational modification and intracellular localization, their possible eukaryotic hosts. The integration of all sequential condensa- inhibition by metabolites or products, and fition functions does not rule out various internally their export. This still uncomplete list of actions with other enzymes involved. In cy- factors illustrates the complexity of tasks to closporin formation D-alanine and the com- be addressed. Many of the details are discussed in Chapter 6. For an overview of the plex amino acid (4R)-4-[(E)-2-butenyl]-4-methyl-L-threonine (Bmt) have to be supplied as state of metabolic flow studies in Penicillium direct precursors. Thus a respective alanine chrysogenum the reader is referred to the reracemase and an enzyme system forming a cent publications of NIELSEN et al. (JoRpolyketide precursor which has to be trans- GENSEN et al., 1995a, b; NIELSENand JOR1995). Comparable work on the feraminated are involved in the process as well GENSEN, (OFFENZELLER al., 1996). Likewise, acyl- mentation of other peptides is scarce, but preet ated peptides can be assumed to have specific cursor-directed feeding work has been repolyketide synthases required for the respec- viewed recently (THIERICKE and ROHR, tive acyl-CoA starter unit contained in or cor- 1993). Some of the results are discussed in egulated with their respective peptide biosyn- Sect. 3.2.2.1. thetic genes. Biosynthetic systems can be identified by the respective alignments identifying the var- 3.2.1 Fermentation Procedures ious modules. For isolation of genes conveIt is beyond the scope of this chapter to disnient use can be made of the highly conserved core sequences to PCR-specific gene seg- cuss data on fermentation procedures which ments (BORCHERT al., 1992). Adjacent re- actually are similar in secondary metabolite et gions are explored then, and the correlation is production in general. Therefore, just a few attempted by gene disruption and agreement remarks on recent developments are added: of the peptide structure with the predicted The control of parameters in metabolite fermodular arrangement (BERNHARD al., mentations has been improved by measuring et 1996; MEISSNER al., 1996). These proce- not only dissolved oxygen but also redox poet dures are in complete analogy to polyketide tential, dissolved carbon dioxide (DAHOD, systems (SCHWECKE al., 1995). In the case 1993), and culture fluorescence (NIELSENet et
3.1.3 Biosynthetic Enzyme Systems
294
Tab. 4. Peptide Synthetases Currently Studied Peptide Linear Peptides: ACV Ergopeptines Gramicidin Cyclopeptides: HC-toxin Gramicidin S Tyrocidine Cyclosporin Lactones: Destruxin Actinomycin SDZ215-104 Surfactin Organism Structural Type" P-3' R-P-3d P-15-M' Multi-Enzymes (size kDa)b 425 415b 420 140 370 300 700 570b 127b 512b 123b 400 800 17Wb 800 49 280 480 1650 402 401 144b 450 900 700 350 750 347b 2 4 1 4 1 3 6 11 6 1 2 3 11 3 3
1
Activation Modules
Aspergillus nidulans Acremonium chrysogenum Streptomyces clavuligerus Claviceps purpurea Bacillus brevis
Helminthosporium carbonum Bacillus brevis Bacillus brevis Beauveria niveum Metarhizium anisopliae Streptomyces chrysomallus Cylindrotrichum oligospermum Bacillus subtilis
P-4' C-(P-5)zg C-lOh

c-lli
L-6-MJ R-(L-5)zk L-111 R-L-7"
Branched Cyclopeptides: Lysobactin Lysobacter sp. Bacitracin Bacillus licheniformis
P-11-L-9h P-12-C-7"
Cyclodepsipeptides: Enniatin Fusarium scirpi

a
C-(D-2)3"
'
c(DPhe-Pro-Phe-DPhe-Asn-Gln-Val-Orn-Leu) ' c(DAla-NMeLeu-NMeLeu-NMeVal-NMeBmt-Abu-Sar-NMeLeu-Va1-NMeLeu-Ala) 1 c(DHimv-Pro-Ile-NMeVal-NMeAla-PAla)

Mha-c(Thr-DVal-Pro-Sar-Val)
I
The abbreviations used are: P peptide, C cyclopeptide, L lactone, D depsipeptide, R acyl, M modified. The number of residues activated by each (mu1ti)enzyme is indicated. 3 x 2 stands for the trimerization of 2 residues. Size verified by gene sequencing. a-Aad-Cys-DVal DLYSA-Ala-Phe-Pro f-Val-Gly-Ala-DLeu-Ala-DVal-Val-DVal-T~-(DLeu-T~)3-ethanolamine c(DPro-Ala-DAla-Aeo) c(DPhe-Pro-Val-Om-Leu)2
c(NMeVal-DHiV)3
c(DHiv-NMeLeu-Leu-NMeVal-NMeBmt-Thr-Sar-NMeLeu-Leu-NMeLeu-Ala) ~(3hC~~-Glu-Leu-DLeu-Val-Asp-DLeu-Leu) DLeu-Leu-c(2hPhe-2hLeu-Leu-DArg-Ile-aThr-Gly-2hAsn-Ser) (Ile-Cys)-Leu-DGlu-Ile-c(Lys-DOrn-Ile-Phe-Asn-DAsp-His)
295
al., 1994). Monitoring of the state of mycelia defined targets under conditions of enhanced in fermentations permits more detailed inves- recombination could become a valuable tool. tigations of morphology and vacuolation et (PAULand THOMAS,1996; VANHOUTTE et and 3.2.2.1 Family Exploitations al., 1995; NIELSEN al., 1995; NIELSEN 1995). Oils have been used with KRABBEN, The concept of peptide families is illusadvantage as carbon sources in cephamycin fermentations (PARKet al., 1994a, b). Solid- trated in Fig. 5 using the examples of gramistate fermentation for the production of sur- cidin S, gramicidin, cyclosporin, and aureobafactin and iturin by Bacillus subtilis has been sidin. Analogs of gramicidin S have not been exploited (OHNOet al., 1995, 1996). The sep- detected for a long time. Modern mass specaration of products is a promising approach trometric techniques have considerably ento avoid feedback problems. Attempts with hanced analytical resolution (NOZAKI and 1987; THIBAULT al., 1992). et an aqueous two-phase system have been MURAMATSU, made in subtilin fermentation with Bacillus Both bacterial peptides show quite limited subtilis (KUBOIet al., 1994). Alternatively, a structural variations. The only remarkable microfiltration module was used in nisin pro- feature is that the intiating valine in linear duction with Lactococcus lactis (TANIGUCHI gramicidin has been found to be replacable by isoleucine. Likewise, only tryptophan in et al., 1993). The discovery of the role of autoregulators position 11 of gramicidin has been exchanged in metabolite production (CHATER and BIBB, by other aromatic amino acids. From largeChapter 2, this volume) has led to production scale fermentations of the fungal metabolites processes now. Product yields are enhanced cyclosporin and aureobasidin 32 and 29 anaby the controlled addition of an autoregula- logs have been purified and characterized. In tors at high cell concentrations (YANGet al., general, the number of tolerated substitutions is not more than two. This indicates a sieving 1996). mechanism presumably exerted by the biosynthetic process. The combinatorial game as played in chemical synthesis of analogs thus 3.2.2 Structural Alterations meets only limited success in the enzymatic system. In fact, only a small amount of preIn order to achieve structural alterations dictable analogs is detected. 18 variable positwo approaches are available, either to screen tions should permit the formation of more for analog producers or to manipulate the than lo4 cyclosporin analogs, but only 32 peptide composition by precursor feeding. have been detected (FLIRIet al., Chapter 12, Genetic manipulations are less obvious be- this volume). The respective abortion prodcause the respective selection processes are ucts, however, have not been investigated yet. still unclear. It seems obvious that peptides Likewise, 14 replacements of aureobasidins encountered in various sources show a unique permit the synthesis of more than lo3 anadesign with respect to structural variations. logs, 29 of which have been found (IKAIet al., These variations are evident from the obser- 1991a, b; AWAZUet al., 1995). Within a notably small range of variations vation of peptide families. It is fascinating to observe variations in certain residues while analogs have been detected by screening others are invariant. Remarkable invariance methods. The scope of this approach is illusand in can be achieved solely by selection within the trated by LANCINI CAVALLERI Chapenzymatic steps of peptide synthesis and may ter 9 on dalbaheptides. Screening for cycloreflect proof-reading mechanisms known sporin analogs gave rise to the peptidolacfrom the ribosomal machinery. It can be as- tone analog SDZ204-125 (Tab. 4), where D-alasumed that structural selections are con- nine is replaced by D-hydroxyisovalerate, the nected with biological targets which still need variable position 2 is dominated by threonine, to be identified. Once biological selection and one N-methylation is missing. In the case processes are understood selection against of the cyclodepsipeptide enniatin different
296
fOm Leu Cit Abu Lys 'DPhe +2Pro +Val +*Om +'Leu "Leu +*'Om +"Val t2'Prot"DPhe
. 1
flle N a l 4 l y +'Ma -+DLeu+Ma +DVal +Val +'DVal-+
Trp +DLeu -+"Trp +DLeu +Trp +DLeu +Trp +EA Phe TYr
DVaY MeLeu MeAOC* DSer Leu Val MedeoxyBmt 8DAla+YeLeu+MeLeu+MeVal+1MeBmt
. 1
Ala+MeLeutVal+MeLeu+MeGly+2Abu Abu Leu Nva Val Gly Ala Leu lle Thr Melle Val Nva
(3)
2,5h3mP 2,4h3mP 2,3h3mP Dhiv
Val
MeTyr Me2hPhe
'DHmp+Me2VaI+'Phe~Me4Phe
t
Meg2hVal+aLeu+Me7Val+6alletsPro
MeVal N2MeAsp MeLeu Melle Val Met Leu
. 1
MeTyr mFPhe MemFPhe oFPhe MeoFPhe 'DHmp+Me2Val+'Phe+Me*Phe Ye82hVal+8Leu+Me7Valt6alle+6Pm alle Nle Hyp Nva Met Spro
. 1
(4 b)
(3) cyclosporins, and (4) aureobasidins. Only two positions are variable in the Bacillus peptides (1) and (2). Note that only one of the four tryptophans has been found to be exchanged. The fungal peptides show conserved and less restricted positions. In all analogs found usually one of the variations is present, and in rare cases two positions are exchanged compared to the main products. More exchanges apparently reduce the rate of synthesis so that such products become extremely rare. Aureobasidin analogs (4b) were obtained by directed biosynthesis.
Fig. 5. Peptide analogs isolated from fermentation broths: (1) gramicidin S analogs, (2) linear gramicidins,
4 Future Prospects
297
Fusaria have been shown to have altered profiles of branched-chain analogs. A study of the respective peptide synthetases revealed et altered substrate binding sites (PIEPER al., 1992).
DOHREN, 1982; LAWEN TRABER, and 1993). Rates of synthesis of gramicidin S analogs have been studied with respect to multiple
D-CHA D-T~A' D-Trp D-OmTyr D-mTyr D-oTyr D-Tyr D-IPhe D-BrPhe D-CIPhe D-oFPhe D-mFPhe D-pFPhe
3.2.2.2 Biosynthetic Manipulations

The potential of biosynthetic manipulations has to address either the precursor situation or the enzymatic template. A convenient procedure includes feeding of substrates or substrate analogs. Template alterations are still in their infancy and they will become of special importance when selection against new targets not encountered in natural or ecological systems becomes available.
Sar 4s-Pro 3s-Pro hPro Aze 3,4APro
Leu Nva Nle alle Nle lle
Lys Arg
Nle alle lle -'Leu
'DPhe -'Pro
t
"Leu
-Val
-*Om
+*'Om $ ' V a l" P r o " D P h e

MeCys MeaThr MeSer Me2a4m4HEA MecyclodihydroBmt Me2a3h4buOA MeNva+ Me2a3h4,8mNA MetbuGly Me2a3h60EA MetbuAla Me2a3h4m20A+ MeallylGly Me2a3h4mOA MeCPG+ MeNle Mealle+ Me3hCHA+ Melle+ MeCHA
(1)
DLys DPhe DVal+ Work in directed biosynthesis has been suc- DCys cessful in various cases including tyrocidines, DcyclopropylGly+ actinomycins, viridogriseins (where the natu- 1-CI-D-vinylGly ral D-hydroxyproline is replaced by feeding of DtbuAla proline to the nonhydroxylated analog which 2-F-DAla+ 2-CI-DAla+ is a more effective antibiotic; OKUMURA,R-Ala+ 1990), ergot peptides, enniatins, cyclosporins, DAbu+ and aureobasidins (for a review, see THIER- Gly+ ICKE and ROHR,1993). The remarkable ex-
Directed biosynthesis
tension of products is illustrated with aureobasidin analogs which are not available without feeding (Fig. 5). It is possible to replace proline by hydroxyproline and thioproline (TAKESAKO al., 1996). However, systems et may be surprisingly restrictive. Attempts to replace 4-hydroxyproline in the echinocandin analog L-671329 by feeding a culture of the producer Zalerion arboricola were not sucet cessful (ADEFARATI al., 1991). It has been established later that the hydroxy group is essential for the antifungal activity. Such results imply a target adaptation of biosynthetic systems. An extension of in vivo feeding approaches is the in vitro enzymatic synthesis of peptides employing the isolated multienzyme systems (KLEINKAUF and VON DOHREN, 1997). The potential of the method is illustrated in Fig. 6 using gramicidin S, cyclosporin, and the pepand tolide SDZ214-125 (KLEINKAUF VON
MedihydroBmt MeLeu vinylGly MeAOC' DSer Leu Leu Val MedeoxyBmt 'DAla+MeLeu+MeLeu+Me%al+'MeBmt t 1 (2) AlatMeLeut~altMeLeui-MeGlytEAbu Abu Leu Nva Val Gly Ala Leu Ile Thr Melle Val Leu Nva Gly+ Nva Nle vinylGly Val CYS Phe aAla Ile+ alle+ CPG allylGly Abu tbuAla tbuGly allylGly aThr CYS Ile PPT
Fig. 6 Peptides formed by in vitro synthesis with . supply of amino acids, ATP, and SAM if required (1) gramicidin S, (2) cyclosporin, and (3) SDZ214125. For abbreviations see p. 278.
298
MetbuAla MeallylGly MeCPG Mealle+ MeAbu+
2hiCap+ D2h3mVa+ D2hVa+ D2hBu+ vinylGIy DLac
hydration, the reprogramming approach is quite limited. In the peptide field a much Me-dihydroBmt larger reservoir of structures seems to be Me2a3h4mOA available. Using a double recombination apMeLeu proach STACHELHAUS al. (1995, 1996b) et MeCHA succeeded in replacing the last amino acid D H i v ~ M e L e u ~ L e u ~ M e l l a l - r M e B ~ module of surfactin synthetase by substituting J(3) the terminal leucine domain with bacterial orAlatMeLe~t~LeutMeLeutMeGIy+~Thr nithine and phenylalanine domains and with fungal cysteine and valine domains. This apAbU+ Leu 3hNva proach offers exciting opportunities of varyAbu vinylGly ing residues within a given structural type. FiNva Nva Nle nally, even new peptide structures should be CYS available from designed enzyme systems.
+ molecular mass by FAB-MS

Fig. 6. Continued.
substitutions. It has been shown that rates are roughly additive and overall rates decrease drastically when poor substrates are combined. In the evaluation of properties of cyclosporin and SDZ213-125 analogs a large number of cyclopeptides have been produced with these giant multienzymes in the mg range. A comparison revealed that the peptolide synthetase had a much more restricted substrate profile (LAWEN and TRABER, 1993). In v i m investigation of enzyme properties is essential for the efficient utilization of substrate feeding and the evaluation of structural alterations of enzymes. Genetic approaches Metabolic engineering could approach both substrate pools and composition as well as the substrate selection properties of enzymes involved. Should the latter approach attempt to change the amino acid sequence of the considered peptides the term reprogramming has been introduced. The idea of reprogramming is to exchange biosynthetic modules in order to arrive at new structures. This exchange has been demonstrated in several cases for the polyketide forming systems. As polyketide building blocks are restricted to acetate, propionate, or butyrate residues which may be modified by reductions and de-
4 Future Prospects
4.1 Peptides of Ribosomal Origin
Impressive examples of antimicrobial properties of peptides of ribosomal origin have been demonstrated. So far uses have been limited to topical applications (a magainin analog is in clinical trial for cutaneous application in skin and eye infections), and no data are available on the uptake in tissues. An obvious limitation seems to be the degradation, especially by trypsin-like proteases. Promising fields of research include the use of precursor peptides and sequence alterations to improve proteolytic stability. Efforts for the exploitation of peptide structures are justified by the ease of engineering these peptide structures compared to enzymatic systems and research has been directed towards multidrug resistant bacteria, sepsis, and even anticancer drug development (BOMAN, 1995). A fascinating area of research aims at the in vivo production of peptide factors to counteract insect-borne diseases like malaria, trypanosomiasis, and filariasis. It has been proposed to modify the insect immune system of the respective vectors or symbionts in order to eliminate the parasites (HAMet al., 1994).
5 Compilation of Compounds
299
retrieved best from the available data banks (Chemical Abstracts; Kitasato Microbial Origin Chemistry Database, Usako Corp., Tokyo; Antibase, Database, Chemical Concepts, HeiPeptides of enzymatic origin continue to delberg). Data on ribosomal peptides are easemerge from various screening programs. ily accessed from the standard gene banks. New promising sources are, e.g., marine or- For retrieval of additional information the ganisms like sponges as a rich source of pep- given names of compounds, organisms, autidic structures, harboring various kinds of thors, and years of publication are sufficient. Peptides of ribosomal origin have been microorganisms. The exploitation of such microbes that are often surface-associated has grouped in Tab. 5. This compilation illusjust begun and is mainly hampered by the trates the historically grown pathways of the studies which originated mainly in the insect lack of experience in cultivation. The extensive understanding of biosynthet- field, then achieved spectacular results with ic reactions permits various structural modifi- frog peptides, and now identify many antimications to be introduced into the genetic crobial peptides from mammalian sources. background, thus facilitating the biotechno- For details, the reader is referred to recent reet logical production of compounds. New at- views (HANCOCK al., 1995; LEHRERet al., et 1995). Mitempts will be made by the implementation of 1993; BOMAN al., 1994; BOMAN, strategies from ribosomal antibiotic induction crobial peptides of ribosomal origin are the approaches to various microbial, plant, and subject of Chapter 8 on lantibiotics (JACKet animal sources. A lot can be learned from the al., this volume) while the fields of neuropepinteraction of organisms in ecosystems, and tides and hormones are not within the scope genome sequencing will certainly enhance of the chapter. Peptides of nonribosomal origin are listed this understanding considerably. The advancing analytical techniques will permit a more in Tab. 6. A number of observations can be rapid access to even more complicated natu- made concerning the sources especially suited ral products. The production of such com- for peptides. Thus the actinomycetes are still pounds will remain a challenge for biotechnol- the leading group for new products, and here extensive variations in the structural types of %Y* compounds are found. The groups of dalbaheptides and thiopeptides in particular have benefitted from directed screening approaches. Bacilli have received less attention, but the potential of various cyclopeptides from well-known strains and also from marine Bacilli with slightly deviating structures has not yet been explored in depth. A surprisIt is the purpose of this compilation covering group are the blue-green algae, mainly ing most peptides and related structures pub- cyanobacteria, which show an impressive varlished during the last five years (1992-1996) iability of peptide structures with various to illustrate trends in sources, screening tar- properties. This variability implies frequent gets, variability of structures within groups of horizontal transfer of biosynthetic genes, a organisms, and to address some problems of promising field to be exploited in order to source identification and stability. This list make use of natural combinatorial apupdates and extends the table of peptides proaches. Fungi continue to provide interestpresented in Vol. 4 of the First Edition of ing compounds, most peptides being rather Biotechnology (KLEINKAUFand VON small. Large fungal peptides are mainly asDOHREN,1986). A more extensive compila- sembled in the family of peptaibols. Plant peptides of presumably nonribosomal tion of peptide structures including publications until 1989 can be found in VON DOH- origin often are not modified and could as well originate from ribosomal precursors by REN (1990). Data on specific compounds are
4.2 Peptides of Nonribosomal
5 Compilation of Compounds
Tab. 5. Compilation of Peptides of Ribosomal Origin Source Sequencea Properties Reference
Peptide
Microbial Sources Leukocin A Leuconostoc gelidum Ual 187 Nisin Lactococcus lactis Staphylococcus epidermidis Lactobacillus plantarum Rhizomucor pusillus Bacillus subtilis GKNGCK AY SLQMGATAIKQVKKLFKKW
HASTING al. (1991 et HURST(1981) KALETTA al. (198 et
KYYGNGVHCTKSGCSVNWGEAFSAGVHRLANGGNGFW ITSISLCTPGCKTGALMGCNMKTATCHCSIHVSK
Pep-5
TAGPAIRASVKQCQKTLKATRLFTVSCK-
Plantaricin A Sillucin
antibacterial (G -) antibacterial (G*) antibacterial (G*) antibacterial
ACLPNSCVSKGCCCGBSGYWCRQCGIKYTC
Subtilin
MSKFDDFDLDVVKVSKQDSKITPQWKSESLCTPGCVTGALQTCFLQTLTCNCKISK
antibacterial (G*) antibacterial
NISSEN-MEYER al. et (1993) BRADLEY SOMand kuti (1979) BANERJEE and HANSEN (1988)
Plants AFPl Rape (Brassica napus)
QKLCERPSGTWSGVCGNNNACKNQCINLEKARHGSCNWFPAHK QKLCERPSGTXSGVCGNNNACKNQCIR VGECVRGRCPSGMCCSQFGYCGKGPKYCGR
antifungal antifungal antibacterial, antifungal
TERRAS al. (1992) et
AFP2 Ac-AMP
Crambin
TTCCPSIVARSNFNVCRIPGTPEAICATYTG-
TERRAS al. (1992) et BROEKAERT al. et (1992) TEETERet al. (1981) antifungal DUVICK al. (1992) et CAMMUE al. (1992 et QKLCERPSGTWSGVCGNNNACKNQCINLETERRAS al. (1990) et
MBP-1
Turnip (Brassica rapa) Amaranth (Amaranthus caudatus) Crambe plants (Crambe abyssinica) Maize (Zea mays) Mirabilis lalapa Radish (Raphanus sativus)
Mj-Amp1 (2 types)
CIIIPGATCPGDYAN RSGRGECRRQCLRRHEGQPWETOECMRRCR QCIGNGGRCNENVGPPYCCSGFCLRQPGQYGYCKNR
Rs-AFP1
antibacterial (G*); antifungal antibacterial
Insects Abaecin
Honey bee (Apis mellificu) YVPLPNVPQPGRRPFPTFPGQGPFNPKIKQP QG Y LINPK GNNRPVYIPQPRPPHPRI
antibacterial
CASTREELS al. et (1990) antibacterial antibacterial, cytotoxic antibacterial, cytotoxic antibacterial cytotoxic antibacterial antibacterial (G+)
Andropin
VFIDILDKVENAIHNAAQVGIGFAKPFEKWNPFKELERAGQRVRDAVISAAPAVATVGQAAA IAR G IKITTMLAKLGKVLAHVa
Fruit fly (Drosophilu melunoguster) Apedaecin Honey bee (Apis rnellificu, IA lymph fluid) Bactericidin Tobacco hornworm (Munducu B2 sextu, larvae hemolymph) Bombolitin Bumblebee (Megubombus pennsylvunicus, venom) Cecropin Silk moth (Bombyx mori) Hornet (Vespu crubro, venom) QAATI FLPLILRKIVTALa
(G-)
RWKIFKKIEKVGQNIRDGIVKAGPAVAVVG-
CASTREELS al. et (1989) DICKINSON al. et (1988) ARGIOLAS and PISANO (1985) Qu et al. (1987) ARGIOLAS and PISANO (1984) BULETet al. (1993) BULETet al. (1992) TESHIMA al. (1987 et
Crabrolin
Drosocin
GKPRPYSPRPTSIHPRPIRV
Fruit fly (Drosophilu melunoguster) Dragonfly (Aeschnu cyuneu, larvae) Silkworm (Bombyx rnori)
GFGCPLDQMQCHRHCQTITGRSGGYCSGPLKLTCTCYR
Insect defensin Lepidopteran C Mastoparan Wasp (Vespulu lewisii, venom) Bee (Apis rnellificu, venom) Blowfly (Phormiu terrunovu) Bee, royal jelly (Apis mellificu) Fleshfly (Surcophugu peregrinu) Fleshfly (Surcophugu peregrinu)
RWKLFKKIEKVGRNVRDGLIKAGPAIAVIGQAKSL INLKALAALAKKIL
GIGAVLKVLTTGLPALISWIKRKRQQ
Melittin
BERNHEIMER and RUDI(1986) TOSTESON and TOSTESON (1984)
ATCDLLSGTGINHSACAAHCKLRGNRG-
LAMBERT al. (198 et GYCNGKGVCVCRN
Phormicin A (B) Royalisin
VTCDLLSFKGQVNDSACAANCLGKAGGHCEKGVCICRKTSFKDLWDKYF
Sapecin
ATCDLLSGTGINHSACAAHCLLRGNRG-
antibacterial (G+) antibacterial, cytostatic, antifungal antibacterial (G+) antibacterial (G+) antibacterial GYCNGKAVCVCRN
FUJIWARA al. (199 et
Sarcotoxin
GWLKKIGKKIERVGQHTRDATIQGLGIAQQAANVAATARa
antibacterial
HANZAWA al. et (1990) OKADA NATOR and
Tab. 5. Continued
Peptide RRWCFRVCYRGFCYRKCRa KWCFRVCYRGICYRRCR
Source
Sequencea
Properties antibacterial antibacterial; antifungal
Reference LAMBERT al. (198 et NAKAMURA al. et (1988) BONTEMS al. (199 et
PolypheHorseshoe crab (Limulus musin polyphemus) Tachyplesin Horseshoe crab (Limulus (3 types) polyphemus) Toxin 2 Sahara scorpion (Androctonus australis Hector)
VKDGYIVDDVNCTYFCGRNAYCNEECTKLKGESGYCQWASPY GNACYCKLPDHVRTKGPGRCH AMPMLRLa AAPKFVGRRGAPYFV GAPAFV
Molluscs CARP Mytilus edulis
HIRATA al. (1987) et IKEDAet al. (1992) IKEDAet al. (1992)
MIP Land snail (Achatina fulica) (10 types) MIP Land snail (Helix pomatia) (12 types)
catch-relaxing peptide (muscle) contraction inhibitor contraction inhibitor
Frogn oads GLWSKIKEVGKEAAKAAAKAAGKAALGAVSEAV GIGALSAKGALKGLAKGLAEHFANa antibacterial, antifungal anibacterial DALYet al. (1992)
Adenoegulin Bombinin
Two-colored leaf frog (Phyllomedusa bicolor) Yellow-bellied toad (Bombina variegata) Asian toad (Bombina orientalis) European frog (Rana esculenta)
GIGASELSAGKSALKGLAKGLAEHFANa
FLPLLAGLAANFLPKIFCKITRKC
BLP-1 (4 types) Brevinin
antibacterial antibacterial, cytotoxic
CSORDAS MICH and (1970) GIBSON al. (1991) et SIMMACO al. (199 et
Dermaseptin 1 (6 types) Esculentin

ALWKTMLKKLGTMALHAGKAALKAAADTISQGTQ
Sauvages leaf frog (Phyllomedusu suuvugii) European frog (Runa esculentu)

Xenopus luevis, skin Xenopus luevis, skin Xenopus luevis, stomach
antibacterial, antifungal antibacterial
MOR et al. (1991) SIMMACO al. (199 et ZASLOFF (1987) KUCHLER al. (198 et MOOREet al. (1991) CLARK al. (1994) et antibacterial SURES and CRIPPA ( 1984)
Magainin I
GIFSKLGKKIKNLLISGLKNVGKEVGMDVVRTGIDIAGCKIKGEC GIGKFLHSAGKFGKAFVGEIMKS GMASKAGAI AGKI AKVALKAL GVLSNVIGYLKKLGTGALNAVLK FLGGLIKIVPAMICAVTKKC GWASKIGQTLGKIAKVGLKELIQPK
PGLa PGQ
Ranalexin
anibacterial, antifungal antibacterial antibacterial; antifungal antibacterial
XPF
Bullfrog ( R u m cutesbeiunu, skin) Xenopus luevis, skin
Snakes Toxin 1 Waglers pit viper (Trimeresurus wugleri, venom)
GGKPDLRPCHPPCHY IPRPKPR
SCHMIDT al. (1992 et
Mammals Bactenecin BacS bovine neutrophils bovine neutrophils
RLCRIWIRVCR RFRPPIRRPPIRPPFYPPFRPPIRPPIFPPIRPPFRPPLRFP
VRNHVTCRINRGFCVPIRCPGRTRQIGTCFGPRIKCCRSW
antibacterial antibacterial (G-) antibacterial
ROMEO al. (1988) et FRANK al. (1990) et SELSTED al. (1993 et
BNBD-2 bovine neutrophils (13 types) Cecropin P1 Pig (Susscrofu, small intestine)
SWLSKTAKKLENSAKKRISEGIAIAIQGGPR
antibacterial (G-) antibacterial antibacterial
LEE et al. (1989) SELSTED al. (192) et AGERBERTH et al. (1993) AGERBERTH al. et
Cryptdin Mouse (Mus musculus, (5 types) intestine) Endozepine Pig (Susscrofu domesticu) Pig (Sus scrofu domesticu)
Gastric
LRDLVCYCRSRGCKGRERMNGTCRKGHLLYTLCCR KQATVGDINTERPDILDKGKAKWDAWNGLKGTSKEDAMKAYINKVEELKKKYGI ISDYSIAMDKIRQQDFVNWL-
Tab. 5. Continued
Peptide
Macaca fascicularis
Source GYRSNYLYDN
Sequencea
Properties antifungal
Reference
Hiastadin Human neutrophils Bovine neutrophils Bovine (N-terminal of lactoferrin) Rabbit (Oryctolagus cuniculus, macrophage; neutrophils) Pig (Susscrofa dornestica) Rat ( R a m s norvegicus)
DSHEERHHGRHGHHKYGRKFHEKHHSHRACYCRIPACIAGERRYGTCIYQGRLWAFCC
Xu et al. (1990)
LEHRER al. (1991) et SELSTED et al. (1992) BELLAMY al. (199 et antibacterial, antifungal, cytotoxic antibacterial SELSSSTED al. et (1983) antibacterial, antifungal, cytotoxic antibacterial antibacterial
HNP-1 (6 types)
ILPWKWPWWPWRR FKCRRWQWRMKKLGAPSITCVRRAF WCACRRALCLPRERRAGFCRIRGRIHPLCCRR

RADTQTYQPYNKDWIKEKIYVLLRRQAQQAGK VICYCRRTRCGFRERLSGACGYRGRIY RLCCR MERSTRELCLNF'IVVLITVILIWLLVRSY QY
Indolicidin Lactoferricin MCPI (2 types; NP-1 (5 types) Peptide 3910 RatNP-1 (4 types) Sarcolipin Rabbit (Oryctolagus cuniculus, skeletal muscle) Bovine (Bos taurus, seminal plasma) Bovine (Bos taurus, tracheal mucosa)
Seminalplasmin
SDEKASPDKHHRFSLSRYAKLANRLANPKLLETFLSKWIGDRGNRSV
AGERBERTH et al. (1993) EISENHAUER al. et (1989) WAWRZYNOW al. et (1992) REDDY and BHARGAVA (1979)
TAP
NPVSCVRNKGICVPIRCPGSMKQIGTCVGRAVKCCRKK
antifungal, antibacterial proteolipid-like, ionophore? antibacterial, antifungal, cytotoxic antibacterial, antifungal
DIAMOND al. (199 et
a Only plain sequences are given, no disulfide links are indicated: a: C-terminal amidated; most accession numbers of the gene sequences can b retrieved from HANCOCK al. (1995). et
Tab. 6. Compilation of Peptides of Nonribosomal Origin (Bacteria/Fungi/Plants/Animals) Organism Structural Type" AA-M R-L-6 L-8 Properties Reference
Compound
Bacteria - Bacilli FR901537 Fusaricidin N-4909 Bacillus sp. Bacillus polymyxa Bacillus sp.
Halobacillin Isohalobacillin B Surfactins Surfactin-like Bacillopeptin Cereulide Homocereulide Bacitracins Bacillus sp. (marine) Bacillus sp. Bacillus subtilis natto Bacillus pumilus (marine) Bacillus subtilis Bacillus cereus Bacillus cereus (marine) Bacillus sp.
antitumor antibacterial stimulates apolipoprotein E secretion cytotoxic ACAT inhibitor antifungal (phytopathol.) K +-complex extremely cytotoxic antibacterials
OOHATA al. (1995) et KAJIMURA KANEDA and (1996) HIRAMOTO al. (1996) et TRISCHMANN (1994) et al. HASUMI al. (1995) et OKAet al. (1993) KALINOVSKAYA al. (1995) et KAJIMURA al. (1995) et SUWAN al. (1995) et WANGet al. (1995) IKAI al. (1995) et
L-8(2) L-8 L-S(2) L-8(2 C-8(2) C-D-12 C-D-12 P-12-C-7M(2) P-2 R-P-2-M C-2-M C-2-M R-P-2-M R-P-3 R-P-3-M(U)
Actinomycetes Monamidocin Streptomyces sp.
KAMIYAMA al. (1995) et

OGITA al. (1992) et
Matylstatin
PhevaIin Thaxtomin Cutinostatin Amonobactins Napsamycin
Actinomadura atramentaria Streptomyces sp. Streptomyces scabies Actinomycete Aeromonas hydrophila Streptomyces sp. Streptomyces halstedii Streptomyces sp.
ALVAREZ al. (1995) et GELINet al. (1993) HIGASHI al. (1996) et TELFORD al. (1994) et CHATTERJEE al. (1994) et R-P-3 R-P-3-M HIRAOet al. (1995) OMURA al. (1993) et
Nerfilin I Pepticinnamins
fibrinogen receptor antagonist type IV collagenase inhibitor calpain inhibitor phytotoxin cutinase inhibitor siderophore antibacterial (Pseudomonas) neurite outgrowth inducer farnesyl-protein
Tab.6 Continued .
Organism Structural Type" Properties HIV-1 protease inhibitor P-4-M(U) P-4-M P-4-M P-4-M R-D-4 R-P4-M(U) (R-P-4)2-M (R-P-4)2-M P-5-(CO) P-5-M R-P-5 (R-p-5)~ R-P-5-AA-M c-5 Streptomyces misakiensis Streptomyces aurantiacus Coleophoma emperri Streptomyces cummerosporus c-5 C-6 R-C-6(4) R-L-6 HIV-1 protease inhibitor calpain inhibitor HIV-1 protease inhibitor aPase inhibitor Reference STELLA al. (1991) et KANETO al. (1993) et ALVAREZ al. (1994) et STEFANELLI al. (1995) et CHUNG al. (1996) et MURAKAMI al. (1996) et ISONO al. (1993) et BLUMet al. (1995) OKADA al. (1994) et AOYAGI al. (1991) et KAMIYAMA al. (1994) et MATSON al. (1993) et FUKUCHI al. (1995) et MIYATA al. (1992) et endothelin receptor antagonist enothelin binding inhibitor antibacterial, cytotoxic antifungal antibacterial. antitumor KOJIRIet al. (1991) GRAFEet al. (1995) IWAMOTO al. (1994) et HANADA al. (1992) et
Compound
~Y-MAPI
Mer-N5075A
MTMTLA GE20372 MR-387
Formobactin
Streptomyces sp. Micromonospora sp. Nocardia sp. Streptomyces chromofuscus Streptomyces griseus Streptomyces sp. Streptomyces neyagawaensis Nocardia sp.
Muredomycins Echinoserine BE-22179 Poststatin Streptomyces flavidovirens Streptomyces tendae Streptomyces gangtokensis Streptomyces viridochromogenes Streptomyces filipinensis Streptomyces willmorei
Cyclot hialidine WS 1279
Sandramycin Rotihibin
radical scavenger, neuronal protecting antibiotic antibacterial (weak) topoisomerase I1 inhibitor pro-endopeptidase inhibitor DNA gyrase inhibitor bone marrow growth stimulating antitumor plant growth regulator
ws-7338
Nocardioides sp. Streptomyces graminofaciens Streptomyces sp.
BE-18257 Aurantimycin WF11899AIBIC Protactin
Salinamide IClOl Streptomyces sp. Streptomyces bottropensis Microbispora sp. Streptomyces lavendofoliae P-6(7)-M P-7-C-4-M P-7-L-5 R-P-7 R-L-7
Streptomyces sp. (marine) Streptomyces sp.
R-L-6-M R-L-6-M
RPI-856 Bottromycin A2 Cochinmicins Piperastatin A
anti-inflammatory extracellular matrix antagonist HIV protease inhibitor endothelin antagonist sprine carboxypeptidase inhibitor tachykinin antagonist antiviral antibacterial antibacterial antibacterial gp120-CD4 binding inhibitor tachykinin inhibitor
TRISCHMANN al. (1994) et UENOet al. (1993) ASANOet al. (1994) KANEDA (1992) LAMet al. (1992) MURAKAMI al. (1996) et SHIGEMATSU al. (1993) et NARUSE al. (1993) et Box et al. (1991) TAKEUCHI al. (1992) et NADKARNI al. (1994) et MATSUZAKI at. (1994) et
WS9326A
Kistamicins MM 5526618 Saccharothrix sp. Amycolatopsis sp. Streptomyces sp.
Streptomyces violaceusniger Microtetraspora parvosata Amycolatopsis sp.
Galacardins
Balhimycin
Chloropeptins
P-7-M R-P-7-M (GLYC) R-P-7-M (GLYC) R-P-7-M (GLYC) P-7-M R-C-7 C-8 (W2 R-(P-5)2-M
WS9326A
HAYASHI al. (1992) et antibacterial antitumor HIV-1,2 reverse transcriptase inhibitor antibiotic tip A promoter inducing SELVA al. (1996) et LEET et al. (1996) LINGHAIN al. (1996) et MINAMI al. (1994) et YUNet al. (1994)
A21459 Himastatin
Quinoxapeptin
Streptomyces violaceusniger Actinoplanes sp. Streptomyces hygroscopicus MA7095 (nocardioform) Streptomyces sp. Streptomyces sp. Streptomyces bernensis Streptomyces sp. Streptomyces sp.
Cypemycin Promotiocin
Berninamycin
LAU and RINEHART (1994) tip A promoter inducing antibacterial YUNet al. (1994) FERRARI al. (1995) et
Geninthiocin
GE37,468
P-11-M P-13-C-10 [PYRI-M P-14-C-12 [PYR]-M P-15-C-13 [PYRI-M P-15-C-10-M
Tab. 6 Continued .
Organism Streptomyces gardneri Streptomyces sp. Structural Type" Properties Reference DEBONO al. (1992) et YUNand SETO (1995)
Compound
A10255BlGlJ
Promoinducin
P-1742-14 [PYRI-M P-18-C-14 [PYRI-M
Cyanobacteria
Aeruginosin Aeruginosin 298A Nostoc muscorum Nostoc sp. Lyngbya majuscula Microcystis sp. Microcystis aeruginosa Lyngbya majuscula Nodularia spumignea Anabaena flos-aquae Nostoc sp. Westiellopsisprolifica Oscillatoria agardhii Microcystis aeruginosa Oscillatoria agardhii Nostoc sp. Microcystis aeruginosa
Microcystis aeruginosa Microcystis aeruginosa
R-P-3 P-4
trypsin inhibitor thrombin and trypsin inhibitor
MURAKAMI al. (1995) et MURAKAMI al. (1994) et NAGATSU al. (1995) et ichthyotoxic angiotensin converting enzyme inhibitor immunosuppressive BARROW al. (1995) et TAKIZAWAal. (1995) et CHOIet al. (1993) OKINO al. (1993) et
Muscoride A
Cryptophycin Antillatoxin Linear peptides Microginin
P-4-M (TJ=P) L-4(4) L-4-M(4) R-P-4(2,3) P-5 R-P-5-M C-5(2,3) P-6-C-5(4) C-6 C-6-M C-6-M(4) R-L-6-M C-7-M(2) R-P-7-L-5-M R-P-7-L-6-M
Microcolins Nodularins Anabaenopeptins Nostocyclamide
Westiellamide Oscillamide Micropeptin 90 DHB-microcystin-RR Nostocyclin
antialgal, anticyanobacterial cytotoxic (moderate) chymotrypsin inhibitor plasminltrypsin inhibitor
KOEHNet al. (1992) NAMIKOSHI al. (1994) et HARADA al. (1995) et TODOROVAal. (1995) et PRINSEP al. (1992) et SANOand KAYA(1995) ISHIDA al. (1995) et SANOand KAYA(1995) KAYAet al. (1996) protein phosphatase-1 inhibitor cell differentiation TSUKAMOTO (1993) et al.
Microcystilide A
Aeruginopeptins Calophycin Puwainaphycins Laxaphycins Hormothamnin Schizotrin Microviridins

Bacteria - Various Epothilon Fluvibactin WS75624
Microcystis aeruginosa Calothrix fusca Anabaena sp. Anabaena laxa Hormothamnion enteromorphoides Schizothrix sp. Microcystis aeruginosa
R-P-8-L-6-M C-10-M R-C-lO(2)
c-11
fungicidal cardioactive antifungal antimicrobial elastase inhibitor
c-11
R-P-13-C1W) R-P-14(C-4L-4)
HARADA MOONet a GREGSO FRANKM GERWIC
PERGAM OKINO et
Mucor hiemalis Vibrio fluvialis Saccharothrix sp. Pseudomonas sp. (marine) Flexibacter sp. Methanobacterium thermoautotrop hicum Lactobacillus sp. Micromonospora sp. Mycobacterium senegalense Polyangium sp. Chromobacterium violaceum Polyangium sp. Alteromonas luteoviolacea Azomonas macrocytogenes Flexibacter sp. Myxococcus virescens Streptosporangium amethystogenes Pseudomonas reactans
AA-M(PK) R3-A R-P-2-M c-2 P-3-M R-P-3-M c-4 L-4-M R-P-4-M (GLYC) R-P-4-M C-R-P-4-M R-PJ-M R-P-6-L-5 R-P-7-M P-7-L-5 R-L-6 R-S-R -P-7 R-P-9-L-6
CI-4 TAN-057 Methanofurans Cyclotetrapeptides Rakicidins Glycopeptidolipid antigen Thiangole FR-901,228 Thiangazole Aletrobactin A Azoverdin YM4714112 Myxochromide A TAN-1511 WLIP
antifungal, cytotoxic siderophore endothelin converting enzyme inhibitor chitinase inhibitor antibacterial ( S . aureus methicillin resistant) natural cofactor metal binding, melanin inhibitor; tyrosinase inhibitor cytotoxic HIV-1 inhibitor antitumor siderophore siderophore elastase inhibitor induction of cytokines white line inducing principle
GERTHet YAMAMO TSURUM
IZUMIDA e KATAYA
SULLION
KURANA et al. (19
MCBRIEN LOPEZMA
BOYCEet LI et al. (1
JANSEN et DENGet a LINGET et ORITAet TROWITZ TAKIZAW
HANet al.
Tab. 6. Continued
Compound Cystobacter violaceus Pseudomonas syringae Pseudomonas jluorescens Pseudomonas sp. unidentified gram-negative
Organism
Structural Type"
Properties antifungal, cytotoxic phytotoxin siderophore antibacterial
Reference SCHUMMER al. (1996) et FUKUCHI al. (1992) et TSUBOTANI al. (1993) et UI et al. (1995) FUKAI al. (1995) et
Vioprolides Syringomycin Ferrocins Pholipeptin L-9-M(2) R-L-9 R-L-10-M R-P-1l-L9(2) R-P-12-L-11
Bu-2841
Fungi Beauveria bassiana Microascus longiirostris Chromelosporium fulvum Aspergillus fumigatus Talaromyces Aspergillus sp. Thielavia minor Nannizia gypsea Aspergillus fumigatus C-R-AA R-AA-A R-AA-M R-AA P-2 P-2-M C-2-M KAGAMIZONO al. (1995) et platelet aggregation inhibitor cysteine protease inhibitor thiol protease inhibitor palmitoyl-transferase inhibitor insecticidal superoxide anion generation inhibitor ACAT inhibitor cell cycle inhibitor C-2-M C-2-M (TERP) (C-2)~-M (C-2)z-M C-3-M L-3-M(2,5) P-3-C-2-M R-P-3-M Aspergillus jlavus Gliocladium sp. Penicillium sp. Ustilaginoidea virens Penicillium citreo-viride Penicillium fellutanum (marine) substance P inhibitor oncogene inhibitor substance P inhibitor antimitotic, cancerostatic nerve growth factor promoter Yu et al. (1996) MORISHITA al. (1994) et MANDALA al. (1994) et POPPet al. (1994) KITAet al. (1994) SHINOHARA al. (1994) et CUIet al. (1996) BARROW SEDLOCK and (1994) CHUet al. (1995) SUNet al. (1994) KOISOet al. (1994) MATSUNAGA al. (1991) et SHIGEMORI al. (1991), TSUJIet al. et (1994)
Bassiatin
Cathestatin AM4299A Lipoxamycin
NK-374200 WIN 64821 OPC-15161
Gypsetin Tryprostatins
PEDT Sch529OOll Benzomalvins Ustiloxins Citreoindole Fellutamide
Stevastelins Beauvericins YF-044P-D F-165 (Fungi Imperfecti) Hapsidospora irregularis Aspergillus niger Pithomyces chartarum Acremonium sp. Zalerion arboricola Aspergillus sydowii Aschersonia sp. Metarhizium anisopliae R-C-6(3) L-6-M L-6 R-C-6(3) L-5-M-M(2) C-5-M C-D-5 R-P-5-L-4(3) c-5
Penicillium sp. Beauveria bassina Candida albicans
L-4(2) C-D-4-M R-P-5
MORINO al. (1996) et GUPTAet al. (1995) SATOet al. (1994) INOUE al. (1996) et HAMANO al. (1992) et KIMet al. (1993) MOUSSA LE QUESNE and (1996) HOCHLOWSKI al. (1994) et
Plactin
Leualacin Malformin Pithomycolide Aselacins
immunosuppressant insecticidal aspartic proteinase inhibitor stimulates fibrinolytic activity Ca*+-blocker phytotoxic endothelin receptor antagonist antifungal antifungal insecticidal suppresses hepatitis B virus surface antigen production nematicidal neurokinin antagonist antihelmintic phytotoxic? antibiotic
Pneumocandins
Deoxymulundocandin Destruxin A415 Desmethyldestruxin perfect fungus Fusarium sp, Fusarium avenaceum Aspergillus sp. Mycelia sterilia Bipolyris zeicola Bipolyris zeicola Paecilomyces sp.
HENSENS al. (1992), MORRIS al. et et (1994) MUKHOPADHAY al. (1992) et KRASNOFF al. (1996) et CHENet al. (1995) KAWAZU al. (1993) et TOMODA al. (1992) et KASTICet al. (1992) BARROW al. (1994) et
Bursaphelocides Enniatins D-F Enniatin A2 WIN 66306
PF1022 BZR-cotoxin IV BZR-cotoxin 1/11 Leucinostatins, P-168
L-6-M C-D-6 C-D-6 C-7-M (TERP) C-D-8 L-8-M C-D-9-M R-P-9-M R-P-10-M P-10-M P-10-M
Helioferins LF'237-F8 Trichogin A
antifungal antibiotic Mycogone rosea Tolypocladium geodes Trichoderma longibrachiatum Stachybotrys chartarum Petriella sordida C-11-M L-13-M immunosuppressive antifungal
SCHERKENBECKal. (1995) et UEDA et al. (1995) UEDA et al. (1992, 1994) KUWATA al. (1992), ISOGAI al. et et (1992) GRAFEet al. (1995) TSANTRIZOSal. (1996) et AUVIN-GUETTE al. (1992) et SAKAMOTO al. (1993) et LEE et al. (1995)
FR901459 Petriellin A
Tab. 6. Continued
Organism Hypocrea peltata Trichoderma polysporum R-P-19-M R-P-19-M Structural Type Properties antibiotic ion channel forming Reference MATSURA al. (1993, 1994) et IIDAet al. (1993), NAGAOKA al. et (1994) MURATA al. (1995) et BANERJEE al. (1993) et BARBONI al. (1994) et AUVIN al. (1996) et antitumor
Compound
Hypelcins Trichosporin B
Plants p-Coumaroyl-tryptophan Aurantiamide R-AA R-P-2-M R-P3-M(E) R-P-CM(C)
Cyclopeptide alkaloids
Mucronine J
Astin A Astins D, E Astin J Asterinins A-C Asterinins D, E Astericins D-F Astins Astins Pseudostellarins A-C
C-5(2) c-5 P-5 P-5 P-5-M P-5 C-S-(U) C-5(2)
antitumor
c-5
R-P-4-M C-6 C-6 C-6/C-5
tyrosinase inhibitor
MORITA al. (1993) et MORITA al. (1993) et MORITA al. (1995) et CHENG al. (1994) et CHENGet al. (1996) CHENG al. (1996) et MORITA al. (1996) et MORITA al. (1994) et MORITA al. (1994) et AUVIN al. (1996) et ZHAOet al. (1995) MORITA al. (1996) et MORITAet al. (1995)
Mucronine J
Stellarins B, C
Dichotomins Segetalin A/B-D
Coffea canephora Piper aurantiacum Clematis hepaulensis Zizyphus mucronata (roots) Zizyphus mucronata (root, bark) Aster tataricus Aster tataricus Aster tataricus (roots) Aster tataricus (roots) Aster tataricus Aster tataricus (roots) Aster tataricus Aster tataricus Pseudostellaria heterophylla Zizyphus mucronata (bark) Stellaria yunnanensis (roots) Stellaria dichotoma Vaccaria segetalis
Cyclogossine A c-7 C-8 C-718 c-9 C-6 C-6 c-7 c-7 P-8-C-5(CN) c-9
HORSTEN al. (1996) et ZHAO al. (1995) et MORITAet al. (1995) BERG et a]. (1996) MORITA al. (1994) et ITOKAWA al. (1993) et
VAN DEN
Stellarins F, G Pseudostellarin DIH
Podacycline A Segetalin A RA VII Stellaria yunnanensis Vaccaria segetalis (seeds) Lycium chinense Leonurus heterophylus
Jatropha gossypifolia (latex) Stellaria yunnanensis Pseudostellaria heterophylla Jatropha podagrica (latex) Vaccaria segetalis (seeds) Rubia cordifolia eukaryotic translation inhibitor
Stellarins D, E Segetalin E Lyciumin A Cycloneoluripeptides
ZHAOet al. (1995) MORITA al. (1996) et MORITAet al. (1996) MORITA al. (1996) et
Sponges Geodiamolide G Cyclotheonamide Microsclerodermins Konbamide Orbiculamide

Cymbastela sp. Theonella sp. Microscleroderma sp. Theonella sp. Theonella sp. Discodertnia sp.
cytotoxic senne protease inhibitor antifungal calmodulin antagonist cytotoxic antifungal (Candida albicans) cytotoxic, antifungal cytotoxic c-7 c-7
COLEMAN al. (1995) et LEE et al. (1993) BEWLEY al. (1994) et SCHMIDT WEINBRENNER and (1996) FUSETANI al. (1991) et GUNASEKERA al. (1994) et KONGet al. (1992) BEWLEY al. (1996) et PETTITet al. (1992) PETTITet al. (1995) PETTITet al. (1993,1995) cancer cell growth inhibitor cytotoxic PETI-ITet al. (1993, 1994), KONATet (1995)
Discobahamins
Pseudoaxinellin Aciculitins Stylostatin Stylopeptide I
L-5-M(7) C-5(C0,3) C-6(2,3) P-6(U)-C-5 R-P-7-C-6M(3) R-P-7-C-6M(3,4) c-7 C-7-M c-7 c-7
Phakellistatin
Axinastatins
Pseudoaxinella massa Acucilites orientalis Stylotella aurantium Stylotella sp.lPhakellia costata Phakellia costata and Stylotella aurantium Axinella spp. Pseudoaxinyssa sp. Didemnum molle
Malaysiatin Cyclodidemnamide
C-7-M
Tab. 6 Continued .
Compound Theonella sp. R-P-8-C-6M(3,CO)
Organism
Structural Type"
Properties superoxide generation response of human neutrophils inhibitor anti-HIV
Reference KOBAYASHI al. (1991, 1995) et D'AURIAet al. (1996) WILLIAMS al. (1993) et ZAMPELLA al. (1996) et ABOU-MANSOUR al. (1995) et ITAGAKI al. (1992) et
Keramamides L-8-C-7-M L-9(3) R-P-9-L-6-M R-P-9-L-7(3)
Callipeltins Majusculamide C Callipeltin A Didemnin B* Callipelta sp. Ptilocaulis trachys Callipelta sp. Trididemnum cyanophorum Theonella sp. Theonella sp. Theonella swinhoei Discoderma sp. Theonella swinhoei Discodermia kiiensis Halichondria cylindrata Theonella swinhoei
Keramamide F
Theonellamides Theonegramide
cytotoxic antifungal
MATSUNAGA FUSETANI and (1995) BEWLEY FAULKNER and (1994) GULAVITA al. (1992) et KOBAYASHI a1 (1994) et antimicrobial antifungal, cytotoxic RYUet al. (1994) LI et al. (1995) HAMADA al. (1994) et
Polydiscamide A
Theonellapeptolide IId
Discodermins F-H Halicylindramides Polytheonamides
R-P-9-C-7M(CO) C-12(2,3) C-l2(2,3)-M (GLYC) R-P-13-L5-M R-P-13-L11(M) R-P-14-L-6 R-P-14-L-6 P-48
Ascidians Bistratamides C/D Patellins
Lissoclinum bistratum Lissoclinum sp. Lissoclinum patella Discoderma kiiensis
C-6-M C-6-M(terp), C-7IC8 C-6-M L-7-M
CNS depressant (D)
FOSTER al. (1992) et CAROLL al. (1996) et cytotoxic cytotoxic ISHIDA al. (1995) et TADAet al. (1992)
Trunkamide A Patellamides Discokilides
Sea Hares
Aurilide Dolastatin H Dolastatins C, D cytotoxic cytotoxic (D)
Dolabella auricularia Dolabella auricularia Dolabella auricularia
C-D-7(5) R-P-4-M R-D-4
SUENAGA al. (1996) et SONEet al. (1996) SONEet al. (1993)
Molluscs
Keenamide Kahalide F Pleurobranchus forskalii Elysia rufescens and Bryopsis sp.
C-6-M(terp) R-P-13-L-5
cytotoxic
WESSON HAMANN and (1996) HAMANN SCHEUER and (1993)
SpiderslScorpions Nephilatoxins Nephilatoxin 7 Clavamine Spidamine
Nephila Nephila Nephila Nephila
clavata clavata clavata clavata
P2-A-AA R-P-ZA P-2-A-P-3 R-P-2-A
neurotoxins neurotoxins insecticidal toxin
MIYASHITA al. (1992) et MATSUSHITA al. (1995) et YOSHIOKA al. (1992) et CHIBAet al. (1996)
~~
Abbreviations used in the description of the structural types are: A amine, AA amino acid, P peptide, C cyclopeptide, L peptidolactone (cyc structure closed with an ester bond), D depsipeptide (compound containing both peptide and ester bonds); the number associated with each str tural type (P-n, C-n, L-n, D-n) gives the number of amino acids or hydroxy acids contained in the structure; branched cyclic compounds are given the total length with the ring size included (P-10-C-5 is thus a branched decapeptide with a pentapeptide cyclic region); acyl compounds, amidati or modifications by aminoalcohols are not included. Additional information is provided by R acylation by aliphatic or aromatic carboxylic acids, modifications (largely N-methylations), (GLYC) glycosylation, (TERP) modification by terpenoids, [PYR] ring formation by a pyridine moi formed by two modified serine residues (thioipeptides), U urea-type of linkage in the peptide chain, CO an additional CO-residue in the pepti chain, E ether bond cyclization, S thioether link, CN direct C-N linkage for cyclization, (n) gives information on cyclization types differing from t common a-carboxyl-a-amino-link; 2, 3, and 4 stand for bonds involving p, y or S positioned amino, carboxylic, or hydroxy groups.
316
yet unidentified cleavage and cyclization reactions. The restricted number of sources indicates that this field investigations have almost begun. Animals as sources have been restricted mainly to marine organisms. The large group of sponges gives rise to the problem of the origin of compounds, since sponges are known to contain large numbers of associated microbes. Recently, the production of a complex sponge metabolite has been traced to an associated actinomycete (BEWLEY et al., 1996). In the future of screenings more attention will be paid to the interactions of organisms. In addition, search for biosynthetic genes involved in peptide formation is a promising pathway to detect, generate, and produce new metabolites.
BAYER, FREUND, JUNG,G. (1995), PostA., S., translational heterocyclic backbone modifications in the 43-peptide antibiotic microcin B17, Eur. J. Biochem. 234,414-426. BENDER,C., PALMER,D., PENALOZAVAZQUEZ, V., A., RANGASWAMY, ULLRICH,M. (1996), Biosynthesis of coronatine, a thermoregulated phytotoxin produced by the phytopathogen Pseudomonas syringae, Arch. Microbiol. 166, 71-75. BERGEY, R., HOWE,G. A., RYAN, A. (1996), D. C. Polypeptide signaling for plant defensive genes exhibits analogies to defense signaling in animals, Proc. Natl. Acad. Sci. 93, 12053-12058. BERNHARD, DEMEL, G., SOLTANI, VON F., K., DOHREN,H., BLINOV, (1996), Identification V. of genes encoding for peptide synthetases in the gram-negative bacterium Lysobacter sp. ATCC 53042 and the fungus Cyclidrotrichum oligospermum, DNA Sequence 6,319-330. BEVINS, L., ZASLOFF, (1990), Peptide from C. M. frog skin, Annu. Rev. Biochem. 59, 395-414. BEWLEY, A., HOLLAND, D., FAULKNER, C. N. D. J. (1996), Two classes of metabolites from Theonelle swinhoei are localized in distinct populaADEFARATI, A., GIACOBBE, A., HENSENS, A. R. tions of bacterial symbionts, Experientia 52,7160. D., TKACZ,J. S. (1991), Biosynthesis of L722. 671329, an echinocandin-type antibiotic proBILLICH, ZOCHER,R. (1987), N-methyltransA., duced by Zalerion arboricola: origins of some of ferase function of the multifunctional enzyme the unusual amino acids and the dimethylmyrisenniatin synthetase, Biochemistry 26, 8417tic acid side chain, J. Am. Chem. SOC. 113,35428423. 3545. A., Y., J. AHARONOWITZ, COHEN,G., MART~N, F. BILLICH, ZOCHER,R. (1990), Biosynthesis of N-methylated peptides, in: Biochemistry of Pep(1992), Penicillin and cephalosporin biosynthetic H., tide Antibiotics (KLEINKAUF, VON DOHREN, genes: structure, organization, regulation, and H., Eds.), pp. 57-80. Berlin: de Gruyter. evolution, Annu. Rev. Microbiol. 46, 461-495. I., T., APARICIO, F., MOLNAR, SCHWECKE, KO- BLANC, GIL,P., BAMAS-JACQUES, LORENJ. V., N., S. L. NIG, HAYDOCK, F., KHAW, E., STAUNA., J., ZON, S., ZAGOREC,M., SCHLEUNIGER, TON,J., LEADLAY, F. (1996), Organisation of P. L., BISCH, D., BLANCHE,F., DEBUSSCHE, the biosynthetic gene cluster for rapamycin in CROUZET, THIBAUT,D. (1997), IdentificaJ., Streptomyces hygroscopicus - analysis of the ention and analysis of genes from Streptomyces zymatic domains in the modular polyketide synthpristinaspiralis encoding enzymes involved in the ase, Gene 169, 9-16. biosynthesis of the 4-dimethylamino-~-phenylalaAWAZU, IKAI,K., YUMAMOTO, NISHIMURA, nine precursor of pristinamycin I, Mol. MicroN., J., K., MIZUTANI, TAKESAKO, KATO, I. S., K., biol. 23, 191-202. (1995), Structures and antifungal activities of BOMAN, G. (1995), Peptide antibiotics and their H. new aureobasidins, J. Antibiot. 48,525-527. role in innate immunity, Annu. Rev. Immunol. BAILEY, M., KERSHAW, J., HUNT, B. A., PAA. M. 13,62-92. I. A. S. TERSON, C., CHARNLEY, K., REYNOLDS, H. E., CLARKSON, M. (1996), Cloning and se- BOMAN, G., MARSH,J., GOODE,J. A. (1994), J. Antimicrobial Peptides. Chichester, New York: quence analysis of an intron-containing domain John Wiley & Sons. from a peptide synthetase of the entomopathoS., M. genic fungus Metarhizium anisopliae, Gene 113, BORCHERT, PATIL, S. S., MARAHIEL, A. (1992), Identification of putative multifunctional 195-197. peptide synthetase genes using highly conserved BALTZ,R. H. (1996), Daptomycin and A54145, in: Biotechnology of Industrial Antibiotics, 2nd Edn. oligonucleotide sequences derived from known synthetases, FEMS Microbiol. Lett. 92, 175-180. (STROHL, Ed.). New York Marcel Dekker. W.,
6 References
6 References
317
BORMANN, C., MOHRLE, V., BRUNTNER,C. (1996), Cloning and heterologous expression of the entire set of structural genes for nikkomycin synthesis from Streptomyces tendae Tu901 in Streptomyces lividans, J. Bacteriol. 178, 12161218. BREIL,B. T., BORNEMAN, TRIPLETT, W. J., E. (1996), A newly discovered gene, TfuA, involved in the production of the ribosomally synthesized peptide antibiotic trifolitoxin, J. Bacteri01. 178,4150456. BUADES, MOYA, (1996), Phylogenetic analyC., A. sis of isopenicillin N synthetase horizontal gene transfer, J. Mol. Evol. 42, 537-542. BURMESTER, HAESE,A., ZOCHER,R. (1995), J., Highly conserved N-methyltransferases as integral part of peptide synthetases, Biochem. Mol. Biol. Znt. 37, 201-207. CHEN,Q., WERTHEIMER, M., TOLMASKY, A. M. J. E., CROSA, H. (1996), The AngR protein and the siderophore anguibactin positively regulate the expression of iron-transport genes in Vibrio anguillarum, Mol. Microbiol, 22, 127-134. CHINERY, COFFEY, J. (1996), Trefoil pepR., R. tides: less clandestine in the intestine, Science 274,204. S., CHONG, P., PODMORE, KIESER, H., REDENP. BACH, M., TURGAY, K., MARAHIEL, A., M. HOPWOOD, SMITH, (1996), Physical idenD., C. tification of a chromosomal locus encoding biosynthetic genes for the calcium-dependent antibiotic of Streptomyces coelicolor A3(2). Abstracts Symp. Enzymol. Biosynth. Nat. Prod., Technical University, Berlin. s., W. CLARK, P., DURELL, MALOY, L., ZASD. LOFF,M. (1994), Ranalexin, J. Biol. Chem. 269, 10849-10 855. COQUE, J. R., ENGUITA, J., MART~N,F., LIJ. F. J. RAS, (1995a), A two protein component 7-cephP. em-methoxylase encoded by two genes of the cephamycin C cluster converts cephalosporin to 7-methoxycephalosporin C, J. Bacteriol. 177, 2230-2235. COQUE, J. R., J. PBREZ-LLARENA, J., ENGUITA, F. F. J., FUENTE, L., MART~N, F., LIRAS,P. J. J. (1995b), Characterization of the cmcH genes of Nocardia lactamdurans encoding a functional 3 hydroxymethylcephem 0-carbamoyltransferase for cephamycin biosynthesis, Gene 162,21-27. COSMINA, RODRIGUEZ, DE FERRA, F., P., F., GRANDI, PEREGO,M., VENEMA, VAN G., G., SINDEREN, (1993), Sequence and analysis of D. the genetic locus responsible for surfactin synthesis in Bacillus subtilis, Mol. Microbiol. 8, 821831. DAHOD,S. K. (1993), Dissolved carbon dioxide measurement and its correlation with operating
I -
parameters in fermentation process, Biotechnol. Progr. 9,655-660. DE CRECY-LAGARD, MARLI~RE, SAURIN v., P., W. (1995), Multienzymatic nonribosomal peptide biosynthesis: identification of the functional domains catalysing peptide elongation and epimerization, C. R. Acad. Sci. Paris 318, 927-936. DIEz, B., GUTI~RREZ, BARREDO, L., VAN S., J. SOLINGEN, VAN DER VOORT, L. H. M., P., MARTfN, J. F. (1990), The cluster of penicillin biosynthetic genes. Identification and characterization of the pcbAB gene encoding a-aminoadipyl-cysteinyl-valine synthetase and linkage to the pcbC and penDE genes, J. Biol. Chem. 265, 16358-16 365. DONADIO, STAVER, J., MCALPINE, B., S., M. J. SWANSON, B., KATZ,L. (1991), Modular orJ. ganization of genes required for complex polyketide biosynthesis, Science 252, 675-679. DSOUZA, NAKANO, M., ZUBER, (1994), C., M. P. Identification of comS, a gene of the srfA operon that regulates the establishment of genetic competence in Bacillus subtilis, Proc. Natl. Acad. Sci. USA 91,9397-9401. ERSPAMER, MELCHIORRI, FALCONIERI-ERV., P., SPAMER, G., NEGRI, L., CORSI, R., SEVERINI, C., BARRA, SIMMACO, KREIL, (1989), D., M., G. Deltorphins: a family of naturally occurring peptides with high affinity and selectivity for 6 opioid binding sites, Proc. Natl. Acad. Sci. USA 86,5188-5192. FAULKNER, J. (1996), Marine natural products, D. Nat. Prod. Rep. 13, 75-125. J. F R ~ C H E T , GUITTON, D., HERMAN,F., D., FAUCHER, HELYNCK, MONEGIER D., G., DE SORBIER, RIDOUX, P., JAMES-SURCOUF, B., J. E., VUILHORGNE, (1994), Solution structure M. of RP-71955, a new 21 amino acid tricyclic peptide active against HIV-1 virus, Biochemistry 33, 42-50. FUJITA, MINAKATA, NOMOTO, FURUK., H., K., KAWA, Y., KOBAYASHIE, (1995), StructureM. activity relations of fulicin, a peptide containing a D-amino acid residue, Peptides 16, 565-568. GEKELER,W., GRILL, E., WINNACKER, E.-L., ZENK,M. H. (1989), Survey of the plant kingdom for the ability to bind heavy metals through phytochelatin, Z. Naturforsch. 44q 361-369. GRILL, E., LOEFFLER,E., WINNACKER, E.-L., ZENK,M. H. (1989), Phytochelatins, a class of heavy-metal binding peptides from plants, are synthesized from glutathione by a specific y-glutamylcysteine dipeptididyl transpeptidase (phytochelatin synthase), Proc. Natl. Acad. Sci. USA 86,6838-6842. GUILVOUT, MERCEREAU-PUIJALON,BONI., O., NEFOY, S., PUGSLEY, P., CARNIEL, (1993), A. E.
318
High-molecular-weight protein 2 of Yersinia enfrom Bacillus subtilis Rb14, J. Ferment. Bioeng. terocolitica is homologous to AngR of Vibrio an76,445450. guillarum and belongs to a family of proteins in- IHARA,M., FUKUMOTO, (1996), Recent proK. volved in nonribosomal peptide synthesis, J. gress in the chemistry of non-terpenoid indole Bacteriol. 175, 5488-5504. alkaloids, Nat. Prod. Rep. 13, 241-261. HAESE, A., SCHUBERT,M., HERRMANN, M., IKAI,K., TAKESAKO, SHIOMI, MORIGUCHI, K., K., ZOCHER, (1993), Molecular characterization R. M., UMEDA, YAMAMOTO, KATO,I., NAY., J., of the enniatin synthetase gene encoding a mulGANAWA, (1991a), Structure of aureobasidin H. tifunctional enzyme catalysing N-methyl-depsiA, J. Antibiot. 44,925-933. peptide formation in Fusarium scirpi, Mol. Mi- IKAI,K., SHIOMI, TAKESAKO, MIZUTANI, K., K., crobiol. 7, 905-914. S., YAMAMOTO, OGAWA,Y., UENO, M., J., HAM,P. J., CHALK, SMITHIES, ALBUQUERR., B., KATO,I. (1991b), Structures of aureobasidins B QUE,c., HAGEN,H. (1994), Antibacterial pepto R, J. Antibiot. 44,1187-1198. tides in insect vectors of tropical parasitic disJACK,R., TAGG,J. R., RAY,B. (1995), Bacterioeases, in: Antimicrobial Peptides (J. GOODE, cins of gram-positive bacteria, Microbiol. Rev. Ed.), pp. 140-150, Ciba Foundation Series. New 59,171-200. York: John Wiley & Sons. H., NIELSEN, J., VILLADSEN, J., HAMOEN, W., ESHUIS, JONGBLOED, VEN- JORGENSEN, L. H., J., MOLLGAARD, (1995a), Metabolic flux distriH. EMA, G., VAN SINDEREN, (1995), A small D. butions in Penicillium chrysogenum during fedgene, designated comS, located within the codbatch cultivations, Biotechnol. Bioeng. 46, 117ing region of the fourth amino acid-activation 131. domain of srfA, is required for competence deJORGENSEN, H., NIELSEN, J., VILLADSEN, J., velopment in Bacillus subtilis, Mol. Microbiol. MOLLGAARD, (1995b), Analysis of penicillin H. 15, 55-63. V biosynthesis during fed-batch cultivations with R. HANCOCK, E. W., FALLA,T., BROWN,M. a high-yielding strain of Penicillium chrysoge(1995), Cationic bactericidal peptides, Adv. Minum, Appl. Microbiol. Biotechnol. 43, 123-130. crob. Physiol. 37, 135-175. K.-D., KELLNER, JUNG, R., HANSEN, N. (1993), Antibiotics synthesized by KALETTA, C., ENTIAN, J. G., REIS,M., SAHL,H.-G. (1989), Pep-5, a new posttranslational modification, Annu. Rev. Milantibiotic: structural gene isolation and prepepcrobiol. 47, 535-564. tide sequence, Arch. Microbiol. 152, 16-19. HECK,S. D., FARACI, S., KELBAUGH, R., W. P. KALLOW, W., VON DOHREN,H., KENNEDY, J., SACCOMANO, A., THADEIO, F., VOLKN. P. TURNER, (1996), Abstract, 3rd Eur. Conf. G. MA", R. A. (1995), Posttranslational amino Fungal Genet., Monster, p. 159. acid epimerization - enzyme catalyzed isomerH., ization of amino acid residues in peptide chains, KELLENBERGER,C., HIETTER, Luu, B. (1995), Regioselective formation of the three disulfide Proc. Natl. Acad. Sci. USA 93,4036-4039. bonds of a 35-residue insect peptide, Peptide HERZOG-VELIKONJA, PODLESEK, GRABB., Z., Res. 8, 321-327. NAR, M. (1994), Isolation and characterization H., H., Y. of Tn917-generated bacitracin deficient mutants KIMURA, MIYASHITA, SUMINO, (1996), Organization and expression in Pseudomonas of Bacillus licheniformis, FEMS Microbiol. Lett. putida of the gene cluster involved in cephalol21, 147-152. sporin biosynthesis from Lysobacter lactamgenus HILTON, D., ALAEDDINOGLU,G., DEMAIN, M. N. YK90, Appl. Microbiol. Biotechnol. 45, 490A. L. (1988), Bacillus subtilis mutant deficient in 501. the ability to produce the dipeptide antibiotic W., I. bacilysin: isolation and mapping of the mutation, KLAPHECK,S., FLIEGNER, ZIMMER, (1994), Hydroxymethyl-phytochelatins (y-glutamylcysJ. Bacteriol. 170, 1018-1021. HODGSON, E., FOSBERRY, P., RAWLINSON, teine),-serine) are metal-induced peptides of the J. A. Pocaceae, Plant Physiol. 104, 1325-1332. N. S., Ross, H. N. M., NEAL,R. J., ARNELL, J. H., E. H. C., EARL,A. J., LAWLOR, J. (1995), Clavu- KLEINKAUF, VON DOHREN, (1982), A surlank acid biosynthesis in Streptomyces clavuligevey of enzymatic biosynthesis of peptide antibiotrus: gene cloning and characterization, Gene 166, ics, in: Trends in Antibiotic Research, pp. 22049-55. 232. Tokyo: Japan Antibiotics Res. Ass. HUANG, C., ANO, T., SHODA,M. (1993), Nu- KLEINKAUF, VON DOHREN, (1986), Peptide C. H. H., cleotide sequence and characteristics of the antibiotics, in: Biotechnology 1st Edn., Vol. 4 gene, Lpa-14, responsible for biosynthesis of the (REHM, H.-J., REED, G., Eds.), pp. 283-307. lipopeptide antibiotics iturin a and surfactin Weinheim: VCH.
6 References
319
binds covalently to cysteine-273 on protein phosKLEINKAUF,H., VON DOHREN, (1996), A nonH. phatase 1, FEBS Lett. 371, 236-240. ribosomal system of peptide biosynthesis, Eur. J. Biochem. 236, 335-351. MAGNUSON, SOLOMON, GROSSMAN, D. R., J., A. (1994), Biochemical and genetic characterization KLEINKAUF, VON DOHREN,H. (1997), EnzyH., of a competence pheromone from B. subtilis, matic generation of complex peptides, Progr. Cell 77, 207-216. Drug Res. 48, 27-51. s. KONDO, IMAI,K., ISOBE,M., GOTO, T., Mu- MARTfN, J. F., GUTIERREZ, (1995), Genes for N., beta-lactam antibiotic biosynthesis, Antonie van RASUGI, WADA-NAKAGAWA,HAYASHI, A., C., Leeuwenhoek Int. J. Gen. Mol. Microbiol. 67, Y. (1984), Cadystin A and B, major unit pep181-200. tides comprising cadmium binding peptides induced in a fission yeast-separation, revision of MEI, B., BUDDE, D., LEONG, A. (1993), sidl, A. S. structure and synthesis, Tetrahedron Lett. 25, a gene intiating siderophore biosynthesis in Usti3869-3872. lago maydis: molecular characterization, regulation by iron, and role in phytopathogenicity, KUBOI, R., MARUKI, TANAKA, KOMASAT., H., Proc. Natl. Acad. Sci. USA 90,903-907. WA, I. (1994), Fermentation of Bacillus subtilis ATCC 6633 and production of subtilin in poly- MEISSNER, D I ~ M A NE.,,BORNER, (1996), K., N T. Toxic and nontoxic strains of the cyanobacterethylene glycollphosphate aqueous two-phase ium Microcystis aeruginosa contain sequences systems, J. Ferment. Bioeng. 78,431-436. homologous to peptide synthetase genes, FEMS LAMBALOT, H., WALSH, T. (1995), Cloning, R. C. Microbiol. Lett. 135, 295-303. overproduction, and characterization of the Echerichia coli holo-acyl carrier protein synth- MEUWLY,P., THIBAULT, SCHWAN,A. L., P., ase, J. Biol. Chem. 270,24658-24661. RAUSER, E. (1995), Three families of thiol W. peptides are induced by cadmium in maize, LAMBALOT, H., GEHRING, M., FLUGEL, R. A. R. Plant J. 7,391-400. S., ZUBER, LACELLE, MARAHIEL, A., P., M., M. G., R., REID,R., KHOSLA, WALSH, T. (1996), A MITTENHUBER, WECKERMANN, MARAC., C. new enzyme superfamily - the phosphopanteHIEL,M. A. (1989), Gene cluster encoding the theinyl-transferases, Chemistry & Biology 3, genes for tyrocidine synthetase 1 and 2 from Bacillus brevis. Evidence for an operon, J. Bacteri923-936. 01. 171,4881-4887. LAWEN, TRABER, (1993), Substrate specifiA., R. M., R. cities of cyclosporin synthetase and peptolide Mo, Y. Y., GEIBEL, BONSALL, F., GROSS, SDZ 214-103 synthetase, J. Biol. Chem. 268, D. C. (1995), Analysis of sweet cherry (Prunus avium L.) leaves for plant signal molecules that 20452-20465. activate the syrB gene required for synthesis of LEHRER, I., LICHTENSTEIN, K., GANZ,T. R. A. the phytotoxin, syringomycin, by Pseudomonas (1993), Defensins: antimicrobial and cytotoxic syringae pv. syringae, Plant Physiol. 107, 603peptides of mammalian cells, Annu. Rev. Immu612. nol. 11, 105-128. J. S. LEWIS, R. (1996), Muscarine, imidazole, oxazole, MOLNAR,I., APARICIO, F., HAYDOCK, F., J. KHAW, L. E., SCHWECKE, KONIG, A., T., thiazole and peptide alkaloids, and other miscelSTAUNTON, LEADLAY, F. (1996), OrganiJ., P. laneous alkaloids, Nat. Prod. Rep. 13, 435-467. sation of the biosynthetic gene cluster for rapaLI, Y. M., MILENE, C., MADISON, L., KOLTER, J. L. mycin in Streptomyces hygroscopicus - analysis R., WALSH, T. (1996), From peptide precurC. of genes flanking the polyketide synthase, Gene sors to oxazole and thiazole-containing peptide 169, 1-7. antibiotics - microcin B17 synthase, Science 274, MOR,A., DELFOUR, NICOLAS, (1991), IdenA., P. 1188-1193. tification of a D-alanine-containing polypeptide LIU, G.-H., CHEN,I.-L., HSIEH,J.-C., LIU, S.-T. precursor for the peptide opioid, dermorphin, J. (1996), Cloning and sequencing of the peptide Biol. Chem. 266,6264-6270. synthetase genes encoding the biosynthesis of N., K., T. fengycin, Abstracts, Symp. Enzymol. Biosynth. MORIKAWA, HAGIWARA, HAKAJIMA, (1992), Brevinin-1 and -2, unique antimicrobial Nat. Prod., Technical University, Berlin. peptides from the skin of the frog, Rana brevipoMACCABE, P., RIACH, B. R., UNKLES, E., A. M. S. da porsa, Biochem. Biophys. Res. Commun. 189, KINGHORN, R. (1990), The Aspergillus niduJ. 184-190. lans npeA locus consists of three contiguous H. genes required for penicillin biosynthesis, MOTAMEDI, (1996), FK506 biosynthetic gene cluster. Abstracts, Symp. Enzymol. Biosynth. EMBO J. 9,279-287. MACKINTOSH, W., DALBY, N., CAMPBELL, Nat. Prod., Technical University, Berlin. R. K. J., H. D. G., COHEN, T., COHEN, MACKINTOSH, NIELSEN, JORGENSEN, S. (1995), Metabolic P. P., control analysis of the penicillin biosynthetic C. (1995), The cyanobacterial toxin microcystin
320
P., pathway in a high-yielding strain of Penicillium PIECQ,M., DEHOTTAY, BIOT, A., DUSART,J. (1994), Cloning and nucleotide sequence of a rechrysogenum, Biotechnol. Progr. 11,29%305. gion of the Kibdelosporangium aridum genome NIELSEN, KRABBEN, (1995), Hyphal growth J., P. homologous to polyketide biosynthetic genes, J. and fragmentation of Penicillium chrysogenum DNA Sequ. Mapping 4,219-229. in submerged cultures, Biotechnol. Bioeng. 46, 588-598. PIEPER, KLEINKAUF, ZOCHER, (1992), R., H., R. Enniatin synthetases from different Fusaria exNIELSEN,J., JOHANSEN, L., VILLADSEN, C. J. hibiting distinct amino acid specificities, J. Anti(1994), Culture fluorescence measurements durbiot. 45, 1273-1277. ing batch and fed-batch cultivations with Penicillium chrysogenum, J. Biotechnol. 38, 51-62. PIERS,K. L., BROWN, H., HANCOCK, E. W. M. R. (1993), Recombinant DNA procedures for proJ., C. C. NIELSEN, JOHANSEN, L., JACOBSEN, L., ducing small antimicrobial cationic peptides in KRABBEN, VILLADSEN, (1999, Pellet forP., J. bacteria, Gene 134,7-13. mation and fragmentation in submerged cultures of Penicillium chrysogenum and its relation to PITKIN, W., PANACCIONE, G., WALTON, D. J. D. J. penicillin production, Biotechnol. Progr. 11, 93(1996), A putative cyclic peptide efflux pump en98. coded by the toxA gene of the plant pathogenic fungus Cochliobolus carbonum, Microbiology NOZAKI,S., MURAMATSU, (1987), Natural hoI. 142,1557-1565. mologs of gramicidin S, J. Antibiot. 37, 689-90. J., OFFENZELLER, SANTER, TOTSCHNIG, M., G., K., POSPIECH,A., BIETENHADER, SCHUPP, T. R., (1996), Two multifunctional peptide synthetases Su, Z., MOSER,H., TRABER, SCHNEIDERSCHERZER, (1996), Biosynthesis of the unE. and an 0-methyltransferase are involved in the usual amino acid (4R)-4-[(E)-2-buteny1]-4-me- biosynthesis of the DNA-binding antibiotic and antitumour agent saframycin Mxl from Myxothyl-L-threonine of cyclosporin A: enzymatic coccus xanthus, Microbiology 142, 741-746. analysis of the reaction sequence including identification of the methylation precursor in a poly- PO~TERAT, STEPHAN, O., H., METZGER,J. W., ketide pathway, Biochemistry 35, 8401-8412. GNAU, ZAHNER,H., JUNG,G. (1994), AboV., rycin-a tricyclic-21-peptide antibiotic isolated OHNO, ANO,T., SHODA, (1995), Production A., M. from Streptomyces griseoflavus, Liebigs Ann. of a lipopeptide antibiotic, surfactin, by recombiChem. 1994,741-743. nant Bacillus subtilis in solid state fermentation, Biotechnol. Bioeng. 47,209-214. K. RAINEY, B., BRODEY,C. L., JOHNSTONE, P. OHNO, ANO,T., SHODA, (1996), Use of soyA., M. (1993), Identification of a gene cluster encoding bean curd residue, okara, for the solid state subthree high-molecular weight proteins, which is strate in the production of a lipopeptide antirequired for synthesis of tolaasin by the mubiotic, iturin a, by Bacillus subtilis NB22, Process shroom pathogen Pseudomonas tolaasii, Mol. Biochem. 31,801-806. Microbiol. 0, 643-652. Y. OKUMURA, (1990), Bioysynthesis of viridogri- RAUSER, E. (1990), Phytochelatins, Annu. Rev. W. sein, in: Biochemistry of Peptide Antibiotics Biochem. 59,61-86. (KLEINKAUF, VON DOHREN,H., Eds.), pp. REICHERT,J., SAKAITANI, H., M., WALSH, C. T. 365-378. Berlin: de Gruyter. (1992), Characterization of EntF as a serine-activating enzyme, Protein Sci. 1, 549-555. K., M. PARK,Y. S., INOUE, YAHIRO,K., OKABE, (1994a), Improvement of cephamycin C produc- REICHHART, J.-M., PETIT, I., LEGRAIN, DIM., tion by a mutant resistant to linoleic acid, J. FerMARCQ, J. L., KEPPI,E., LECOCQ, P., HOWJ. ment. Bioeng. 78, 88-92. M" A, J. A., ACHSTETTER, (1992), ExpresT. PARK,Y. S., MOMOSE, TSUNODA, OKABE, I., K., sion and secretion in yeast of active insect defenM. (1994b), Enhancement of cephamycin C prosin, an inducible antibacterial peptide from the duction using soybean oil as the sole carbon fleshfly Formia terranova, Invert. Reprod. Dev. source, Appl. Microbiol. Biotechnol. 40, 77321, 15-24. 779. RICHTER, EGGER,R., KREIL,G. (1987), D-AlaK., nine in the frog skin dermorphin is derived from PAUL,G. C., THOMAS, R. (1996), A structured C. model for hyphal differentiation and penicillin L-alanine in the precursor, Science 238, 200202. production using Penicillium chrysogenum, Biotechnol. Bioeng. 51, 558-572. RINCE, DUFOUR, LEPOGAM, THUAULT, A., A., S., PFEIFER,E., PAVELA-VRANCIC, VON DOHM., D., BOURGEOIS, M., LE PENNEC, P. (1994), C. J. H. REN,H., KLEINKAUF, (1999, CharacterizaCloning, expression and nucleotide sequence of tion of tyrocidine synthetase 1 (TY1): Requiregenes involved in production of lactococcin DR, ment of post-translational modification for pepa bacteriocin from Lactococcus lactis, Appl. Entide biosynthesis, Biochemistry 34, 7450-7459. viron. Microbiol. 60, 1652-1657.
6 References ROLLEMA,H. S., METZGER,J. W., BOTH, P., 0. KUIPERS, P., SIEZEN,R. J. (1996), Structure and biological activity of chemically modified nisin species, Eur. J. Biochem. 241, 716-722. ROMEO, D., SKERLAVAJ, BOLOGNESI, B., M., R. GENNARO, (1988), Structure and bactericidal activity of an antibiotic dodecapeptide purified from bovine neutrophils, J. Biol. Chem. 263, 9573-9575. SAHL,H.-G., JACK,R. W., BIERBAUM, (1995), G. Lantibiotics - biosynthesis and biological activities of peptides with unique posttranslational modifications, Eur. J. Biochem. 230, 827-853. SAKAJOH, SOLOMON, A., DEMAIN, L. M., N. A. (1987), Cell-free synthesis of the dipeptide antibiotic bacilysin, J. Ind. Microbiol. 2, 201-208. SCHAUWECKER, PFENNIG, F., F., KELLER, U. (1996), Molecular cloning of the actinomycin synthetase genes acmA and acmB from Streptomyces chrysomallus, Abstracts, Symp. Enzymol. Biosynth. Nat. Prod., Technical University, Berlin. SCHNELL, ENGELKE, AUGUSTIN, RoN., G., J., SENSTEIN, R., UNGERMANN, GOTZ, F., ENV., TIAN, K.-D. (1992), Analysis of genes involved in the biosynthesis of the lantibiotic epidermin, Eur. J. Biochem. 204,5743. SCHWARTZ, ALIJAH,R., NUSSBAUMER, D., B., W. PELZER,S., WOHLLEBEN, (1996), The peptide synthetase gene phsA from Streptomyces virdochromogenes is not juxtaposed with other genes involved in nonribosomal biosynthesis of peptides, Appl. Environ. Microbiol. 62, 570577. SCHWECKE, APARICIO, F., MOLNAR, KOT., J. I., S. NIG,A., KHAW,L. E., HAYDOCK, F., OLIYNYK, CAFFREY, CORTES, LESTER,J. M., P., J., B., BOHM,G. A., STAUNTON, LEADLAY, J., P. F. (1995), The biosynthetic gene cluster for the polyketide immunosuppressant rapamycin, Proc. Natl. Acad. Sci. USA 92, 7839-7843. SCOTT-CRAIG, S., PANACCIONE, G., PAJ. D. CARD, A., WALTON, D. (1992), The cyclic J. J. peptide synthetase catalysing HC toxin production in the filamentous fungus Cochliobolus carbonum is encoded by a 15.7 kilobase open reading frame, J. Biol. Chem. 267,26044-26049. SELSTED, E., OUTLETTE, J. (1995), DefenM. A. sins in granules of phagocytic and non-phagocytic cells, Trends Cell. Biol. 5, 114-119. SIMMACO, MIGNOGNA, BARRA,D., BosM., G., SA,F. (1993), Novel antimicrobial peptides from skin secretion of the European frog Rana esculenta, FEBS Lett. 324, 159-161. SKAUGEN, NISSEN-MEYER, JUNG,G., STEM., J., S., C., VANOVIC, SLEITEN, K., INGER, ABILDGAARD, NES,I. F. (1994), In vivo conversion M.,
321
of L-serine to D-alanine in a ribosomally synthesized poylpeptide, J. Biol. Chem. 269, 2718327185. SMITH,D. J., BURNHAM, K. R., BULL,J. H., M. HODGSON, E., WARD,J. M., BROWNE,P., J. BROWN, BARTON, EARL,A. J., TURNER, J., B., G. (1990), /?-Lactam antibiotic biosynthetic genes have been conserved in clusters in prokaryotes and eukaryotes, EMBO J. 9,741-747. SOLOMON, M., MAGNUSON, SRIVASTAVA, J. R., A., GROSSMAN, D. (1995), Convergent sensA. ing pathways mediate response to two extracelMar competence factors in Bacillus subtilis, Genes Dev. 9,547-558. STACHELHAUS, SCHNEIDER, MARAHIEL, T., A., M. A. (1995), Rational design of peptide antibiotics by targeted replacement of bacterial and fungal domains, Science 269, 69-72. STACHELHAUS, HUSER,A., MARAHIEL, A. T., M. (1996a), Biochemical characterization of peptides carrier protein (PCP), the thiolation domain of multifunctional peptide synthetases, Chemistry & Biology 3, 913-921. STACHELHAUS, SCHNEIDER, MARAHIEL, T., A., M. A. (1996b), Engineered biosynthesis of peptide antibiotics, Biochem. Pharmacol. 52, 177186. STEFFENS, C. (1990), The heavy metal-binding J. peptides of plants, Annu. Rev. Plant Physiol. Plant Mol. Biol. 41,533-575. STEIN, VATER,J. (1996), Amino acid activation T., and polymerization at modular multienzymes in nonribosomal peptide biosynthesis, Amino Acids 10, 201-227. STINTZI, A., CORNELIS, P., HOHNADEL, D., MEYER, J.-M., DEAN,C., POOLE,K., KOURAMBAS, S., KRISHNAPILLAI, (1996), Novel pyoV. verdine biosynthesis gene(s) of Pseudomonas aeruginosa PAO, Microbiology 142, 1181-1190. STROHL, R. (Ed.) (1996), Biotechnology of InW. dustrial Antibiotics, 2nd Edn. New York: Marcel Dekker. STROHL, R., FLOSS, G. (1995), Thiopeptides, W. H. in: Genetics and Biochemistry of Antibiotic ProL. duction (VINING, C., STUITARD,C., Eds.), pp. 223-238. Boston, MA: Butterworth-Heinemann. SUBBALAKSHMI, KRISHNAKUMARI,NAGAC., V., RAJ, R., SITARAM, (1996), Requirements for N. antibacterial and hemolytic activities in the bovine neutrophil derived peptide indolicidin, FEBS Lett. 395, 48-52. TAKESAKO, MIZUTANI, SAKAKIBARA, K., S., H., ENDO,M., YOSHIKAWA, MSUDA, SONOY., T., E., KOYAMA, KATO, I. (1996), Precursor directed biosynthesis of aureobasidins, J. Antibiot. 49,676-681.
322
TANIGUCHI, HOSHINO, URASAKI, FuM., K., H., JII,M. (1993), Continuous production of an antibiotic polypeptide (nisin) by Lactococcus lactis using a bioreactor coupled to a microfiltration module, J. Ferment. Bioeng. 78, U122. S., THIBAULT, FAUBERT, KARUNANITHY, P., D., BOYD, R. K., HOLMES,C. F. B. (1992), Isolation, mass spectrometric characterization, and protein phosphatase inhibition properties of cyclic peptide analogs of grarnicidin S from Bacillus brevis (Nagano strain), Biol. Mass Spectrom. 21, 367-379. THIERICKE, ROHR,R. (1993), Biological variaR., tion of microbial metabolites by precursor-directed biosynthesis, Nut. Prod. Rep. 10, 263289. TOGNONI, FRANCHI, MAGISTRELLI, CoA., E., C., LOMBO, E., COSMINA, GRANDI, (1995), A p., G. putative new peptide synthase operon in Bacillus subtilis: Partial characterization, Microbiology 141,645-648. TURGAY, MARAHIEL, A. (1995), The gtcRS K., M. operon coding for two-component system regulatory proteins is located adjacent to the grs operon of Bacillus brevis, DNA Sequence 5, 283290. VANHOUTTE, PONS, M. N., THOMAS,C. R., B., LOUVEL,L., VIVIER,H. (1995), Characteriza-
tion of Penicillium chrysogenum physiology in submerged cultures by color and monochrome image analysis, Biotechnol. Bioeng. 48,l-11. VINING, C., STUITARD,C. (Eds.) (1995), GeL. netics and Biochemistry of Antibiotic Production. Boston, M A Butterworth-Heinemann. VON DOHREN, H. (1990), Compilation of peptide structures - a biogenetic approach, in: BiochemH., istry of Peptide Antibiotics (KLEINKAUF, VON DOHREN, H., Eds.), pp. 411-507. Berlin, New York: de Gruyter. K., WEBER,G., SCHORGENDORFER, SCHNEIDERSCHERZER, LEITNER, (1994), The peptide E., E. synthetase catalyzing cyclosporine production in Tolypocladium niveum is encoded by a giant 45.8 kilobase open reading frame, Curr. Genet. 26,120-125. D., WILLIAMSON, P., GAUVREAU, WILLIAMS, M. D. H., WARING, J. (1982), Structure and conM. formation of fourteen antibiotics of the quinoxaline group determined by 'H-NMR, J. Antibiot. 35, 62-66. YANG, Y. K., MORIKAWA, M., SHIMIZU,H., S., T., Y. SHIOYA, SUGA,K., NIHIRA, YAMADA, (1996), Maximum virginiamycin production by optimization of cultivation conditions in batch culture with autoregulator addition, Biotechnol. Bioeng. 49,437-444.
Biotechnology
8 Lantibiotics
RALPHJACK FRIEDRICH GOTZ JUNG GUNTHER

Tubingen, Germany
1 Introduction 325 2 The Unique Chemistry and Structure of Lantibiotics 325 2.1 Modified Amino Acids Found in the Lantibiotics 327 2.1.1 Lanthionine (Lan) and 3-Methyllanthionine (MeLan) 328 2.1.2 2d-Didehydroalanine (Dha) and 2,3-Didehydrobutyrine (Dhb) 328 2.1.3 D-Alanine 330 2.2 Type A Lantibiotic Primary Structures 330 2.2.1 Nisin 330 2.2.2 Subtilin 331 2.2.3 Epidermin and Gallidermin 333 2.2.4 Pep5 333 2.2.5 Epilancin K7 333 2.2.6 Strepococcin A-FF22, Lacticin 481 (Lactococcin DR), and Salivaricin A 334 2.2.7 Lactocin S 335 2.2.8 Enterococcal Cytolysin/Bacteriocin 335 2.2.9 Carnocin U149 and Mutacin 336 2.3 Type B Lantibiotic Primary Structures 336 2.3.1 Cinnamycin, Duramycin, Duramycin B, Duramycin C, and Ancovenin 336 2.3.2 Mersacidin and Actagardine 338 3 Lantibiotic Structures in Solution 338 3.1 Type A Lantibiotics 339 3.1.1 Gallidermin 339 3.1.2 Nisin and Subtilin 339 3.1.3 Pep5 340 3.2 Type B Lantibiotics 340 3.2.1 Cinnamycin and the Duramycins 340 3.2.2 Mersacidin and Actagardine 341
324
8 Lantibiotics
4 The GenedProteins Involved in Lantibiotic Biosynthesis and Genetic Regulation 341 4.1 The Requirement for Multiple Gene Products in Lantibiotic Biosynthesis 341 4.2 Structural Genes and Pre-Lantibiotics 341 4.2.1 The Location and Features of the Structural Genes and their Transcription 341 4.2.2 The General Structure of Prepeptides 345 4.2.3 The Possible Role(s) of the Leader Peptide 346 4.3 Amino Acid Modifying Proteins 347 4.3.1 LanB/LanC and LanM Proteins 347 4.3.2 EpiD 348 4.4 Transport Proteins 348 4.5 Leader Peptidases (LanP) 348 4.6 Proteins Regulating Lantibiotic Biosynthesis 350 4.7 Producer Self-Protection (Immunity) Mechanisms 351 4.8 The Chain of Events Leading to Lantibiotic Biosynthesis and Maturation 353 5 Biological Activities of Lantibiotics 353 5.1 Type A Lantibiotics 353 5.1.1 Primary Mode of Action 353 5.1.2 Secondary Mode of Action 356 5.2 Type B Lantibiotics 357 5.2.1 Mersacidin and Actagardine 357 5.2.2 Cinnamycin and the Duramycins 357 6 Applications of Lantibiotics 358 6.1 Applications as a Food/Beverage Preservative 358 6.2 MedicaUParamedical and Veterinary Applications 359 7 Conclusions and Future Perspectives 359 8 References 360
2 The Unique Chemistry and Structure of Lantibiotics
325
1 Introduction
The search for novel pharmaceutical compounds with potent new biological activities has only relatively recently begun to look inward and away from complex in-lab synthetic processes toward the huge array of natural compounds produced by bacteria, offering a plethora of new possibilities. Such studies of bacterial-derived proteins in particular have unearthed a vast array of naturally produced compounds, not the least of which is the group of highly modified antimicrobial and enzyme-inhibitory peptides collectively referred to as lantibiotics. Originally coined to describe the rapidly expanding group of antibiotic-like peptides which were found to contain the non-protein amino acids lanthionet ine and 3-methyl lanthionine (SCHNELL al., 1988), the name lantibiotics belies the full extent and complexity of this class of bacterially synthesized peptides. For example, as the number of characterized lantibiotics increases, novel modified amino acids of interest to chemists and biochemists alike are constantly being identified, including those with unusual crosslinks or unsaturated R-groups, while analysis of the solution structure of the peptides concerned is offering new perspectives on the relationship between peptide structure and biological function. Perhaps more strikingly though, a number of recent studies of the biosynthesis of lantibiotics has revealed that they are produced by ribosomal biosynthesis and subsequently modified to generate the mature, biologically et active compound (SCHNELL al., 1988; BANERJEE and HANSEN, 1988; BUCHMANN et al., 1988; KALETTA and ENTIAN, 1989). What this means is that there are proteins within the lantibiotic-producing cells which are able to carry out specific transformations of amino acids, converting them to novel structures which may play a role in the biological activity, stability, etc. of the peptide and, more importantly, may prove useful to those biotechnologists interested in creating similar modifications of other peptides. It is this last feature of ribosomal biosynthesis which sets the lantibiotics apart from the classical peptide antibiotics which are typ-
ically synthesized by large multi-enzyme complexes in the cell and for which there exists no structural gene (KATZ and DEMAIN, 1977; KLEINKAUFand VON DOHREN, 1986, 1987, 1990; NAKANO ZUBER, and 1990). Studies of the lantibiotics have revealed that they are genetically encoded and are uniformly produced on the ribosome as a precursor peptide which is subsequently modified at specific points to give rise to the large number of modified amino acids found in these peptides. In addition, the peptides are produced with a leader peptide which is removed during maturation and are transported by specific transport-related proteins from the cell. Still other lantibiotic-specific proteins are involved in the genetic regulation of biosynthesis and generation of the specific producer cell selfprotection mechanism(s) frequently observed. The mature lantibiotics have also found a number of potential applications including medical and veterinary antibiosis, food, beverage, and cosmetic preservation and as regulators of both human immune function and 1981; JUNG, 1991a, b; blood pressure (HURST, DELVES-BROUGHTON, MOLITOR 1990; and SAHL, 1991; BIERBAUM SAHL, 1993; DE and VUYSTand VANDAMME, 1993; JACK et al., 1995; SAHL et al., 1995). In the following chapter we wish to provide an overview of the novel structures, mechanism(s) of biosynthesis, genetic regulation, biological activities and current as well as potential applications for this fascinating, novel class of bacterialderived, biologically-active peptides.

In order to appreciate the potential of the lantibiotics as biologically active peptides, as well as the possibilities for the application of their respective biosynthetic machinery, it is first necessary to discuss the structure of both the peptides themselves and the modified
Tab. 1. A Compilation of the Currently Described Lantibiotics" Including Several of their Respective Chemical and Physical Characteristics Producing Organism Mass [Da] 3353 3330 3317 2164 2164 2151 3488 3032 3764 2795 2901 2315 4164 2631 4635 3245 2042 2014 1951 2008 1959 1825 1890
0 0 0 -1 0 -1 0
Lantibiotic
Net Chargeb (at pH 7.0) +3 +3 +2 +3 +3 +3 +7 +5 -1 +1 0 0 0 0 N.R. N.R.

5 5 5 4 4 4 3 3 2 3 3 3 N.R." N.R. N.R. 3
Number of Rings'
% Modified Residuesd
Reference
Type A Nisin A Nisin Z Subtilin Epidermin Gallidermin [lV,6L]-Epidermin Pep5 Epilancin K7 Lactocin S SA-FF22 Lactococcin DR Salivaricin A Cytolysin L1 Cytolysin L2 Carnocin U149 Mutacin Lactococcus lactis Lactococcus lactis Bacillus subtilis Staphylococcus epidermidis Staphylococcus gallinarum Staphylococcus epidermidis Staphylococcus epidermidis Staphylococcus epidermidis Lactobacillus sake Streptococcus pyogenes Lactococcus lactis Streptococcus salivarius Enterococcus faecalis Enterococcus faecalis Carnobacterium piscicola Streptococcus mutans Streptomyces cinnamoneus Streptomyces cinnamoneus Streptoverticillium spp. Streptomyces griseoluteus Streptomyces spp. Bacillus subtilis Actinoplanes spp. 4 4 4 4 3 4 4
38 38 40 41 41 41 26 32 24 27 26 27 N.R. N.R. N.R. N.R. 47 47 47 47 37 42 45
GROSSand MORRELL (197 MULDERS al. (1991) et GROSSet al. (1973) ALLGAIER al. (1986) et KELLNER al. (1988) et SAHLet al. (1995) KELLNER al. (1989) et VAN DE KAMPet al. (1995a SKAUGEN al. (1994) et JACKet al. (1994a) RINCEet al. (1994) Ross et al. (1993) GILMORE al. (1994) et GILMORE al. (1994) et STOFFELS al. (1994) et NOVAK al. (1994) et FREDENHAGEN (1991 et al. FREDENHAGEN (1991 et al. FREDENHAGEN al. (1991) et FREDENHAGEN al. (1991 et WAKAMIYA al. (1985) et KOGLERet al. (1991) ZIMMERMANN (19951
Type B Cinnamycin Duramycin Duramycin B Duramycin C Ancovenin Mersacidin Actagardine
Adapted from JACKand SAHL(1995) and SAHLet al. (1995). Includes those amino- and carboxy termini which remain unmodified following posttranslational modification.
327
amino acids which they contain. Tab. 1 lists all of the currently characterized lantibiotics, the bacterium which produces them as well as a number of their physicochemical properties relevant to the proceeding sections (SAHLet al., 1995; JACKand SAHL, 1995). In addition, an earlier review of the lantibiotics has suggested that the peptides can be divided into two distinct groups (JUNG, 1991a, b), a subdivision scheme which has been maintained in the following chapter. The type A lantibiotics can be defined as elongated, helical peptides whose primary mode of action appears to be directed at the depolarization of the bacterial cytoplasmic membrane through the formation of voltagedependent pores. Alternatively, type B lantibiotics are generally somewhat smaller, compact, globular structures which either inhibit bacterial cell wall replication, interact with specific enzymes, or activate T-cell proliferation (see also Sect. 5 ) .
0 0
0
2.1 Modified Amino Acids Found in the Lantibiotics

While the lantibiotics take their name from the observation that they contain lanthionine (SCHNELLet al., 1988), this is not the only modified amino acid that they possess. Currently, a number of modified amino acids and other residues have been found in lantibiotic structures including: meso-lanthionine (Lan), lanthionine sulphoxide, fhreo-p-methyllanthionine (MeLan), S-[(Z)-2-aminovinyl]-~cysteine (AviCys), S-[(Z)-2-aminovinyl1-3methyl-D-cysteine, 2,3-didehydroalanine (Dha), 2,3-didehydrobutyrine (Dhb), D-alanine, 2oxopyruvate, 2-oxobutyrate, hydroxypyruvate, erythro-3-hydroxyaspartateand (2S,8S)lysinoalanine, the structures of which are presented for comparison in Fig. 1. Some of these residues such as Lan, MeLan, Dha, and Dhb occur in most, if not all, lantibiotics so0
II
-HN
-S
II
II
-HN\m,cH--6
\@YC-6-H
a 2
-"\myCH--6
"\iyC
II
-6-H
H&-C
H '
-S
II -w\TIycH--6.
II
-HN\N/c-oH -6-H
I
( ~ 9 3
a 2
H2C -H N-
a 2
n
-HN H
c\@Y
ri
-HN
\Q (
-HN
G - H
- Ca
I
H2C
S -
CH
II
H--6
c\@Y
I
il
-m
\@'- H C.
CH
- \ "
C-H I
2-
II
H3C-C-SH/
II
S-[(Z)-Z-aminovinyl]-o-cysteine
S[(Z)-2-runinovinyl]-3-methyls-cysteinc
rrythm-341ydro~~@c acid
-\/II
HlC
II
- \ "
2-
II
CH H3C
II
- H-C " \
7-
II
II II c-cI
II I1 c-c-
OH 0 H-C-C-
I
I
II
CH3
CH3
P a33
2-oxobutyIyl
-3
z.3diddlydrcahiw
(Z)-2.3-didehydmbutyrine
danine
2-oxopynrvyl
Z-hydroXypyNvyl
Fig. 1. The structures of the modified amino acids and other residues found in lantibiotics. The chirality of various structures is indicated (where appropriate) above the respective a-carbon atom.
328
8 Lantibiotics
far characterized, while others occur only in individual cases. However, due in part to their novel nature and properties, many of these residues may be of great interest to the researchers involved in biotechnology.
2.1.1 Lanthionine (Lan) and 3-Methyllanthionine (MeLan)

It was the pioneering research of E. GROSS and co-workers that finally demonstrated that the peptide antibiotics nisin and subtilin actually contained Lan and MeLan (as well as Dha and Dhb), confirming previously held beliefs (BERRIDGE al., 1952; GROSSand et MORRELL, 1971; GROSSet al., 1973). However, it was not clear at that time how these amino acid structures should arise. Later studies showed that inhibitors of protein biosynthesis prevented nisin and subtilin production (HURST, 1981), whilst the isolation of the structural genes for a number of lantibiotics has finally proved conclusively that these residues arise from posttranslational modification of a ribosomally synthesized precursor peptide (SCHNELL al., 1988; BUCHMANN al., et et 1988; BANERJEE HANSEN, and 1988; KALETTA and ENTIAN,1989). In addition, these studies also showed that Ser, Thr and Cys residues found in the prepeptides but not in the mature lantibiotics, must be the precursors for the formation of Lan, MeLan, Dha and Dhb. While it is not clear how Lan and MeLan are synthesized in the cell, the structure of these thioether-linked, di-carboxy, di-amino acids has been studied in detail and a model (Fig. 2) for their biosynthesis has been proposed (SCHNELL al., 1988; JUNG,1991a, b). et The first step involves the site-specific dehydration of the a-amino-/3-hydroxy acids Ser and/or Thr in the propeptide part of the prelantibiotic; such dehydration results in the formation of Dha or Dhb, respectively. Presumably, this reaction is carried out by a specific enzyme within the lantibiotic-producing cells, however, neither the enzyme nor the molecular mechanism responsible have yet been identified. Also, the enzyme(s) capable of such a reaction should be novel since, cur-
rently characterized serine and threonine dehydratases appear to act on free amino groups via pyridoxal phosphate-dependent Schiff base reactions, suggesting that Ser and Thr residues incorporated into a peptide chain would not be appropriate as substrates for these enzymes. The subsequent step in lanthionine biosynthesis (Fig. 2) involves the addition of the sulphydryl group of a neighboring Cys residue to the a,punsaturated amino acid to form a covalent thioether bridge (SCHNELL al., 1988; et JUNG,1991a, b). Again, while it is not clear how this step occurs in vivo (i.e., spontaneous addition or specific protein-directed addition), it has some interesting ramifications for the stereospecificity of the final product. Since lantibiotics prepeptides are synthesized on the ribosome, they contain only L-amino acids; despite this, Lan and MeLan in the mature lantibiotic are consistently found in the rneso(i.e., D/L-isomer)-configuration with the SerlThr-derived half in the D-form while the Cys-derived half remains in the L-configuration. In addition, in the case of type A lantibiotics the SerKhr-derived half of the Lan and MeLan is always closer to the N-terminus than is the Cys-derived half of the amino acid, however, this is not always true amongst the type B lantibiotics (JUNG, 1991a, b).
2.1.2 2,3-Didehydroalanine (Dha) and 2,3-Didehydrobutyrine (Dhb)

The a$-unsaturated amino acids Dha and Dhb (formed as described above for the first step in Lan/MeLan biosynthesis) are stable within the mature lantibiotic, however, their appearance at the amino terminus of the peptide, such as occurs during amino acid sequencing by Edman-degradation, can prove problematic. The structure elucidation of a number of lantibiotics has previously been hampered by the presence of these amino acids (GROSS and MORRELL, 1971; ALLGAIER et al., 1986; KELLNERet al., 1988, 1989; JUNG,1991a, b; JACKet al., 1994a; VAN DE KAMPet al., 1995b), the problems resulting from the spontaneous oxidative deamina-

(Sert3) [precpidermin(-30to +2)]-N
(CYS+'l)
329
H I
I H-C-OH I H (L)-serine
w C-H
O I1 C-
[Preepidermin(+4to +6)]-N
H I
w C-H
O II C-
[pre-epidermin(+7to +22)]
I 7%
SH (Lkcysteine
Site-specific dehydration
@ha+3)
(CYS+7) [pre-epidermin(Mto +6)] -N
[preepid&n(-30 to +2)]-N
H I
O I 1
II H-C I H
\C /
c-
H I
L I /
O I1 CC-H I SH
[preepidermin(+7to +22)]
YIEL
2,34dehydrodanine
[Precpidermin(-30to +2)]-N
I H-C I H
\ d H-C
H I
O I1 C-
1
S
(L)-cysteine
Stereospecific addition
[pre-epidermin(+4to +6)]-N
H I
b I /
O I1 CC-H I
[pre-epidermin(+7to +22)]
7-"
meso-lanthionine
Fig. 2. Proposed pathway for the chemical reactions occurring during the biosynthesis of lanthionine from the precursor amino acids serine and cysteine using, as an example, the formation of ring A of epidermin. The chirality of the respective compounds is indicated (where appropriate) above the a-carbon atom.
tion of N-terminally situated Dha or Dhb to produce a sequence-blocking pyruvyl or butyryl group, respectively (GROSSand MORELL, 1971; JUNG, 1991a, b). During the elucidation of the structure of Pep5 (KELLNERet al., 1989) this problem was circumvented by first reducing the two Dhb residues with benzyl-
thioalcohol, however this has proved unsuccessful with other lantibiotics such as SAFF22 (JACKet al., 1994a). Recently, MEYER (1994) has developed a convenient, new amino acid sequencing regimen for the detection of suitably derivatized dehydroamino acids and thioethers which involves thiol addition,
330
8 Lantibiotics
oxidation with per-trifluoroacetic acid and a second round of thiol addition. Similar oxidative deaminations of N-terminal Dhb and Dha leads to the formation of the novel sequence blocking 2-oxobutyryl and 2-oxopyruvyl residues found at the N-terminus of Pep5 and lactocin S, respectively (KELLNERet al., 1989; SKAUGEN al., 1994) and will be diset cussed in subsequent sections.
\ad
F-H
~-6-0~
I H (+seine
2.1.3 D-Alanine
Amino acids in bacterially-synthesized proteins and peptides are normally present as Lisomers, however, the type A lantibiotic lactocin s, which is produced by Lactobacillus sake strain Lb45, contains D-alanine (SKAUGEN et al., 1994). Interestingly, analysis of the prepeptide sequence showed that the precursor for this amino acid was L-serine, rather than alanine as might be expected. Thus, it seems that specific serine residues are dehydrated to form Dha and then hydrogenated in a stereospecific fashion to yield D-alanine during lactocin S biosynthesis (Fig. 3). Since the Dha residues found in other lantibiotics are generally stable, it seems likely that such a reaction should be catalyzed by a hitherto unknown enzyme. In addition, the identification of an enzyme or enzyme system able to create D-isomers of alanine in peptides from serine precursors offers considerable hope for new approaches in biotechnology and peptide engineering.
H-C
I H
2,3-&dehydroalanine
2.2 Type A Lantibiotic Primary Structures

2.2.1 Nisin
Since the structure of nisin was the first lantibiotic described (GROSS and MORRELL, 1971) and nisin was probably the first lantibiotic ever identified (ROGERSand WHITTIER, 1928), it is fitting to begin any treatise of lantibiotic structure with this particular example. Nisin is produced by the cheese starter culture organism Lactococcus lactis ssp. lactis and is antimicrobial for a broad variety of gram-positive bacteria (HURST,1981; MOLI-
[ I d n S(l to 611 -N
H I
H-i H-C-H
I H
O I1 C-
[ l a d n S(8-3711
(Dkalanine
Fig. 3. Proposed pathway for the chemical reactions occurring during the biosynthesis of D - a h i n e from a precursor serine residue at position 7 of the type A lantibiotic lactocin S. The chirality of the respective compounds is indicated above the a-carbon atom.
331
and SAHL,1991). The peptide itself (Fig. 4) is a 34 amino acid, 3353 Da, pentacyclic structure formed from its content of one Lan and 4 MeLan residues (GROSS and MORRELL,1971). While rings A, B, and C are separate the C-terminal pair of rings (D and E) form a bicycle (in all cases rings have been assigned ascending alphabetic characters, starting with the A ring closest to the N-terminus). In addition, nisin contains one Dha and 2 Dhb residues. In addition to nisin, a number of nisin fragments have also been isolated and characterized, many of which have altered antimicrobial, activity. BERRIDGE al. (1952) first reet ported the isolation of nisin fragments generated by long-term storage under acidic conditions which have subsequently been isolated and characterized by CHANet al. (1989a, b). Nisinl-32(amide), which results from the hydrolysis of the Va133-Dha34 peptidyl bond, has similar antimicrobial activity to native nisin, suggesting that the C-terminal two amino acids are dispensable and that Dha33 plays no particular role in the biological activity of the peptide (ROLLEMA al., 1991). However niet sinl-32(amide), in which the DhaS-Leu6 peptidy1 bond had also been hydrolyzed and ring A was therefore opened, was virtually inactive. Subsequently these studies have been confirmed by site-directed mutagenesis (DODD and GASSON, 1994); the exchange Dha33Ala resulted in only slightly decreased activity while the double exchange DhaSAla/ Dha33Ala essentially destroyed the antimicrobial activity of the peptide. Similarly, the presumably conservative exchange DhaSDhb led to a substantial decrease in activity (between 2- and 10-fold) against some indicators (KUIPERSet al., 1992), perhaps suggesting that these residues are not only important for antimicrobial activity but may also alter the spectrum of bacterial strains affected. In addition, these authors also altered ring C of nisin with the dual exchange Metl7Gln/GlylSThr. Two products were resolved from this exchange; intact nisin containing either Thrl8 or Dhbl8. The latter product not only contains a ring with the same sequence as the structurally similar lantibiotic subtilin (Fig. 4), but also demonstrates that the biosynthetic
TOR
machinery of the cell was able to cope with forming a novel dehydrated residue at a position not normally occupied by such an amino acid. Interestingly, [Met17Gln/GlylSDhb]-nisin also had improved antimicrobial activity against Bacillus cereus and decreased activity against Micrococcus flavus, while [Metl7Gln/ Glyl8/Thr]-nisin had heightened activity against M . flavus and reduced activity against B. cereus (KUIPERS al., 1992). et Recently, MULDERS al. (1991) identified et a naturally occurring analog of nisin which they named nisin Z. The peptide was characterized by a single amino acid exchange (His27Asn), resulting from a single base change in the 3rd base of the 27th codon of the structural gene nisZ. Interestingly, while this exchange appears to have little (if any) effect on the biological activity of the mutant nisin, it seems to make the peptide more soluble at or around pH 7, presumably as a result of the increased hydrophilicity of Asn at neutral pH. Thus, nisin Z might prove to be useful for applications where the low solubility of nisin at neutral pH (HURST,1981) has traditionally proved problematic (see also Sect. 6).
2.2.2 Subtilin
Like nisin, subtilin is a pentacyclic type A lantibiotic (Fig. 4); in fact, the arrangement and nature of all of the thioethers are conserved between nisin and subtilin, even though the amino acid sequences of the two peptides varies significantly (GROSS and MORRELL,1971; GROSS et al., 1973). First described by JANSEN and HIRSCHMANN (1944), subtilin is produced by Bacillus subtilis and is active against both the vegetative cells of a number of gram-positive bacteria as well as the outgrowth of endospores of Bacillus spp. and Clostridium spp. (HURST,1981). Interestingly, the two activities are distinguishable, apparently dependent on saturation of Dha5 which leads to loss of subtilin antispore activity, the half-life of which is only 0.8 d (HANSEN al., 1991; LIUand HANSEN, et 1992; HANSEN, 1993). Using site-directed mutagenesis, the exchange Dha5Ala was shown to result in the loss of antispore activity, confirm-
Nisin
Subtilin
- l r
c -co
S -
Pep5
ii
II
I -
'
'
Epilancin K7
~17- - m
dH
Epidennin
Lactocin S
II
7 c--co
SA-FF22
2 The Uniqrxe Chemistry and Structure of Lantibiotics
333
4 Fig. 4. The structure of several type A lantibiotics. Subsequently, a structural analog of epiderAmino acid residues involved in posttranslational min, called gallidermin and produced by Stamodifications have been identified by highlighting phylococcus gallinarum, has also been isowhile additional modified residues are indicated by lated and characterized (KELLNER et al., their respective chemical structures. Dha, 2,3-dide- 1988). While gallidermin appears to have a hydroalanine; Dhb, 2,3-didehydrobutyrine;Abu, aaminobutyric acid; Ala-S-Ala, (2S,6R)-lanthionine; slightly different spectrum of biological activiAbu-S-Ala (2S,3S,6R)-3-methyllanthionine. ar- ty to epidermin (especially against PropioniThe rangement of the thioether bonds in the lantibiotics bacterium acnes), it differs only by the exSA-FF22, lacticin 48l/lactococcin DR, and salivari- change of Ile6 in epidermin for Leu, an excin A remains unreported. change which is conservative and does not al-
ter the molecular mass. In all other respects, the two structures are identical. ing that saturation of Dha5 was the primary cause of subtilin instability (LIU and HANSEN, 1992). In addition, these authors also showed that the exchange Glu4Ile (i.e., the same as is found in this position of nisin; Fig. 4) not only increased the half-life of the antispore activity by 57-fold but also increased the sporicidal activity by a factor of 3. From these results, they suggest that Glu4 may be involved in the spontaneous saturation of subtilin, probably acting as a Michael-type acceptor. In similar structure-function studies, subtilin with a succinylated N-terminus has been isolated; this peptide also had reduced biological activity (CHANet al., 1993).
2.2.4 Pep5
The type A lantibiotic Pep5 was first isolated from Staphylococcus epidermidis strain 5 (SAHLand BRANDIS, 1981) and has subsequently been thoroughly characterized (KELLNERet al., 1989, 1991). The peptide (Fig. 4) is the largest characterized type A lantibiotic (3488 Da) and consists of 34 amino acids, 8 of which are basic. In addition, Pep5 is tricyclic as a result of its containing one MeLan and two Lan residues; ring A is separate while, rings B and C overlap. The N-terminus of Pep5 is occupied by a 2-oxobutyryl residue (KELLNERet al., 1989) and analysis of the sequence of the structural gene, pepA (KALEITA et al., 1989), shows that the +1 codon of the prepeptide encodes a threonine residue which is subsequently converted to Dhb during modification and maturation (WEIL et al., 1990). It appears that once this Dhb residue is located at the N-terminus (i.e., following removal of the leader peptide) it undergoes oxidative deamination to form 2oxobutyryl (Fig. 5), probably by a spontaneous mechanism (KELLNERet al., 1989).
2.2.3 Epidermin and Gallidermin

Elucidation of the structure of the Staphylococcus epidermidis lantibiotic epidermin (ALLGAIER al., 1986) has proved imporet tant in terms of our understanding of lantibiotic biosynthesis, as it assisted in the isolation of the structural gene (epiA), proving conclusively that lantibiotics were ribosomalet ly synthesized (SCHNELL al., 1988). Epidermin (Fig. 4) is a 22 amino acid (2164 Da), tetracyclic peptide which contains one residue each of Dhb and MeLan along with two residues of Lan. The 4th cyclic structure results from the novel C-terminal mono-carboxy, diamino acid, Cys(Avi) (Fig. 1) between residues 19 and 22, the formation of which results from oxidative decarboxylation of the C-terminal Cys residue by the enzyme EpiD (see also Sect. 4.3.2). In addition, ring B is identical to ring B of both nisin and subtilin, as is the arrangement of both rings A and B among these 3 lantibiotics.
2.2.5 Epilancin K7
Also produced by Staphylococcus epidermidis, epilancin K7 is somewhat similar to Pep5 (Fig. 4). Recently, VAN DE KAMPet al. (1995a, b) solved the structure of the peptide by a variety of chemical and homonuclear NMR-based strategies and showed that it is a 3052 Da, tricyclic, type A lantibiotic containing one Lan and two MeLan residues. In addition, epilancin contains two residues each of Dha and Dhb as well as 6 lysine residues (and
334
8 Lantibiotics
Pre-Pep5
f7-C-l lf
FHz
CH3
pepS(+2 to +34)]
1
HzNT-C-
Cleavage of leadex peptide
HP fl
7%
CH3
[pepS(+210+34)]
+H 0 &
bN-C-C-
FepS(+2 to +34)1
bC4-H
2-oxobutyryl-PepS
Fig. 5. Proposed pathway for the spontaneous oxidative deamination of dehydrobutyrine located at the N-terminus of Pep5 after cleavage of the leader peptide.
et no acidic amino acids), accounting for its ex- RELL, 1971), TAGG al. (1973a, b) reported tremely basic nature. Again, like Pep5 that a clinical isolate of Streptococcus pyo(KELLNER et al., 1989), the N-terminus of genes designated strain FF22 produced a difepilancin is blocked; however, while sponta- fusible inhibitor of bacterial growth which neous deamination of the N-terminal Dha they called streptococcin A-FF22. Subsequent should result in the formation of 2-oxopyru- partial purification and characterization vyl (in analogy to that shown for Pep$ Fig. showed that the inhibitory agent was essen5), VAN DE KAMPet al. (1995a) suggest that tially proteinaceous, of low molecular weight et the N-terminal residue is occupied by 2-hy- (TAGG al., 1973a, b), and was rather similar 1978; droxypyruvate (Fig. 1). The formation of this to nisin (TAGGand WANNAMAKER, 1992). Recently, streptococgroup at the N-terminus probably suggests JACK and TAGG, that an additional, novel enzyme function is cin A-FF22 was purified to homogeneity and required for epilancin biosynthesis. In addi- chemically characterized (JACK and TAGG, tion, this study also isolated a deletion pep- 1991); the lantibiotic was shown to be a 27 tide resulting from hydrolysis of the Ala2- amino acid, 2795 Da type A lantibiotic (Fig. 4) Dha3 peptidyl bond; epilancin K7(3-33) is which contains one Lan and two MeLan resialso blocked, in this case by the expected 2- dues as well as one residue of Dhb (JACK et oxopyruvyl group, presumably formed by oxi- al., 1994a). Using the sequence information dative deamination of the N-terminal Dha re- gained from these studies, the structural gene sidue. Interestingly, this peptide appears to be scnA has been cloned and sequenced and approximately as active as native epilancin confirms the correct sequence for the peptide K7, suggesting these two N-terminal amino (HYNESet al., 1993). Unfortunately, because acids are not essential for its biological activi- of the overlapping nature of the thioether rings in streptococcin A-FF22, assignment of tY. the ring order by enzymic and chemical digestions and modifications has proved impossible (JACK et al., 1994a). 2.2.6 Streptococcin A-FF22, In addition, JACK and TAGG(1991) also Lacticin 481 (Lactococcin DR), isolated a naturally-occurring derivative of and Salivaricin A streptococcin A-FF22 devoid of detectable antibacterial activity, apparently as a result of At around the same time as the structure of the loss of the N-terminal 4 amino acids nisin was being reported (GROSS and MOR- which are not involved in formation of ring
335
structures. While C-terminal amino acid-deficient derivatives have not yet been found, it would seem that the antibacterial activity of streptococcin A-FF22 is dependent on (at least) an intact N-terminus, in direct contrast to observations made with the type A lantibiotic epilancin K7 (VAN DE KAMP et al., 1995a). Another type A lantibiotic which shows striking similarities to the structure of streptococcin A-FF22 has been isolated from Lactococcus lactis ssp. lactis and called lacticin 481 (PIARD al., 1992). In addition, RINCEet al. et (1994) have recently isolated a lantibiotic from Lactococcus lactis ssp. lactis DR which proved to be identical to lacticin 481. Preliminary amino acid sequence analysis of lacticin 481 has allowed the isolation of the structural gene, lcfA (PIARDet al., 1993) and complete analysis of the primary structure of the peptide has shown it to be a 28 amino acid, 2901 Da type A lantibiotic with one MeLan, one Dhb and two Lan residues (Fig. 4). Interestingly the peptide shares both sequence and structural similarities with streptococcin AFF22 (JACK et al., 1994a), especially in the presence of the 3 overlapping thioether rings and the single residue of unmodified serine at the C-terminus (PIARD al., 1993). et Salivaricin A (Fig. 4) is a lantibiotic produced by Streptococcus salivarius strain 20P3 which acts against a number of pathogenic streptococci including the principal human pathogen, Streptococcus pyogenes (DEMPSTER and TAGG, 1982). Purification and microcharacterization of the peptide responsible showed that it is a small (2315 Da) type A lantibiotic with one Lan and two MeLan residues (Ross et al., 1993). In addition, this particular lantibiotic is apparently the only one so-far characterized which does not contain either Dha or Dhb. Also, although no attempts appear to have been made to determine the order of the thioether bridges, it seems likely that all three should overlap.
terized (M0RTVEDT and NES, 1990; M0RTet al., 1991). In addition, the structural gene encoding the prepeptide has been isolated and characterized in order to gain an insight into the likely structure of this lantibiotic (SKAUGEN al., 1994). Taken together, et these studies have demonstrated that the peptide is the largest lantibiotic so-far characterized (3764 Da) and probably contains two Lan residues. In addition, in a situation analogous to that observed in Pep5 (Fig. 5; KELLNER et al., 1989) the N-terminus is blocked, probably by a 2-oxopyruvyl group which would arise from the spontaneous oxidative deamination of a N-terminally located Dha residue. However, the nature of the blocking group as well as the arrangement of the thioether bridges has yet to be resolved. Finally, lactocin S has also been shown to contain D-alanine at positions which contain Ser in the prepeptide sequence, prompting speculation that these residues arise following stereospecific hydrogenation of Dha residues at these sites in the propeptide domain (see also Sect. 2.1.3).
VEDT
2.2.8 Enterococcal Cytolysin/Bacteriocin

The cytolysin/bacteriocin of Enterococcus faecalis has long-since been shown to be a virulence determinant which also has the ability to kill certain gram-positive bacterial strains (BROCKand DAVIE,1963), however, the recent isolation of the genetic elements responsible for its biosynthesis have demonstrated the relationship between this toxin and the lantibiotics (GILMORE al., 1994). The cytoet lysin/bacteriocin system appears to comprise two peptides both of which are lantibiotics and both of which are required for antibacterial activity; Cyll is a 4163 Da peptide while Cy12 has a mass of only 2631 Da, however, while lanthionine has been identified in these peptides the exact number of residues etc. is not yet clear (SAHLet al., 1995). From the derived amino acid sequences obtained through analysis of the genes, it appears that the two peptides have regions of identity, suggesting that they may have arisen through gene dupli-
2.2.7 Lactocin S
Lactocin S (Fig. 4) is a large, 37 amino acid lantibiotic produced by Lactobacillus sake which has been purified and partially charac-
336
8 Lantibiorics
cation and partial deletion (GILMORE al., et 1994).
2.2.9 Carnocin U149 and Mutacin

STOFFELS al. (1992) have reported the et isolation of a large lantibiotic from Curnobacterium piscicolu which they named carnocin UI49. The peptide is particularly hydrophobic as judged by its chromatographic behavior, has a mass of 4635 Da and a novel N-terminal amino acid sequence of G-S-E-I-Q-P-R which is blocked to further sequencing at the 8th residue. While the exact number and nature of amino acids present in this lantibiotic remains to be determined, preliminary results suggest it is somewhat similar to lactocin S (SAHLet al., 1995). In addition, the peptide appears to have a propensity for membrane perturbation and seems to interact with nisin-producing strains in a specific way, suggesting there may be an interaction of carnocin U149 with both the cytoplasmic membrane and parts of the nisin-synthesizing machinery of the cell (STOFFELS al., 1994). et Mutacin, a lantibiotic from Streptococcus mutuns T8, has also been isolated and characterized (NOVAKet al., 1994). This lantibiotic has a mass of 3245 Da, contains one MeLan and two Lan residues and has the unique Nterminal amino acid sequence N-R-W-W-QG-V-V-X, where X indicates a sequence blocking group.
2.3 Type B Lantibiotic Primary Structures 2.3.1 Cinnamycin, Duramycin, Duramycin B, Duramycin C, and Ancovenin
Many of the type B lantibiotics described so-far have gained considerable interest, not so much because of their unusual structures nor their antimicrobial activity (which in most cases is only moderate), but because they act as inhibitors of essential enzymes involved in both the human immune and circulatory
blood pressure regulatory systems, including: phospholipase A2 and angiotensin-converting enzyme (see also Sect. 5). Over a period of a number of years, searches for such inhibitors of immune function and blood pressure regulation have identified a number of these compounds (some of which have subsequently been shown to be identical) including: cinnamycin (formerly also known as lanthiopeptin or Ro09-0198), duramycin (formerly also known as leucopeptin), duramycin B, duramycin C, and ancovenin (BENEDICT al., et 1952; SHOTWELL al., 1958; GROSS, 1977; et WAKAMIYA al., 1985,1988; KESSLER al., et et 1987, 1988, 1991; NARUSEet al., 1989; FREDENHAGEN et al., 1990,1991; HAYASHI al., et 1990), the structures of which are shown in Fig. 6. Each of these compounds has been isolated from different strains of Streptomyces spp. or Streptoverticillium spp. and all show limited antimicrobial activity, principally directed against Bacillus spp. (see also Sect. 5). Since the structures of cinnamycin, duramycin, duramycin B, duramycin C, and ancovenin are so similar, they will be regarded here as the cinnamycin-group of type B lantibiotics and described together. Indeed, SAHL et al. (1995) have suggested that since these type B lantibiotics share such an extent of structural and sequence similarity (only 7 conservatively exchanged amino acids overall), that they ought to be viewed as structural variants of the same compounds. As would be expected, the type B lantibiotics of the cinnamycin group all contain lanthionine; indeed, the ring structures and arrangements are conserved among all these peptides (Fig. 6). However, while each contains one Lan and two MeLan residues (FREDENHAGEN et al., 1991; KESSLER al., 1991; et JUNG,1991a, b), there are some notable differences between type A and type B lantibiotics. Firstly, these type B lantibiotics all contain a head-to-tail bridge, formed by a MeLan residue spanning residues 1-18 and this Melan, along with that formed between residues 5 and 11 is in the opposite orientation to that observed in all type A lantibiotics (i.e., the Cys-derived half of the di-amino acid is located toward the N-terminus). In addition, cinnamycin-group type B lantibiotics (with the exception of ancovenin) also con-
337
Fig. 6 The structure of several type B lantibiotics. Amino acid residues involved in posttranslational mod. ifications have been identified by highlighting while additional modified residues are indicated by their acid Alarespective chemical structures. Abu, a-aminobutyric acid; Asp-OH, erythro-3-hydroxy-aspartic NH-Lys, (2S,8S)-lysinoalanine;Ala-S-Ala, (2S,6R)-lanthionine; Ala-SO-Ala, lanthionine sulphoxide; AbuS-Ala (2S,3S,6R)-3-methyllanthionine.
tain a single residue of hydroxy-aspartic acid (Figs. 1 and 6), however, it is not clear from whence or by what mechanism(s) this residue arises (FREDENHAGEN a . 1991; JUNG, et l , 1991a, b; SAHLet al., 1995). Similarly these type B lantibiotics (again with the exception
of ancovenin) contain the aminoether linked, di-carboxy, di-amino acid, (2S,8S)-lysinoalanine (Fig. l),formed between residues 6 and 19 (Fig. 6). It has been proposed (JUNG, 1991a, b) that such a residue could form by the addition of the -amino group of Lysl9
338
8 Lantibiotics
across the unsaturated Dha6, in a manner rather analogous to that observed for the formation of lanthionine shown in Fig. 2. What is not clear about these additional modifications is what role (if any) they play in the biological activity of their respective lantibiotic; indeed, ancovenin contains neither hydroxyaspartic acid nor lysinoalanine, yet is functionally similar to other type B lantibiotics of this group (WAKAMIYA al., 1985; SHIBAet al., 1986), et suggesting that they may not be essential for biological activity.
2.3.2 Mersacidin and Actagardine

By contrast to the cinnamycin-like type B lantibiotics described in the preceding section, mersacidin and actagardine have been studied extensively, in part, because of their strong antimicrobial activities (see also Sect. 5). Furthermore, as can be seen in Fig. 6 they also differ significantly in their structural characteristics, neither containing the headto-tail bridging pattern described above. Mersacidin (Fig. 6) is produced by Bacillus spp. (strain HIL Y-85,54728), is the smallest lantibiotic (1825 Da) and consists of 20 amino acids arranged to give a tetracyclic structure with rings C and D forming a bicycle (CHATERJEE et al., 1992a, b). It is hydrophobic with no net charge, contains a single residue of Dha and 3 MeLan residues and is the only lantibiotic to contain the C-terminal unsaturated amino acid (S)-[(Z)-2-aminovinyl]-3methyl-D-cysteine. This last residue is probably formed by a similar process as that responsible for Cys(Avi) formation in epidermin (Fig. 1; ALLGAIER al., 1986; KUPKEet et al., 1992), except that the dehydrogenated decarboxylated C-terminal Cys residue should react with a Dhb residue, rather than a Dha. In addition, KOGLER al. (1991) have shown et that mersacidin also contains a residue of MeLan (between residue 1 and 2) formed by linkage in the opposite direction to that observed with all type A lantibiotics (i.e., the Cys-derived "half" of the first ring of mersacidin is located to the N-terminal side with respect to the Dhb-derived "half"). Actagardine (formerly gardimycin) was isolated from Actinoplunes spp. ATCC 31048 and 31049 by PARENTI al. (1976) and the et
correct structure of the peptide has been determined by ZIMMERMANN JUNG(1995) and after previous attempts (MALABARBA al., et 1985, 1990 KETTENRING al., 1990). Actaet gardine contains 4 sulphide rings (1 Lan and 3 MeLan residues) with the rings B, C and D forming a tricycle. Interestingly, with the exception of a single, conservative amino acid exchange, ring B of actagardine and ring C of mersacidin are identical (Fig. 6), a feature which may be related to their similar biological activity (see also Sect. 5 ) , which is the inhibition of peptidoglycan biosynthesis (SOMMA et al., 1977; BROTZ et al., 1995; SAHLet al., 1995). In addition, actagardine is the only lantibiotic shown to contain 3-methyllanthionine sulphoxide (Fig. 6), however, it is not yet clear whether this residue has arisen as an artefact of the purification process (KETTENRING et al., 1990; ZIMMERMANN and JUNG, 1995).
3 Lantibiotic Structures in Solution

The lantibiotics, be they either type A or type B, contain a vast number of modified residues, many of which form bridging structures along various portions of the respective peptides length (c.f. Figs. 5, 6 and Tab. 1). Obviously then, these modified residues and unusual bridging patterns should have some ramifications on the overall structure adopted by the peptides in solution. Initially, unsuccessful attempts were made to obtain crystals of several lantibiotics, in order to determine their spatial structures by X-ray crystallography (JUNGet al., unpublished data). Subsequently, the advent of sensitive, high resolution 2-D, 3-D and 4-D NMR techniques has allowed a number of laboratories to look at the solution conformations of various lantibiotic structures (SLIJPERS al., 1989; CHAN et et al., 1989b; PALMER et al., 1989; VAN DE VENet al., 1991a,b; LIANet al., 1991; GOODMAN et al., 1991; FREUND al., 1991a, b, c; et KESSLERet al., 1991; ZIMMERMANN al., et 1993; ZIMMERMANN and JUNG, 1995; VAN DE KAMPet al., 1995a). In addition, a number of
3 Lantibiotic Structures in Solution
339
these studies have also been able to take advantage of the nature of NMR and assess the effects of solutions of differing lipophilicity, micelles, perturbants (e.g., urea) and phospholipid vesicles on the solution structures obtained.
3.1 Type A Lantibiotics
3.1.1 Gallidermin
The structure of gallidermin (Fig. 7) has been obtained both in trifluoroethanol (TFE)/water (95 :5) and dimethylsulphoxide (DMSO) where it adopts an extended, corkscrew-like conformation (FREUND et al., 1991a, b). Ring B (residues 8-11), which is identical to ring B of nisin, forms a p-turn type 11, while the central domain (residues 11-15) shows the greatest degree of flexibility over the whole molecule, due to its content of turn-like motifs. Interestingly, this region incorporates a potential trypsin cleavage site (Lysl3-Dhbl4) which appears to be exposed by this flexibility and has been shown by molecular modelling to fit the active site of the enzyme (FREUNDet al., 1991b). Residues
Overall, gallidermin adopts a distorted helix-like structure which has some degree of flexibility in the central region. In addition, it is amphiphilic with the hydrophobic C-terminal residues aligned on one face, while the N-terminal part contains the hydrophilic residues also oriented towards the opposite face of the cork-screw (FREUNDet al., 1991a, b). In TFE/water the peptide has an overall length of about 3 nm, a diameter of approximately 1 nm and a net dipole moment of around 75 Debye (JUNG,1991a, b). The amphiphilicity, high dipole moment and membrane spanning length observed for gallidermin in solution may help to explain its biological activity, which is the formation of voltage-dependent pores in biological membranes (SAHL,1991; BENZet al., 1991).
3.1.2 Nisin and Subtilin

A number of research groups have independently investigated the solution structure of nisin (and, to some extent, subtilin) and found that in aqueous solution the peptide adopts a relatively random, unordered strucDE ture (VAN VEN et al., 1991a, b; GOODMAN et al., 1991; LIANet al., 1991, 1992; CHANet al., 1992). Some degree of restraint is shown in the structures of both ring B and the overlapping rings D and E, however, ring A and C, several of the residues located at the Nand C-termini of the peptide as well as those in the central part show considerable conformational freedom. By contrast, in mixed, lipophilic solvents such as TFE/water or DMSO the structure of nisin becomes somewhat more stabilized and shows that the peptide adopts overall a-helical structure. Taken together, nisin can be viewed as a pair of helical structures (residues 3-19 and 23-28), separated by a flexible hinge region between amino acids 20-22 (VAN DE VEN et al., 1991a, b; LIANet al., 1992); the remaining residues at the N-terminus (amino acids 1-2) and C-terminus (amino acids 2934) are extremely flexible and could not be defined, even in the stabilizing, lipophilic solutions. In addition, while it is not so obvious in aqueous solution, nisin, like gallidermin, is amphiphilic. Furthermore, a study by GOODMAN et al. (1991) of nisin in DMSO showed
Fig. 7. Stereorepresentation (backbone ribbon) of the solution structure of the type A lantibiotic gallidermin; c.f. Fig. 4 for the primary structure. The dots represent sulphur atoms, and the N-terminus is at the right side.
Asnl8, Ala19, Ala21, and the C-terminal Cys(Avi) are oriented inward forming a hydrophilic core to an otherwise hydrophobic cage-like structure consisting of the overlapping rings at the C-terminus which incorporate the outwardly oriented residues Phel7 and Tyrl8 (FREUND al., 1991a, b). et
340
8 Lantibiotics
that it formed a kinked rod-like structure with an overall length of ca. 5 nm, diameter of ca. 2 nm and calculated dipole moment of at least 80 Debye, all features which are consistent with its mode of action (see also Sect. 5).
3.1.3 Pep5
Although they are not yet complete, studies of the lantibiotic Pep5 in solution suggest that it has a similar structure to the type A lantibiotics gallidermin, nisin, and subtilin previously characterized (FREUND et al., 1991c; FREUND JUNG,1992). Analysis of and the circular dichroism of Pep5 in water suggests it forms a random, disordered structure while, on the addition of lipophilic solvents there is an increasing propensity toward formation of helical elements. Similarly NMR studies have shown that in aqueous solution Pep5 has no defined structure except in the region of the four-membered ring B and that the unbridged regions (amino acids 1-8) and the central region (amino acids 14-23) are particularly flexible and unstructured and (FREUND al., 1991c; FREUND JUNG, et 1992). However, as with the CD experiments, NMR of Pep5 in 90% TFE shows that the peptide is able to adopt an overall amphiphilic helical structure with most of the charged, hydrophilic amino acids oriented to one face of the rod. In addition, the central region is not bridged however, the presence of the two dehydroamino acids appear to stqbilize local structure in this region (FREUND and JUNG,1992).
Fig. 8. Stereorepresentation (backbone ribbon, sulphur atoms represented as dots) of the type B lantibiotic duramycin B. For the primary structure see Fig. 6.
3.2 Type B Lantibiotics 3.2.1 Cinnamycin and the Duramycins

In many respects the type B lantibiotics cinnamycin, duramycin, duramycin B, duramycin C, and ancovenin can be considered structural variants of one another (DE Vos et al., 1991) as they have the same principal bridging pattern (apart from ancovenin which lacks the lysinoalanine bridge) and amino acid exchanges are relatively conservative (FREDENHAGEN 1991; SAHL et al., et al., 1995). Therefore, it is not unreasonable to ex-
pect that their solution structures should be somewhat similar. In addition, all of these lantibiotics contain a head-to-tail MeLan bridge which reduces the possibilities for conformational freedom when compared to the type A lantibiotics (c.f. Figs. 4 and 6). The currently published structures of cinnamycin, duramycin B, and duramycin C (Fig. 8) as well as the preliminary structure of ancovenin are all in general agreement as to the overall structure of these lantibiotics (KESSLER et al., 1991; ZIMMERMANN al., 1993; et NISHIKAWA al., 1988). The peptides are et bent into a U-shape by a turn induced by Pro9 and stabilized by the three thioether rings. In addition, the structure is further stabilized by antiparallel P-sheets in the N- and C-terminal regions and the planar nature of the backbone is slightly distorted by the lysinoalanine ring between residues 6 and 19 (ZIMMERMANN al., 1993). Furthermore, et the amino acid exchanges have little or no influence on the overall structures, but do appear to influence the degree of mobility of local structural elements and the overall hydrophobicity of the peptides. Like the type A lantibiotics, the type B lantibiotics are highly amphiphilic with all of the hydrophobic amino acids clustered into the bend of the U and the hydrophilic residues localized at the termini. Interestingly, the C-terminal region of cinnamycin has considerable structural similarity to cyclo-TP5, a thymopoietin analog able to stimulate T-cell maturation (KESSLERet al., 1991). The possibility that cinnamycin could stimulate T-lymphocytes may help to explain the previous observation that cinnamycin has in vivo anti-Herpes simplex activity (WAKAMIYA al., 1988). et
4 The Genes/Proteins Involved in Lantibiotic Biosynthesis and Genetic Regulation
341
biosynthesis are shown for epidermin in Fig. 9 (see also Sect. 4.8). In light of these observations, it is not surSo-far at least, the spatial structure of mersacidin has not been reported. However, pre- prising that following the discovery of the liminary results for actagardine (ZIMMER- first specific genes encoding lantibiotic strucMANN and JUNG, unpublished data) show et tural genes (SCHNELL al., 1988; BUCHthat it is a very rigid peptide, with well-de- MANN et al., 1988; BANERJEE HANSEN, and 1989), subsefined structure, even though it lacks the head- 1988; KALETTA and ENTIAN, to-tail bridge of the other type B lantibiotics. quent analysis of the DNA flanking these In addition, the structure appears to have an genes has revealed the presence of a number of other open reading frames, often clustered amphiphilic nature. together, which seem to be involved in the biosynthesis, genetic regulation and immunity required during lantibiotic production. A number of these lantibiotic-synthesizing gene clusters are compared in Fig. 1 0 wherever possible, geneslgene products with similar function have been given the same letter designation (DEVos et al., 1991) and the prefix fanLan (lantibiotic-related) has been used to indicate groups of genes (SAHLet al., 1995). Thus, e.g., the general term fanA indicates all lantibiotic structural genes while the general 4.1 The Requirement for Multiple term LanA indicates all pre-lantibiotics.
3.2.2 Mersacidin and Actagardine
4 The GenedProteins Involved in Lantibiotic Biosynthesis and Genetic Regulation

Gene Products in Lantibiotic Biosynthesis
Lantibiotics are encoded by specific gene sequences and are synthesized on the ribosome hence, the prepeptides are limited to containing only the 20 amino acids allowed et for in the genetic code (SCHNELL al., 1988; SAHLet al., 1995; JACK et al., in press; JACK and SAHL,1995). Therefore, in order to mature into the biologically active forms which have been isolated outside the producing cells, a number of posttranslational modification events must occur including: specific amino acid modifications, formation of intrapeptide thioether and/or aminoether rings, removal of the leader peptide from the N-terminus and transport of the peptide out of the cell to the extracellular matrix. In addition to these functions other controlled events are necessary for lantibiotic production, such as the regulation of their biosynthesis and the generation of immunity mechanism(s) to protect the producing cell from the action of its own lantibiotic. Although it is not possible to state a definitive order of events, the generalized chain of reactions leading to lantibiotic
4.2 Structural Genes and Pre-Lantibiotics 4.2.1 The Location and Features of the Structural Genes and their Transcription
Much debate has surrounded the location of the structural gene encoding pre-nisin, with various research groups suggesting that it was located on either the chromosome or a plasmid (KOZAK al., 1974; TSAIand SANDINE, et 1987; BUCHMANN al., 1988; KALETTA and et ENTIAN, 1989; GIREESH al., 1992). Howevet er, it is now clear that the entire operon responsible for nisin production, consisting of nisABTCIPRKFEG (KUIPERS al., 1993a; et VAN DER MEERet al., 1993; ENGELKE al., et and 1995) is encoded 1994; SIEGERS ENTIAN, on a 70 kbp conjugative transposon, either called Tn5301 (DODD et al., 1990, 1991; HORNet al., 1991) or Tn5276 (RAUCH al., et and DE Vos, 1992), de1990, 1991; RAUCH pending on the strain of Lactococcus factis
342
8 Lantibiotics
&-
cleavage site
Pre-epidermin
* Site-specificdehydration (restricted to propeptide domain) * Formation oflanthionine and 3-methyllanthioninenngs * Reductive earboxylation of C-terminal cysteine residue * Formation of uninovinylcysteine thioether nng * Cleavage of leader peptide
r
Epidermin
Fig. 9. General schema for the events occurring during the biosynthesis of the type A lantibiotic epidermin.
from which it was isolated. Furthermore, these authors have shown that these transposons also encode several of the genes necessary for sucrose metabolism in lactococci. Therefore, the recent observations that sucrose metabolism-related genes can be found immediately proceeding the nisA BTCIPRKFEG gene cluster suggests that this represents all of the genes necessary for the production of nisin (SIEGERS ENTIAN, and 1995). The nisin structural gene, nisA (Fig. lo), is the first gene in the cluster and is located very close to the 5' terminus of the transposon (DODDet al., 1990). BUCHMAN al. (1988) et identified a putative pindependent terminator immediately following the structural gene and subsequent studies have mapped the transcription start-site and identified an appropriate promoter for transcription of nisA (KUIPERS et al., 1993a; ENGELKEet al., 1992). In addition, it has been suggested that
nisA and nisB may be co-transcribed and appropriate sized transcripts have been identified along with a terminator after nisB (STEEN et al., 1991; ENGELKE al., 1992; KUIPERS et et al., 1993a). Little is known of the events surrounding transcription of the genes downstream of nisAB, however, a tandem promoter immediately preceding nisT could suggest that the remaining genes are transcribed as part of a polycistronic message (STEENet al., 1991). The genes involved in the biosynthesis of epidermin have also been studied extensively. The gene cluster responsible for epidermin production consists of epiT'TABCDQP and is carried on the 54 kbp plasmid pTu32 (SCHNELL al., 1991,1992; AUGUSTIN al., et et 1991, 1992). The structural gene, epiA (Fig. lo), is very likely transcribed along with epiBCD, since epiB does not appear to possess its own promoter and transcripts of ap-
343
E G
nir
B
I
x x
x x n +
epi
T TA K Hw
T
B
X
I X K
l-
Pep
IA
H m
n
T
n
W
lar
UVP
n;r
LI L2
M
>I
-001
>
3kb
Fig. 10. The arrangement of the gene clusters involved in the biosynthesis of the the lantibiotics nisin (nis), subtilin (spa), epidermin (epi), Pep5 (pep), lactocin S (las),lactococcin DR (lcn), and the enterococcal cytolysin/bacteriocin (cyl). In general, homologous genes have been given the same alphabetic suffix. Thus, A-genes are the structural genes encoding the prepeptides (with the exception of spas), B, C, and M-genes code for putative modification enzymes, T, E, and F-genes encode ABC superfamily translocator proteins, P-genes are responsible for the production of leader peptidases, R and K-genes produce histidine kinasel response regulator proteins (with the exception of epiQ which is equivalent to the R-genes of other gene clusters), and I-genes encode specific proteins involved in producer self-protection. In addition, the function of the alternatively named genes in the lactocin S-producing gene cluster are unknown. The arrows represent the relative direction of transcription of the respective genes.
propriate size for co-transcription (ca. 5 kb) can be found (SCHNELL al., 1992; AUGUSet TIN et al., 1992). Furthermore, these studies have speculated that a putative terminator between epiA and epiB may be leaky, allowing some readthrough and thereby regulating the levels of transcription of epiB, epic, and epiD. In addition, the start-site for epiA transcription along with an appropriate promoter have been identified; the promoter is activated by the regulatory protein EpiQ which probably binds to an inverted repeat
located immediately preceding the -35 region (PESCHEL al., 1993). et Alternatively, the subtilin structural gene, spas (Fig. lo), is located in the middle of the subtilin-producing gene cluster spaBTCSIRKFG, preceded.by an appropriate transcriptional promoter (BANERJEE and HANSEN, 1988). These authors were also able to map the transcriptional start-site and identify a 0.5 kb mRNA transcript corresponding to spas which had an unusually long half-life of approximately 45 min. In addition, a pair
344
8 Lantibiotics
of typical pindependent terminator struc- cin DR, lcnDRl (Fig. lo), was located on a 70 tures have been identified immediately pro- kbp plasmid in Lactococcus lactis ssp. lactis et ceeding the structural gene (BANERJEE and DR along with two other genes (RINCE al., HANSEN, 1988; HANSEN al., 1991). While 1994). Interestingly, this study showed that et transcriptional data for the genes downstream only lcnDRl and lcnDR2 were required for of spas has not been reported, the identifica- the production of lactococcin DR, while the tion of a transcriptional start-site for spaB 3rd gene lcnDR3 which has strong homology (the first gene in the subtilin-synthesizing with transport proteins was dispensable. Simigene cluster) has prompted speculation that lar genes to lcnDR2, designated lasM and upstream genes are produced as a large poly- cylM have been located in the gene clusters cistronic mRNA transcript (CHUNG and responsible for production of lactocin S and HANSEN, 1992; CHUNG al., 1992; HANSEN, cytolysin/bacteriocin, respectively (SKAUGEN, et 1994; GILMOREet al., 1994). In addition, 1993). PepA, the structural gene for Pep5 (Fig. 10) lasM is surrounded by a number of additional can be localized to the 18 kbp plasmid ORFs which do not appear to have homology pED503, harbored within Staphylococcus epi- with known proteins, one (or more) of which dermidis strain 5 (ERSFELD-DREBEN al., may encode the novel enzyme responsible for et 1984; KALETTA al., 1989). The gene cluster catalyzing the stereospecific conversion of et 1994). Analysis of responsible for Pep5 production consists of Dha to D-Ala (SKAUGEN, pepTIAPBC, all of which, with the exception the DNA surrounding scnA, the structural of the putative transporter gene (pepT), ap- gene encoding pre-streptococcin A-FF22, has pear to be essential for Pep5 production and revealed that scnA is preceded by a putative immunity (BIERBAUM al., 1994; MEYERet terminator and proceeded by an inverted reet al., 1995). Both pepl (responsible for immuni- peat which might act either as a transcriptionty) and pepA are preceded by promoters, al terminator or as an mRNA processing site et however, while both can be transcribed inde- (HYNES al., 1993). So-far at least, there is a distinct paucity of pendently, Pep1 is not produced in the absence of pepA (REISet al., 1994). In addition, information concerning the genetic elements REISet al. (1994) were able to identify a weak responsible for the production of type B lantipindependent terminator immediately after biotics and only the structural genes for cinpepA and suggested that, analogous to the sit- namycin and mersacidin have yet been reuation observed in the epidermin-synthesiz- ported (ENTIANand KALETTA,1991; KAet ing gene cluster, read-through may occur and LETTA et al., 1991a; BIERBAUM al., 1995). regulate downstream transcription. This view In the case of cinnamycin, the structural gene would be further supported by the observa- (cinA) was identified, cloned and sequenced tion that the proceeding ORF, pepP, appears from the chromosome of Streptoverticillium AN and KALETTA, to lack a promoter. Recent analysis of the griseoverticillatum (ENTI et genes responsible for production of epilancin 1991; KALETTA al., 1991a) while, the merK7 have suggested that it also has a similar sacidin structural gene (mrsA) was identified, arrangement of genes to that observed for the cloned and sequenced from the producing Pep5-producing gene cluster except that no strain Bacillus subtilis HIL Y-85,54728 et immunity gene (elk4 could be identified be- (BIERBAUM al., 1995). In both cases it is tween elkT and elkA, the structural gene apparent that the products (pre-cinnamycin and pre-mersacidin) are quite different from et (VAN DE KAMP al., 1995b). The structural genes for a number of other the structural gene products of the type A lantibiotics including the bacteriocidcytolysin lantibiotics; both have extremely long leader of enterococci, lactococcin DR (also called peptides with cleavage sites resembling those lacticin 481), SA-FF22 and lactocin S have found in signal sequences for secreted proalso been identified and sequenced (GIL- teins. So-far at least, no information concernMORE et al, 1994; RINCEet al., 1994; PIARD ing additional genes involved in the biosynet al., 1993; HYNESet al., 1993; SKAUGEN, thesis of the type B lantibiotics appears to 1994). The structural genes for pre-lactococ- have been reported.
345
Physical evidence for the structure of the pre-type A lantibiotics has come from the isolation of pre-Pep5 from the producing strain Staphylococcus epidermidis strain 5 (WEILet In general, the information currently avail- al., 1990). In this study, pre-Pep5 was isolated, able concerning the structure of the lantibiot- albeit in very small quantities due to the apic prepeptides has been deduced from their parent raid turnover within the cell, and gene sequences. Overall, lantibiotic prepep- shown to contain modified amino acids only tides (Fig. 11) consist of two domains: a lead- within the propeptide domain, even though er peptide and a propeptide domain, sur- the leader peptide contains hydroxyl amino rounding an appropriate processing site. In acids which could potentially be dehydrated. addition, the leader peptide domains are gen- Similar results have subsequently been oberally acidic while the propeptide domains tained with pre-nisin (VAN DER MEER et al., are generally basic or neutral, both domains 1993). In addition, using a mutant S. epiderhave predicted helical propensity and are sep- midis strain 5 it has subsequently been possiarated by a predicted turn structure encom- ble to isolate prepeptides in different stages passing the cleavage site which may allow the of dehydration, however, the leader peptide processing protease access to this region alone was never detected, suggesting it is rap(JUNG, 1991a, b; JACK et al., 1995; SAHLet idly destroyed after cleavage (SAHLet al., al., 1995). Exceptions to these observations 1991). Chemical synthesis of appropriate pre-lancan be found in the two type B lantibiotic prepeptide sequences so-far identified (EN- tibiotics has also provided useful insight into the structure of their fully unmodified form. TIAN and KALETTA, 1991; KALETTA et al., et 1991a; BIERBAUM al., 1995); the leader Both the propeptide and leader peptide dopeptides are considerably longer and do not mains (as well as segments of these) of gallicontain the structural features identified for dermin and Pep5 have been synthesized, as well as the complete pre-gallidermin (BECKthe type A lantibiotics (Fig. 11).
4.2.2 The General Structure of Prepeptides
Type A
pre-nisin prc-subtilin prcPep5 pre-epidermin
cleavlge site
MSTKDFNLDLVSVSKKDSGASPR MSKFDDFDLDWKVSKQDSKITPQ MKNNKNLFDLEIKKETSQNTDELEPQ

MEAVKEKNDLFNLWKVNAKLSNDSGAEPR
1 -
ITSISLCTPGCKTGCHCSIWSK
WKSESLCTPGCVTWQTCFLQLTCNCKISK
- TAGPAIRASVKQCQKTLKATRLFTVSCKGI(NGCK
IASKFICTPGCAKTGSFNSYCC
GKNGVFKTISHECHLNTWAFIATCCS
KGGSGVIHTISHECNNNSWQRIFTCCS
pre-SA-FE2 pre-lactmccin DR prc-salivarichA pre-laetocin S
NEKNNEVINSIQEVSLEELDQIIGA MKEQNSFNLLQEVTESELDLILGA
MNAMKNSKDILNNAIEEVSEKELMEVAGG
MKTEKKVLDELSLHRSWGARDVESSMNAD
KRGSGWIATITDDCPNSVFVCC
- STPVLASVAVSMEUPTASVLYSWAGCmSAKHHC
Type B
p-innunycin pre-masacidin
~ ~ I L ~ S W D A D F R A A L L E N P A A F G A S A A A L P T P V E R Q D a CRQSCSFGPETMCDGNTK
MSQEAIIRSWKDPFSRENSTQNPAGNPFSELK!?AGXDKLVGAGm
CTFTLPGGGGVCTLTSECIC
Fig. 11. The primary structure of several representative lantibiotic prepeptides, i.e., the primary translation product of the respective structural genes. The sequences are shown centered around the cleavage site for leader peptide processing.
346
8 Lantibiotics
SICKINGER and JUNG, 1991). Circular dichroism analysis of these segments confirmed that the leader peptide formed predominantly helical structures. In addition, both synthetic pre-gallidermin and naturally isolated unmodified pre-Pep5 showed stronger helical propensity than either of their respective domains alone, suggesting that the prepeptide helix may be stabilized by interaction between the pro- and leader peptide domains (BECK-SICKINGER JUNG,1991; SAHLet and al., 1991). Similarly, both pre-nisin and a large number of their fragments have been synthesized and studied (BYCROFT al., 1991; SURoVOY et et al., 1992). Initial results from the NMR analysis of pre-nisin in water suggests that it is highly flexible with little or no preference for particular conformations (FREUND and JUNG,1992). In addition, synthetic pre-nisin has been shown to stoichiometrically bind Zn2+ ions, an event which may act to stabilize the conformation of the prepeptide and may have ramifications in other biosynthetic reactions (SUROVOY al., 1992). et
4.2.3 The Possible Role(s) of the Leader Peptide

A number of roles for the leader peptide have been suggested and should be considered. Firstly, the leader peptide may serve to keep the prepeptide in an inactive form within the cell. This view may be supported by the observations that leader peptide cleavage is generally the last step in lantibiotic biosynthesis and that fully modified but unprocessed pre-nisin and pre-pep5 are not biologically active (WEILet al., 1990; SAHLet al., 1991; VAN DER MEERet al., 1994). Alternatively, it may be that the leader peptide signals transport of the lantibiotic out of the cell. However, the leader peptides of lantibiotics share little similarity with the signal peptides of the sec-dependent export systems (PUGSLEY, 1993) and the identification of specific transport systems of the ABC-superfamily (FATH and KOLTER, 1993) within lantibiotic-producing gene clusters would suggest that sec-dependent transport is not used, although the leader peptide
might still direct the immature lantibiotic to the identified transporters. Finally, the conserved properties of the lantibiotic leader peptides might suggest that they contain specific sequences or motifs which direct the biosynthetic enzymes to the appropriate site for amino acid modification. Alternatively, they may stabilize the conformation of the propeptide domain such that the biosynthetic machinery can access it and carry out specific posttranslational modifications (JUNG, 1991a, b). Two recent studies may support this last proposal. Firstly, VAN DER MEER et al. (1994) have shown that a number of lantibiotics have conserved motifs at particular positions among their respective leader peptides. In this study, they identified the conserved residues Phe-18, Asn or Asp-17, Leu-16, Asp or Glu-15, Ser-10, Asp-7, Ser-6, Pro-2, and Arg or Gln-1 and systematically set about altering these residues in the structural gene for nisin by site-directed mutagenesis. Alterations at Arg-1 resulted in failure of the protease to cleave the leader peptide while, somewhat surprisingly, alterations to the highly conserved Pro-2 had no effect on either biosynthesis or processing. Similarly, changing the Ala-4 resulted in production of unprocessed, mature nisin, suggesting that the peptide no longer fit the active site of the leader peptidase. Alternatively, exchanges with Asp-7 had little effect while those at Ser-10 resulted in improved production of nisin. However, exchanges in positions -6, -15, -16 and -18 could all be shown to completely inhibit nisin biosynthesis, suggesting that (at least) these residues may be essential for the biosynthesis of this lantibiotic. Secondly, gene fusion of the nisin propeptide domain to the leader domain for subtilin production resulted in the production of fully modified but unprocessed nisin, suggesting that the biosynthetic machinery of the cell (with the exception of the processing protease) recognized the hybrid protein as a suitable substrate (KUIPERS al., 1993b). Alteret natively, a similar hybrid produced by fusion of the nisin leader domain to the subtilin propeptide region was neither matured nor processed in Bacillus subtilis and only a leader peptide domain which contained the first 7
347
amino acids of the subtilin leader fused to the remaining 17 amino acids from the nisin leader sequence was sufficient to get production et of subtilin (RINTALA al., 1993). Taken together, these results seem to suggest that there may be structural motifs in the leader peptide which aid in creating a suitable substrate for subsequent modifications.
4.3 Amino Acid Modifying Proteins 4.3.1 LanB/LanC and LanM Proteins
The most obvious and striking feature of the structure of the lantibiotics is their content of the non-protein amino acids lanthionine and 3-methyllanthionine. Since neither codons nor tRNAs for these amino acids are found in the cell, they must arise by the action of specific enzymes acting on the precursor amino acids found in the prepeptide. Isolation and genetic analysis of the gene clusters responsible for the production of a number of lantibiotics has revealed the presence of a number of open reading frames; although the role of most can be predicted by homology with other proteins, several genes have been identified which have no homology with proteins of known function and may be the gene products responsible for the formation of the novel amino acids found in the lantibiotics (SCHNELL al., 1991; AUGUSTIN al., 1991; et et KALETTA et al., 1991). Indeed it has been clearly shown by gene disruption and complementation of both spaB and spaC as well as with epiB and epic that lantibiotic biosynthesis is dependent on the production of these proteins (KLEIN al., 1992). In addition, the et appearance of subtilin in the culture supernatant correlates directly with the expression of SpaB (GUTOWSKI-ECKEL 1994). et al., The lanB genes so-far characterized encode proteins of approximately lo00 amino acids (Fig. 10) which are principally hydrophilic, but which also have several hydrophobic regions which might indicate membrane spanning domains. Indeed, using antibodies raised
et to NisB, ENGELKE al. (1992) were able to demonstrate that NisB associates with the membranes of fractionated nisin-producing cells, however they were not able to determine whether NisB was an integral membrane protein or merely associated loosely with this fraction. Similar results have subsequently been obtained with SpaB analysis, suggesting that the site of lantibiotic biosynthesis might be organized and localized to the cytoplasmic membrane (GUTOWSKI-ECKEL et al., 1994). The deduced size of the proteins produced by the fanC genes (Fig. 10) is somewhat smaller at around 450 amino acids (SAHLet al., 1995) and a number of conserved residues can be identified within their sequences. Although localization studies have not yet been reported, the LanC proteins appear to consist of regularly alternating regions of hydrophobic and hydrophilic nature (ENGELKE al., et et al., 1992; GUTOWSKI-ECKEL 1994). Recently a number of lantibiotic-producing gene clusters have been identified which do not contain either the lanB or lanC genes, including lactococcin DR (also known as lacticin 481), lactocin S and the enterococcal cytolysinhacteriocin (RINCEet al., 1994; PIARD et al., 1993; SKAUGEN, 1994; GILMORE al., et 1994). Interestingly, these gene clusters contain instead a different gene, designated lanM (Fig. lo), which share homology in their Cterminal region with the lunC genes. AIthough they do not appear to have any homology with the lunB gene products, it is possible that the LanM proteins represent hybrid modifying proteins, able to carry out the functions of both LanB and LanC. This hypothesis may be further supported by the observation that the leader peptides of the respective lantibiotics synthesized by lanM-containing gene clusters are quite different to those synthesized by lanB/LanC-containing operons (Fig. 10). What must still be investigated is the molecular mechanism(s) involved in the catalysis of dehydration and thioether ring formation, regardless of whether or not it is carried out by a LanB/LanC or LanM system or even by some hitherto unidentified protein.
348
8 Lantibiotics
4.3.2 EpiD
10). Interestingly, all belong to the superfamily of transport complexes known as the ATPThe gene cluster responsible for the biosyn- binding cassette (ABC) transporters (HIGand 1993). In gen1992; FATH KOLTER, thesis of epidermin contains the gene epiD GINS, (Fig. lo), which is not found in any of the oth- eral terms, ABC transporters consist of a er lantibiotic-producing operons (AUGUSTIN dimeric complex consisting of two duplicated et al., 1991, 1992; SCHNELL al., 1991, 1992); domains, one domain of each monomer conet similarly, only epidermin and its structural taining the cytoplasmic ATP-binding region analog gallidermin possess the C-terminal characterized by a consensus Gly-Xaa-Glymodified amino acid (S)-[(Z)-2-aminovinyl]- Lys-Ser-Thr sequence (where Xaa indicates D-cysteine (Figs. 1 and 4; ALLGAIERet al., any amino acid) and the other a membrane 1986 KELLNERet al., 1988). Furthermore, spanning region. In some cases, each domain expression of epiD has been shown to be es- may be encoded by separate genes, although, sential for epidermin production (SCHNELL et with the exception of epiTT,all of the charal., 1991, 1992; AUGUSTIN al., 1991, 1992). acterized lantibiotic transporters are encoded et Subsequently, the protein EpiD has been ex- by a single gene encoding both domains tensively studied and represents the first ex- (FATH and KOLTER, 1993; SAHL et al., ample of a novel enzyme responsible for ami- 1995). In the case of epidermin production, it is no acid modification isolated from a lantibiotic-producing strain. KUPKE al. (1992) were not clear that epiTT is a functional transloet able to overexpress EpiD as a fusion protein cator since, a frame shift would be required to with maltose-binding protein (MBP) and re- transcribe both proteins. Furthermore, hetercover the EpiD after removal of the MBP ologous expression of epidermin in Staphyloanalysis of EpiD revealed that is a 21 kDa fla- coccus carnosus suggests that transport is not voprotein enzyme which requires the cofactor dependent on epiTT and that host-encoded flavinmononucleotide (FMN) and is able to transport systems can efficiently substitute for et catalyze the oxidative decarboxylation of the this putative transport system (SCHNELL et C-terminal cysteine residue of pre-epidermin al., 1992). Similarly, BIERBAUM al. (1994) (Fig. 12). Subsequently, both purified pre-epi- were able to show that disruption of pepT redermin (KUPKE et al., 1993) and synthetic duced yields of Pep5 by about 90% compared pro-epidermin have been used as substrates to the wild type. The observations of both of to characterize the catalytic activity of this en- these studies indicate that the gene clusterzyme by mass spectrometry (KUPKEet al., encoded transporters in these staphylococci 1994). From this study it could be concluded may well be complemented, at least in part, that EpiD is responsible for, at least, the oxi- by other cellular transporters. Similar results dation of the C-terminal Cys residue of epi- have been obtained with other transporters of dermin (the decarboxylation shown in Fig. 12 the ABC-superfamily, which are generally might occur simultaneously and that, since thought to have some degree of flexibility in EpiD acted both on pre-epidermin and the their substrate specificity (FATH and KOLTsynthetic propeptide domain, its activity is ER, 1993). In direct contrast, interruption of not directed via the leader peptide region the spaT gene results in abolition of the production of subtilin by Bacillus subtilis of the prepeptide form. (CHUNGet al., 1992; KLEIN and ENTIAN, 1994).
4.4 Transport Proteins

Genes which are likely to encode proteins responsible for the transport of lantibiotics out of the cell (designated lanT genes) have so-far been identified in each of the characterized, lantibiotic-producing operons (Fig.
4.5 Leader Peptidases (LanP)

Three genes encoding putative peptidases (Fig. 10) which are probably responsible for the cleavage of the leader peptide of the respective lantibiotics have so far been se-
349
[precpidermin(-30to +21)]-N
H
I
(CYS+W
II
\@,/c-oH
C-H I YZ H SH
[pre-epidermin(-30to +21))-N
H I
\_7--OH
O II
@ha+l9) [pre-epidermin(-30to +18)] -N
H I
\ /
H O I II C- [praepidemin(+20to +21)]-N
\(Q
C II CH2
CH CH I SH
II
Fig. 12. Proposed pathway for the formation of the aminovinyl cysteine residue at the CH H O I I II terminus of epidermin by the C- [precpidemin(+20 to +21)]-N [pre-epidemin(-30to +18)] -N flavoprotein enzyme E ~ ~ D . \m \ d The chirality of the respective CH H-C II residues is indicated (where I CH H-C S appropriate) above the a-carI bon atom. FMN, flavinmonoH nucleotide; FMNH2, reduced (S)-[(Z)-2-aminovinyl]-o-cysteine FMN.
Stereospecific addition
quenced nisP (VANDER MEER et al., 1993; ENGELKE al., 1994), epiP (SCHNELL al., et et 1992), and pepP (MEYERet al., 1995), and VAN DE KAMPet al. (1995b) have identified the partial sequence of elkP, probably re-
sponsible for cleavage of the leader peptide of epilancin K7. Each of the proteins shares homology with subtilisin-like serine proteases, showing conserved active-site residues and a putative oxyanion hole.
350
8 Lantibiotics
NisP (Fig. 10) encodes a putative 74.7 kDa protein which, when expressed in E. coli, can be detected as a 54 kDa protein, consistent with processing of a prepropeptide to generate a mature protease (VAN DER MEERet al., 1993). Furthermore, NisP contains a C-terminal extension not found on most other proteases of this class and which could serve as a membrane anchor, suggesting that NisP is both exported from the cell and anchored to the outside of the cytoplasmic membrane. This observation would suggest that cleavage of the nisin leader peptide should be the ultimate step in maturation, a conclusion confirmed since modified but unprocessed nisin can be found in the culture media after growth of a NisP-deficient mutant of a nisinproducing Lactococcus lactis (VAN DER MEER et al., 1993). In addition, this study showed that NisP produced in E. coli was able to cleave this accumulated unprocessed nisin to release the mature, biologically active form. The putative protease required for processing of epidermin (EpiP) is also encoded within the epidermin-synthesizing gene cluster (Fig. 10) and is a 461 amino acid protein with et calculated mass of 51 kDa (SCHNELL al., 1992). Interestingly, heterologous expression of epidermin in Staphylococcus carnosus is not dependent on the presence of a functional epiP gene, suggesting that host-encoded protease(s) are capable of substituting for EpiP (AUGUSTIN al., 1992). Furthermore, no et equivalent protease has been found in the subtilin-synthesizing gene cluster, suggesting that an intrinsic Bacillus subrilis-encoded protease is able to process immature subtilin (SAHL al., 1995). et By contrast, pepP (Fig. lo), encoding the leader peptidase involved in Pep5 maturation, encodes a smaller protease which is devoid of a preprosequence and is therefore not exported from the cell (MEYERet al., 1995). Thus the site of leader peptide cleavage for Pep5 must be intracellular and is therefore not the ultimate step in lantibiotic maturation in Staphylococcus epidermidis 5. In addition PepP lacks some of the conserved residues typical of subtilisin-like serine proteases however, this study showed, using both gene disruption and site-directed mutagenesis of spe-
cific residues, that PepP is essential for processing of the immature Pep5 Similarly, the partial sequence of elkP suggests that epilancin K7 processing is carried out inside the cell, since the putative peptidase lacks a preprosequence required for export (VAN DEN KAMP et al., 1995b).
4.6 Proteins Regulating Lantibiotic Biosynthesis

Except in the cases of subtilin and mersacidin production, lantibiotics are generally produced most abundantly during logarithmic growth when energy sources are at their maximum, suggesting constitutive production as and opposed to regulated biosynthesis (SAHL BRANDIS, 1981; HORNER et al., 1990, DE VUYST and VANDAMME, 1991; JACK and TAGG, 1992); however, this may not be the case since the lantibiotic-producing gene clusters appear to encode specific regulatory elements (Fig. 10). Like the transport proteins which are members of a larger superfamily of transport proteins, so too the gene products responsible for the regulation of lantibiotic biosynthesis appear to part of a large group of proteins with similar function, that of the twocomponent response/regulatory elements. In general (MSADEKet al., 1993), these regulatory elements are made up of two proteins the first of which is a membrane-bound histidine kinase able to respond to an extracellular signal by autophosphorylation of a specific histidine residue in its cytoplasmic domain. Subsequently, the phosphate group is transferred to the second component, an intracellular regulatory protein which is usually a transcriptional activator, able to bind DNA. Genes with homology to these two-component regulators have been found in the gene clusters encoding biosynthesis of both nisin and subtilin (KLEINet al., 1993; KUIPERSet al., 1993a; VAN DER MEER et al., 1993; ENGELKE et al., 1994). Both the insertional inactivation of spaRK followed by complementation of the sapR (KLEINet al., 1993) as well as deletion of nisR (VAN DER MEERet al., 1993) have been used to demonstrate that both of these genes are essential for regulation of subtilin and nisin biosynthesis, respectively.
351
However, so-far at least, none of the respective proteins have been studied in detail, nor have potential activator binding regions been identified in the gene clusters. In contrast, considerable study of the regulation of epidermin production, both at the DNA and protein level, has been carried out. The gene cluster responsible for epidermin biosynthesis (Fig. 10) encodes EpiQ, which appears to be equivalent to the regulator element and shares significant homology with both SpaR and NisR (AUGUSTIN al., 1992; et SCHNELL al., 1992; PESCHEL al., 1993). et et While a respective histidine kinase has not been identified, overexpression of EpiQ leads to increased epidermin production and the gene appears to be essential for heterologous expression of epidermin in Staphylococcus curnosus, suggesting that this heterologous host may provide a suitable response element. In addition, the purified EpiQ has also been studied; EpiQ is a 25 kDa DNA-binding transcriptional activator able to bind to one or more of several putative operator sites upet stream of the epiA promoter (PESCHEL al., 1993). Interestingly, no response/regulator gene pair has yet been identified associated with the gene cluster responsible for Pep5 biosynthesis which is carried on the plasmid pED503 in Stuphylococcus epidermidis 5 (ERSFELD-DREBEN al., 1984; MEYERet al., et 1995). However, a vector containing the genes epiABCDQ, which are insufficient for epidermin production in Staphylococcus curnosus, converted a strain of S. epidermidis 5 devoid of pED503 into an epidermin producer (AUGUSTIN, 1991). This gene cluster is insufficient for epidermin production in S. curnosus since epiQ is transcribed from the same promoter as one of the deleted genes (epiP; Fig. 10). In addition, a chromosomal fragment isolated from S. epidermidis 5 and containing a response/regulator could complement this vector and give epidermin production in S. curnosus (AUGUSTIN, 1991). Thus, it may be that since these chromosomal genes from S. epidermidis 5 can regulate heterologous epidermin production, they may also be capable of directing the biosynthesis of Pep5 What remains unclear is: What is the external signal to which the response regulatory
elements are responding? In the case of nisin, DE VUYST and VANDAMME (1991, 1992) have suggested that nisin production may be associated with carbon source control, since nisin production and sucrose metabolism phenotypes are both encoded on the same transposon and that high phosphate levels could stimulate nisin biosynthesis (DE VUYSTand VANDAMME, 1993). However, HORNER al. et (1990) demonstrated that high phosphate was detrimental for production and that the carbon source had not effect on biosynthesis of both epidermin and Pep5 Thus, the signal remains enigmatic however, lantibiotic biosynthesis seems, in general, programmed to occur when there is an abundance of energyproviding substrate in the growth medium.
4.7 Producer Self-Protection (Immunity) Mechanisms

Clearly, lantibiotic synthesizing cells are producing substances which are potentially detrimental to their own continued well-being, however, such strains generally show a high degree of resistance to the action of their own lantibiotic. Recent studies have shown that this resistance is provided by specific producer self-protection (immunity) mechanisms. A number of early studies using coelimination or co-transfer demonstrated that lantibiotic production and immunity were linked (ERSFELD-DREBEN al., 1984; GASet and WANNAMAKER, 1978). SON,1984; TAGG More recently, the genes encoding specific immunity-related polypeptides have been identified (Fig. 10) at least in the producers of nisin, subtilin, and Pep5 (KUIPERSet al., 1993a; KLEIN and ENTIAN, 1994; REIS and SAHL,1991; REIS et al., 1994). Interestingly, these studies have shown that, while there is a great deal of similarity between Spa1 and NisI, the producing strains do not share cross immunity with each other, nor are they immune to the effects of Pep5 Indeed, cross immunity between lantibiotic producers has only been demonstrated where the lantibiotics produced can be considered natural variants of one another, such as in the case of
352
8 Lantibiotics
nisin A and nisin Z or epidermin and [VallIleI-epidermin (DE Vos et al., 1993; SAHLet al., 1995). PepI, the immunity protein responsible for protection against the action of Pep5, is a 69 amino acid peptide with a strongly hydrophilic C-terminal region and a strongly hydrophobic N-terminal domain; expression of PepI of which is absolutely essential for the immune phenotype (REIS and SAHL, 1991; REIS et al., 1994). These studies also showed that production of the immune phenotype is dependent on co-expression of both pepA and pepZ and that even partial deletions in pepA, the structural gene, resulted in complete abolition of immunity, although the reasons for this are not apparent. In addition, since PepI appears to be accessible to exogenously added proteases, it is likely to be located outside the membrane, perhaps loosely associated with it (REISet al., 1994). Furthermore, in this study it was observed that the membranes of cells expressing PepI were not depolarized by PepS, suggesting that PepI and Pep5 might interact with PepI preventing formation of transmembrane pores. However, analysis of the possible in vitro interactions between synthetic PepI and Pep5 was unable to demonstrate any direct interaction between these two peptides (SUROVOYand JUNG,unpublished data; SAHLet al., 1995). In contrast to PepI, the immunity proteins NisI and SpaI (Fig. 6) are considerably larger at 245 and 163 amino acids, respectively et (KUIPERSet al., 1993a; ENGELKE al., 1994). In addition, both are typical of bacterial lipoproteins, possessing both an N-terminal signal sequence and a membrane-anchoring Cys residue immediately proceeding the cleavage site. Moreover, both have relatively hydrophilic C-terminal domains, suggesting that this region might be exposed outside the cytoplasmic membrane. At least in the case of NisI, partially immune phenotypes are possible, however, as was observed for PepI, full immunity to nisin was only achieved when both nisA and nisZ were co-expressed (KUIPERS et al., 1992; REISand SAHL, 1991; REISet al., 1994). In addition, expression of NisI in E. coli with disrupted outer membranes protected the cells from the action of exogenous nisin (KUIPERS al., 1993a). et
Potential additional mechanism(s) involved in immunity to lantibiotics have been identified in both nisin-producing and subtilin-producing bacterial strains. Recently, SIEGERS (1995) have identified the genes and ENTIAN nisE, nisF and nisG in Lactococcus lactis 6F3 and shown that disruption of these genes leads to increased sensitivity of the producing strain to exogenous nisin. Analysis of the gene sequences obtained suggest that NisE and NisF together form another ABC-transport system but, this transporter is apparently not involved in the transport of nisin during maturation. In addition, NisENsF shares homology with McbEIMcbF, the transporter reported to be responsible for immunity to microcin B17 in certain strains of E. coli (GARRIDO et al., 1988). NisG on the other hand is an hydrophobic protein with significant similarities to many of the colicin immunity proteins, which are thought to interact directly with the channel-forming colicins (PUGSLEY, 1988; SONGand CRAMER, 1991). Similar results have also been obtained with analysis of subtilin immunity determinants (except that the NisE/NisF homolog in Bacillus subtilis is called spaF/SpaG), prompting speculation that the E/F and G-proteins of these two strains are involved in providing an additional degree of protection to that generated by the immunity proteins NisI and SpaI (KLEIN and ENTIAN, 1994; SIEGERS and ENTIAN, 1995). However, how these proteins might achieve this is anything but clear. Observations of similarities between NisENsF and the microcin B17 immunity proteins McbE/McbF further confuse the issue because this bacteriocinkolicin apparently has an intracellular target, specifically inhibiting DNA gyrase (GARRIDO al., 1988). Howevet er, the primary mode of action of nisin and subtilin involves depolarization of the cytoplasmic membrane through the formation of voltage-dependent pores (see also Sect. 5). Thus, some authors have speculated that the additional transport systems might act either to transport external nisin into the cell for degradation or transport internalized subtilin out of the cell, thus generating the observed additional degree of immunity (KLEIN and ENTIAN, 1994; SIEGERSand ENTIAN, 1995). In any case, it is now apparent that multiple
5 Biological Activities of Lantibiotics
353
systems are involved in self-protection of Lactococcus lactis and Bacillus subtilis against nisin and subtilin, respectively. So-far no reports of such additional immunity mechanism(s) have been given for other type A lantibiotic-producing strains.
1991). Thus, it could be concluded that the primary translation product has a very short half-life, site-specific dehydration is the first step in biosynthesis and that thioether rings are formed in a separate step. In the case of nisin biosynthesis VAN DER MEER et al. (1993) showed that nisin secretion is the penultimate step in processing and that the last step involves cleavage of the leader peptide. 4.8 The Chain of Events Leading However, this sequence of events may not to Lantibiotic Biosynthesis and hold for all lantibiotics since Pep5 (at least) Maturation appears to be processed inside the cell with transport the ultimate step in production of From an historical point of view it is inter- this lantibiotic (MEYERet al., 1995). Several modification reactions are likely to esting to note that even at the same time as the structures of nisin and subtilin were being occur spontaneously, including the formation determined (GROSS and MORRELL, 1971; of the N-terminal2-oxopyruvyl and 2-oxobutGROSSet al., 1973) other researchers were yryl groups of lactocin S and Peps, respectiveproposing that these lantibiotics were synthe- ly (MORTVEDT et al., 1991; SKAUGEN et al., et sized by posttranslational modification. Ini- 1994; KELLNER al., 1989). These reactions would require the removal of the leader peptially, HURST(1966) showed that nisin biosynthesis was prevented by inhibitors of pro- tide and should therefore proceed processing. tein synthesis. On the basis of the incorpora- Similarly, since the N-terminal hydroxypyrution of radiolabeled amino acids INGRAM vyl group at the N-terminus of epilancin K7 (1969, 1970) suggested that didehydroamino probably requires enzymatic reduction, this acids should result from dehydration of Ser reaction must proceed processing and may and Thr and that the thioether rings might ar- constitute one of the last steps in biosynthesis ise from addition of the sulphydryl groups of of this lantibiotic (VAN DE KAMP et al., Cys residues. Later, NISHIOet al. (1983) used 1995a, b; SAHL et al., 1995). Alternatively, anti-subtilin antibodies to isolate precursor since EpiD is able to act on the C-terminal subtilin and then used cell extracts from sub- Cys residue of both pre-epidermin, synthetic tilin-producing B. subtilis to convert this pre- pro-epidermin, and other synthetic peptides cursor into a substance with the same activity not related to epidermin, oxidative decarband electrophoretic mobility as natural subtil- oxylation may well precede thioether ring in. However, the isolation of the structural formation during Cys(Avi) biosynthesis et genes encoding the precursor peptides for (SCHNELL al., 1988 KUPKEet al., 1992, epidermin, nisin, and subtilin provided the fi- 1993,1994, 1995). nal evidence that both didehydroamino acid and thioether rings arose from modifications to genetically encoded Ser, Thr, and Cys residues (SCHNELL al., 1988; BANERJEE et and HANSEN, 1988; BUCHMANN al., 1988; KALet ETTA and ENTIAN, 1989). Subsequently the isolation of pre-peptides of Pep5 and nisin has shed a great deal of light on the sequence and subcellular localiza- 5.1 Type A Lantibiotics tion of modification events required for lantibiotic biosynthesis. Cytoplasmic pre-Pep5 has 5.1.1 Primary Mode of Action been shown to consist of a mixture of dehyAs earlier mentioned, the type A lantibiotdrated forms (6-fold dehydrated, 5-fold dehydrated, etc.) which do not contain L a d ics have been defined as those lantibiotic pepMeLan rings (WEILet al., 1990, SAHLet al., tides whose mode of action is principally di-
354
8 Lantibiotics
rected towards the killing of bacterial cells (JUNG,1991a, b). In general their activity is confined to gram-positive bacteria. However, provided that the outer membrane is first disrupted, several type A lantibiotics have been shown to also act against a number of gramnegative bacteria, including E. coli and Sulmonellu spp. (STEVENS al., 1991). Thus, it et would appear that the lipid-rich outer membrane protects gram-negative bacteria from type A lantibiotic action, probably by preventing access to the inner membrane, the site of their action. In addition, the range of bacteria affected by the different type A lantibiotics varies considerably. Some, such as salivaricin A exhibit a relatively limited antimicrobial spectrum (Ross et al., 1993), while others such as nisin inhibit a broad range of gram-positive bacteria (e.g., many strains of micrococci, streptococci, lactococci, pediococci, staphylococci, lactobacilli, Listeriu spp., and mycobacteria) as well as both the vegetative cells and spores of Buciffusspp. and Cfostridium spp. (HURST, 1981; DELVESBROUGHTON, 1990, MOLITOR and SAHL, 1991; BIERBAUM SAHL, 1993; DE VUYST and and VANDAMME, 1993). Historically, it is interesting that while nisin has been used in biopreservation for more than 35 years, the mechanism by which it and other type A lantibiotics kill bacterial cells has only recently been clarified (HURST, 1982; MOLITORand SAHL,1991; JACKet al., in press; SAHLet al., 1995). In early studies, RAMSEIER (1960) suggested that nisin might be acting as a detergent both because of its highly basic pZ and the observation that nisin treatment of cells induced leakage of UV-absorbing intracellular constituents. Later, GROSSand MORRELL (1971) determined the structure of nisin and observed that the unsaturated amino acids Dha and Dhb found in nisin might be able to interact with the sulphydry1 groups of specific enzymes within a cell, while other studies suggested that nisin could interfere with the biosynthesis of the bacterial cell wall (LINNETTand STROMINGER, 1973; REISINGER al., 1980). Subsequently, it has et been suggested that type A lantibiotics such as nisin kill cells by interfering with energy transduction (SAHL, 1985); this conclusion was based on the observation that nisin treat-
ment of sensitive cells resulted in the cessation of most macromolecular biosyntheses (e.g., protein, DNA, RNA, and polysaccharides) and that nisin activity was dependent on external factors such as pH, temperature and the phase of growth of the target cells (HURST, 1981; SAHL and BRANDIS,1981; SAHL, 1991). Based on these studies, subsequent analysis of the mechanism of action of the type A lantibiotics has revealed much about the way in which they exert their antimicrobial activity. Treatment of susceptible bacterial cells with micromolar concentrations of type A lantibiotics such as nisin, Pep5, subtilin, epidermin, gallidermin, or streptococcin A-FF22 arrests amino acid uptake and induces rapid efflux of preaccumulated amino acids (SAHL and BRANDIS, 1983; RUHRand SAHL,1985; SCHULLER al., 1989; SAHL,1991; BENZet et al., 1991; JACKet al., 1994b) and, at least in some cases, has also been shown to cause the efflux of the potassium analog R b + (SAHL and BRANDIS, 1983; RUHR and SAHL,1985). In addition, following Pep5-treatment of cells, ATP could be found in the external medium (SAHL and BRANDIS, 1983); since there are no known transport systems for ATP, these results, along with the observed efflux of other low molecular-weight intracellular macromolecules, suggest that type A lantibiotics form discrete pores in the cytoplasmic membrane of susceptible bacteria. This mechanism is in contrast to the generalized disruption that might be expected from the action of a surfactant. In addition, both ATP and R b + efflux as well as the arrest of macromolecule biosynthesis is effected by the growth phase of the target cells (SAHLand BRANDIS, 1983; SAHL, 1991; JACKet al., 1994b), suggesting that the pore formation occurs in an energydependent manner. Further evidence for the mode of action of type A lantibiotics come from experiments with cytoplasmic membrane vesicles. Artificially-energized vesicles, which have been treated with type A lantibiotics, rapidly efflux preaccumulated radiolabeled amino acids; by contrast, pre-treatment of the vesicles (prior to energization) induced little or no efflux until after the cells were sufficiently energized (SAHL,1985; RUHRand SAHL,1985; SAHL et
355
al., 1987; KORDEL et al., 1988; SCHULLER et al., 1989; JACK et al., 1994b). These results further demonstrate that type A lantibiotics form transmembrane pores in an energy-dependent fashion and allow efflux of preaccumulated intracellular components. However, these same studies also showed that lantibiotic treatment of either whole cells, artificial vesicles which had been energized with valinomycin-induced potassium gradients or artificially energized liposomes (GAO et al., 1991; ABEEet al., 1991) resulted in dissipation of the membrane potential, suggesting that the pores formed are non-specific and allow influx of extracellularly accumulated protons and (probably) other ions and small molecules. The dissipation of the membrane potential accounts for the observed arrest of energy-dependent macromolecular biosynthesis and is different to that of the protonophores since pore formation is energy-dependent, potentiated by the membrane potential (SAHL, 1991; GARCIA-GARCERA al., et 1993). Further support for the prediction that type A lantibiotics kill cells by disruption of energy transduction through the formation of energy-dependent pores in the cytoplasmic membrane have been provided by analysis of the pores formed in artificial bilayers such as the black-lipid membranes (BLM); in addition these studies have allowed the measurement of several of the physical properties of the formed pores (BENZ et al., 1978, 1991). Artificial BLMs can be formed across a small whole separating two chambers or wells in a teflon block, both of which are filled with a conducting salt buffer solution. Using electrodes, it is then possible to apply a defined potential difference across an artificial bilayer and measure lantibiotic-induced current flows through the normally insulative model membrane. In such a system, the type A lantibiotics can be shown to induce discrete pores since, application of a sufficiently high potential induces current flow; reduction of the potential allows the pores to close and restores the insulative properties of the bilayer, indicating that membrane disruption is not generalized (SAHL et al., 1987; KORDELet al., 1988; SCHULLER al., 1989; BENZ et al., et 1991; JACK et al., 1994b). In addition, these
studies have also shown that nisin and Pep5 form pores only in the presence of a truns-negative membrane potential, while subtilin, epidermin, gallidermin, and streptococcin A-FF22 form pores irrespective of the orientation of the applied potential. Analysis of the current voltage curves obtained from these studies also allowed determination of the threshold potential for lantibiotic-induced pore formation; epidermin and gallidermin required ca. 50 mV, Pep5, nisin and subtilin require ca. 80 mV, while streptococcin A-FF22 required ca. 100 mV. Further physical properties, including the mean diameter and the lifetime of type A lantibiotic channels can be determined from the analysis of single channels formed in BLMs. If the pore is assumed to be a cylinder with a length equivalent to the thickness of the membrane which is filled with the same solution as bathes the membrane (of known conductance), then from the observed conductance it is possible to calculate the diameter of the pore. Streptococcin A-FF22 pores appear to be relatively unstable, appearing as millisecond time scale bursts and a with a mean diameter of about 0.5-0.6 nm (JACKet al., 1994b). Alternatively, nisin and Pep5 pores are somewhat more stable (tens to hundreds of milliseconds) and have a diameter of ca. 1 nm (SAHLet al., 1987; KORDEL et al., 1988), while subtilin pores are larger et again at ca. 2 nm diameter (SCHULLER al., 1989). Epidermin and gallidermin pores are somewhat different; their mean lifetimes may extend up to 30 s and their diameter appears to increase with the applied potential (BENZ et al., 1991). The determination of many of the physical properties of the type A lantibiotic peptides has allowed some understanding of how they might form pores in the bacterial cytoplasmic membrane. In general, type A lantibiotics have both sufficient length and a sufficiently high dipole moment to be consistent with voltage-dependent channel formation in phospholipid bilayers (FREUND et al., 1991a b; BENZet al., 1991) as well as the central flexible region thought to be essential for stabilization of the transmembrane structure (VOGEL al., 1993). In addition, NMR studet ies have shown that the peptides themselves
356
8 Lantibiotics
appear to form amphiphilic helical conformations in solution and present their hydrophobic residues on one "face" of the helix and the charged, hydrophilic residues on the opposite "face" (FREUNDet al., 1991a, b, c; GOODMAN al., 1991; LIAN et al., 1991; et PALMER al., 1989; SLIJPER al., 1989; VAN et et DE VEN et al., 1991a, b). Clearly, a single lantibiotic molecule is insufficient to induce pore formation and multiple peptides must somehow coalesce in the bilayer to form an aggregate with the properties observed above. In addition, since type A lantibiotics can form pores in artificial vesicles and bilayers, it is clear they have no requirements for specific membrane-associated "receptors" as has been suggested for several other channel-forming antibacterial peptides including lactococcin A and pediocin PA-1 (VAN BELKUMet al., et 1991; CHIKINDAS al., 1993). Although the exact mechanism remains unclear, current models (Fig. 13) for type A lantibiotic-induced pore formation suggest that the peptides accumulate at the cytoplasmic membrane, perhaps attracted to the bilayer by ionic interactions (SAHL, 1991; BENZ et al., 1991; JACK et al., 1994b). In the absence of a membrane potential they should remain oriented lateral to the membrane but apparently with their hydrophobic "face" intimateet ly associated with the bilayer (SCHULLER al., 1989; SAHL,1991; BENZet al., 1991; JACK et al., 1994b; DRIESSENal., 1995); on appliet
cation of a sufficiently high membrane potential the peptides apparently adopt a transmembrane orientation and create a pore. However, a number of questions remain unanswered; e.g., it remains to be seen whether the peptides aggregate before or after adoption of the transmembrane orientation, how many pores are the minimum required for transmembrane conductance and which terminus of the peptide inserts through the membrane. In addition, the observations that pore diameters fluctuate and the short pore lifetimes observed with most type A lantibiotics suggest that the number of peptide involved in pore formation is not static, but rather more dynamic.
5.1.2 Secondary Mode of Action

In addition to forming pores in phospholipid bilayers, both Pep5 and nisin have been shown to induce autolysis in Staphylococcus simuluns cells (BIERBAUM SAHL,1991). and Characterization of the mechanism by which they achieve this suggests that cationic lantibiotics such as nisin and Pep5 are able to competitively release cell wall autolytic enzymes normally bound to, and regulated by, polyanionic cell wall constituents such as lipoteichoic-, teichoic-, and teichuronic acids, probably by an ion exchange-like mechanism (SAHL, 1985; BIERBAUM and SAHL, 1987,
Fig. 1 . Model for the mode of 3 action of the type A lantibiotics, involving the formation of voltage-dependent pores in the cytoplasmic membrane of sensitive bacteria.
357
1988, 1991). In addition, it has been shown diates also failed to be incorporated into that activation of the autolytic enzymes oc- cross-linked peptidoglycan in the actagardinecurred most markedly in the area of the septa treated bacilli and isolation of a labeled interbetween dividing daughter cells (BIERBAUM mediate (UDP-acetylmuramyl-N-acetylgluand SAHL,1985,1991) and that the activation cosamine pentapeptide linked to a C55-isowas not specific since similar results could be prenylphosphate carrier) suggested that actaobtained using synthetic cationic peptides, gardine treatment was able to inhibit the such as poly-lysine or poly-arginine (BIER- transfer of cell wall precursor to the peptidoBAUM and SAHL,1991). Since only small so- glycan receptor during cell wall biosynthesis. lutes may pass outward (Fig. 13) pores When staphylococci are incubated in the formed by these lantibiotics in the cytoplas- presence of mersacidin their growth is first inmic membrane should allow an influx of wa- hibited, following which the cells begin to ter into the cell, increasing osmotic pressure. lyse; cessation of growth is accompanied by This increase in intracellular pressure com- an inability to incorporate D-alanine into the bined with the interference in energy trans- cell wall and by a marked decrease in the duction (resulting from depolarization of the thickness of the cell wall (BROTZet al., 1995). membrane) which should prevent repair of In addition, these authors observed that merthe weakened cell wall, probably results in sacidin treatment had no effect on other mathe observed lysis (BIERBAUM and SAHL, cromolecular biosynthetic processes such as DNA, RNA, and protein biosynthesis, in di1991). rect contrast to the effects observed with similar cells treated with type A lantibiotics. Furthermore, whereas both mersacidin and the 5.2 Type B Lantibiotics glycopeptide antibiotic vancomycin have approximately the same MIC and both act on 5.2.1 Mersacidin and Actagardine cell wall biosynthesis, mersacidin was not inhibited by the tripeptide L-Lys-D-Ala-D-Ala The type B lantibiotics mersacidin and ac- which is a potent inhibitor of vancomycin actagardine also kill bacterial cells rather effi- tion. This last result suggests that, while the ciently, however, their primary mechanism of molecular target of mersacidin is unclear, it activity appears to be vastly different to that differs from that of vancomycin. described above for the type A lantibiotics, directed primarily at the cell wall rather than the cytoplasmic membrane. Mersacidin acts 5.2.2 Cinnamycin and the principally against streptococci and staphylococci; this particular type B lantibiotic has de- Duramycins monstrated in vivo activity against methicilAmong other type B lantibiotics, at least cilin-resistant Staphylococcus aureus strains and could therefore offer an alternative to namycin and duramycin have been shown to vancomycin treatment of such infections inhibit the growth of Bacillus spp.; treated (CHATERJEE al., 1992b). Alternatively, ac- cells show increased membrane permeability, et tagardine is primarily effective against obli- undergo a reduction in ATP-dependent progate anaerobes and streptococci and has been tein translocation and ATP-dependent calused experimentally to treat Streptococcus cium uptake, show marked slowing in the rate pyogenes infections (ARIOLI et al., 1976; of chloride uptake and show significantly reMALABARBA al., 1990). et duced potassium and sodium ATPase activiet In the late 1970s, SOMMAet al. (1977) ties (RACKER al., 1983; STONEet al., 1984; et showed that low concentrations of actagar- NAVARRO al., 1985; CHENand TAI, 1987). dine inhibited the incorporation of N-acetyl- Duramycin-resistant bacilli remain unaffected glucosamine and L-alanine into Bacillus sub- and have been shown to possess cytoplasmic tilis cell wall peptidoglycan. In these experi- membranes with altered phospholipid compoments it was observed that other interme- sitions; whereas sensitive cells contain phos-
358
8 Lantibiotics
phatidyl ethanolamine, resistant cells do not (NAVARROet al., 1985; DUNKLEY al., et 1988). These results suggest that duramycin recognizes and interacts with specific phospholipids in the cytoplasmic membrane of sensitive cells and that alterations to this composition, while conferring duramycin resistance, do not seem to effect other membraneassociated cellular activities (DUNKLEY al., et 1988; CLEJAN al., 1989). et In addition to inhibiting the growth of bacilli, duramycin also inhibits a number of metabolic properties of isolated mitochondria and specifically increases the permeability of the inner membrane, the major component of which is phosphatidyl ethanolamine (SOKOLOVE et al., 1989). Similarly, cinnamycin has been shown to induce hemolysis in isolated erythrocytes and phosphatidyl ethanolaminecontaining liposomes; cinnamycin had no effect on erythrocytes if it was first incubated with phosphatidyl ethanolamine or on liposomes which were prepared from phosphatidyl inositol, phosphatidyl serine, or cardiolipin (CHOUNG al., 1988a, b). The latter reet sults suggest that cinnamycin interacts with phospholipids containing both a glyceryl backbone and a primary amino group. Using electron-spin resonance, these studies also showed that the interaction of cinnamycin and phosphatidyl ethanolamine results in reorganization of the membrane bilayer. Taken together, these results have further confirmed the direct interaction of type B lantibiotics with specific phospholipids present in the membrane bilayer. Recently, SHETH et al. (1992) have suggested that the increases in membrane permeability associated with duramycin treatment could result from the formation of ion channels. They observed that both the membranes of cultured epithelial cells and artificial BLMs treated with duramycin showed complex conductance states, some of which were discrete, and that pores were weakly anion selective. In addition, this study showed that pore diameter increased with time, perhaps explaining duramycin-induced increases in membrane permeability. Duramycin, duramycin B, duramycin C, and cinnamycin may also act as immunoregulators as they are able to influence the activity
of phospholipase A2, an important enzyme of the immune system involved in the production of prostaglandins and leucotrienes (MARKIand FRANSON, 1986; FREDENHAGEN et al., 1990, 1991; MARKI et al., 1991). The principal substrate of phospholipase A2 is phosphatidyl ethanolamine; interaction of these type B lantibiotics with the substrate renders it unavailable for lipolysis, thus interfering with immune function.
6 Applications of Lantibiotics
6.1 Applications as a Food/Beverage Preservative
Since the type A lantibiotic nisin has been used as a biopreservative for at least 35 years, there is currently a great deal of information regarding its application to foods and beverages for which readers are referred to the excellent reviews of HURST(1981), DELVESBROUGHTON (1990), MOLITOR and SAHL (1991), and DE VUYST and VANDAMME (1993). However, it is worth briefly mentioning that already this peptide has been utilized in a number of areas including: fish and lowtemperature processed canning, brewing, winemaking, cheese and other fermented dairy product manufacture as well as in the preservation of cosmetics and deodorants. Nisin has proved particularly useful in a number of these areas because of its characteristics. Nisin, e.g., demonstrates good activity against a number of food-borne pathogens and spoilage bacteria including Clostridium botulinum, Listeria monocytogenes, Lactobacillus spp. and Leuconostoc spp. In addition, the peptide is particularly heat and acid stable making it particularly suitable for addition to products with low pH (e.g., fermented products) and products which subsequently undergo some sort of pasteurization.
7 Conclusions and Future Perspectives
359
6.2 MedicaVParamedical and Veterinary Applications

Interestingly, some of the lantibiotics may also prove to be useful in medicinal preparations for both human and animal application. The type A lantibiotics epidermin and gallidermin have previously been shown to be active against Propionibacterium acnes, the causative agent of the acnes disease. These observations have led to speculation that these lantibiotics might be used in the external treatment of acnes as a replacement therapy for the currently used erythromycinhitamin A creams. Such application of a lantibiotic should have several advantages: (1) so-far at least, acquired resistance to these lantibiotics has not been observed, (2) because of their size and nature, the peptides should not be absorbed into the body and they should be of low toxicity, and (3) it should be possible to develop large-scale, low cost production schemes for these bacterially-derived pepet tides (ALLGAIERal., 1991; JUNG, 1991a, b; UNGERMANN 1991). Similarly, the notet al., able acid stability of nisin (HURST, 1981) has led several recent studies to suggest that it may have efficacy in the treatment of gastric ulcers, due to its antimicrobial activity against Helicobacter pylori (DE VUYST and VANDAMME, 1993). In addition, alternative suggested applications of nisin include: as a mouthrinse for the prevention of gingivitis and plaque or as a germicidal preparation for the prevention of mastitis in dairy cattle (SEARSet al., 1992; HOWELL al., 1993). et More significantly perhaps, the cinnamycingroup type B lantibiotics have been shown to effect human immune function, leucocyte proliferation and may play a role in protection against Herpes simplex virus (MARKI and FRANSON, 1986; FREDENHAGEN al., 1990, et 1991; MARKI et al., 1991); thus the potentials for application of these compounds in human health care are obvious. In addition, the observation that mersacidin displays antibacterial activity against methicillin-resistant Staphylococcus aureus strains as well as multidrug resistant enterococci by novel mechanisms offers the greatest hope for the development of new therapeutic agents to combat
the threat of these potentially fatal microorganisms (BROTZ et al., 1995).
7 Conclusions and Future Perspectives

From the biotechnological standpoint then, the lantibiotics represent an exciting class of peptides for two reasons. Firstly, as antimicrobial compounds they may offer not only new opportunities in areas such as food and beverage preservation (HURST, 1981; DELVES-BROUGHTON, 1990; MOLITORand SAHL, 1991; DE VUYST and VANDAMME, 1993), but may also represent (at least in the case of mersacidin and gallidermin) novel therapeutic agents with potential for medicinal use (BROTZ et al., 1995; SAHLet al., 1995). Secondly, it is clear that the lantibiotics are produced by novel biosynthetic mechanisms; the enzymes which are capable of transforming the 20 protein amino acids into such highly modified forms may well prove useful for the generation of new peptide compounds and mimetics, previously only produced through time-consuming and costly in-lab chemical syntheses (SAHLet al., 1995; JACK et al., in press; JACK and SAHL, 1995). Thus, greater understanding of the biosynthetic mechanisms involved in lantibiotic production may one day allow us to introduce non-peptide amino acids and enantiomeric amino acids into useful proteins, perhaps improving their stability, resistance to proteolytic activity or even their biological and catalytic activity. The continued study of novel peptides and biosynthetic enzymes, such as those involved in lantibiotic production, give the highest prospects for the development of new, useful, peptide-based compounds through biotechnology.
360
8 Lantibiotics
Escherichia coli, Biochim. Biophys. Acta 511, 305-319. BENZ,R., JUNG,G., SAHL,H.-G. (1991), Mechanism of channel formation by lantibiotics in black lipid membranes, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 359372. Leiden: ESCOM Scientific Publishers BV. BERRIDGE, J., NEWTON, G., ABRAHAM, N. G. E. P. (1952), Purification and nature of the antibiotic nisin, Biochem. J. 52, 529-535. BIERBAUM, SAHL,H.-G. (1985), Induction of G., autolysis of staphylococci by the basic peptide antibiotics Pep5 and nisin and their influence on the activity of autolytic enzymes, Arch. Microbiol. 141, 249-254. BIERBAUM, SAHL,H.-G. (1987), Autolytic sysG., tem of Staphylococcus simulans 22: influence of cationic peptides on activity of N-acetylmuramoyl-L-alanine amidase, J. Bacteriol. 169, 54525458. BIERBAUM, SAHL,H.-G. (1988), Influence of G., cationic peptides on the activity of the autolytic endo-PN-acetylglucosaminidaseof Staphylococcus simulans 22, FEMS Microbiol. Rev. 58, 223228. BIERBAUM, SAHL,H.-G. (1991), Induction of G., autolysis of Staphylococcus simulans 22 by Pep5 and nisin and influence of the cationic peptides on the activity of the autolytic enzymes, in: Nisin G., and Novel Lantibiotics (JUNG, SAHL,H.-G., Eds.), pp. 386-396. Leiden: ESCOM Scientific Publishers BV. BIERBAUM, SAHL,H.-G. (1993), Lantibiotics G., unusually modified bacteriocin-like peptides from Gram-positive bacteria, Zentralbl. Bakteriol. 278, 1-22. BIERBAUM, REIS,M., SZEKAT, SAHL, G., C., H.-G. (1994), Construction of an expression system for engineering of the lantibiotic Peps, Appl. Environ. Microbiol. 60, 4332-4338. BIERBAUM, BROTZ,H., KOLLER, G., K.-P., SAHL, H.-G. (1995), Cloning, sequencing and production of the lantibiotic mersacidin, FEMS Micro121-126. biol. Lett. U7, BROCK, D., DAVIE, M. (1963), Probable idenT. J. tity of a group D hemolysin with a bacteriocine, J. Bacteriol. 86, 708-712. BROTZ, H., BIERBAUM, MARKUS, MOLIG., A., TOR,E., SAHL,H.-G. (1995), Mode of action of mersacidin - inhibition of peptidoglycan synthesis via a novel mechanism? Antimicrob. Agents Chemother. 39,714-719. BUCHMAN, B., BANERJEE, HANSEN, N. W. S., J. (1988), Structure, expression and evolution of a gene encoding the precursor of nisin, a small protein antibiotic, J. Biol. Chem. 263, 1626016266.
8 References
ABEE,T., GAO,F. H., KONINGS, N. (1991), The W. mechanism of the lantibiotic nisin in artificial membranes, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL, H.-G., Eds.), pp. 373-385. Leiden: ESCOM Scientific Publishers BV. ALLGAIER,H., JUNG, G., WERNER, R. G., SCHNEIDER, ZAHNER,H. (1986), EpiderU., min: sequencing of a heterodet tetracyclic 21peptide amide antibiotic, Eur. J. Biochem. 160, 9-22. ALLGAIER,H., WALTER, J., SCHLUTER,M., WERNER, G. (1991), Strategy for the purificaR. tion of lantibiotics, in: Nisin and Novel Lantibiotics (JUNG, SAHL,H.-G., Eds.), pp. 422G., 433. Leiden: ESCOM Scientific Publishers BV. ARIOLI, BERTI,M., SILVESTRI, G. (1976), V., L. Gardimycin, a new antibiotic from Actinoplanes. 111. Biological properties, J. Antibiot. 29, 511515. AUGUSTIN, (1991), Identifizierung und RegulaJ. tion der Biosynthesegene des Lantibiotikums Epidermin aus Staphylococcus aureus Tii 3298. PhD Thesis, University of Tiibingen, Germany. R., KUPKE, T., AUGUSTIN,J., ROSENSTEIN, N., G., SCHNEIDER, SCHNELL, ENGELKE, U., ENTIAN, K.-D., GOTZ, F. (1991), Identification of epidermin biosynthetic genes by complementation studies and heterologous expression, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL, H.-G., Eds.), pp. 277-286. Leiden: ESCOM Scientific Publishers BV. AUGUSTIN, ROSENSTEIN, WIELAND,B., J., R., SCHNEIDER, SCHNELL, ENGELKE, U., N., G., ENTAIN, K.-D., GOTZ, F. (1992), Genetic analysis of epidermin biosynthetic genes and epidermin-negative mutants of Staphylococcus epidermidis, Eur. J. Biochem. 204,1149-1154. BANERJEE, HANSEN, N. (1988), Structure and S., J. f expression o a gene encoding the precursor of subtilin, a small peptide antibiotic, J. Biol. Chem. 263,9508-9514. BECK-SICKINGER, G., JUNG, (1991), SyntheA. G. sis and conformational analysis of lantibiotic leader-, pro- and pre-peptides, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 218-230. Leiden: ESCOM Scientific Publishers BV. BENEDICT, G., DVONCH, SHOTWELL, L., R. W., 0. PRIDHAM, G., LINDENFELSER, A. (1952), T. L. Cinnamycin, an antibiotic from Streptomyces cinnamonensis nov. sp, Antibiot. Chemother. 2, 591-594. BENZ, R., JANKO,K., Boos, W., LAUGER,P. (1978), Formation of large ion-permeable membrane Channels by the matrix protein (porin) of
8 References
361
B. W. G. BYCROFT, W., CHAN, C., ROBERTS, C. K. (1991), Synthesis of pro- and prepeptides related to nisin and subtilin, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 204217. Leiden: ESCOM Scientific Publishers BV. CHAN,W. C., LIAN,L.-Y., BYCROFT,B. W., RoBERTS, G. c. K. (1989a), Isolation and characterization of two degradation products derived from the peptide antibiotic nisin, FEBS Lett. 252,29-36. CHAN,W. C., LIAN,L.-Y., BYCROFT, W., RoB. BERTS,G. c. K. (1989b), Confirmation of the complete structure of nisin by complete 'HNMR resonance assignments in aqueous and dimethylsulphoxide solution, J. Chem. SOC. Perkin Trans. 1,2359-2367. CHAN, C., BYCROFT, W., LEYLAND, L., W. B. M. LIAN,L.-Y., YANG,J. C., ROBERTS, C. K. G. (1992), Sequence-specific resonance assignment and conformational analysis of subtilin by 2DNMR, FEBS Lett. 300,56-62. B. M. CHAN,W. C., BYCROFT, W., LEYLAND, L., LIAN,L.-Y., YANG,J. C., ROBERTS, C. K. G. (1993), A novel posttranslational modification of the peptide antibiotic subtilin: isolation and characterization of a natural variant from Bacillus subtilis ATCC 6633, Biochem. J. 291,23-27. CHATERJEE, CHATERJEE, LAD, S. J., PHANS., S., SALKAR, M. S., RUPP, R. H., GANGULI, N. B. H.-W., KOGLER, (1992a), MerH. FEHLHABER, sacidin, a new antibiotic from Bacillus: fermentation, isolation, purification and chemical characterization, J. Antibiot. 45,832-838. CHATERJEE, CHATERJEE, K., JANI,R. H., S., D. BLUMBACH, GANGULI, N. KLESEL,N., J., B. LIMBERT, SEIBERT, (1992b), Mersacidin, M., G. a new antibiotic from Bacillus: in vitro and in vivo antibacterial activity, J. Antibiot. 45, 839845. CHEN, L., TAI,P. C. (1987), Effects of antibiotics L. and other inhibitors on ATP-dependent protein translocation into membrane vesicles, J. Bacteri01. 169,2372-2379. CHIKINDAS, L., GARCIA-GARCERA, J., M. M. DRIESSEN, J. M., LEDERBOER, M., NIA. A. SEEN-MEYER, NES, I. F., ABEE,T., KONING, J., W. N., VENEMA, (1993), Pediocin PA-1, a G. bacteriocin from Pediococcus acidilactici PAC1.O forms hydrophilic pores in the cytoplasmic membrane of target cells, Appl. Environ. Microbiol. 59, 3577-3584. CHOUNG, S.-Y., KOBAYASHI, INOUE, TAKET., J., MOTO,K., ISHITSUKA, INOUE,K. (1988a), H., Haemolytic activity of a cyclic peptide Ro 090198 isolated from Streptoverticillium, Biochim. Biophys. Acta 940, 171-179. CHOUNG, S.-Y., KOBAYASHI, TAKEMOTO, T., K.,
ISHITSUKA, INOUE,K. (1988b), Interaction H., of a cyclic peptide, Ro 09-0198, with phosphatidylethanolamine in liposomal membranes, Biochim. Biophys. Acta 940, 180-187. CHUNG, Y.-J., HANSEN, N. (1992), DeterminaJ. tion of the sequence of spaE and identification of a promoter in the subtilin (spa) operon in Bacillus subtilis, J. Bacteriol. 174, 66994702. CHUNG, Y.-J., STEEN, T., HANSEN, N. (1992), M. J. The subtilin gene of Bacillus subtilis ATCC 6633 is encoded in an operon that contains a homologue of the hemolysin B transport protein, J. Bacteriol. 174, 1417-1422. A. CLEJAN,S., GUFFANTI, A., COHEN, M. A., KRULWICH, A. (1989), Mutations of Bacillus T. jirmus OF4 to duramycin resistance results in substantial replacement of membrane lipid phosphatidylethanolamine by its plasmalogen form, J. Bacteriol. 171, 1744-1746. DELVES-BROUGHTON, J. (1990), Nisin and its uses as a food preservative, Food Technol. 44, 100112. DEMPSTER, P., TAGG, R. (1982), The producR. J. tion of bacteriocin-like substances by the oral bacterium Streptococcus salivarius, Arch. Oral Biol. 27, 151-157. DE Vos, W. M., JUNG, G., SAHL, H.-G. (1991), Appendix: definitions and nomenclature of lantibiotics, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 457463. Leiden: ESCOM Scientific Publishers BV. DE VOS, W. M., MULDERS, W. M., SIEZEN, J., J. R. 0. HUGENHOLTZ, KUIPERS, P. (1993), PropJ., erties of nisin Z and distribution of its gene nisZ in Lactococcus lactis, Appl. Environ. Microbiol. 59,213-218. DE VUYST, VANDAMME, J. (1991), Microbial L., E. manipulation of nisin biosynthesis and fermentation, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL, H.-G., Eds.), pp. 397-409. Leiden: ESCOM Scientific Publishers BV. DE VUYST, VANDAMME, J. (1992), Influence L., E. of the carbon source on nisin production in Lactococcus Iactis subspp. lactis batch fermentations, J. Gen. Microbiol. 138, 571-578. DE VUYST, L., VANDAMME, J. (1993), Nisin, a E. lantibiotic produced by Lactococcus Iactis subspp. lactis: properties, biosynthesis, fermentation and applications, in: Bacteriocins of Lactic Acid Bacteria: Microbiology, Genetics and ApplicaL., E., tions (DEVUYST, VANDAMME, Eds.), pp. 151-221. London: Chapman and Hall. DODD, H. M, GASSON, J. (1994), Bacteriocins M. of lactic acid bacteria, in: Genetics and BiotechM. nology of Lactic Acid Bacteria (GASSON, J., DE Vos, W. M., Eds.), pp. 211-251. London: Chapman and Hall.
362
8 Lantibiotics
structural revision of duramycin and cinnamycin, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 131-140. Leiden: ESCOM Scientific Publishers BV. S., G. FREUND, JUNG, (1992), Lantibiotics: an overview and conformational studies on gallidermin and Pep5, in: Bacteriocins, Microcins and LantiR., C., F., biotics (JAMES, LAZDUNSKI, PAT~US, Eds.), pp. 75-92. Berlin: Springer-Verlag. FREUND, JUNG,G., GUTBROD, FOLKERS, S., O., G., GIBBONS, A., ALLGAIER, WERNER, W. H., R. (1991a), The solution structure of the lantibiotic gallidermin, Biopolymers 31, 803-811. S., O., FREUND, JUNG,G., GUTBROD, FOLKERS, G., GIBBONS, A. (1991b), The three-dimenW. sional solution structure of gallidermin determined by NMR-based molecular graphics, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL, H.-G., Eds.), pp. 91-102. Leiden: ESCOM Scientific Publishers BV. FREUND, JUNG, GIBBONS, A., SAHL, S., G., W. H.G. (1991c), NMR and circular dichroism studies on Pep5, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL,H.-G., Eds.), pp. 103-112. Leiden: ESCOM Scientific Publishers BV. GAO,T., ABEE,T., KONINGS, N. (1991), MechW. anism of the peptide antibiotic nisin in liposomes and cytochrome c oxidase-containing proteoliposomes, Appl. Environ. Microbiol. 57, 2164-2170. GARCIA-GARCERA, G. J., ELFERINK, G. L., M. M. DRIESSEN, J. M., KONINGS, N. (1993), In A. W. vitro pore-forming activity of the lantibiotic nisin: role of proton motive force and lipid composition, Eur. J. Biochem. 212,417422. GARRIDO,M. C., HERRERO, M., KOLTER,R., MORRENO, (1988), The export of the DNA P. replication inhibitor microcin B17 provides immunity for the host cell, EMBO J. 7, 18531862. GASSON, J. (1984), Transfer of sucrose fermentM. ing ability, nisin resistance and nisin production into Lactococcus Iactis 712, FEMS Microbiol. Lett. 21, 7-10. GILMORE, S., SEGARRA, A., BOOTH, C., M. R. M. D. BOGIE,C. P., HALL, L. R.,CLEWELL, B. (1994), Genetic structure of the Enterococcus faecalis plasmid pAD1-encoded cytolytic toxin system and its relationship to lantibiotic determinants, J. Bacteriol. 176, 7335-7344. GIREESH,T., DAVIDSON, E., HILLIER, J. B. A. (1992), Conjugal transfer in Lactococcus lactis of a 68-kilobase-pair chromosomal fragment containing the structural gene for the peptide bacteriocin nisin, Appl. Environ. Microbiol. 58, 16701676. GOODMAN, PALMER, E., MIERKE, Ro, M., D. D., S., NUNAMI, WAKAMIYA, FUKASE, K., T., K.,
DODD,H. M., HORN,N., GASSON, J. (1990), M. Analysis of the genetic determinant for production of the peptide antibiotic nisin, J. Gen. Microbiol. 136,555-566. DODD,H. M., HORN,N., SWINDELL, GASSON, S., M. J. (1991), Physical and genetic analysis of the chromosomally located transposon Tn5301, responsible for nisin biosynthesis, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 231-242. Leiden: ESCOM Scientific Publishers BV. DRIESSEN, J. M., VAN DEN HOOVEN, w., A. H. KUIPER, VAN DE KAMP,M. M., SAHL, H.W., W. G., KONINGS, N. H, KONINGS, N. (1995), R. Mechanistic studies of lantibiotic-induced permeabilization of phospholipid vesicles, Biochemistry 34, 1606-1614. DUNKLEY, A., JR., CLEJAN, GUFFANTI, E. S., A. A., KRULWICH, A. (1988), Large decreases in T. membrane phosphatidylethanolamine and diphosphatidylglycerol upon mutation to duramycin resistance do not change the protonophore resistance of Bacillus subtilis, Biochim. Biophys. Acta 943,13-18. ENGELKE, GUTOWSKI-ECKEL, HAMMELG., Z., MA", M., ENTIAN, K.-D. (1992), Biosynthesis of the lantibiotic nisin: genomic organization and membrane localization of the NisB protein, Appl. Environ. Microbiol. 58, 3730-3743. ENGELKE, GUTOWSKI-ECKEL, KIESAU, G., Z., P., SIEGERS, HAMMELMANN, ENTIAN,K.K., M., D. (1994), Regulation of nisin biosynthesis and immuntiy in Lactococcus lactis 6F3, Appl. Environ. Microbiol. 60, 814-825. ENTIAN, K.-D., KALEITA,C. (1991), Isolation and characterization of the cinnamycin structural G., gene, in: Nisin and Novel Lantibiotics (JUNG, SAHL, H.-G., Eds.), pp. 303-308. Leiden: ESCOM Scientific Publishers BV. ERSFELD-DREBEN, SAHL,H.-G., BRANDIS, H., H. (1984), Plasmid involvement in production of and immunity to the staphylococcin-likepeptide Pep5, J. Gen. Microbiol. 130,3029-3035. FATH,M. J., KOLTER, (1993), ABC transportR. ers: bacterial exporters, Microbiol. Rev. 57, 9951017. FREDENHAGEN, FENDRICH, MARKI,F., A., G., MARKI,W., GRUNER, RASCHDORF, PEJ., F., TER, H. H. (1990), Duramycins B and C, two new lanthionine-containing antibiotics as inhibitors of phospholipase A2, J. Antibiot. 43, 14031412. FREDENHAGEN, MARKI, F., FENDRICH, A., G., MARKI,W., GRUNER, VAN OOSTRUM, J., J., RASCHDORF, PETER,H. H. (1991), DuramyF., cin B and C, two new lanthionine-containing antibiotics as inhibitors of phospholipase A2 and
8 References
363
W. J. HORIMOTO, KITAZAWA, S., M., FUJITA, H., HYNES, L., FERREITI,J. J., TAGG, R. (1993), Cloning of the gene encoding streptococcin AKUBO, A., SHIBA, (1991), Conformations of T. FF22, a novel lantibiotic produced by Streptonisin and its fragments using synthesis, NMR coccus pyogenes and determination of its nucleoand computer simulations, in: Nisin and Novel tide sequence, Appl. Environ. Microbiol. 59, Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 1969-1 971. 59-75. Leiden: ESCOM Scientific Publishers INGRAM, (1969), Synthesis of the antibiotic niL. BV. sin: formation of lanthionine and p-methyllanGROSS, E. (1977), a,PUnsaturated and related thionine, Biochim. Biophys. Acta 184, 216-219. amino acids in peptides and proteins, Adv. Exp. INGRAM, (1970), A ribosomal mechanism for L. Med. Biol. 86, 131-153. synthesis of peptides related to nisin, Biochim. GROSS, MORELL, L. (1971), The structure of E., J. Biophys. Acta 224, 263-265. nisin, J. Am. Chem. Soc. 93, 4634-4635. R. H.-G. (1995), Unique postransGROSS, KILTZ,H. H., NEBELIN, (1973), Sub- JACK, W., SAHL, E., E. lational modifications involved in lantibiotic biotilin. VI. Die Struktur des Subtilins. Hoppe-Seysynthesis, Trends Biotechnol. 13,269-278. lers 2. Physiol. Chem. 354, 810-812. R. GUTOWSKI-ECKEL, KLEIN,C., SIEGERS,K., JACK, W., TAGG,J. R. (1991), Isolation and parG., tial structure of streptococcin A-FF22, in: Nisin BOHM,K., HAMMELMANN, ENTIAN, M., K.-D. G., and Novel Lantibiotics (JUNG, SAHL,H.-G., (1994), Growth phase-dependent regulation and Eds.), pp. 171-179. Leiden: ESCOM Scientific membrane localization of SpaB, a protein inPublishers BV. volved in biosynthesis of the lantibiotic subtilin, JACK, W., TAGG, R. (1992), Factors affecting R. J. Appl. Environ. Microbiol. 60, 1-11. the production of the group A Streptococcus HANSEN, N. (1993), Antibiotics synthesized by J. bacteriocin SA-FF22, J. Med. Microbiol. 36,132posttranslational modification, Ann. Rev. Micro138. biol. 47, 535-564. A., J., Y. M. JACK,R. W., CARNE, METZGER, STEVAHANSEN, N., CHUNG, J., LIU,W., STEEN, J. N O V I ~ ,S., SAHL,H.-G., JUNG, G., TAGG, R. J. T. (1991), Biosynthesis and mechanism of action (1994a), Elucidation of the structure of SAof nisin and subtilin, in: Nisin and Novel LantiFF22, a lanthionine-containingantibacterial pepbiotics (JUNG, SAHL, G., H.-G., Eds.), pp. 287tide produced by Streptococcus pyogenes strain 302. Leiden: ESCOM Scientific Publishers BV. FF22, Eur. J. Biochem. 220,455-462. HAYASHI, NAGASHIMA,K., TERUI, KAWAF., Y., MURA, Y., MATSUMOTO, K., ITAZAKI, H. JACK,R. W., BENZ,R., TAGG,J. R., SAHL,H.-G. (1994b), The mode of action of SA-FF22 a lanti(1990), The structure of PA48009: the revised biotic isolated from Streptococcus pyogenes structure of duramycin, J. Antibiot. 43, 1421strain FF22, Eur. J. Biochem. 219, 699-705. 1430. R. J. HIGGINS, F. (1992), ABC-transporters: from mi- JACK, W., TAGG, R., RAY,B. (1995), BacteriC. ocins of Gram-positive bacteria, Microbiol. Rev. croorganisms to man, Ann. Rev. Cell Biol. 8, 6759,171-200. 113. R. G., C., HORN,N., SWINDELL, DODD,H. M., GASSON, JACK, W., BIERBAUM, HEIDRICH, SAHL, S., H.-G. (in press), The genetics of lantibiotic bioM. J. (1991), Nisin biosynthesis genes are ensynthesis, BioEssays. coded by a novel conjugative transposon, Mol. JANSEN, F., HIRSCHMANN, J. (1944), Subtilin E. D. Gen. Genet. 228, 129-135. - an antibacterial product of Bacillus subtilis: HORNER, T., UNGERMANN, ZAHNER, H., V., culturing conditions and properties, Arch. BioFIEDLER, H.-P., UTZ, R., KELLNER, JUNG, R., chern. 4,297-304. G. (1990), Comparative studies on the fermentative production of lantibiotics by staphylococci, JUNG,G. (1991a), Lantibiotics - ribosomally synthesized biologically active polypeptides conAppl. Microbiol. Biotechnol. 32, 511-517. T. J. HOWELL, H., FIORELLINI, P., BLACKBURN, taining sulphide rings and a$-didehydroamino acids, Angew. Chem. (Int. Edn. Engl.) 30,1051S. J., P., PROJAN, J., DE LA HARPE, WILLIAMS, 1068. R. C. (1993), The effect of a mouthrinse based on nisin, a bacteriocin, on developing plaque JUNG,G. (1991b), Lantibiotics: a survey, in: Nisin G., and Novel Lantibiotics (JUNG, SAHL,H.-G., and gingivitis in beagle dogs, J. Clin. PeriodontEds.), pp. 1-34. Leiden: ESCOM Scientific Pub01. 20,335-339. lishers BV. HURST, (1966), Biosynthesis of the antibiotic niA. K.-D. (1989), Nisin, a pepsin by whole Streptococcus lactis organisms, J. KALEITA,C., ENTIAN, tide antibiotic: cloning and sequencing of the Gen. Microbiol. 44,209-220. nisA gene and posttranslational processing of its HURST, (1981). Nisin, Adv. Appl. Microbiol. 27, A. peptide product, J. Bacteriol. 171, 1597-1601. 85-123.
364
8 Lantibiotics
KALETTA, ENTIAN, C., K.-D., KELLNER, JUNG, R., produced by Bacillus subtilis ATCC 6633, Appl. G., REIS,M., SAHL,H.-G. (1989), Peps, a new Environ. Microbiol. 60, 2793-2801. lantibiotic: structural gene isolation and prepep- KLEIN, C., KALETI-A, SCHNELL, ENTIAN, C., N., tide sequence, Arch. Microbiol. 152, 16-19. K.-D. (1992), Analysis of the genes involved in the biosynthesis of the lantibiotic subtilin, Appl. KALETI-A, ENTIAN, K.-D., JUNG,G. (1991a), C., Environ. Microbiol. 58, 132-142. Prepeptide sequence of cinnamycin (Ro 090198): the first structural gene of a duramycin- KLEIN,C., KALETTA,C., ENTIAN,K.-D. (1993), type lantibiotic, Eur. J. Biochem. 199, 411415. Biosynthesis of the lantibiotic subtilin is regulated by a histidine kinaselresponse regulator sysKALETI-A, KLEIN,C., SCHNELL, ENTIAN, C., N., tem, Appl. Environ. Microbiol. 59, 296-303. K.-D. (1991b), An operon-like structure of the genes involved in subtilin biosynthesis, in: Nisin KLEINKAUF, VON D ~ H R E N , (1986) Peptide H. H., and Novel Lantibiotics (JUNG,G., SAHL,H.-G., antibiotics, in: Biotechnology, 1st Edn., Vol. 4 Eds.), pp. 309-319. Leiden: ESCOM Scientific (REHM,H. J., REED, G., Eds.), pp. 283-307, Publishers BV. Weinheim: VCH. KATZ,E., DEMAIN, L. (1977), The peptide anti- KLEINKAUF, VON DOHREN, (1987), BiosynA. H. H., biotics of Bacilllus: chemistry, biogenesis and thesis of peptide antibiotics, Ann. Rev. Micropossible functions, Bacteriol. Rev. 41, 499474. biol. 41,259-289 KELLNER, R., JUNG,G., HORNER,T., ZAHNER, KLEINKAUF, VON DOHREN, (1990), Non-riH. H., H., SCHNELL, ENTIAN,K.-D., GOTZ, F. N., bosomal biosynthesis of peptide antibiotics, Eur. (1988), Gallidermin, a new lanthionine-containJ. Biochem. 192,l-15. ing polypeptide antibiotic, Eur. J. Biochem. 177, KOGLER,H., BAUCH,M., FEHLHABER, H.-W., 53-59. GRIESINGER, SCHUBERT, C., W., TEETZ, V. M., KELLNER, JUNG, JOSTEN, KALETTA, C., R., G., (1991), NMR-spectroscopic investigations on ENTIAN,K.-D., SAHL, H.-G. (1989), Peps: strucmersacidin, in: Nisin and Novel Lantibiotics ture elucidation of a large lantibiotic, Angew. (JUNG,G., SAHL, H.-G., Eds.), pp. 159-170. Chem. 28,616-619. Leiden: ESCOM Scientific Publishers BV. KELLNER, JUNG, SAHL, R., G., H.-G. (1991), Struc- KORDEL, BENZ,R., SAHL, M., H.-G. (1988), Mode ture elucidation of the tricyclic lantibiotic Pep5 of action of the staphylococcin-like peptide containing eight positively charged amino acids, Peps: voltage-dependent depolarization of bacin: Nisin and Novel Lantibiotics (JUNG, G., terial and artificial membranes, J. Bacteriol. 170, SAHL, H.-G., Eds.), pp. 141-158. Leiden: 84-88. ESCOM Scientific Publishers BV. KOZAK,W., RAJCHERT-TRZPIL, ZAJDEL,J., M., S., D., KESSLER, STEUERNAGEL, GILLESSEN, H., DOBZANSKI, T. (1974), The effect of proflavW. KAMIYAMA, (1987), Complete sequence deT. in, ethidium bromide and elevated temperature termination and localization of one imino and on the appearance of nisin-negative clones in nithree sulphide bridges of the nonadecapeptide sin-producing strains of Streptococcus lactis, J. Ro 09-0198 by homonuclear 2D-NMR spectrosGen. Microbiol. 83,295-302. copy: the DQF-RELAYED-NOESY-experi- KUIPERS, P., ROLLEMA, S., YAP, W. M. G. 0. H. ment, Helv. Chim. Acta 70, 726-741. J., BOOT, H. J., SIEZEN, J., DE Vos, W. M. R. S., KESSLER, STEUERNAGEL, WILL,M., JUNG, H., (1992), Engineering dehydrated amino acid resiG., KELLNER, GILLESSEN, KAMIYAMA, R., D., dues in the antimicrobial peptide nisin, J. Biol. T. (1988), The structure of the polycylic nonadeChem. 267,24340-24346. capeptide Ro 09-0198, Helv. Chim. Acta 71, KUIPERS, P., BEERTHUYZEN, M., SIEZBN, 0. M. R. 1924-1929. J., DE Vos, W. M. (1993a), Characterization of KESSLER, SEIP, S., WEIN,T., STEUERNAGEL, the nisin gene cluster nisABTCIPR of LactococH., S., WILL, M. (1991), Structure of cinnamycin cus lactis: requirement of expression of nisA and (Ro 09-0198) in solution, in: Nisin and Novel nisl genes for development of immunity, Eur. J. G., Lantibiotics (JUNG, SAHL,H.-G., Eds.), pp. Biochem. 216,281-292. 141-158. Leiden: ESCOM Scientific Publishers KUIPERS, P., ROLLEMA, S., DE Vos, W. M., 0. H. BV. SIEZEN,R. J. (1993b), Biosynthesis and secreKEITENRING, MALABARABA, VEKEY,K., J., A., tion of a precursor of nisin Z by Lactococcus lacCAVALLERI, (1990), Sequence determination B. tis directed by the leader peptide of the homoloof actagardine, a novel lantibiotic, by homonugous lantibiotic subtilin from Bacillus subtilis, FEBS Lett. 330,23-27. clear 2D NMR spectroscopy, J. Antibiot. 43, 1082-1088. KUPKE, T., STEVANOVI~,SAHL,H.-G., GOTZ, S., KLEIN,C., ENTIAN, K.-D. (1994), Genes involved F. (1992), Purification and characterziation of in self-protection against the lantibiotic subtilin EpiD, a flavoprotein involved in the biosynthe-
8 References
sis of the lantibiotic epidermin, J. Bacteriol. 174, 5354-5361. KUPKE,T., STEVANOVI~, OTTENWALDER, S., B., METZGER,J. W., JUNG,G., GOTZ, F. (1993), Purification and characterization of EpiA, the peptide substrate for posttranslational modifications involved in epidermin biosynthesis, FEMS Microbiol. Lett. 112, 4348. KUPKE, T., KEMPTER,C., GNAU,V., JUNG, G., GOTZ, F. (1994), Mass spectroscopic analysis of a novel enzymatic reaction: oxidative decarboxylation of the lantibiotic precursor peptide EpiA catalyzed by the flavoprotein EpiD, J. Biol. Chem. 269,5653-5659. KUPKE, T., KEMFTER, C., JUNG, G., GOTZ, F. (1995), Oxidative decarboxylation of peptides catalyzed by flavoprotein EpiD: determination of substrate specificity using peptide libraries and neutral loss mass spectrometry, J. Biol. Chem. 270, 11282-11289. LIAN, L.-Y., CHAN,W. C., MORLEY,S. D., RoD. BERTS, c. K., BYCROFT,B. w., JACKSON, G. (1991), NMR studies of the solution structure of nisin A, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 43-58. Leiden: ESCOM Scientific Publishers BV. LIAN, L.-Y., CHAN,W. C., MORLEY,S. D., RoB. D. BERTS,G. c. K., BYCROFI-, w., JACKSON, (1992), Solution structures of nisin and its two major degradation products determined by NMR, Biochem. J. 283,413-420. LINNETT, E., STROMINGER, L. (1973), AddiP. J. tional inhibitors of peptidoglycan synthesis, Antimicrob. Agents Chemother. 4, 231-236. LIU, W., HANSEN,J. N. (1992), Enhancement of the chemical and antimicrobial properties of subtilin by site-directed mutagenesis, J. Biol. Chem. 267,25078-25085. MALABARABA, LANDI, M., PALLANZA, A., R., CAVALLERI, (1985), Physicochemical and B. biological properties of actagardine and some acid hydrolysis products, J. Antibiot. 38, 15061511. MALABARABA, PALLANZA,R., BERTI, M., A., CAVALLERI, (1990), Synthesis and biological B. activity of some amide derivatives of the lantibiotic actagardine, J. Antibiot. 43, 1089-1097. MARKI,F., FRANSON, (1986), Endogenous supR. pression of neutral-active and calcium-dependent phospholipase A2 in human polymorphonuclear leucocytes, Biochim. Biophys. Acta 879, 149-156. MARKI, F., HANNI,E., FREDENHAGEN, VAN A., OOSTRUM, (1991), Mode of action of the lanJ. thionine-containing peptide antibiotics duramycin, duramycin B, duramycin C and cinnamycin as direct inhibitors of phospholipase A2, Biochem. Pharmacol. 42,2027-2035.
365
MEYER,H. E. (1994), Analyzing posttranslational protein modifications, in: Microcharacterization of Proteins (KELLNER,R., LOTTSPEICH,F., MEYER,H. E., Eds.). Weinheim: VCH. C., MEYER,C., BIERBAUM, HEIDRICH, REIS, G., M., SOLING,J., IGLESIAS-WIND, I., KEMPM. TER, C., MOLITOR, E., SAHL,H.-G. (1995), Nucleotide sequence of the lantibiotic Pep5 biosynthetic gene cluster and functional analysis of PepP and PepC: evidence for a role of PepC in thioether formation, Eur. J. Biochem. 232, 478489. MOLITOR, SAHL,H.-G. (1991), Applications of E., nisin: a literature survey, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 434439. Leiden: ESCOM Scientific Publishers BV. MBRTVEDT, C. I., NES, I. F. (1990), Plasmid-associated bacteriocin production by Lactobacillus sake, J. Gen. Microbiol. 136, 1601-1607. MBRTVEDT, c. I., NISSEN-MEYER, SLETTEN, K., J., NES, I. F. (1991), Purification and amino acid sequence of lactocin S, a bacteriocin produced by Lactobacillus sake L45, Appl. Environ. Microbiol. 57, 1829-1834. MSADEK,T., KUNST, F., RAPOPORT,G. (1993), Two component regulatory systems, in: Bacillus subtilis and Other Gram-Positive Bacteria (SONENSHEIN, HOCH. L., LOSICK, Eds.), pp. A., R., 729-746. Washington, D C ASM. MULDERS, W. M., BOERRIGTER, J., ROLLEJ. I. MA,H. S., SIEZEN,R. J., DE Vos, W. M. (1991), Identification and characterization of the lantibiotic nisin Z, a natural nisin variant, Eur. J. Biochem. 201,581-584. NAKANO, M. M., ZUBER, P. (1990), Molecular biology of antibiotic production in Bacillus, Crit. Rev. Biotechnol. 10, 223-240. NARUSE,N., TENMYO, TOMITA,K., KONISHI, O., M., MIYAKI, KAWAGUCHI, FUKASE,K., T., H., WAKAMIYA, SHIBA,T. (1989), LanthiopepT., tin, a new peptide antibiotic. Production, isolation and properties of lanthiopeptin, J. Antibiot. 42, 837-845. NAVARRO, CHABOT, SHERRILL, ANEJA, J., J., K., R., ZAHLER, A., RACKER, (1985), InteracS. E. tion of duramycin with artificial and natural membranes, Biochem. 24,4645-4650. NISHIKAWA, TESHIMA,T., WAKAMIYA, M., T., SHIBA,T., KOBAYASHI, OKUBO,T., KYOY., GOKU,Y., KIDO, Y. (1988), Chemistry of lantibiotics, in: Peptide Chemistry 1987 (SHIBA,T., S., SAKAKIBARA, Eds.), pp. 71-74. Osaka: Protein Research Foundation. NISHIO,C., KOMURA, KURAHASHI, (1983), S., K. Peptide antibiotic subtilin is synthesized via precursor proteins, Biochem. Biophys. Res. Comm. 116,751-758.
366
8 Lantibiotics
REIS, M., SAHL,H.-G. (1991), Genetic analysis of the producer self-protection mechanism (immunity) against Peps, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 320331. Leiden: ESCOM Scientific Publishers BV. M., REIS, M., ESCHBACH-BLUDAU, IGLESIASWIND, M. I., KUPKE,T., SAHL,H.-G. (1994), Producer immunity towards the lantibiotic Peps: identification of the immunity gene pepl and localization and functional analysis of its gene product, Appl. Environ. Microbiol. 60, 43324338. REISINGER, P. SEIDEL, H., TSCHESCHE,H., HAMMES, P. (1980), The effect of nisin on W. murein synthesis, Arch. Microbiol. 127, 187193. RINCE, DUFOUR, LE POGAM, THUAULT, A., A., S., D., BOURGEOIS, M., LE PENNEC, P. (1994), C. J. Cloning, expression and nucleotide sequence of genes involved in production of lactococcin DR, a bacteriocin from Lactococcus lactis, Appl. Environ. Microbiol. 60, 1652-1657. RINTALA, GRAEFFE, PAULIN, KALKKIH., T., L., NEN,N., SARIS,P. E. J. (1993), Biosynthesis of nisin in the subtilin producer Bacillus subtilis ATCC 6633, Biotechnol. Lett. 15,991-996. ROGERS, A., WHITTIER, 0. (1928), Limiting L. E. 6 factors in lactic fermentation, J. Bacteriol. 1 , 211-214. ROLLEMA, S., BOTH, P., SIEZEN,R. J. (1991), H. NMR and activity studies of nisin degradation products, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 123-130. Leiden: ESCOM Scientific Publishers BV. C. Ross, K. F., RONSON, W., TAGG,J. R. (1993), Isolation and characterization of the lantibiotic salivaricin A and its structural gene salA from Streptococcus salivarius 20P3, Appl. Environ. Microbiol. 59, 2014-2021. RUHR,E., SAHL,H.-G. (1985), Mode of action of the peptide antibiotic nisin and influence on the membrane potential of whole cells and on artificial membrane vesicles, Antimicrob. Agents Chemother. 27,841-845. SAHL,H.-G. (1985), Influence of the staphylococcin-like peptide Pep5 on membrane potential of bacterial cells and cytoplasmic membrane vesicles, J. Bacteriol. 162, 833-836. SAHL, H.-G. (1991), Pore formation in bacterial membranes by cationic lantibiotics, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 347-358. Leiden: ESCOM Scientific Publishers BV. SAHL,H.-G., BRANDIS, (1981), Production, puH. rification and chemical properties of an antistaphylococcal agent produced by Staphylococcus epidermidis, J. Gen. Microbiol. 127, 377-384.
NOVAK,J., CAULFIELD, W., MILLER, E. J. P. (1994), Isolation and biochemical characterization of a novel lantibiotic mutacin from Streptococcus mutans, J. Bacteriol. 176, 4316-4320. PALMER,D. E., MIERKE, F., PAITARONI,C., D. GOODMAN, M., WAKAMIYA, FUKASE,K., T., FUJITA,H., SHIBA, (1989). Interactive NMR T. and computer simulation studies of lanthioninering structures, Biopolymers 28, 397408. PARENTI, PAGANI, BEREITA, G. (1976), F., H., Gardimycin, a new antibiotic from Actinoplanes. I. Description of the producer strain and fermentation studies, J. Antibiot. 24, 501-506. PESCHEL, AUGUSTIN, KUPKE,T., STEVAA., J., N O V I ~ , GOTZ, F. (1993), Regulation of epiS., dermin biosynthetic genes by EpiQ, Mol. Microbiol. 9, 31-39. PIARD,J.-C., MURIANA, M., DESMAZEAUD P. M. J., KLAENHAMMER, R. (1992), Purification T. and partial characterization of lacticin 481, a lanthionine-containing bacteriocin produced by Lactococcus lactis subspp. lactis CNRZ 481, Appl. Environ. Microbiol. 58, 219-284. PIARD,J.-C., KUIPERS,0. P., ROLLEMA, S., H. DESMAZEAUD, J., DE VOS, W. M. (1993), M. Structure, organization and expression of the lct gene for lacticin 481, a novel lantibiotic produced by Lactococcus lactis, J. Biol. Chem. 268, 16361-16368. PUGSLEY, P. (1988), The immunity and lysis A. genes of COINplasmid pCHAP4, Mol. Gen. Genet. 211, 335-341. PUGSLEY, P. (1993), The complete general seA. cretory pathway in Gram-negative bacteria, Microbiol. Rev. 57, 50-108. RACKER,E., RIEGLER,C., ABDEL-GHANY, M. (1983), Stimulation of glycolysis by placental polypeptides and inhibition by duramycin, Cancer, Res. 44,1364-1367. RAMSEIER, R. (1960), Die Wirkung von Nisin H. auf Clostridium butyricum, Arch, Microbiol. 37, 51-94. RAUCH,P. J. G., DE Vos, W. M., (1992), Characterization of the novel nisin-sucrose conjugative transposon Tn5276 and its insertion in Lactococcus lactis, J. Bacteriol. 174, 1280-1287. RAUCH, J. G., BEERTHUYZEN, M., DE Vos, P. M. W. M. (1990), Nucleotide sequence of IS904 from Lactococcus lactis subspp. lactis NIZO R5, Nucleic Acids Res. 18, 4253. RAUCH, J. G., BEERTHUYZEN, M., DE Vos, P. M. W. M. (1991), Molecular analysis and evolution of conjugative transposons encoding nisin production and sucrose metabolism in Lactococcus lactis, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 243-249. Leiden: ESCOM Scientific Publishers BV.
8 References
SAHL, H.-G., BRANDIS, (1983), Efflux of low Mr H. substances from the cytoplasm of sensitive cells caused by the staphylococcin-like agent Pep5, Zentralbl. Bakteriol. Hyg. I. Abt. Orig. A . 252, 166-175. SAHL,H.-G., KORDEL, BENZ,R. (1987), VoltM., age-dependent depolarization of bacterial membranes and artificial lipid bilayers by the peptide antibiotic nisin, Arch. Microbiol. 149, 120-124. SAHL,H.-G., REIS, M.,ESCHBACH, SZEKAT, M., C., BECK-SICKINGER, G., METZGER, STEA. J., V A N O V I ~ , JUNG, (1991), Isolation of Pep5 S., G. prepeptides in different stages of modification, in: Nisin and Novel Lantibiotics (JUNG, G., SAHL, H.-G., Eds.), pp. 332-346. Leiden: ESCOM Scientific Publishers BV. SAHL,H.-G., JACK,R. W., BIERBAUM, (1995), G. Lantibiotics: Biosynthesis and biological activities of peptides with unique posttranslational modifications, Eur. J. Biochem. 230, 827-853. SCHNELL,N., ENTIAN,K.-D., SCHNEIDER, U., GOTZ, F., ZAHNER, KELLNER, JUNG, H., R., G. (1988), Prepeptide sequence of epidermin, a ribosomally-synthesized antibiotic with four sulphide rings, Nature (London) 333, 276-278. SCHNELL, ENGELKE, AUGUSTIN, RoN., G., J., SENSTEIN, R., GOTZ, F., ENTIAN, K.-D. (1991), The operon-like organization of lantibiotic epiderrnin biosynthesis genes, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 269-276. Leiden: ESCOM Scientific Publishers BV. SCHNELL, ENGELKE, AUGUSTIN, RoN., G., J., SENSTEIN, R., UNGERMANN, v., GOTZ, F., ENTIAN, K.-D. (1992), Analysis of genes involved in the biosynthesis of the lantibiotic epidermin, Eur. J. Biochem. 204, 57-68. SCHULLER, BENZ,R., SAHL,H.-G. (1989), The F., peptide antibiotic subtilin acts by formation of voltage-dependent multi-state pores in bacterial and artificial membranes, Eur. J. Biochem. 182, 181-186. SEARS, M., SMITH, S., STEWART, K., GONP. B. W. ZALEZ, R. N., RUBINO, D., GUSIK, A., KUS. S. LISEK, E. s., PROJAN,s. J., BLACKBURN, P. (1992), Evaluation of a nisin-based germicidal formulation on teat skin of live cows, J. Dairy Sci. 75,3185-3190. SHETH, R., HENDERSON, M., HLADKY, B., T. R. S. CUTHBERT, W. (1992), Ion-channel formaA. tion by durarnycin, Biochim. Biophys. Acta 1107, 179-185. SHIBA, WAKAMIYA, FUKASE, SANO, T., T., K., A., SHIMBO, UEKI,Y. (1986), The chemistry of K., lanthionine-containing peptides, Biopolymers 25, Sll-S19.
367
SHOTWELL, L., STODOLA, H., MICHAEL, 0. F. W. R., LINDENFELSER, A., DWORSCHAK, L. G., PRIDHAM,T. G. (1958), Antibiotics against plant disease. 111. Duramycin, a new antibiotic from Streptomyces cinammonensis forma azacoluta, J. Am. Chem. SOC.80,3912-3915. SIEGERS,K., ENTIAN,K.-D. (1995), Genes involved in immunity to the lantibiotic nisin produced by Lactococcus lactis 6F3, Appl. Environ. Microbiol. 61, 1082-1089. SKAUGEN, NISSEN-MEYER, JUNG,G., STEM., J., K., V A N O V I ~ ,SLETTEN, MORTVEDT-ABILDS., GAARD, I., NES, I. (1994), In vivo conversion C. of L-serine to D-alanine in a ribosomally-synthesized polypeptide, J. Biol. Chem. 269, 2718327185. SLIJPERS, HILBERS, W., KONINGS, N. H., M., C. R. VAN DE VEN, F. J. M. (1989), NMR studies of lantibiotics: assignment of the 'H-NMR spectrum of nisin and identification of interresidual contacts, FEBS Lett. 252, 22-28. SOKOLOVE, M., WESTPHAL, A., KESTER,M. P. P. B., WIERWILE, VAN METER, K. S. (1989), R., Duramycin effects on the structure and function of heart mitochondria. I. Structural alterations and changes in membrane permeability, Biochim. Biophys. Acta 983, 15-22. SOMMA, MERATI, PARENTI, (1977), GarS., W., F. damycin, a new antibiotic inhibiting peptidoglycan synthesis, Antimicrob. Agents Chemother. 11,396-401. SONG,H. Y., CRAMER, A. (1991), Membrane W. topography of ColEl gene products: the immunity protein, J. Bacteriol. 173, 2935-2943. STEEN, T., CHUNG, J., HANSEN, N. (1991), M. Y. J. Characterization of the nisin gene as part of a polycistronic operon in the chromosome of Lactococcus lactis ATCC 11454, Appl. Environ. Microbiol. 57, 1181-1188. STEVENS, A., SHELDON, W., KLAPES, A., K. B. N. KLAENHAMMER, R. (1991), Nisin treatment T. for inactivation of Salmonella species and other Gram-negative bacteria, Appl. Environ. Microbiol. 537,3613-3615. STOFFELS,G., NISSEN-MEYER, GUDMUNDSJ., DOTTIR, SLETTEN, HOLO,H., NES, I. F. A., K., (1992), Purification and characterization of a new bacteriocin isolated from a Carnobacterium spp., Appl. Environ. Microbiol. 58, 1417-1422. STOFFELS,G., GUDMUNDSDOTTIR, ABEE, T. A., (1994), Membrane-associated proteins encoded by the nisin gene cluster may function as a receptor for the lantibiotic carnocin UI49, Microbiology 140, 1443-1450. STONE, K., XIE, X. S., RACKER, (1984), InhiD. E. bition of clathrin-coated vesicle acidification by duramycin, J. Biol. Chem. 259, 2701-2703.
368
8 Lantibiotics
VAN DE
SUROVOY, WEIDELICH, JUNG,G. (1992), A., D., Electrospray mass spectroscopic analysis of metal-peptide complexes, in: Peptides 1992, Proceedings of the 22nd European Peptide SympoC A. sium (SCHNEIDER, . H., EBERLE, N., Eds.), pp. 563-564. Leiden: ESCOM Scientific Publishing BV. L. TAGG,J. R., WANNAMAKER, W. (1978), Streptococcin A-FF22 nisin-like antibiotic substance produced by a group A Streptococcus, Antimicrob. Agents Chemother. 14,31-39. TAGG,J. R., DAJANI, S., WANNAMAKER, W., A. L. GRAY, E. D. (1973a), Group A streptococcal bacteriocin: production, purification and mode of action, J. Exp. Med. 138, 1168-1183. TAGG, J. R., READ, R. S. D., MCGIVEN,A. R. (1973b), Bacteriocin of group A Streptococcus: Partial purification and properties, Antimicrob. Agents Chemother. 4,214-221. TSAI, H.-J., SANDINE,W. E. (1987), Conjugal transfer of nisin plasmid genes from Streptococcus lactis 7962 to Leuconostoc dextranicum 181, Appl. Environ. Microbiol. 53, 352-357. UNGERMANN, GOEKE, K., FIEDLER, V., H.-P., ZAHNER,H. (1991), Optimization of fermentation and purification of gallidermin and epidermin, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL, H.-G., Eds.), pp. 410-421. Leiden: ESCOM Scientific Publishers BV. VAN BELKUM, J., KOK, J., VENEMA, M. G., HOLO, W. H., NES, I. F., KONINGS, N., ABEE,T. (1991), The bacteriocin lactococcin A specifically increases the permeability of lactococcal cytoplasmic membranes in a voltage-independent, protein-mediated manner, J. Bacteriol. 173, 79347941. VAN DE KAMP, M., HORSTINK, M., VAN DEN L. HOOVEN,H., KONING,R. N. H., HILBERS,C. W., FREY,A., SAHL,G.-G., METZGER,J., VAN DE VEN, F. J. M. (1995a), Sequence analysis by NMR spectroscopy of the peptide lantibiotic epilancin K7 from Staphylococcus epidermidis K7, Eur. J. Biochem. 227,757-771. VAN DE KAMP,M., VAN DE VEN, F. J. M., KONINGS,R. H. H., HILBERS, W., METZGER,J. C. G., W., JUNG,G., KUIPERS,0. P., BIERBAUM, SAHL,H.-G. (1995b), Elucidation of the primary structure of the peptide lantibiotic epilancin K7 from Staphylococcus epidermidis: cloning of the epilancin K7-encoding gene and Edman degradation of the mature peptide, Eur. J. Biochem. 230,587-600. VAN DE VEN, F. J. M., VAN DEN HOOVEN, W., H. KONINGS,R. N. H., HILBERS,C. W. (1991a), NMR-studies of lantibiotics: the structure of nisin in aqueous solution, Eur. J. Biochem. 202, 1181-1188.
VEN,F. J. M., VAN DEN HOOVEN, W., H. KONINGS,R. N. H., HILBERS,C. W. (1991b), The spatial structure of nisin in aqueous solution, in: Nisin and Novel Lantibiotics (JUNG,G., SAHL,H.-G., Eds.), pp. 3542. Leiden: ESCOM Scientific Publishers BV. VAN DER MEER, J. R., POLMAN, J., BEERTHUYZEN, M. M., SIEZEN,R. J., KUIPERS, P., DE 0. Vos, W. M. (1993), Characterization of the Lactococcus lactis nisin A operon genes nisP encoding a subtilisin-like serine protease involved in precursor processing and nisR encoding a regulatory protein involved in nisin biosynthesis, J. Bacteriol. 175, 2578-2588. VAN DER MEER, J. R., ROLLEMA, H. S., SIEZEN, R. J., BEERTHUYZEN, M., KUIPERS,0. P., M. DE Vos, W. M. (1994), Influence of amino acid substitutions in the nisin leader peptide on biosynthesis and secretion of nisin by Lactococcus lactis, J. Biol. Chem. 269, 3555-3562. VOGEL,H., NILSSON, RIGLER,R., MEDER, S., L., BOHEIM,G., BECK, W., KURTH,H.-H., JUNG, G. (1993), Structural fluctuations between two conformational states of a transmembrane helical peptide are related to its channel-forming properties in planar lipid membranes, Eur. J. Biochem. 212, 305-313. WAKAMIYA, UEKI, Y., SHIBA, T., KIDO, Y., T., MOTOKI,Y. (1985), The structure of ancovenin, a new peptide inhibitor of angiotensin I converting enzyme, Tetrahedron Lett. 26, 665-668. WAKAMIYA, FUKASE,K., NARUSE,N., KONIT., SHI,M., SHIBA,T. (1988), Lanthiopeptin, a new peptide effective against Herpes simplex virus: structural determination and comparison with R o 09-0198, an immunopotentiating peptide, Tetrahedron Lett. 29, 47714772. WEIL, H.-P., BECK-SICKINGER, G., METZGER, A. J., STEVNANOVI~, JUNG, G., JOSTEN, M., S., SAHL, H.-G. (1990), Biosynthesis of the lantibiotic Pep5: Isolation and characterization of a prepeptide containing dehydroamino acids, Eur. J. Biochem. 194,217-223. ZIMMERMANN, (1995), Raumstrukturaufklarung N. der Lantibiotika Duramycin B, Duramycin C und Actagardin durch mehrdimensionale Kernresonanzspektroskopie, PhD Thesis, University of Tubingen, Germany. ZIMMERMANN, JUNG,G. (1995), The tetracyclic N., lantibiotic actagardine. 'H-NMR and I3C-NMR assignments and revised primary structure, Eur. J. Biochem. 228,786-797. ZIMMERMANN, FREUND,S., FREDENHAGEN, N., A., JUNG,G. (1993), Solution structures of the lantibiotics duramycin B and C, Eur. J. Biochem. 216,419-428.
Biotechnology
9 Glycopeptide Antibiotics (Dalbaheptides)
GIANCARLO LANCINI
BRUNOCAVALLERI
Gerenzano, Italy
1 Introduction 371 2 Descriptive Chemistry 371 2.1 Physicochemical Properties 376 3 Biological Activity 377 3.1 In v i m Antibacterial Activity 377 3.2 Mechanism of Action 378 3.3 Resistance 379 3.4 In vivo Efficacy and Pharmacology 379 4 Producing Organisms 380 5 Methods of Screening 381 6 Fermentation 381 6.1 Fermentation Media - Carbon Sources 381 6.2 Fermentation Media - Nitrogen Sources 382 6.3 Fermentation Media - Effect of Phosphates 382 6.4 Inhibition by the Final Product 382 6.5 Control of Complex Composition 383 7 Recovery and Purification 383 8 Biosynthesis 384 8.1 Origin of the Uncommon Amino Acids 384 8.2 Origin of Fatty Acids of Lipoglycopeptides 386 8.3 Glycosylation and Final Modifications 387 9 Chemical Modifications 387
310
10 Biotransformations 389 10.1 Deglycosylation 389 10.2 Glycosylation 389 10.3 Deacylation 389 10.4 Other Biotransformations 390 11 References 390
2 Descriptive Chemistry
371
1 Introduction
The term glycopeptide antibiotics is commonly used to indicate a family of microbial metabolites, active on gram-positive bacteria, and closely related in their chemical structure and biological activity to the antibiotics vancomycin and ristocetin - the first members of the family isolated in the early 1950s. The expression glycopeptide antibiotics of the vancomycin-ristocetin family is more precise and is also often used. The characteristics common to all members of the family are: - a structure composed of a linear heptapeptide in which at least five of the amino acid residues are aromatic, the rings linked to form a triphenyl ether moiety and a diphenyl group; - a unique mechanism of action, i.e., inhibition of bacterial growth by binding to the D-Ala-D-Ala terminus of peptidoglycan precursors, thus inhibiting cell wall formation. Taking into account their chemical characteristics and unusual mechanism of action, the name Dalbaheptides, from Dal (anyl-D-alanine) B(inding) A(ntibiotics with) Hept(apept)ide (structure), was proposed for these antibiotics .by PARENTI and CAVALLERI (1989). Several reviews have been published reporting different aspects of glycopeptide antibiotics: discovery, isolation, and purification (CASSANI,1989; SITRIN and FOLENA-WASSERMAN, 1989; CAVALLERI and PARENTI, 1992), mechanism of action (BARNA and WILLIAMS, 1984; REYNOLDS, 1989), fermentation and biosynthesis (LANCINI, 1989; LANCINI and CAVALLERI, 1990), chemistry and chemical derivatives (MALABARBA al., et 1993a), structure-activity relationship (NAGARAJAN, 1993), pharmacology (CASSETTA et al., 1991; PHILLIPS and GOLLEDGE, 1992; BROGDEN PETERS,1994). Biological and and chemical aspects are comprehensively discussed in a multiauthor book published recently (NAGARAJAN, 1994a). Although many natural and semisynthetic glycopeptides are provided with relevant antibacterial activity, only vancomycin and teico-
planin are used in human medicine, due to their effect on gram-positive pathogens refractory to established antibiotics, such as multiresistant Staphylococcus aureus, coagulase-negative staphylococci, clostridia, and enterococci. Eremomycin, a relatively recent natural product is under clinical evaluation. Avoparcin is commercially available as growth promoter in animal feed. Finally, ristocetin is a diagnostic agent for a particular disorder of genetic origin (von Willebrands disease) due to the ability to aggregate blood platelets.
The first glycopeptide antibiotic isolated from microbial fermentation was ristocetin in 1953, followed by vancomycin in 1955. Almost immediately their composition of sugar carrying small peptides was defined, but the determination of their structures was completed only 20 years later, mainly by studies with high-field NMR. In the 1960s and 1970s the isolation of a small number of new glycopeptides including avoparcin and teicoplanin was reported, but only in the 1980s the introduction into screening programs of specific methods for the detection of these antibiotics in fermentation broth resulted in the discovery of the majority of glycopeptides now available. Modern techniques of isolation and purification such as reverse-phase chromatography and affinity chromatography and availability of powerful spectroscopic techniques such as 2D-NMR, Fast Atom Bombardment Mass Spectrometry (FAB-MS), and Ion Spray MS have now made it possible to determine the structure of new products in a relatively short time. A list of natural glycopeptides is shown in Tabs. 1-4. More detailed references and the chemical structures are provided in the previously cited reviews by LANCINI and CAVALLERI (1990) and CAVALLERI and PARENTI (1992). It should be noted that in some cases different code numbers or names were assigned to identical chemical entities produced by different strains, before their
372
9 Glycopeptide Antibiotics (Dalbaheptides}
structure had been elucidated. A striking example is that of eremomycin that was successively reported as A82846A, MM45289, and LY264826. In addition to the products listed in Tabs. 1-4 the following compounds with unknown structure were reported AM374 from Streptomyces ebureosporeus (KUNSTMANN and PORTER, 1974), A477 from Actinoplanes sp. (HAMILL al., 1973), AB65 et from Saccharomonospora viride (TAMURA 1975). and TAKEDA, Many of the microorganisms listed produce families (complexes) of strictly related compounds. The components (factors) of a complex usually differ in the level of methylation or chlorination of the peptidic skeleton or in the presence of additional sugars. Analogs lacking some or all sugar units present in the parent antibiotic (therefore designated as "pseudoaglycones" or aglycones, respectively) are either the result of incomplete glycosylation or generated by chemical or enzymatic deglycosylation during fermentation or recovery and purification. In several cases, further studies on complex producing strains have revealed the presence of additional minor components in the fermentation broth that had
initially been overlooked (BORGHI et al., 1989; NAGARAJAN, 1993). The glycopeptide skeleton is shown in Fig. 1. The numbering of the amino acid residues is that proposed by BARNAet al. (1984); it has been widely used because it allows an easy comparison of NMR data of glycopeptides having different core structures. It should be
OH
Fig. 1 General lieptapeptide structure of dalbahep. tides and interaction with the D-alanyl-D-alanine peptide terminus. Hydrogen bonds are indicated by dottet lines.
Tab. 1 Naturally Occurring Dalbaheptides - Ristocetin Type .

Name Ristocetin A. B Ristomycin A, B Producing Strain" Company, Yearb References SZTARICSKAI BOGNAR, and 1984 KATRUKHA SILAEV, and 1986 SZTARICSKAI BOGNAR, and 1984 KATRUKHA SILAEV. and 1986 DEBONO al., 1984 et HUNTet al., 1984a MICHEL al., 1980 et DEBONO al., 1980 et BOECK al., 1985 et HUNTet al., 1985 BOECKand MERTZ,1986 SKELTON WILLIAMS, and 1990 HOLDOM al., 1988 et HOLDEN al., 1991 et SKELTON al., 1991 et
Nocardia lurida, NRRL 2430 Abbott. 1953 Proactinomyces fructiferi Inst. New Antibiotics, Moscow, 1962 Lilly, 1971 Lilly, 1976 Lilly, 1982 Lilly, 1982 Pfizer, 1987 Pfizer, 1991
Actinoplanes missouriensis, Actaplanin (A4696 complex) ATCC 23342 A35512 complex Streptomyces candidus, NRRL 8156 A41030 complex Streptomyces virginiae, NRRL 15156 A47934 Streptomyces toyocaensis, NRRL 15009 Actinoplanes sp., UK-68597 ATCC 53533 UK-69542 Saccharothrix aerocolonigenes, ATCC 53829
a
The name of the producing strain is according to the original publication; note that several strains have been subsequently reclassified into different genera. Year of first paper or patent publication.
373
HO
CH, OH
pz x
OH OH
Fig. 3. Vancomycin.
OH
Fig. 2. Ristocetin A.
noted that the residues are numbered from the amino terminus, in contrast to the usual system of peptide numbering starting from the carboxyl terminus. Phenylamino acids 2,4,5,6, and 7 are present in all glycopeptides and a classification based on the remaining amino acids 1 and 3 is commonly used. The compounds wherein the phenylic moieties of amino acids 1 and 3 are linked by an oxygen belong to the ristocetin type (Tab. 1); their structure is exemplified by that of ristocetin A in Fig. 2. Vancomycin type includes the glycopeptides listed in Tab. 2, where amino acids 1 and 3 are aliphatic. Synmonicin, also included in Tab. 2, is an exception, since amino acid 3 is a thioamino acid and amino acid 1 is aromatic. The structure of vancomycin is shown in Fig. 3. In the actinoidin type (Tab. 3) amino acid 1 is always a p-hydroxyphenylglycine whereas amino acid 3 is a phenylalanine or p-hydroxyphenylglycine (Fig. 4). Within a group other characteristics account for the large variety of glycopeptide structures with chlorine atoms, methyl and hydroxy groups at different positions of the phenyl residues. Amino acid 6 and occasionally amino acid 2 carry a hydroxyl group in @position. Common or unusual sugars are linked at different positions through glyco-
OH
Fig. 4. Actinoidin A.
sidic bonds. The sugars vancosamine (and the related epivancosamine and 4-ketovancosamine), ristosamine, actinosamine, and acosamine were first isolated from glycopeptides. Only A41030A, B, E and A47934 do not contain any sugar. In a few compounds a phenolic hydroxyl is esterified as a sulfate monoester (A47934, UK-68597, UK-69542). The terminal carboxyl is often a methyl ester and the terminal amino group can be methylated. In some ristocetin-type glycopeptides linear or branched (exceptionally unsaturated) fatty
374
Tab. 2. Naturally Occurring Dalbaheptides - Vancomycin Type
Name Vancomycin
Producing Straina
Company, Yearb Lilly, 1955 Tohoku University, 1961 Otsuka Pharm., 1978 Lilly, 1982 Lilly, 1984 Kitasato Institute, 1986 Smith, Kline & French, 1986 Shionogi, 1987 Inst. New Antibiotics, Moscow, 1987 Lepetit, 1987 Lilly, 1987 Shionogi, 1988 Lilly, 1989 Beecham, 1989 Beecham, 1989 Hoechst, 1990 Pfizer, 1990 Beecham, 1990 Hoechst, 1992 Lilly, 1993
References HARRIS al., 1983 et SZTARICSKAI BOGNAR, and 1984 MATSUMOTO, 1961 KAMOGASHIRAal., 1983 et ANGet al., 1988 BOECK al., 1984 et HUNTet al., 1984b HIGGINS al., 1985 et SPIRI-NAKAGAWA 1986 et al., ARJUNARAOet al., 1986 TSUJIet al., 1988b BRAZHNIKOVA 1989 et al., GAUSE al., 1989 et RIVAet al., 1989 HAMILL al., 1988 et NAGARAJANet al., 1989a TSUJIet al., 1988a NAGARAJAN al., 1989b et DOOLIN al., 1989 et GOODet al., 1990 Box et al., 1990 FRANCO al., 1990 et SKELTON al., 1990 et COATES al.. 1990a et
NADKARNI al.. 1994 et
Streptomyces orientalis, NRRL 2450 K-288 Streptomyces haranomachiensis OA-7653 A, B Streptomyces hygroscopicus ssp. hiwasaensis, ATCC 31613 Nocardia orientalis, A51568A (N-demethyl vanco- NRRL 15232 mycin) N. orientalis, NRRL 2450 M43 complex Nocardia sp., Izupeptin A, B FERM P-8656 Synmonicin A, B, C Synnemomyces mamno(CWI-785) orii, ATCC 53296 N. orientalis, PA-42867 Orienticin (PA-42867-A, B, C, D) Actinomyces sp., INA-238 Eremomycin A42867 Nocardia sp., ATCC 53492 Amycolatopsis orientalis, A82846 A, B, C NRRL 18098, NRRL 18099 Chloroorienticin A, A. orientalis. PA-45052 B, C, D, E (PA-45052) Kibdelosporangium philA80407 A. B ippinensis, NRRL 18198 A . orientalis. NCIB 12531 MM 45289, MM 47756 A . orientalis. NCIB 12608 MM 47761, MM 49721 Kibdelosporangium decDecaplanin caensis, DSM 4763 (M 86-1410) UK-72 051 A . orientalis MM 55270, Arnycolatopsis sp., NCIB 40086 MM 55271, MM 55272 Amycolatopsis sp., Balhimycin DSM 5908 Amycolatopsis albus, A83850 A, B NRRL 18532 See footnotes in Tab. 1
HAMILL YAO, 1993 and
acids are linked as amides to the amino group of a glucosamine (or 2-aminoglucuronic acid) moiety. These antibiotics listed in Tab. 4 are thus complexes the components of which dif-
fer in the nature of the aliphatic side chains as illustrated by the structure of teicoplanin (Fig. 5). The aliphatic moieties cause a certain lipophilic character of these lipoglycopep-
Tab. 3. Naturally Occurring Dalbaheptides - Actinoidin Type Name Actinoidin A, B Avoparcin (LL-AV290 comP W Chloropolysporin A, B, C Actinoidin A2 Helvecardin A, B MM47766, MM47767, MM55256, MM 55260 Galacardin A. B Producing Straina
Proactinomyces actinoides Streptomyces candidus, NRRL 3218 Faenia interjecta, FERM BP-583 Nocardia sp., SKF-AAJ-193 Pseudonocardia compacta ssp. helvetica, SANK 65185 Amycolatopsis orientalis, NCIB 40011 Saccharothrix sp., SANK 64289
375
Company, Yearb Inst. New Antibiotics, Moscow, 1956 American Cyanamid, 1966 Sankyo, 1983 Smith, Kline & French, 1987 Sankyo, 1988 Beecham, 1989
References BERDNIKOVAal., 1982 et MCGAHREN al., 1980 et MCGAHREN al., 1983 et OKAZAKI al., 1987 et TAKATSU al., 1987a et DINGERDISSENal., 1987 et HEALDet al., 1987 TAKEUCHI al., 1991a et TAKEUCHI al., 1991b et ATHALYE al., 1989 et
Sankyo, 1990
TAKEUCHI al., 1992 et
See footnotes in Tab. 1
Tab. 4. Naturally Occurring Dalbaheptides - Lipoglycopeptides Name Teicoplanin Producing Straina Company, Yearb Lepetit, 1975 Smith, Kline & French, 1983 Lepetit, 1984 Smith, Kline & French, 1985 Smith, Kline & French, 1986 Beecham. 1990 References PARENTI al., 1978 et CORONELLI al., 1987 et SHEARER al., 1985 et SITRIN al., 1985 et SELVA al., 1986 et GOLDSTEIN al., 1987 et SHEARER al., 1986 et FOLENA-WASSERMAN 1986 et al., CHRISTENSEN al.. 1987 et COATES al., 1990b et Box et al., 1991 MICHEL YAO, 1991 and COATES al., 1991 et
Actinoplanes teichomyceticus, ATCC 31121 Kibdelosporangium ariArdacin (aridicin, AAD-216 dum. ATCC 39323 complex) A40926 complex Actinomadura sp., ATCC 39727 K. aridum, ATCC 39922 Kibdelin (AAD-609 comP W Actinomadura parvosata, Parvodicin ATCC 53463 (AAJ-271) Amycolatopsis sp., MM 49728, NCIB 40089 MM 55266, MM 55267, MM 55268 Streptosporangium carA84575 complex neum, NRRL 18437, NRRL 18505 Amycolatopsis sp., MM56597, NCIB 40089 MM 56598
Lilly, 1991 Beecham, 1991
See footnotes in Tab. 1
376
-NH,
H;OH k OH
T&-I:R=
T&-2:R=
T-A?-3R I
T-M-4:Rr
T-M-5: R I
0
0
Fig. 5. Teicoplanin (teichomycin complex).
tides that directly influences their biological properties and in particular their pharmacokinetic behavior.
Tab. 5. Isoelectric Points (I. P.) of Some Dalbaheptidesa
Antibiotic A47934 A40926 Ardacin A41030 OA-7653 Teicoplanin Chloropolysporin A35512B Actaplanin AM374 Avoparcin Vancomycin Ristocetin A42867 A477
a
I. P.
2.1 Physicochemical Properties

Glycopeptides are colorless or whitish powders, generally water-soluble, that usually strongly retain water or crystallization solvents. Their molecular weights range from 1150-2300 Da. Some of them as, e.g., teicoplanin can be isolated as an internal salt or as a partial monoalkaline (sodium) salt, depending on the pH value of the aqueous medium in the final purification step. Others can be isolated as acidic salts: vancomycin and actaplanin as hydrochlorides, and ristocetin A, avoparcin, and eremomycin as sulfates. Many pharmaceutically acceptable basic and acidic addition salts are claimed in patent applications. The net charge of several glycopeptides has been determined by electrofocusing. The isoelectric points reported in Tab. 5 range
3.2 3.7-3.9 3.9 4.9 4.9 5.0 7.5 7.5-7.8 7.6-8.5 7.7-8.0 7.7-8.1 7.7 8.1 8.1 7.9-8.1
Determined by electrofocusing (RIVA,E., SOFFIENTINI, A., personal communication)
3 Biological Activity
377
from 3.2 (A47934) to 8.1 (ristocetin, A477, and A42867). Retention times from reverse-phase HPLC give an approximate indication of the relative lipophilicity. In Tab. 6 the values obtained with representative compounds are listed. All glycopeptides exhibit similar UV absorption spectra with maxima at about 280 nm in acidic or neutral media that shift to 292-300 nm under alkaline conditions. NMR experiments led to complete assignment of each hydrogen, carbon, and nitrogen atom for the majority of the glycopeptides and greatly contributed to structure elucidation of both natural and semisynthetic compounds. Although almost 1000 natural and semisynthetic glycopeptides are presently known no crystalline forms suitable for X-ray analysis have been reported, except for a degradation product of vancomycin (designated as CDP-I) that allowed the determination of the absolute configuration of vancomycin (SHELDRICK et al., 1978). The configuration of the seven ami-
Fig. 6. Stereo model of teicoplanin aglycone.
Tab. 6. Reverse-Phase HPLC Retention Times (tR) of Some Dalbaheptides" Antibiotic AM374 Vancomycin Ristocetin Chloropolysporin A35512B Actaplanin Avoparcin A47934 OA-7653 A41030 A477 Ardacin Teicoplanin A40926
tR [min]
7 10.5 11 12 12.5 14 13 17 20 21 23 24-26 24-28 30
no acids is lR, 2R, 3S, 4R, 5R, 6S, and 7S, respectively. The NMR spectra of vancomycin and of glycopeptides discovered later show that the stereochemistry and the overall three-dimensional conformation is the same in all compounds except for a few epimers at amino acid residue 1. In teicoplanin five of the amide bonds are trans, the 5-6 bond is cis: modification of the bond conformation often occurs during chemical reactions. As an example, the three-dimensional structure of teicoplanin aglycone is presented in Fig. 6.
3.1 In vitro Antibacterial Activity
Glycopeptide antibiotics are active on most gram-positive aerobic and anaerobic bacteria (CAMFOLI-RICHARDS 1990). Gram-neet al., gative bacteria are generally insensitive except for Neisseria, Gardnerella, and Branhamella strains on which several natural or semisynthetic glycopeptides exert a weak effect. However, most strains of the genera Lactobacillus, Leuconostoc, and Pediococcus - although gram-positive - are insensitive (JOHNSON al., 1990). The minimal inhibitoet ry concentrations (MIC) of some glycopep-
a On Ultrasphere ODS 5 pm (250-4.6 mm) column (Beckman); mobile phases: (A) 0.02 M aq NaH2P04/CH3CN9: 1, pH 6; (B) 0.02 M aq NaH2P04/CH3CN3 :7, pH 6; gradient from 5%-60% of eluent (B) in 40 min; flow rate 1.6 mL.min-'. (RIVA, and SOFFIENTINI, personal communiE. A., cation)
378
Tab. 7. Comparison of Antibacterial Activity (MIC, pg.mL-') of Some Dalbaheptides (GOLDSTEIN al., et 1987) Antibiotic
Staphylococcus aureus TOUR StaphyloStrepto- Enterococcus coccus coccus epidermidis pyogenes faecalis ATCC C 203 ATCC 7080 12228 Propionibacterium acnes ATCC 6919
0.016 nd 0.13 nd nd nd nd nd nd nd
Bacteroides Neisseria fragilis gonorrhoeae ATCC ISM 681126 23745

64 nd 128 nd nd nd nd nd nd nd 2 8 32 32 64 64 64 64 128 > 128
A40926 A47934 Teicoplanin Vancomycin Ristocetin A35512B A41030A Aridicin Avoparcin Actaplanin
0.06 0.06 0.13 0.25 4 1 0.03 1 2 1
0.06 0.03 0.13 0.5 2 0.5 0.008 4 2 2
0.06 0.13 0.06 0.13 0.25 0.13 0.13 0.13 0.25 0.13
0.06 0.13 0.13 0.5 1 0.5 0.13 2 0.25 0.5
nd: not determined
tides on representative pathogens are reported in Tab. 7. Since - as discussed later the activity of these antibiotics is due to their ability to bind peptidoglycan precursors that are similar in gram-positive and gram-negative bacteria, the insensitivity of the latter is attributed to a lack of penetration through the outer membrane. It is, however, noteworthy that some semisynthetic derivatives exhibit a certain activity against Escherichia coli, Proteus vulgaris, and Pseudomonas aeruginoet sa (MALABARBA al., 1993a). The effect of glycopeptides on staphylococci and streptococci is clearly bactericidal at concentrations slightly higher than the MICs. Their bactericidal effect is weaker in Enterococcus faecium or E. faecalis (CAMPOLIRICHARDS al., 1990). Lethality can be acet companied by cell lysis depending on the strain and on environmental conditions.
3.2 Mechanism of Action

A most interesting property of glycopeptide antibiotics is their ability to block cell wall formation in sensitive bacteria by binding to precursors of peptidoglycan synthesis (REYNOLDS, 1989; NICASand ALLEN,1994). Inhibition of cell wall synthesis in growing
bacterial populations is supported by several experimental data (WALLASand STROMINGER,1963; SOMMER al., 1984) the most relet evant of which are: - Addition of a glycopeptide to a bacterial culture blocks the uptake of peptidoglycan precursors such as acetyl glucosamine or diamino pimelic acid several minutes before any effect can be noticed on the uptake of precursors of DNA, RNA, or proteins. - Inhibition of uptake is accompanied by accumulation of intermediates of peptidoglycan biosynthesis, notably UDP-muramylpentapeptide, indicating that these antibiotics do not interfere with early steps of peptidoglycan synthesis but affect a later stage of the process. The observation that vancomycin activity is reversed by cell wall fragments or by late intermediates of the peptidoglycan biosynthetic pathway gave the first indication of the action of glycopeptides on a molecular level. The classical work of NIETO and PERKINS (1971a, b) established clearly that glycopeptides bind with relatively high affinity to peptides having a D-Ala-D-Ala carboxyl terminus. A lower binding affinity is observed when other amino acids substitute for one of these residues. The tripeptide N,N-diacetyl-LLys-D-Ala-D-Ala was used by several authors
379
as a model of the natural ligand to determine the binding affinity of different antibiotics. An association constant of 1.5.10" L-mol-' was calculated for vancomycin. The values obtained with ristocetin and teicoplanin were 5.9-105 and 2.6.10" Lsmol-', respectively (SOMMA al., 1984). et Experiments performed to establish which biochemical reaction is specifically inhibited in the peptidoglycan biosynthetic pathway have not achieved decisive results so far. Glycopeptides are most probably unable to cross the bacterial cytoplasmic membrane as was experimentally shown with iodinated vancomycin (PERKINS NIETO,1970); therefore, and their action appears to be limited to events occurring at the membrane surface, such as transglycosylation or transpeptidation reac1982). Vancomycin and teitions (PERKINS, coplanin were reported to inhibit transglycosylation in a cell free membrane system. However, the inhibitory effect was observed only at antibiotic concentrations substantially higher than the MICs of these drugs (SOMMA et al., 1984). Another aspect difficult to be interpreted is the often observed poor correlation between the degree of affinity to model tripeptides and the antimicrobial activity of the antibiotic. An extreme case is that of erernomycin, consistently more active than vancomycin but demonstrating a significantly lower affinity to Ac2-Lys-D-Ala-D-Ala (GOODet al., 1990).
3.3 Resistance
To date, despite many years of clinical use, vancomycin resistance has not been reported for Streptococcus sp. or Staphylococcus aureus clinical isolates. In vitro selection for vancomycin or teicoplanin resistance in S. aureus is difficult and strains with only modest increases in MIC are obtained (JOHNSON al., et 1990). A decreased susceptibility to vancornycin and more frequently to teicoplanin has 3.4 In vivo Efficacy and been observed with coagulase-negative staphylococci, such as S. epidermidis or S. hae- Pharmacology molyticus. The biochemical mechanism deterThe efficacy of the principal glycopeptides mining this low-level resistance has not been elucidated yet (ARDUINOand MURRAY, in curing experimental infections has been assessed in septicemia models in mice (PAL1993).
Enterococci resistant to vancomycin or teicoplanin have been isolated with increasing frequency in the last years, especially in intensive care units. Three major phenotypes termed VanA, VanB, VanC can be distinguished in resistant Enterococcus species (ARTHUR COURVALIN, and 1993; NICASand ALLEN, 1994). Enterococci carrying the VanA gene are characterized by high-level resistance to vancomycin and teicoplanin. A cluster of genes the expression of which is associated with the VanA phenotype has been identified in these strains on a transposon designated Tn2546. Biochemical characterization of the gene products has revealed that the resistance is due to an unusual pathway of peptidoglycan biosynthesis in which UDPmuramyl-tetrapeptide-D-lactateis synthesized instead of the normal precursor UDP-muramyl-pentapeptide. Apparently, all enzymes catalyzing the subsequent reactions of the pathway accept as substrates the intermediates carrying this modification. The result is a peptidoglycan structure in which the residue D-Ala-D-Lac substitutes for D-Ala-D-Ala. As glycopeptides do not form complexes with peptides ending with D-Ala-D-Lac the strain results resistant to their action. Phenotype VanB is characterized by moderate or high-level resistance to vancomycin and sensitivity to teicoplanin. The mechanism of resistance is the same as of the VanA phenotype. However, in VanB strains expression of resistance genes is inducible by vancomycin but not by teicoplanin which explains the different susceptibility to the two antibiotics. The VanC phenotype comprises species such as Enterococcus gallinarum in which the pentapeptides of peptidoglycan and its biosynthetic precursors constitutively possess a D-Ala-D-Ser ending. These species show lowlevel resistance to vancomycin and sensitivity to teicoplanin.
380
Tab. 8. Experimental Septicemia in Mice" Antibiotic Infecting Organism (Clinical Isolates)

S. aureus L 165
S. pyogenes L 49
S. pneumoniae L 44
MIC ED50 [pg * mL - '1 [mg- kg -'I Teicoplanin Vancomycin Ampicillin Cephaloridine Erythromycin
a
MIC [pg - mL -'I

0.05 0.8 0.02 0.01 0.05
ED50
[mg - kg -'I
MIC [pg-mL-'I
0.1 0.8 0.02 0.02 0.01
ED50 [mg .kg - '1

0.41 1.9 4.1 0.93 26
0.4 0.8 0.1 0.05 0.5
0.72 7.2 8.1 2.8 28
0.11 0.58 0.1 0.03 0.44
Animals treated subcutaneously once daily for three days (PALLANZA al., 1983). et
et al., 1983). The results obtained with teicoplanin and vancomycin are reported in Tab. 8 in comparison with other antibiotics. In general, a higher in vivo efficacy corresponds to a higher in vitro activity although no simple correlation is apparent. Eremomycin appears to be more active than vancomycin both in vitro and in experimental infections. Ardacins are less active in vivo than vancomycin although having a comparable antibacterial activity. This also holds for A40926 when compared to teicoplanin. A few products have been tested in more complex models of experimental infection. Teicoplanin and vancomycin were found effective in reducing the bacterial load in experimentally induced endocarditis in rats and et rabbits (GOLDSTEIN al., 1994). A clinical overview of vancomycin has been published by ZECKELand WOODWORTH (1994). Toxicology, pharmacokinetics, pharmacology, and therapeutic use of teicoplanin have been extensively reviewed by GOLDSTEIN et al. (1994) and BROGDEN and PETERS (1994).
LANZA
4 Producing Organisms
All known glycopeptides are produced by microorganisms of the order of Actinomycetales isolated from soil samples collected from almost everywhere. Surprisingly, only a few
glycopeptides are produced by streptomycetes and practically all of them are of the ristocetin type, although ristocetin itself was isolated from strains originally considered as Nocardia or Proactinomyces and later classified as Amycolatopsis (LECHEVALIER al., et 1986). A large proportion of known glycopeptide is generated by this genus. In fact, the vancomycin producing strain has been in turn classified as Streptomyces, then as Nocardia, and at present as Amycolatopsis; several glycopeptides of the vancomycin type are produced by A . orientalis and most probably several other strains should be reclassified into the same genus. Actinoidin producers are probably also Amycolatopsis although originally considered as Proactinomyces or Nocardia. Lipoglycopeptides are produced by a variety of rare actinomycetes such as Actinomadura, Actinoplanes, Streptosporangium, and the new genus Kibdelosporangium. Interestingly, glycopeptides that are minor components of complexes produced by a microbial strain are the main products of other soil isolates. Partially purified vancomycin produced by A . orientalis NRRL 2450 contains minor quantities of at least nine structurally related compounds (NAGARAJAN, 1993). One of them, A51586A, is the main product of N. orientalis NRRL 15232 (HUNT et al., 1984b).
6 Fermentation
381
5 Methods of Screening
The first glycopeptide antibiotics were discovered by conventional screening procedures such as growth inhibition of a test organism on agar plates. Later, specific tests were devised to detect inhibitors of cell wall synthesis. In the Lepetit Laboratories, fermentation broths were tested for differential . inhibition of a S aureus strain and an L-form derived from it. In this way teicoplanin was discovered. Similarly, izupeptin was found on the basis of its differential activity on cell wall possessing bacteria and on mycoplasma (SPIRI-NAKAGAWA al., 1986). et Subsequently, very efficient screening tests were implemented exploiting the unique mechanism of action of the antibiotics. One of them was based on affinity chromatography; matrix bound D-Ala-D-Ala was prepared by forming a peptide bond between aminocaproylSepharose and the amino group of the dipeptide (CORTIand CASSANI, 1985). A different antimicrobial activity of culture filtrates before and after passage on the affinity resin was taken as evidence for the presence of a glycopeptide that could easily be isolated by elution. By this method 72 strains (about 0.3% of the fermentation broth tested) were identified as producers of glycopeptides in a screening campaign (CASSANI, 1989); about 60% were ristocetin producers. Among the others teicoplanin, avoparcin or actaplanin producing strains were identified, and two new glycopeptides (A40926 and A42867) were discovered as well. RAKEet al. (1986) applied an assay based on the affinity of glycopeptides to the tripepto 1936 tide N,N-diacetyl-L-Lys-D-Ala-D-Ala cultures. Reversion of antibacterial activity by addition of this receptor analog indicated the presence of a glycopeptide. The test proved to be very specific: 42 glycopeptides were isolated, 6 of which were novel ones including kibdelin, parvodicin, and actinoidin A*. A colorimetric assay (SPERA) was proposed (CORTIet al., 1985) based on the competition of horseradish peroxidase bound teicoplanin and the putative glycopeptide for a D-Ala-D-Ala peptide attached to the test tube. Glycopeptide A82846 was discovered
by testing fermentation broths against vancomycin in a polyclonal antibody assay (ELISA) (YAO et al., 1988).
6 Fermentation
Culture media and fermentation conditions reported for production of glycopeptide antibiotics either at the laboratory level or in industrial manufacturing do not substantially differ from those described for other antibiotics. As usual, precultures (vegetative cultures) are prepared in one or more stages using rich media composed of soluble proteins and rapidly utilizable carbon sources. Fermentation conditions must be adjusted individually for the different producing strains. In general, optimal temperatures are between 28 and 30C, and the optimal pH is around 7.0. All producing organisms are strictly aerobic and a consistent oxygen supply must be provided by adequate aeration and agitation systems.
6.1 Fermentation Media - Carbon Sources

Carbon sources frequently used in fermentation media include glucose, glycerol, starch, or dextrin. The use of oleate or other lipids is less common. Antibiotic production is often repressed by the presence of rapidly utilizable carbon sources in the production medium. A few studies of this effect have been reported for glycopeptide fermentation. There is evidence that antibiotic production by Amycolatopsis species is subjected to this repression, since higher yields of ristomycin (ristocetin) are obtained when galactose or glycerol are used as carbon sources in Amycolatopsis lurida fermentations (TOROPOVA al., 1982). Similaret ly the production of demethyl vancomycin is increased when galactose or dextrin substitutes for glucose in the fermentation medium (BOECKet al., 1984); dextrin or starch are the preferred carbon sources in the fermentations of vancomycin and of the related antibiotic
382
izupeptin (MERTZand DOOLIN,1973; SPIRINAKAGAWA al., 1986). All these antibiotet ics are produced by Amycolatopsis strains. A marked negative effect of glucose is observed in the production of the antibiotic complex A41030 by cultures of Streptomyces virginiae (BOECKet al., 1985). In contrast, Actinoplanes species appear to be less affected by the nature of carbon sources. Only minor differences in actaplanin yields are observed when dextrin or other slowly metabolized carbon sources substitute for glucose in Actinoplanes missouriensis fermentations. A medium composed of glucose and yeast extract appears suitable for teicoplanin production by cultures of A . teichomyceticus. A substantial increase in yields of ardacins and kibdelins, complexes produced by Kibdelosporangium strains, is observed up on addition of oleic acid (as methyl ester) to the et fermentation medium (SHEARER al., 1985). Oleic acid is a good source of acetyl CoA one of the starting materials of these antibiotic biosyntheses. Higher availability of acetyl CoA was therefore proposed to explain the oleate effect on yields. However, since ardacins and kibdelins are lipoglycopeptides oleic acid possibly is a direct precursor of their fatty acid chain as it was shown for teicoplanin, a lipoglycopeptide (see Sect. 8.2).
6.3 Fermentation Media - Effect of Phosphates

Phosphate concentration is an important parameter in antibiotic fermentation processes as phosphates are necessary for optimal growth, but high concentrations inhibit production. In vancomycin fermentations antibiotic production is depressed by phosphate concentrations above 0.1 g.L-' (MERTZ and DOOLIN, 1973). Even lower concentrations are sufficient to inhibit production of demethy1 vancomycin or ristocetin (BOECKet al., 1984; TOROPOVA al., 1974). The fact that et these antibiotics belong to different structural types but are all produced by Amycolatopsis species suggests that phosphate control is characteristic for this genus. Production of ardacins by Kibdelosporangium aridum and of A47934 by S. toyocaensis is also inhibited by relatively low concentrations of phosphates (BOECK and MERTZ, 1986). In contrast, production of the complex A41030 by S. virginiae was enhanced by increasing phosphate concentrations up to 1 g.L-' (BOECKet al., 1985).
6.4 Inhibition by the Final Product

Several examples of inhibition of antibiotic biosynthesis by the final fermentation product are reported in the literature. This phenomenon has also been observed in some glycopeptide fermentations. Ristocetin production, e.g., is inhibited by low concentrations of et this antibiotic (TOROPOVA al., 1974). It was observed in the authors' laboratory that A40926 inhibits its own production when present in the growth phase of cultures. Well documented is the effect of actaplanin on its producing strain. A. missouriensis cultures start producing actaplanin about 40 h after inoculation at the end of mycelial growth. Depending on the fermentation medium 60-90% of the antibiotic produced is tightly bound to the mycelium. When mycelium was centrifuged and resuspended in saline no further growth was observed but pro-
6.2 Fermentation Media Nitrogen Sources

Soybean meal or similar soy products are almost universally used as nitrogen providing ingredients in production media; yeast extract and corn steep liquor are also used frequently. Ammonium salts are never found among the suitable nitrogen sources. Although specific studies are lacking it can be assumed that - similarly to other antibiotics - glycopeptide production is repressed by high concentrations of ammonium ions. Nitrates can be utilized; at least in the case of S. virginiae fermentations addition of sodium nitrate to a soybean and corn steep containing medium increased production of A41030 by 80% (BOECKet al., 1985).
7 Recovery and Purification
383
duction continued normally. When mycelium was centrifuged, washed to completely eliminate soluble actaplanin, and resuspended, growth was resumed and no production was observed. Thus, even low concentrations of actaplanin in the medium apparently inhibit growth and also have a killing effect as demonstrated by viable colony counting in a parallel experiment. Surprisingly the cell bound antibiotic that can be as much as 50 mg per gram of cells has no effect on either growth or production (HUBERet al., 1987).
7 Recovery and Purification

Although glycopeptides are relatively water-soluble, often a large proportion of the product is found in the harvest broth bound to mycelium. Solubilization and release can normally be obtained by adjusting the pH to an appropriate value (according to the isoelectric point of the product acidic or alkaline conditions are required). Alternatively extraction with acetone or methanol has been used. Recovery from the filtered broth may include adjustment of pH and use of water immiscible organic solvents (butanol for teicoplanin), or ion-pair extraction (avoparcin). Other isolation schemes described use ion exchange matrices (Dowex, Amberlite IR9), acidic alumina, cross-linked polymeric adsorbents (Diaion HP, Amberlite XAD), cation exchange dextran gel (Sephadex) and polyamides in various sequences (SITRINand FOLENA-WASSERMAN, 1994). Reverse-phase chromatography with semipreparative and preparative columns, packed with silanized silica gel, allowed purification and separation of glycopeptides and of the single factors of complexes on the basis of differential hydrophobicity. Isocratic and gradient elutions were carried out. Examples are given by SITRIN and FOLENA-WASSERMAN (1989) and in almost every publication listed in Tab. 1-4. The elucidation of the mechanism of action at a molecular level allowed the development of affinity chromatography both as a specific discovery tool and as a powerful method of isolation and purification of glycopeptides from fermentation broths (CORTI and CASSANI, 1985; FOLENA-WASSERMAN al., et 1987). The affinity adsorbent is prepared by immobilizing D-Ala-D-Ala on commercially available activated supports. The affinity constant for the sepharose-D-Ala-D-Ala resin, as determined by equilibrium binding experiments, is 8.08.105 Lemol-' for teicoplanin, 1.57.105 L.mo1-l for vancomycin and 3.89.105 Lamol-' for ristocetin A (CORTI and CASSANI, 1985). An aqueous solution of the glycopeptide is contacted with the adsorbent in batch mode
6.5 Control of Complex Composition

An antibiotic composed of several factors has to be produced during fermentation in constant relative amounts to ensure uniformity of the finished product. Alternatively, the relative amount of the major component can be increased to obtain as a final product a single substance rather than a complex. The factors composing the lipoglycopeptide antibiotics teicoplanin and A40926 differ in the structure of their acyl chains, in which the branched type, either is0 or anteiso, predominates. In biosynthesis of branched fatty acids the initiator molecule determines the type of the final product: isobutyric acid gives rise to is0 chains with an even number of carbon atoms, 2-methylbutyric acid and isovaleric acid give rise to chains with an odd number of carbon atoms of the anteiso and the is0 type, respectively. Addition of one of these precursors to the fermentation broth can selectively increase the amount of the fatty acid that it is initiating and thus of the corresponding component of the complex. It has been shown that in this way the complex composition of both teicoplanin and A40926 can be substantially altered (BORGHI al., 1991a; SELVA et et al., 1992). In practice, better results could be achieved by adding their natural precursors, i.e., valine for isobutyric acid, isoleucine for 2-methylbutyric acid, and leucine for isovaleric acid, to the cultures instead of the acids mentioned above.
384
or loaded on columns. After washing to remove impurities the glycopeptides are eluted with a small volume of an aqueous buffer solution and an organic solvent (acetonitrile, methanol, ethylene glycol) to disrupt the interaction. By varying the pH of the buffer, mixtures of different glycopeptides or components of a complex can be resolved, with the additional advantage of obtaining a concentrated solution. The final pure material can be obtained after desalting by lyophilization.
8 Biosynthesis
Glycopeptides are complex molecules the biosynthesis of which must include several steps: (1) synthesis of the uncommon amino acids constituting the heptapeptide, (2) polymerization of the amino acids, (3) formation of the ether and carbon-carbon bonds between the phenylic groups, (4) synthesis of the unusual sugars and, if present, of fatty acid chains, (5) glycosylation and other final modifications. The sequence of these steps may, to some extent, differ from that indicated; chlorination of the phenyl rings, e.g., may occur before or after the amino acid assembly and glycosylation may precede the oxidative ring linking.
Experimental studies to elucidate the biosynthetic pathway are scanty. In fact, although it appears to be most probable that assembly is performed through the well known thiotemplate system, no experimental evidence is available on this most important aspect. The origin of the unusualsugars and the reactions leading to the formation of the triphenyl ether and diphenyl groups have never been studied. This lack of information is, in part, due to the fact that several laboratories failed to isolate blocked mutants accumulating biosynthetic intermediates. In addition, the most intensely studied antibiotics, i.e., vancomycin, teicoplanin and ristocetin, are produced by Acfinomyces strains for which a system of transformation with exogenous DNA is not available. This has severely hampered studies based on molecular genetic methods.
8.1 Origin of the Uncommon Amino Acids

A few uncommon amino acids are found as building blocks of almost all glycopeptide antibiotics: fl-hydroxytyrosine, 3-chloro-p-hydroxytyrosine, p-hydroxyphenylglycine and m-dihydroxyphenylglycine (Fig. 7). Analysis of 13C NMR spectra of avoparcin produced by fermentations of Sfrepfomyces candidus which had been supplied with 2-13C-
HO
&o
HO
0
$
OH
0
OH
o &
$ o
/
HO
OH
OH
OH
Fig. 7. Amino acids composing teicoplanin heptapeptide.
8 Biosynthesis
385
tyrosine demonstrated that tyrosine is the precursor of the p-hydroxytyrosine and p-hydroxyphenylglycine residues of the peptide (MCGAHREN al., 1980). These results were et confirmed and extended by studies on vancomycin. Experiments with 13C and 3H labeled tyrosine showed that both L- and D-tyrosine are precursors of the two 3-chloro-phydroxytyrosine units, although these differ in their configuration at C-2. Moreover, it was found that phydroxylation occurs with retention of configuration at C-3 (HAMMOND al., 1982). et Similar results were obtained by adding D,L2- 13C-tyrosineto ristocetin fermentations. As expected, the label was incorporated into C-2 of the phydroxytyrosine and C-1 of the p-hydroxyphenylglycine residues (HAMMOND et al., 1983). Although the origin of p-hydroxyphenylglycine from tyrosine is clearly demonstrated, the sequence of reactions and the intermediates of this conversion can be deduced only from indirect evidence. p-Hydroxyphenylglyoxylic acid appears to be the direct precursor of p-hydroxyphenylglycine, since demethy1 vancomycin and A47934 yields are increased by addition of either of these compounds to A. orientalis or S. toyocaensis cultures, respectively (BOECK et al., 1984; BOECKand MERTZ,1986). In K. aridurn fermentations, addition of cold p-hydroxyphe-
nylglyoxylic acid or p-hydroxy mandelic acid depresses the incorporation of radioactivity from L-U14C-tyrosine into ardacin indicating dilution of labeled intermediates (CHUNG et al., 1986a). It is noteworthy that p-hydroxyphenyl acetic acid has no effect, providing evidence that this compound is not an intermediate in the conversion. These results and the proof that the producing organisms are able to convert tyrosine into p-hydroxytyrosine are consistent with the following reaction sequence (Fig. 8): Tyrosine + p-hydroxytyrosine pp-dihydroxyphenylpyruvic acid p-hydroxymandelicacid p-hydroxyphenylglyoxylicacid + p-hydrox yphenylglycine. The origin of m-dihydroxyphenylglycine units has been demonstrated by experiments in which 1,2-13C-acetatewas added either to vancomycin or ristocetin fermentations (HAMMOND al., 1982). The carbons of the et m-dihydroxyphenylglycine residues resulted specifically labeled, indicating that this amino acid backbone originates from cyclization of a polyketomethylene chain. A tetraketide chain was originally suggested as the immediate precursor. However, this requires a ring closure involving the initial methyl group of the chain, a feature never observed in biosynthesis of polyketide antibiotics. It was, therefore,
+ + +
0
HOJ
COOH
OH tyrosine
OH
OH
HoY
Fig. 8. Presumptive pathway of p-hydroxyphenyl glycine biosynthesis from tyrosine.
OH
Q-
6
OH
COOH
COOH
-H2$
OH
p-hydroxyphenylglycine
386
3.5-dihydroxyphenylacetic add
Fig. 9. Hypothetical polyketide chains that can give rise to 3,5-dihydroxyphenyl acetic acid and to 3,S-dihydroxyphenyl glycine.
proposed that a longer chain is formed first, conclusion is based on the following experipart of which could be degraded after cycliza- mental evidence: et tion (HAMMOND al., 1983). An alternative - Production of teicoplanin factor T-A2-1 by A. teichomyceticus characterized by a 4-dehypothesis can be considered assuming that cenoyl moiety is entirely dependent on the malonate rather than acetate is the initiator presence of linoleic acid in the fermentamolecule of the polymerization process. The resulting product, 6-carboxy-3,5-dihydroxy- tion medium. phenylacetic acid, could be easily converted - The relative amount of factor T-A2-3 prointo m-dihydroxyphenylacetic acid by decarduced that is characterized by a linear decaboxylation (Fig. 9). noyl chain is substantially increased by addition of oleic acid to the medium. - Factors T-A2-2, T-A2-4, and T-A2-5 bear 8.2 Origin of Fatty Acids of branched acyl chains, namely S-methylnonanoic acid (iso-C10: 0), 8-methyldecanoic Lipoglycopeptides acid (anteiso-C11:0 ) and 9-methyldecanoic All glycopeptides listed in Tab. 4 are comacid (iso-C11 :O). Analysis of fatty acid conplexes of factors characterized by the presstituents of cell lipids revealed the presence ence of different acyl chains linked, as amof three major components, 14-methylpenides, to amino sugars. The origin of the fatty tadecanoic acid (iso-C16:0), 14-methylacids constituting the acyl moieties of the teihexadecanoic acid (anteiso-C17:0 ) and 13methyltetradecanoic acid (iso-Cl5 :0). coplanin components (see Fig. 5) has been extensively investigated (BORGHI al., 1991a). et These appear to be the logical precursors of the iso-C10:0, anteiso-Cll:O, and isoAltogether the results obtained indicate that C11 : O moieties of T-A2-2, T-A2-4, and Tthese chains are not synthesized de novo but derived from degradation of long-chain fatty A2-5, respectively, assuming the loss of acetate units by the common p-oxidation acid components of cell lipids or from those present in the fermentation medium. This mechanism of fatty acid degradation.
9 Chemical Modifications
387
- A mutant strain of A . teichomyceticus produces a novel teicoplanin factor characterized by a n-nonanoic moiety. Correspondingly, cell lipids contain heptadecenoic acid not present in parent strain cells. - Addition of 14C-acetate to the culture medium at the time of inoculation resulted in substantial labeling of fatty acid moieties. When 14C-acetatewas added to grown mycelium resuspended in saline, radioactivity of acyl chains was negligible in comparison to that of teicoplanin aglycone, demonstrating that the fatty acid moieties derive from molecules formed during the growth phase. A similar correspondence between acyl chains of antibiotic factors and cell lipid composition was found for A40926 and its proet ducer Actinomadura strain (ZERILLI al., 1992).
cule was never achieved; the aglycone was rapidly transformed into mannosyl aglycone but no further glycosylation step was observed (BORGHI al., 1991b). et Antibiotic A47934 produced by S. toyocaensis has a chemical structure similar to that of ardacin aglycone from which it differs in the presence of a sulfate ester on an aromatic ring. Extensive experiments with labeled substrates, blocked mutants, and biochemical inhibitors indicate that the sulfate is added prior to the formation of intermediates that possess antimicrobial activity. These results exclude that sulfate esterification of the aglycone that is antimicrobially active is the last reaction of the biosynthetic pathway (ZMIJEWSKI et al., 1987).
9 Chemical Modifications
8.3 Glycosylation and Final Modifications
The last steps of the biosynthetic pathway were studied in some detail in ardacin production by K. aridum. Sugar substituents in these antibiotic molecules are mannose and a N-acyl-2-aminoglucuronic unit. A related complex of antibiotics, kibdelins, differs from ardacins in the presence of N-acylglucosamine instead of acylaminoglucuronic acid. Kibdelins are readily converted into ardacins by cultures of K. aridum indicating that oxidation of C-6 of glucosamine could be the last biosynthetic step (CHUNG al., 1986a). et Less clear are the results of experiments aimed at defining the glycosylation sequence. When 14C labeled ardacin aglycone or mannosy1 aglycone were added to K. aridum cultures, labeled ardacins were produced (CHUNG al., 1986a). However, since transet formation required a long incubation period and only a small fraction of the added radioactivity was incorporated, the possibility of a degradation of the molecule and recycling of the fragments cannot be ruled out. Similar attempts were carried out by adding teicoplanin aglycone to A. teichomyceticus cultures. Conversion into the final moleEarly investigations on glycopeptide chemistry were carried out mainly for structural studies. Hydrolysis reactions yielded the aglycones and pseudoaglycones, some of which exhibited a higher antimicrobial activity than the parent antibiotic (PHILIPet al., 1960). Later, knowledge of the mechanism of action on a molecular level, elucidation of the structure of several compounds, and establishment of some correlations between pharmacokinetic and physicochemical properties (PITKIN et al., 1986) allowed a more rational design of chemical derivatives. In particular, the efficacy of teicoplanin in experimental infections pointed out the importance of lipophilic side chains and opened the way to the oriented synthesis of a large number of derivatives both of teicoplanin and other glycopeptides. Products with improved antimicrobial activity or better pharmacokinetics have been reported, but none of them is in clinical use so far. Structure-activity relationships of natural and semisynthetic glycopeptides have been reviewed by NAGARAJAN (1994b). The ideal glycopeptide has been outlined in a recent article by FELMINGHAM (1993). An extensive review of chemical modifications is out of the scope of this chapter, therefore only a condensed summary is given.
388
Reaction conditions are those conventional molecule with physicochemical properties in peptide chemistry including suitable pro- similar to those of teicoplanin, and the peptection of active functions not involved in the tidic carboxyl group was condensed with a reaction. However, their application some- number of amines. Among the derivatives obtimes requires specific conditions since differ- tained the -NH(CH2)3N(CH3)2 amide (MDL ent glycopeptides treated with the same rea- 63,246) exhibits a higher antimicrobial activigent under relatively similar conditions yield ty than the parent compound on several padifferent results depending on the nature of thogens, including vancomycin and teicoplanamino acids 1 and 3, on steric factors, and on in resistant strains (MALABARBAet al., 1993b) . the presence of sugars or chlorine atoms. A common modification of glycopeptides is It was suggested that a protonated amino the selective hydrolysis of sugars leading to group plays an important role in the binding the preparation of aglycones and pseudoagly- process through an electrostatic interaction cones. Reports are available, e.g., on avopar- with the carboxyl group of the target peptide et cin (MCGAHRENet al., 1983), teicoplanin (WILLIAMSON al., 1984). The terminal ami(MALABARBAet al., 1984), parvodicin no group o several glycopeptides such as teif (CHRISTENSEN al., 1987), ardacin (SITRIN coplanin, vancomycin, ristocetin, and eremoet et al., 1986), and eremomycin (KOBRIN al., mycin has been variously alkylated, and acyet 1988). The hydrolyses were performed mainly lated. In vancomycin the higher basicity of for investigational purposes but also provided the vancosamine amino group compared to compounds of interest for their intrinsic activ- that of the terminal N-methylleucine allowed ity or as substrates for further chemical trans- a selective acylation of the former (KANNAN et al., 1988; NAGARAJAN al., 1988). et formations. To understand the role of the terminal Since some components of certain glycopeptide complexes differ only in the number carboxylate binding pocket in glycopepor position of C1 atoms, partial or complete tide-substrate interactions binding constants removal of chlorine converts one glycopep- of glycopeptide fragments (produced by contide into another, e.g., A82846B into orienti- trolled degradations) were determined with et and cin A (NAGARAJAN al., 1989a). Selective peptide models. NAGARAJAN SCHABEL dechlorination of vancomycin (HARRIS al., (1988) prepared the vancomycin hexapepet 1985) and teicoplanin (MALABARBA al., tide by removal of the terminal N-methylet 1989b) and its effect on the affinity of the an- leucine, and MALABARBA and CIABATTI tibiotics for model peptides and on their anti- (1992) reported a tetrapeptide fragment from bacterial activity have been reported. Iodov- teicoplanin. ancomycin (HARRISet al., 1986) and 12I-laOn the other hand, relevant peptide fragbeled ristocetin (KIM et al., 1989) were also ments were obtained by stereospecific syntheses with a view of total synthesis of vancodescribed. The terminal carboxyl group which may be mycin (RAMARAO et al., 1992) and teicofree as in teicoplanin or methylated as in ris- planin (CHAKRABORTY VENKATREDand tocetin has been submitted to esterification, DY, 1992). Both these fragments and those reduction to alcohol, or amidation. Teico- prepared by controlled degradation may be planin, its pseudoaglycones, and its aglycone used as building blocks for the synthesis of have been amidated with various amines and non-natural glycopeptides. alkyldiamines (MALABARBA al., 1989c), aet The various approaches to the synthesis of amino acids (MALABARBA al., 1989a), or vancomycin and ristocetin aglycones have et been reviewed by EVANS and DEVRIES polyamines (MALABARBA al., 1992). et A40926, an antibiotic characterized like ar- (1994). dacins by an N-acylglucuronic moiety shows a good antibacterial activity on several pathogens, including N. gonorrhoeue (GOLDSTEIN et al., 1987). The glucuronic carboxyl group was reduced to hydroxymethyl to obtain a
10 Biotransformations
389
10 Biotransformations
10.1 Deglycosylation
Selective hydrolysis of certain sugar substituents that could not be performed chemically was carried out enzymatically. Using naranginase chloropolysporin B and a-or pavoparcin were converted to the corresponding derhamnosyl derivatives with a 40% and 80% yield, respectively. Although the commercial enzyme used contained pglucosidase the glucose moiety was not hydrolyzed, probably because of steric hindrance. a-Mannosidase converted chloropolysporin B and C to the corresponding demannosyl pseudoaglycones with a 71-50% yield. With this enzyme it was also possible to obtain the demannosyl derivative of p-avoparcin which was found identical with e-avoparcin (TAKATSU et al., 1987b). Enzymatic deglycosylation clarified the relationship between galacardin A and p-avoparcin. Treatment of galacardin A with a-galactosidase selectively removed the galactose units linked by a-glycosidic bonds. After incubation either the galactose at the hydroxyl of amino acid 3 or the galactose linked to the L-rhamnose unit at amino acid 1 were removed leaving the sugar units linked by pglycosidic bonds unaffected (TAKEUCHI et al., 1992). The acyl glucosamine moiety of teicoplanin is selectively hydrolyzed by mild acid treatment. More acidic conditions are required to remove the other sugar substituents of the molecule. Therefore, a selective, chemical hydrolyzation of the mannose moiety was not possible without removing the acyl glucosamine unit. The single components of the teicoplanin complex were converted into the corresponding demannosyl derivatives by treatment with cultures of N. orientalis NRRL 2450 (the producer of vancomycin-type M43 complex) or S. cundidus NRRL 3218 (the producer of actinoidin-type avoparcin). The time course of the transformation with washed mycelium of N. orientalis showed a 40% conversion in 72 h at a concentration of 0.25 g+L-' (BORGHIet al., 1991b).
10.2 Glycosylation
Cultures of A. teichomyceticus, the producer of teicoplanin, and of K. aridum, the producer of ardacins, easily 'transform the aglycones of their antibiotics into the mannosyl derivatives (CHUNGet al., 1986a; BORGHI et al., 1991b). Transformation is almost quantitative after 24 h of incubation. Mannosylation is independent of antibiotic biosynthesis, since with K. aridum it occurs also in the presence of glyphosate, a biosynthesis inhibitor, and with A. teichomyceticus it is performed by washed mycelium resuspended in saline. The reaction is not very specific with respect to the substrate: pseudoaglycones are also efficiently transformed by both microorganisms. Protoplast preparations of K. aridum were tested for their ability to convert ardacin aglycone into the complete antibiotic. In this case the prevalent transformation compound was also mannosyl aglycone, accompanied by smaller amounts of poorly defined derivatives carrying other neutral sugars (CHUNG al., et 1986b).
10.3 Deacylation
At present, microbial deacylation seems to be the only method to remove the fatty acid residues from the N-acylamino sugars of lipoglycopeptides. The main component of A40926 complex is characterized by the presence of a 10-methylundecanoyl chain that acylates the amino group of the glucuronic moiety. A few hundred microorganisms were tested for their ability to selectively hydrolyze the amide bond. Positive results were obtained with cultures of A. teichomyceticus ATCC 31121 (the producer of teicoplanin), A. missouriensis NRRL 15646 and NRRL 15647 (SELVAet al., 1988). Ardacins and the components of AAD 609 complex are also deacylated by A. teichomyceticus cultures (CHUNG al., 1986b). Teicoet planin is not deacylated by its producer which is somewhat surprising in view of the structure similarity with kibdelins.
390
10.4 Other Biotransformations

Actaplanin single components or its pseudoaglycone are bioconverted by pure cultures of A. missouriensis NRRL 15646 or A. missouriensis NRRL 15647 into the corresponding CUC/CSV glycopeptides in which the -CO-CH(NH2)- group on the terminal amino acid 1 is oxidatively deaminated into a -CO-CO- group (CLEM et al., 1985). Antibiotic CUCKSV is also produced by cofermentation of A. missouriensis NRRL 15646 and NRRL 15647. They are blocked mutants of A. missouriensis ATCC 31683 (producer of actaplanin components B3, El, and pseudoaglycone) which in turn derives from mutation of Actinoplunes ATCC 23342 (actaplanin producer). The two strains NRRL 15646 and NRRL 15647 do not produce actaplanin (HERSHBERGER, 1985). Bromoactaplanins carrying a Br atom instead of the C1 on the phenyl residue of amino acid 6 were obtained by supplying a spontaneous mutant of the producing strain with NaBr (HUBERet al., 1988). As discussed in Sect. 8.2 the acyl chains of the teicoplanin factors originate from degradation of long-chain fatty acids. This notion was exploited to biosynthesize new antibiotic derivatives. Addition of ricinoleic acid (12-hydroxyoleic acid) to cultures of A. teichomyceticus resulted in the formation of a novel teicoplanin component characterized by a 4-hydroxydecanoyl chain. Similarly, addition of linolenic acid (which has an 18 carbon linear chain with three double bonds) gives raise to a teicoplanin derivative with a 4,7-decadienoyl acyl chain (unpublished data from the authors laboratory). Selective removal of amino acid 1 of vancomycin (to obtain vancomycin hexapeptide) was carried out by biotransformation, as an alternative to the chemical method. Screening of soil microorganisms for their ability to grow on a model compound (D-leucyl-D-tyrosine) as the sole carbon source led to the isolation of a strain (Actinomuduru citreu) that converted vancomycin B into the hexapeptide in 90% yields (ZMIJEWSKI al., 1989). et
11 References
ANG,S-G., WILLIAMSON, P., WILLIAMS, H. M. D. (1988), Structure elucidation of a glycopeptide antibiotic, OA-7653,J. Chem. SOC. Perkin Trans. I, 1949-1956. ARDUINO, C., MURRAY, E. (1993), VancoR. B. mycin resistance in Gram-positive organisms, Curr. Op. Infect. Dis. 6, 715-724. ARJUNA RAO, V., RAVISHANKAR, SADHUKD., HAN, A. K., AHMED, M., GOEL, K., PRABS. A. HU, N. S., VERMA, A. K., VENKATESWARLU, A., ALLAUDEEN, S., HEDDE, D., NISBET, H. R. L. J. (1986), Synmonicins: a novel antibiotic complex produced by Synnemomyces mamnoorii gen. et sp. nov. I. Taxonomy of the producing organism, fermentation and biological properties, Abstract No. 939, 26rd Interscience Conference on Antimicrobial Agents and Chemotheraphy, New Orleans: Am. SOC.Microbiol. ARTHUR, COURVALIN,(1993), Genetics and M., P. mechanisms of glycopeptide resistance in enterococci, Antimicrob. Agents Chemother. 37, 1563-1571. ATHALYE, M., COATES, J., MILNER,P. H. N. (1989), Eur. Patent Appl. 339982. J. D. BARNA, C. J., WILLIAMS, H. (1984), The structure and mode of action of glycopeptide antibiotics of the vancomycin group, Ann. Rev. Microbiol. 38,339-357. J. D. D. BARNA, C. J., WILLIAMS, H., STONE, J. M., LEUNG,T.-W. C., DODDRELL, M. (1984), D. Structure elucidation of the teicoplanin antibiotics, J. Am. Chem. SOC. 106,48954902. BERDNIKOVA, F., LOMAKINA, N., POTAT. N. POVA, N. P. (1982), Structure of actinoidins A and B, Antibiotiki 27,252-258. L. BOECK, D., MERTZ,F. P. (1986), A47934, a novel glycopeptide-aglycone antibiotic produced by a strain of Streptomyces toyocaensis. Taxonomy and fermentation studies, J. Antibiot. 39, 15331540. BOECK, D., MERTZ, P., WOLTER, K., HIGL. F. R. GENS, C. E. (1984), N-demethylvancomycin, a novel antibiotic produced by a strain of Nocardia orientalis. Taxonomy and fermentation, J. Antibiot. 37,446-453. F. G. BOECK, D., MERTZ, P., CLEM, M. (1985), L. A41030, a complex of novel glycopeptide antibiotics produced by a strain of Streptomyces virginiae. Taxonomy and fermentation studies, J. Antibiot. 38, 1-8. BORGHI, ANTONINI, ZANOL, FERRARI, A., P., M., P., ZERILLI, F., LANCINI, C. (1989), IsolaL. G. tion and structure determination of two new analogs of teicoplanin, a glycopeptide antibiotic, J. Antibiot. 42,361-366.
11 References
391
VENELLA, A. J., HEALD, L., KILLMER, B., s. L. BORGHI, EDWARDS, ZERILLI, F., LANA., D., L. G. L., CINI, C. (1991a), Factors affecting the normal MICO,B. A., MUELLER, PAN,C. H., POEHand branched-chain acyl moieties of teicoplanin LAND, B. L., RAKE,J. B., ROBERTS, G. D., components produced by Actinoplanes teichoSHEARER, C., SITRIN, D., NISBET,L. J., M. R. myceticus, J. Gen. Microbiol. 137, 587-592. JEFFS,P. W. (1987), Parvodicin, a novel glycopeptide from a new species, Actinomadura parBORGHI, FERRARI, GALLO,G. G., ZANOL, A., P., vosata: discovery, taxonomy, activity and strucL. G. M., ZERILLI, F., LANCINI, G. (1991b), Mi0 ture elucidation, J. Antibiot. 4 ,970-990. crobial demannosylation and mannosylation of teicoplanin derivatives, J. Antibiot. 44, 1444- CHUNG, K., TAYLOR, OH, Y. K., DEBROSSE, S. P., C., JEFFS,P. W. (1986a), Biosynthetic studies on 1451. aridicin antibiotics. I. Labeling patterns and Box, S. J., ELSON, L., GILPIN, L., WINSTANA. M. overall pathways, J. Antibiot. 39,642-651. LEY (1990), MM 47761 and MM 49721, glycoS. P., peptide antibiotics produced by a new strain of CHUNG, K., OH, Y. K., TAYLOR, GERBER, R., NISBET,L. J. (1986b), Biosynthetic studies Amycolatopsis orientalis. Isolation, purification on aridicin antibiotics. 11. Microbial transformaand structure determination, J. Antibiot. 43,931tions and glycosylations by protoplasts, J. Anti937. biot. 39,652-659. Box, S. J., COATES, J., DAVIS, J., GILPIN, N. C. M. M. P. L., HOUGE-FRYDRYCH,V. S., MILNER, H. CLEM,G. M., BOECK,L. D., ANDERSON, T., C. MICHEL,F. H. (1985), Eur. Patent Appl. (1991), MM-55266 and MM-55268, glycopeptide 142285. antibiotics produced by a new strain of Amycolatopsis: isolation, purification and structure de- COATES, J., ELSON,A. L., ATHALYE, CURN. M., termination, J. Antibiot. 44,807-813. TIS,L. M., MOORES, V. (1990a), Eur. Patent L. Appl. 375213. N. BRAZHNIKOVA, G., LOMAKINA, N., BERDM. T. NIKOVA, F. (1989), Eremomycin: a new glyco- COATES,N. J., SYKES,R., CHRISTOPHER, J., D. peptide antibiotic, in: Bioactive Metabolites from CURTIS, L. M. (1990b), Eur. Patent Appl. M. U., Microorganisms (BUSHELL, E., GRAEFE, 375448. Eds.), Vol. 27, pp. 163-165. Amsterdam: Else- COATES, J., DAVIS,C. J., CURTIS, M., SIKES, N. L. vier. R. (1991), WO Patent 91/16346. BROGDEN, N., PETERS, H. (1994), Teicoplan- CORONELLI, GALLO,G. G., CAVALLERI, R. D. C., B. in. A reappraisal of its antimicrobial activity, (1987), Teicoplanin: chemical, physico-chemical pharmacokinetic properties and therapeutic effiand biological aspects, I1 Farmaco, Ed. Scient. cacy, Drugs 47,823-854. 42,767-786. CAMPOLI-RICHARDS, M., BROGDEN, N., CORTI, CASSANI, (1985), Synthesis and charD. R. A., G. FAULDS, (1990), Teicoplanin. A review of its D. acterization of D-alanyl-D-alanine-agarose: a antibacterial activity, pharmacokinetic propernew bioselective adsorbent for affinity chroma0 ties and therapeutic potential, Drugs 4 , 449tography of glycopeptide antibiotics, Appl. Bio486. chem. Biotechnol. 11,101-109. CASSANI, (1989), Glycopeptides: antibiotic dis- CORTI, RURALI, BORGHI, CASSANI, G. A., L., A., G. covery and mechanism of action, in: Bioactive (1985), Solid-phase enzyme-receptor assay M. Metabolites from Microorganisms (BUSHELL, (SPERA): A competitive binding assay for glyU., E., GRAEFE, Eds.), pp. 221-235. Amsterdam: copeptide antibiotics of the vancomycin class, Elsevier. Clin. Chem. 31, 1606-1610. CASSETTA, BINGEN, LAMBERT-ZECHOVSA., E., DEBONO, M., MOLLOY, M., BARNHART, R. M., KY, N. (1991), La vancomycine en 1991: actualid DORMAN, E. (1980), A35512, a complex of D. et perspectives, Path. Biol. 39,700-708. new antibacterial antibiotics produced by StrepCAVALLERI, PARENTI, (1992), GlycopepB., F. tomyces candidus. 11. Chemical studies on A35512B, J. Antibiot. 33,1407-1416. tides (dalbaheptides), in: Kirk-Othmer Encyclopedia of Chemical Technology, 4le, Vol. 2, pp. DEBONO,M., MERKEL, E., MOLLOY, M., K. R. 995-1018. New York J. Wiley and Sons. BARNHART, PRESTI, HUNT,A. H., HAMM., E., ILL,R. L. (1984), Actaplanin, new glycopeptide CHAKRABORTY, K., VENKAT REDDY, G. T. antibiotics produced by Actinoplanes missou(1992), Studies directed toward the synthesis of riensis. The isolation and preliminary chemical glycopeptide antibiotic teicoplanin: first synthecharacterization of actaplanin, J. Antibiot. 37, sis of the N-terminal 14-membered ring, J. Org. 85-95. Chem. 57,5462-5469. J. R. CHRISTENSEN, B., ALLAUDEEN, S., BURKE, DINGERDISSEN,J., SITRIN, D., DEPHILLIPS, S. H. P. A., GIOVENELLA, J., GRAPPEL,S. F., A. S. S. M. R., CARR, A., CHUNG, K., DEPHILLIPS, MEHTA, J., OH, Y. K., PAN,C. H., ROBERTS, R. P., DINGERDISSEN,J., DIPAOLO,M., GIOJ.
392
On the biosynthesis of the antibiotic vancomyG. D., SHEARER, C., NISBET, J. (1987), M. L. cin, J. Chem. SOC., Chem. Commun. 1982, 344Actinoidin AZ,a novel glycopeptide: production, 346. preparative HPLC separation and characteriza0 tion, J. Antibiot. 4 , 165-172. S. D. R. HAMMOND, J., WILLIAMS, H., NIELSEN, V. (1983), The biosynthesis of ristocetin, J. JR., DOOLIN, E., GALE,R. M., GODFREY 0. W., L. Chem. SOC., Chem. Commun. 1983,116-117. HAMILL, L., MAHONEY, F., YAO,R. C-F. R. D. (1989), Eur. Patent Appl. 299 707. HARRIS,C. M., KOPECKA, HARRIS,T. M. H., K. EVANS, E., DEVRIES, M. (1994), Approaches D. (1983), Vancomycin: Structure and transformato the synthesis of the vancomycin aglycones, in: tion to CDP-I, J. Am. Chem. SOC. 105, 6915Glycopeptide Antibiotics (NAGARAJAN, R., 6922. Ed.), pp. 63-104. New York: Marcel Dekker. HARRIS, M., KANNAN, KOPECKA, HARC. R., H., FELMINGHAM, (1993), Towards the ideal glycoD. RIS,T. M. (1985), The role of chlorine substipeptide, J. Antimicrob. Chemother. 32,663-666. tuents in the antibiotic vancomycin: preparation FOLENA-WASSERMAN, POEHLAND,B. L., G., and characterization of mono- and didechloroYEUNG, W-K., STAIGER, KILLMER, B., E. D., L. vancomycin, J. Am. Chem. SOC. 107,6652-6658. SNADER, DINGERDISSEN, JEWS,P. W. HARRIS,C. M., FESIK,S. W., THOMAS, M., K., J. J., A. (1986), Kibdelins (AAD-609), novel glycopepT. KANNAN, HARRIS, M. (1986), Iodination R., tide antibiotics. 11. Isolation, purification and of vancomycin, ristocetin A, and ristocetin pseustructure, J. Antibiot. 39, 1395-1406. doaglycon, J. Org. Chem. 51,1509-1513. R., F., HEALD,S . L., MUELLER, JEFFS,P. W. (1987), FOLENA-WASSERMAN, G., SITRIN, CHAPIN, L., K. SNADER, (1987), Affinity chromatography of Actinoidins A and AP:Structure determination glycopeptide antibiotics, J. Chromatogr. 392, 0 using 2D NMR methods, J. Antibiot. 4 , 630225-238. 645. FRANCO, M. M., CHA-ITERJEE, VIJAKUMAR, HERSHBERGER, L. (1985), Eur. Patent Appl. C. S., C. E. K. S., CHA-ITERJEE, K., GANGULI, N., D. B. 159436. RUPP,R. H., FEHLHABER, H-W., KOGLER, H., HIGGINS, M., JR., MCCORMICK, H., MERH. M. SELBERT, TEETZ, V. (1990), Eur. Parent G., K. KEL,K. E., MICHEL, H. (1985), Eur. Patent Appl. 356894. Appl. 159180. GAUSE, F., BRAZHNIKOVA,G., LOMAKINA, HOLDEN, S., RUDDOCK, C., TONE,J., MAEG. M. K. J. N. N., BERDNIKOVA, F., FEDOROVA, B., T. G. DA, H. (1991), Brit. Patent 2243610. TOKAREVA, L., BORISOVA, N., BA-ITA, N. V. G. K. J. Y (1989), Eremomycin - New glycopeptide anti- HOLDOM, S., MAEDA,H., RUDDOCK, C., TONE,J. (1988), Eur. Patent Appl. 265 143. biotic: chemical properties and structure, J. AntiHUBER, M., PIEPER, L., TIETZ,A. J. (1987), F. R. biot. 42, 1790-1799. Characterization of the process of the biosyntheGOLDSTEIN, P., SELVA,E., GASTALDO, B. L., sis of the actaplanin complex by Actinoplanes BERTI, PALLANZA, RIPAMONTI, FERM., R., F., missouriensis, J. Ferment. Technol. 65,85-89. RARI, P., DENARO, ARIOLI, CASSANI, M., V., G. F. K. (1987), A40926, a new glycopeptide antibiotic HUBER, M., MICHEL, L., HUNT,A. H., MARTIN,J. W., MOLLOY, M. (1988), Preparation R. with anti-Neisseria activity, Antimicrob. Agents and characterization of some bromine analogs of Chemother. 31,1961-1966. the glycopeptide antibiotic actaplanin, J. AntiB. GOLDSTEIN, P., ROSINA, R., PARENTI,F. biot. 41,798-801. (1994), Teicoplanin, in: Glycopeptide Antibiotics K. (NAGARAJAN, Ed.), pp. 273-307. New York: HUNT, A. H., ELZEY,T. K., MERKEL, E., DER., BONO, M. (1984a), Structures of actaplanins, J. Marcel Dekker. Org. Chem. 49,641-645. GOOD,V. M., GWYNN, N., KNOWLES, J. C. M. D. G. (1990), MM 45289, a potent glycopeptide anti- HUNT, A. H., MARCONI, G., ELZEY,T. K., M. HOEHN, M. (1984b), A51568A: N-demethylbiotic which interacts weakly with diacetyl-L-lyvancomycin, J. Antibiot. 37, 917-919. syl-D-alanyl-D-alanine,J. Antibiot. 43,550-555. D. M., HAMILL,R. L., YAO, R. C. (1993), U.S. Patent HUNT,A. H., DORMAN, E., DEBONO, MOLLOY, R. M. (1985), Structure of antibiotic 5187082. A41030A, J. Org. Chem. 50,2031-2035. M. W. HAMILL, L., HANEY, E., JR., STARK, M. R. JOHNSON, P., U-ITLEY, H. C., WOODFORD, A. A. (1973), Ger. Patent Appl. 2252937. R. HAMILL, L., MAHONEY, F., NAKASUKASA, N., GEORGE, C. (1990), Resistance to vancoR. D. mycin and teicoplanin: an emerging clinical W. M., YAO, R. C-F. (1988), Eur. Patent Appl. problem, Clin. Microbiol. Rev. 3,280-291. 265 701. T., T., M. HAMMOND, J., WILLIAMSON, P., WILLIAMS, KAMOGASHIRA, NISHIDA, SUGAWARA, S. M. (1983), A new glycopeptide antibiotic, OA-7653, D. H., BOECK, D., MARCONI, G. (1982), L. G.
I1 References
393
produced by Streptomyces hygroscopicus ssp. hi- MALABARBA, SPREAFICO, FERRARI, A., F., P., wasaensis, Agric. Biol. Chem. 47, 499-506. J., P., G., KETTENRING, STRAZZOLINI, TARZIA, PALLANZA, BERTI, M., CAVALLERI, R., B. KANNAN, R., HARRIS,C. M., HARRIS,T. M., (1989b), Dechloro teicoplanin antibiotics, J. AnJ. N. D. WALTHO, P., SKELTON, J., WILLIAMS, tibiot. 42, 1684-1697. H. (1988), Function of the amino sugar and Nterminal amino acid of the antibiotic vancomycin MALABARBA, TRANI,A., STRAZZOLINI, A., P., in its complexation with cell wall peptides, J. CIETTO, FERRARI, TARZIA, PALLANG., P., G., Am. Chem. SOC.110,2946-2953. ZA,R., BERTI, (1989c), Synthesis and biologM. ical properties of N63-carboxamidesof teicoplanKATRUKHA, S., SILAEV,A. B. (1986), The G. in antibiotics. Structure-activity relationships, J. chemistry of glycopeptide antibiotics of the vanMed. Chem. 32,2450-2460. comycin group, in: Chemistry of Peptides and W., E., Proteins (VOELTER, BAKER, OVCHINNI- MALABARBA, CIABATTI, KETTENRING, A., R., J., KOV,Y. A., IVANOV, T., Eds.), Vol. 3., pp. V. SCOTTI, R., CANDIANI, PALLANZA, G., R., 289-306. Berlin: de Gruyter. BERTI,M., GOLDSTEIN, P. (1992), Synthesis B. and antibacterial activity of a series of basic amKIM,K-H., MARTIN, OTIS,E., MAO,J. (1989), Y., ides of teicoplanin and deglucoteicoplanin with Inhibition of '"I-labeled ristocetin binding to polyamines, J. Med. Chem. 35,4054-4060. Micrococcus luteus cells by the peptides related A., R., B. to bacterial cell wall mucopeptide precursors: MALABARBA, CIABATTI, CAVALLERI, quantitative structure-activity relationships, J. (1993a), Semisynthetic dalbaheptides: chemistry Med. Chem. 32,84-93. and biological activity, Curr. Topics Med. Chem. 1,359-376. KOBRIN, B., FEDOROVA, KATRUKHA, M. J. B., G. S. (1988), Study on deglycosylation of certain an- MALABARBA,A., CIABATTI, PANZONE, R., G., tibiotics belonging to vancomycin group, AntiMARAZZI, (1993b), WO Patent 93103060. A. biotiki i Chemother. 33,331-335. MATSUMOTO, (1961), A vancomycin-related anK. KUNSTMANN, P., PORTER, N. (1974), US.PaM. J. tibiotic from Streptomyces sp. K-288. Studies on Streptomyces antibiotics, XXXVIII, J. Antibiot., tent 3 803 306. Ser. A 14,141-146. LANCINI, C. (1989), Fermentation and biosynG. W. J. G. thesis of glycopeptide antibiotics, in: Bioactive MCGAHREN, J., MARTIN, H., MORTON, M. O., HARGREAVES, T., LEESE,R. A., LovR. Metabolites from Microorganisms (BUSHELL, ELL, F. M., ELLESTAD, A., O'BRIEN, E., G. E., GRAEFE, Eds.), pp. 283-296. Amsterdam: U., Elsevier. HOLKER, S. E. (1980), Structure of avoparcin J. components, J. Am. Chem. SOC. 102,1671-1674. LANCINI, C., CAVALLERI, (1990), GlycopepG. B. W. R. F., tide antibiotics of the vancomycin group, in: Bio- MCGAHREN, J., LEESE, A., BARBATSCHI, N. G. MORTON, O., KUCK, A., ELLESTAD, A. G. chemistry of Peptide Antibiotics, Recent Ad(1983), Components and degradation comvances in the Biotechnology of PLactams and pounds of the avoparcin complex, J. Antibiot. 36, H., Microbial Bioactive Peptides (KLEINKAUF, 1671-1682. VON DOHREN, Eds.), pp. 159-178. Berlin: de H., Gruyter. MERTZ,F. P., DOOLIN, E. (1973), The effect of L. LECHEVALIER, P., PRAUSER, LABEDA, M. H., D. . inorganic phosphate on the biosynthesis of vancomycin, Can. J. Microbiol. 19,263-270. P., RUAN,J.-S. (1986), Two new genera of nocardioform actinomycetes: Amycolata gen. nov. MICHEL,K. H., SHAH, R. M., HAMILL, L. R. and Amycolatopsis gen. nov., Int. J. Syst. Bacteri(1980), A35512, a complex of new antibacterial 01. 36,29-37. antibiotics produced by Streptomyces candidus. I. Isolation and characterization, J. Antibiot. 33, MALABARBA, CIABATTI, (1992), Int. Patent A., R. 1397-1406. WO 92/10517. MALABARBA, STRAZZOLINI, DEPAOLI, A., P., A., MICHEL,K. J., YAO, R. C-F. (1991), Eur. Patent Appl. 424051. LANDI,M., BERTI,M., CAVALLERI, (1984), B. Teicoplanin, antibiotics from Actinoplanes tei- NADKARNI, R., PATEL,M. V., CHATTERJEE, S. S., chomyceticus nov. sp. VI. Chemical degradation: VIJAYAKUMAR, K. S., DESIKAN,K. R., E. physico-chemical and biological properties of J., B. M. BLUMBACH, GANGULI, N., LIMBERT, acid hydrolysis products, J. Antibiot. 37, 988(1994), Balhimycin, a new glycopeptide antibiot999. ic produced by Amycolatopsis sp. Y-86-21022. Taxonomy, production, isolation and biological MALABARBA, FERRARI, CIETTO,G., PALA., P., activity, J. Antibiot. 47, 334-341. LANZA, R., BERTI, M. (1989a), Synthesis and R. biological activity of N63-carboxypeptidesof tei- NAGARAJAN, (1993), Structure-activity relacoplanin and teicoplanin aglycone, J. Antibiot. tionships of vancomycin-type glycopeptide anti42,1800-1816. biotics, J. Antibiot. 46, 1181-1195.
394
H. NAGARAJAN, (Ed.) (1994a), Glycopeptide Anti- PERKINS, R. (1982), Vancomycin and related R. antibiotics, Pharmacol. Ther. 16, 181-197. biotics. New York Marcel Dekker. H. NIETO,M. (1970), The preparation NAGARAJAN, (1994b), Structure-activity rela- PERKINS, R., R. of iodinated vancomycin and its distribution in tionships of vancomycin antibiotics, in: Glycobacteria treated with the antibiotic, Biochem. J. R., peptide Antibiotics (NAGARAJAN, Ed.), pp. 116,83-92. 195-218. New York: Marcel Dekker. J. J. A. NAGARAJAN, SCHABEL, A. (1988), Selective P ~ I L I P , E., SCHENCK, R., HARGIE,M. P., R., J. W. HOLPER, C., GRUNDY, E. (1960), The incleavage of vancosamine, glucose, and N-methyl creased activity of ristocetins A and B following leucine from vancomycin and related antibiotics, acid hydrolysis, Antimicrob. Agents Ann. 1960, J. Chem. SOC., Chem. Comrnun. 1988, 130610-16. 1307. G., C. NAGARAJAN, SCHABEL, A., OCCOLOWITZ, PHILLIPS, GOLLEDGE, L. (1992), VancomyR., A. cin and teicoplanin. Something old, something F. J. L., COUNTER, T., O n , J. L. (1988), Synthenew, Med. J. Aust. 156,53-57. sis and antibacterial activity of N-acyl vancomycins, J. Antibiot. 41, 1430-1438. PITKIN, H., MICO,B. A., SITRIN, D., NISBET, D. R. L. J. (1986), Charge and lipophilicity govern the NAGARAJAN, BERRY, M., HUNT,A. H., OcR., D. pharmacokinetics of glycopeptide antibiotics, COLOWITZ, L., SCHABEL, A. (1989a), ConJ. A. Antimicrob. Agents Chemother. 29,440-444. version of antibiotic A82846B to orienticin A and structural relationships of related antibiot- RAKE, B., GERBER, METHA, J., NEWMAN, J. R., R. ics, J. Org. Chem. 54,983-986. D. K., OH, Y. K., PHELEN, SHEARER, C., C., M. SITRIN, D., NISBET,L. J. (1986), GlycopepR. NAGARAJAN, BERRY, M., SCHABEL, A. R., D. A. tide antibiotics: a mechanism based screen em(1989b), The structural relationships of ploying a bacterial cell wall receptor mimetic, J. A82846B and its hydrolysis products with chloAntibiot. 39, 58-67. roorienticins A, B and C, J. Antibiot. 42, 1438RAMARAO, A. V., CHAKRABORTY,K., JOSHI, T. 1440. S. P. (1992), The first synthesis of C-terminal biNICAS, T. I., ALLEN, E. (1994), Resistance and N. phenyl moiety of vancomycin, Tetrahedron Lett. mode of action, in: Glycopeptide Antibiotics 33,4045-4048. (NAGARAJAN, Ed.), pp. 219-241. New York R., REYNOLDS, E. (1989), Structure, biochemistry P. Marcel Dekker. and mechanism of action of glycopeptide antiNIETO,M., PERKINS, R. (1971a), Modification H. biotics, Eur. J. Clin. Microbiol. Infect. Dis. 8, of the acyl-D-alanyl-D-alanine terminus affecting 943-950. complex-formation with vancomycin, Biochem. J. 1239789-803. RIVA,E., GASTALDO, BEREITA, M. G., FERL., L. G., RARI, ZERILLI, F., CASSANI, GOLDP., NIETO,M., PERKINS, R. (1971b), The specificity H. STEIN, P., BERTI,M., PARENTI, DENARO, B. F., of combination between ristocetins and peptides M. (1989), A42867, a novel glycopeptide antirelated to bacterial cell wall mucopeptide prebiotic, J. Antibiot. 42, 497-505. cursors, Biochem. J. l24,845-852. E., L., OKAZAKI, ENOKITA, MIYAOKA, TA- SELVA, GASTALDO, BEREITA,G., BORGHI, T., R., H., A., GOLDSTEIN, P., ARIOLI, CASSANI, B. V., G., KATSU, TORIKATA, (1987), ChloropolyT., A. PARENTI, (1986), Eur. Patent Appl. 177882. F. sporins A, B and C, novel glycopeptide antibiotE., G., A., ics from Faenia interjecta sp. nov. I. Taxonomy SELVA, BERETTA, BORGHI, DENARO, M. (1988), WO Patent Appl. 88102755. of producing organism, J. Antibiot. 40,917-923. E., L., G., PALLANZA, BERTI, GOLDSTEIN, P., MA- SELVA, GASTALDO, CASSANI, PARENTI, R., M., B. F. (1992), Eur. Patent Appl. 259781. PELLI, E., RANDISI, SCOITI, R., ARIOLI,V. E., (1983), Teichomycin: in vitro and in vivo evalua- SHEARER, C., ACTOR, P., BOWIE, A., GRAPM. B. tion in comparison with other antibiotics, J. AnPEL, S. F., NASH,C. H., NEWMAN, J., OH, Y. D. timicrob. Chemother. 1 , 419-425. 1 K., PAN,C. H., NISBET,L. J. (1985). Aridicins, novel glycopeptide antibiotics. I. Taxonomy, PARENTI, CAVALLERI, (1989), Proposal to F., B. production and biological activity, J. Antibiot. name the vancomycin-ristocetin like glycopep38,555-560. tides as dalbaheptides, J. Antibiot. 42, 18821883. SHEARER, C., GIOVENELLA, J., GRAPPEL, M. A. S. F., HEDDE, R. D., MEHTA,R. J., OH, Y. K., PARENTI, BERETTA, BERTI,M., ARIOLI, F., G., V. PAN,C. H., PITKIN, H., NISBET, J. (1986), D. L. (1978), Teichomycins, new antibiotics from ActiKibdelins, novel glycopeptide antibiotics. I. Disnoplanes teichomyceticus nov. sp. I. Description covery, production, and biological evaluation, J. of the producer strain, fermentation and biologiAntibiot. 39, 1386-1394. cal properties, J. Antibiot. 31, 276-283.
11 References
395
NOSHITA, (1987a), Chloropolysporins A, B T. SHELDRICK, M., G. JONES,P. G., KENNARD, O., and C, novel glycopeptide antibiotics from FaeWILLIAMS, H., SMITH, H. (1978), StrucD. G. nia interjecta sp. nov. 111. Structure elucidation ture of vancomycin and its complex with acetylof chloropolysporins,J. Antibiot. 40, 933-940. D-alanyl-D-alanine, Nature (London) 271, 223225. TAKATSU, TAKAHASHI, TAKAMATSU, T., S., Y., SITRIN, D., FOLENA-WASSERMAN,(1989), R. G. SHIOIRI, IWADO, HANEISHI, (1987b), T., S., T. Affinity and HPLC purification of glycopeptide Chloropolysporins A, B and C, novel glycopepantibiotics, in: Natural Products Isolation (WAGtide antibiotics from Faenia interjecta sp. nov. MAN,G. H., COOPER,R., Eds.), pp. 111-152. IV. Partially deglycosylated derivatives, J. AntiAmsterdam: Elsevier. biot. 40,941-945. SITRIN,R. D., FOLENA-WASSERMAN, (1994), TAKEUCHI, ENOKITA, OKAZAKI, KAG. M., R., T., Separation methodology, in: Glycopeptide AntiGASAKI, INUKAI, (1991a), Helvecardins A T., M. R., biotics (NAGARAJAN, Ed.), pp. 29-61. New and B, novel glycopeptide antibiotics. I. TaxonoYork: Marcel Dekker. my, fermentation, isolation and physico-chemiSITRIN, D., CHAN, W., DINGERDISSEN, R. G. J. J., cal properties, J. Antibiot. 44,263-270. J. HOLL,W., HOOVER, R. E., VALENTA, R., TAKEUCHI, J. M., TAKAHASHI, INUKAI,M. S., K. WEBB, L., SNADER, M. (1985), Aridicins, (1991b), Helvecardins A and B, novel glycopepnovel glycopeptide antibiotics. 11. Isolation and tide antibiotics. 11. Structural elucidation, J. Ancharacterization, J. Antibiot. 38,561-571. tibiot. 44, 271-277. SITRIN, D., CHAN, W., CHAPIN, GIOVENR. G. F., TAKEUCHI, TAKAHASHI, ENOKITA, SAM., S., R., ELLA, A. J., GRAPPEL, F., JEFFS, W., PHILS. P. KAIDA, HARUYAMA, NAKAMURA, Y., H., T., K. NISBET,L. J. (1986), LIPS,L., SNADER, M., M. KATAYAMA, INUKAI, (1992), Galacardins T., Aridicins, novel glycopeptide antibiotics. 111. A and B, new glycopeptide antibiotics, J. AntiPreparation, characterization, and biological acbiot. 45, 297-305. tivity of aglycone derivatives, J. Antibiot. 39,68TAMURA, TAKEDA, (1975), Antibiotic ABA., I. 75. 65, a new antibiotic from Saccharornonospora SKELTON, J., WILLIAMS, H. (1990), Structure N. D. viride, J. Antibiot. 28,395-397. elucidation of the novel glycopeptide antibiotic TOROPOVA, G., EGOROV, S., OBRAZTSOVA, E. N. UK-68597, J. Org. Chem. 55,3718-3723. A. Y. (1974), Effect of ristomycin and excess of SKELTON, J., WILLIAMS, H., RANCE,M.J., N. D. phosphorus in the medium on activity of enRUDDOCK, C. (1990), Structure elucidation of J. zymes of Proactinomyces fructiferi var. ristomyUK-72051, a novel member of the vancomycin cini, Antibiotiki 19, 418-422. group of antibiotics, J. Chem. SOC. Perkin Trans. 1 77-81. , TOROPOVA, G., EGOROV, S., TKHAKER, E. N. V., SKELTON, J., WILLIAMS, H., RANCE, J., N. D. M. BURAKAEVA, D. (1982), Effect of various A. sources of carbon and nitrogen on biosynthesis RUDDOCK, C. (1991), Structure elucidation of J. a novel antibiotic of the vancomycin group. The of ristomycin, protease and pigment by Nocardia influence of ion-dipole interactions on peptide fructiferi var. ristomycini, Antibiotiki 27, 749backbone conformation, J. Am. Chem. SOC. 113, 753. 3757-3765. TSUJI, KAMIGAUCHI, KOBAYASHI, TEN., T., M., SOMMA, GASTALDO, CORTI,A. (1984), TeiS., L., RUI,Y. (1988a), New glycopeptide antibiotics. coplanin, a new antibiotic from Actinoplanes tei11. The isolation and structures of chloroorientichomyceticus nov. sp., Antimicrob. Agents Checins, J. Antibiot. 41, 1506-1510. mother. 26,917-923. TSUJI, KOBAYASHI, KAMIGAUCHI, YoN., M., T., SPIRI-NAKAGAWA, FUKUSHI, MAEBASHI, SHIMURA, Y., TERUI, (1988b), New glycopepP., Y., Y. N., Y., K., IMAMURA, TAKAHASHI, TANAKA, tide antibiotics. I. The structures of orienticins,J. Y., TANAKA, OMURA, (1986), Izupeptins H., S. Antibiot. 41, 819-822. A and B, new glycopeptide antibiotics produced WALLAS, H., STROMINGER,L. (1963), RistoC. J. by an actinomycete, J. Antibiot. 39, 1719-1723. cetins, inhibitors of cell wall synthesis in StaphySZTARICSKAI,BOGNAR, (1984), The chemisF., R. lococcus aureus, J. Biol. Chem. 238,2264-2266. try of the vancomycin group of antibiotics, in: WILLIAMSON, P., WILLIAMS, H., HAMM. D. Recent Developments in the Chemistry of Natural S. R., Carbon Compounds (BOGNAR, SZANTAY, MOND, J. (1984), Interactions of vancomycin and ristocetin with peptides as a model for proCs.,Eds.), Vol. X, pp. 91-201. Budapest: Akatein binding, Tetrahedron 40, 569-577. demiai Kiad6. D. D. TAKATSU, TAKAHASHI, NAKAJIMA, HA- YAO, R. C., MAHONEY, F., BAISDEN, K., T., S., M., W. MERTZ,F. P., MABE,J. A., NAKATSUKASA, NEISHI, NAKAMURA, KUWANO, KIT., T., H.,
396
M. (1988), A82846, a new glycopeptide antibiotmoieties of the antibiotic complex A40926 and ic complex, produced by Arnycolatopsis orientaltheir relation with the membrane lipids of the producer strain, Rapid Cornrnun. Mass. Specis. I. Discovery, taxonomy and fermentation trorn. 6, 109-114. studies, Abstract No. 974,28th Interscience ConB., R., ference on Antimicrobial Agents and Chemother- ZMIJEWSKI, J., BRIGGS, LOGAN, BOECK, M. Microbiol. apy, Los Angeles: Am. SOC. L. D. (1987), Biosynthetic studies on antibiotic R. ZECKEL,M. L., WOODWORTH, (1994), VancoA47934, Antirnicrob. Agents Chernother. 31, 1497-1501. mycin, a clinical overview, in: Glycopeptide Antibiotics (NAGARAJAN, Ed.), pp. 309409. ZMIJEWSKI, J., LOGAN,R. M., MARCONI, R., M. G., New York: Marcel Dekker. R. F., DEBONO, MOLLOY, M., CHADWELL, M., ZERILLI,L. F., EDWARDS, M., BORGHI,A., D. BRIGGS, (1989), Biotransformation of vancoB. E., GALLO,G. G., SELVA, DENARO,M., LANmycin B to vancomycin hexapeptide by a soil miCINI, G. C. (1992), Determination of the acyl croorganism, J. Nut. Prod. 52,203-206.
Biotechnology
10 Aminoglycosides and Sugar Components in Other Secondary Metabolites
WOLFGANGPIEPERSBERG, JURGEN DISTLER

Wuppertal, Germany
1 Introduction 399 2 Isolation, Distribution, Ecology, and Fermentation 399 3 Biogenesis and Basic Pathways of Cyclitols and Sugar Components 402 3.1 Cyclitols 403 3.2 Sugar Components 405 3.2.1 6-Deoxy- and Other Deoxyhexoses 407 3.2.2 Other Sugar Components 408 4 Genetics and Biochemistry of the Biosynthesis and Functions of Aminoglycosides 409 4.1 Streptomycins and Related Ca Aminoglycosides 415 4.1.1 Streptomycins 416 4.1.1.1 Biosynthesis of Streptidine and Bluensidine 417 4.1.1.2 The L-Dihydrostreptose Pathway 423 4.1.1.3 Hexosamine Pathway 423 4.1.1.4 Condensation of Subunits, Processing, and Export 425 4.1.2 Streptomycin-Related Ca Aminoglycosides 427 4.2 Fortimicins, Istamycins 429 4.3 2-Deoxystreptamine-ContainingAminoglycosides 432 4.4 Other Aminoglycosides 436 4.5 Resistance in Aminoglycoside Producers 440 4.6 Regulation in Streptomycetes 443 4.7 Overview of the Aminoglycoside Pathways 448 4.8 Aminoglycoside-Target Site Interactions and General Effects on Bacterial and Eukaryotic Cells 449
398
5 Related Sugar Components in Other Secondary Metabolites 451 5.1 Lincosamides 451 5.2 Cyclitols 454 5.3 6-Deoxyhexoses 455 5.4 Other Pentose, Hexose, and Heptose Derivatives 456 6 Evolutionary Aspects 457 7 Involvement of Primary Metabolism in the Delivery of Carbohydrate Precursors 459 8 Pathway Engineering and Other Types of Application in Biotechnology 459 9 Conclusions and Perspectives 462 10 Appendix: Chemical Structures 462 11 References 476
2 Isolation, Distribution, Ecology, and Fermentation
399
1 Introduction
There is no clear chemical or biochemical definition of the term aminoglycoside which is traditionally reserved for mono- to oligosaccharidic sugar and/or cyclitol derivatives containing amino nitrogen. Therefore, an attempt is made in this chapter to demonstrate that in biological and especially genetic terms there are many parallels between origins and uses of strongly derived sugar components in secondary metabolites in general and with respect to the aminocyclitol aminoglycosides in particular. A list of the major chemicalgroups of aminoglycosides (see Tab. 1) shows that some new structures have been described in the past decade. Since the 1st Edition of Biotechnology (cf. UMEZAWA al., 1986) the field of amiet noglycoside research has undergone a significant change characterized by the following divergent aspects:
(1) ceasing interest in the search for and development of new antibiotics from this particular group of compounds accompanied by stagnant market figures and clinical use; (2) increasing interest in basic research on the (molecular) biological fundamentals of the production of aminoglycosides and other secondary metabolic carbohydrates, especially initiated by the new methods of genetic engineering in actinomycetes, bacilli, and other groups of biotechnological interest; (3) evaluation of new classes of related natural products for other types of applications, e.g., inhibitors of glycosidases and other enzymes. However, this situation might change again due to the currently increasing problems with antibiotic resistance and the appearance of new and reappearence of old infectious deseases (see discussion in DAVIES, 1992; HOTTA et al., 1995). Examples are methicillin-resistant staphylococci, bacterial infections in HIV patients, and the worldwide renaissance of tuberculosis as a major health concern. In line with this development the authors of this
chapter focus mainly on new aspects of the molecular genetics, biochemistry, and physiology of production of aminoglycosides and the similarly modified sugar components of other chemical classes of secondary metabolites. In addition, future biotechnological applications of this knowledge are discussed. Many details of the chemistry, mode of action, classical strain improvement, fermentation technology, and pharmacology, e.g., of 2deoxystreptamine-containing, streptomycinlike and other related aminocyclitol-aminoglycosides are well known from many reviews published in the past two decades (DAVIES and YAGISAWA,1983; GRAFE, 1992; LoRIAN, 1980; MALLAMS, 1988; UMEZAWA and HOOVER 1982; UMEZAWA al., 1986; WALet LACE et al., 1979) and will not be extensively dealt with here. For the progress made in related fields, such as the chemical and enzymic synthesis of carbohydrate components and their analogs, e.g., cyclitols, hexose and pentose derivatives, sugar mimics, and their biological activity and application, the reader is referred to the following publications: KENNEDY,1988; HORTON et al., 1989; THIEM, 1990 LUKACS OHNO,1990 YAMAGUCHI and and KAKINUMA, 1992a; OGURAet al., 1992; BILLINGTON, 1993; HUDLICKY and CEBULAK, 1993; COLEMANand FRASER,1993; LOOKet al., 1993; KROHN al., 1993; TESTA et et al., 1993; COOKet a1 1994.
2 Isolation, D stribution, Ecology, and Fermentation

In the last decade new metabolites which clearly belong to the aminoglycoside group were detected and described. Tab. 1 gives an overview of the chronology of publications and the biochemical grouping of the most important aminoglycosides. Most sugar-containing secondary metabolites relevant to this chapter are formed by bacteria, mainly actinomycetes, bacilli, and pseudomonads (Fig. 1). However, the bioactive glycosides
400
Tab. 1 Chronology of the Detection and Biochemical Classification of the Major Aminoglycosides" . Aminoglycoside Streptomycin (Al) Neomycins (All) 5 ' -Hydroxystreptomycin(Al) Paromomycins (catenulin) (All) Kanamycins (A12) Trehalosamines (A19) Hygromycin B (A16) Streptozotocin (AM) Spectinomycin (A2) Bluensomycin (glebomycin) (Al) Gentamicins (A14) Kasugamycin (A4) Destomycins (A16) Nojirimycin (AN) Apramycin (A15) Tobramycin (A12) Sisomicin (A14) Ribostamycins (A13) Validamycins (A21) Lividomycins (A11) Butirosin (A13) Acarbose (A22) Myomycin (AS) SS-56-C (A16) Minosaminomycin (A5) Amylostatins (A22) Siastatin (A18) Verdamicin (A14) Sorbistins (A24) Seldomycins (A17) LL-BM123a (A6) Fortimicins (A10) 1-Desoxynojirimycins (AM) Adiposins (A22) Trestatins (A22) Istamycins (A10) Sporaricins (A10) Sannamycin (A10) Dactimicin (A10) Oligostatins (A22) Lysinomicin (A10) Spenilomycin (A3) Valiolamine (A21) 5 '-Hydroxy-N-demethyldihydrostreptomycin (Al) 1-N-Amidino-1-N-demethyl-2-hydroxydestomycin (A16) Neotrehalosadiamine (A19) AC4437 (Al) Allosamidin (A25) CV-1 (A18) Boholmycin (A7) Ashimycins (Al) Trehazolin (A23)
a
Year of First Description 1944 1949 1950 1952 1957 1957 1958 1960 1961 1962 1964 1965 1965 1966 1968 1968 1970 1970 1970 1971 1971 1972 1973 1973 1974 1974 1974 1975 1976 1977 1977 1977 1978 1978 1978 1979 1979 1979 1979 1979 1981 1982 1984 1985 1985 1986 1986 1987 1987 1988 1989 1991
Pathway Formula' Ca(4)-6DOH(2)-HA HA-(4)Cb(S)-P(3)-HA Ca(4)-6DOH(2)-HA HA-(4)Cb(S)-P(3)-HA HA-(4)Cb(6)-HA HA( 1)-H Cb(S)-H(2,3)-HepA HA Ca(4,5)-6DOH Ca(4)-6DOH(2)-HA HA-(4)Cb(6)-PA Ca(4)-6DOH Cb(S)-H(2,3)-HepA HA Cb(4)-OctA(d)-HA HA-(4)Cb(6)-HA HA-(4)Cb(6)-PA HA-(4)Cb(S)-P Cb(l)-Cb(4)-H HA-(4)Cb(S)-P(3)-HA HA-(4)Cb(5)-P Cb(l)-6DOH(I)-(H), Ca(4)-HA Ca(S)-H(2,3)-HepA Ca(4)-6DOH
HA HA-(4)Cb(6)-PA Ca(2)-HA HA-(4)Cb(6)-PA Ca(4)-H(4)-HA Ca(6)-HA HA (H)n-(4)cb(l)-H(l)-(H)n (H)n-(4)Cb(l)-6DOH(l)-(H)n Cb(6)-HA Cb(6)-HA Cb(6)-HA Ca(6)-HA (H)n-(4)Cb(1)-6DOH(1)-(H)n Ca(6)-HA Ca(4,5)-6DOH Cb Ca(4)-6DOH(2)-HA Ca(S)-H(2,3)-HepA HA(1)-HA Ca(4)-6DOH Ca(4)-HA(4)-HA HA HA-(4)Ca(6)-PA(4)-Hep Ca(4)-6DOH(2)-HA(4)-(H), Ca(l,2)-HA
()-4C()sO()() Hn()bl-DHl-H"
For references before 1986 see UMEZAWA HOOPER(1982) and UMEZAWA al. (1986), others are and et referenced in the text. Numbers in brackets refer to the figures showing the formulae in Sect. 10. ' See text in Sect. 3; only the basic formula is given if a group of compounds is characterized.
2 Isolation, Distribution, Ecology, and Fermentation
401
their ecological role in the biotopes of the producing organisms. There, these substances might be involved in the communication bez al=Iz---ou mcm+cp l a ws~-ilr~ 51 tween different organisms and in the hormmYuzaYuI>I mone-like crosstalk between the differenStrept omyces tiated cells of the producer itself (CHADWICK Strepto verticillium and WHELAN, 1992; PIEPERSBERG, 1993). Streptosporangium Also, the routes of dissemination of aminoMicromonospora glycoside resistance mechanisms found in Dactylosporangium both producers and clinically relevant nonAmycolatopsis producing bacteria (cf. Sect. 4.5) and the drivActinoplanes Saccharopolyspora ing selective pressures under natural condiCorynebacterium tions might become accessible to direct exBacillus perimental research. Glycosidic or cyclitol-containing compofig. 1 Aminocyclitol aminoglycoside-producing nents similar to the so-called secondary . bacterial genera (extended according to HASEGA- metabolites of microbes and plants occur in WA, 1991). the structures of many pro- and eukaryotic A C acarbose and related glycosidase inhibitors; extracellular polymers, such as polysaccharBM: bluensomycin; BU: butirosins; DM: destomycins; FM: fortimicins (astromicins); GM: gentami- ides, lipopolysaccharides, glycolipids, and cins; HM-A,B: hygromycins A or B; IM: istamycins; other glycoconjugates. So far, in gram-negaK G kasugamycins; KM: kanamycins; LM: livido- tive unicellular bacteria, only very little evimycins; NM: neomycins; PM: paromomycins; RM: dence has been found for the biosynthesis ribostamycins; SE: seldomycins; SI: sisomicins; SM: and secretion of aminoglycosides or other streptomycins; SP: spectinomycins; TM: tobramy- carbohydrate-containing diffusible substances cin; VM: validamycins. (e.g., pseudomonads produce sorbistins). However, these organisms produce a variable and abundant range of carbohydrate-based from plants should also be considered. In extracellular substances, lipopolysaccharides general, it may be assumed that these glyco- (LPS), and other heteropolymeric polysacsides share a common general biochemical charides (e.g., enterobacterial common antibasis with those of bacteria and that an equi- gen: ECA). There is also emerging evidence valent gene pool is used for their cellular pro- that the same gene pool is used for the production. However, since little is known about duction of these polymers, which in many intheir biosynthesis, this chapter concentrates stances resemble polymerized secondary metabolites, and for the production of the low on prokaryotic production systems. Research covering all aspects of the biology molecular weight end products of, e.g., actiof secondary carbohydrate metabolism in or- nomycete secondary metabolism itself (see ganisms producing secondary extracellular below). Two recent examples of non-eubacterial products is still in its initial stages. In contrast, the resistance mechanisms of antibiotic pro- aminoglycoside-like compounds are (1) ducers for self-protection can be regarded as the glucosaminyl archaetidyl-myo-inositols having been extensively investigated (see (Fig. 2) produced by some methanogenic ArSect. 4.5). Even the basis of physiology in nat- chaea (KOGAet al., 1993) and (2) the very ural biotopes and the ecological role of ami- similar core structure of the glycosyl noglycosides and other secondary metabolites phosphatidylinositol protein anchors for outare unknown (PIEPERSBERG, 1993; MARSH er-surface-attached glycoproteins in eukaryotes (Fig. 3). These are found in a variety of and WELLINGTON, 1994). Aminoglycosides have been shown to be organisms, from trypanosomes to mammals 1993). Interestingly, they have produced in soils (WELLINGTON al., 1993). (ENGLUND, et Therefore, one of the future aims of amino- the same content of nonacetylated glucosamglycoside aminocyclitol research is to study ine and share the 1-phosphoryl-6-glucosami-
ZJ
G I
%z:
& I
402
used for different purposes in the metabolism of mainly extracellularly targeted compounds and that they occur in practically all groups of organisms.
NH*
3 Biogenesis and Basic Pathways of Cyclitols and Sugar Components
Fig. 2. An aminoglycoside-like glycolipid structure in Archaea. R archaetidyl residues ( = glyceryl ethers).
It is unknown what is the general purpose of sugar modification and incorporation into homogenous (sugar-based) or mixed-type (glycoconjugate-like) oligomers - mostly nyl myo-inositol core unit which is synthe- diffusible or membrane-associated substances sized in the mammalian system from phos- outside the cells - and polymers, which are phoinositides and UDP-N-acetylglucosamine mostly coating materials for cell surfaces or at the inner side of the cytoplasmic mem- proteins outside the cells. It might be necesbrane. Following deacetylation and acylation sary for cell-cell interactions which could be at the inositol moiety it is transported to the called communication, special purpose, outer surface where it is condensed to other or individualization metabolism and which glycosidic residues (mannosyl, and in some might be a common feature of all types of orand WHELAN,1992; cases galactosyl, N-acetyl-galactosaminyl, ganisms (CHADWICK 1992, 1993). At least there is phosphorylethanolamine). The phosphoryl- PIEPERSBERG, ethanolamine unit connects the short manno- good evidence that most of the sugar metabsyl chain to the C-terminal carboxyl group of olism beyond glucose, fructose, and some the protein via formation of a new amide basic pentoses and their incorporation into bond (reviewed by ENGLUND, 1993). These storage and cell wall materials is specific for examples might suggest that part of the bio- the cell type or strain, rather than for a spechemistry of aminoglycoside biosynthesis cies or higher taxa, and it seems to be based and, hence, part of the relevant gene pool are on the same common and highly variable and
0-7-0-R
OH
t?
(protein) Asp-EA-Ph8-Man-a(l->P)Man-a(l (glycosidic or other residues)
(glycosidic or other residues)
Fig. 3. Structure in glycosyl phophatidylinositols resembling to aminoglycosides. Such structures are protein anchors of proteins bound to the outer surface of eukaryotic cytoplasmic membranes. Asp: aspartyl residue in the protein chain; EA: ethanolamine; Man: mannosyl residues; P phosphate; R: diacylglycerol.
403
fluid gene pool. This is reflected by high variations in glycosylation reactions in all organisms and especially in differentiating organisms. This gives rise to the high antigenic variation in microbial and higher eukaryotic cells. Examples of this are the different blood group markers in man and the capsular materials, lipopolysaccharides, and various forms of excreted glycosylated products in bacteria, which are described in this chapter.
amine (Fig. 6; WIDLANSKI al., 1989; GODA et and AKHTAR,1992; RINEHART al., 1992; et YAMAUCHI and KAKINUMA 1992c, 1993). This mechanism involves removal of electrons at C-4 of hexosephosphates (C-5 of heptulose derivatives); - elimination of the phosphoester hydroxyl as a leaving group would create a potential for the formation of a carbanion at C-6 (C-7); - finally, after the condensation reaction the keto group at C-4 (C-5) is reduced to restore the hydroxy group (cf. Fig. 4).
-
3.1 Cyclitols
According to our present knowledge there are two different (amino-)cyclitol pathways (Fig. 4).
This step is similar to that which occurs on a C7 sugar acid in the ring closure reaction during the initial biosynthetic pathway lead(I) The myo-inositol pathway. The first route ing to the aromatic amino acids. The main dif(designated Ca in this chapter) is via D-myo- ference in the intitial intermediates of the Ca inositol-3-phosphate which is synthesized by and Cb pathways is that in Cb a first transamithe NAD +-dependent myo-inositol synthase; nation step can follow immediately whereas end products are, e.g., D-chiro-inositol, strep- in Ca a phosphatase and an oxidase (dehytidine, bluensidine, actinamine, and D-myo-l- drogenase) reaction has to first create a suitadeoxy-l-amino-inositol (D-myo-inosamine). ble keto intermediate for a primary aminoA compilation of Ca cyclitols is given in Fig. 5 transfer (see below, Sect. 4.1.3). The further (FLOSS and BEALE, 1989 WALKER,1975a; processing of cyclitols to aminocyclitols might SIPOSand SZABO,1989). The reactions cata- even be at least in part very similar, i.e., the lyzed by the bacterial and the eukaryotic D- first aminotransfer and other later intermemyo-inositol-3-phosphate synthases follow diate steps, and might be catalyzed by related that of a typical intramolecular aldol conden- or even almost identical enzymes. This can be predicted although the enzymes involved are sation. This involves not yet known for most of representative - opening of the pyranose ring and rotation groups of (amino-)cyclitol pathways. Howevof the C-4/C-5 bond to place the C-6 group er, there is one detail in which also the later stages of Ca and Cb cyclitol conversion differ: in an active position, in 2-deoxystreptamine (2DOS) formation the - the reversible oxidation and reduction of incorporation of amino groups stereochemithe C-5 position by NAD+, - the removal of the pro-R-H atom from C-6 cally proceeds in a direction opposite to that shown to occur in streptidine biosynthesis and its transfer to the C-1 carbonyl, and 1975a; see Sects. 4.1 and 4.3). - the use of the p-anomer of ~-glucose-6- (WALKER, The chemical concept of aliphatic and arophosphate as the preferred substrate (WONG and SHERMAN, 1985; FLOSS and matic C7 units which basically constitute cyclohexane derivatives with a nitrogenous BEALE,1989). group in the meta position (m-C7N) domi(2) The dehydroquinate pathway. The second nates the literature on the biogenesis of buildroute to cyclitols (Cb) follows a NAD+-de- ing blocks in many secondary metabolites et pendent dehydroquinate-synthase-like mech- (RINEHART al., 1992; FLOSSand BEALE, anism on open chain hexose-6- or heptulose- 1989; CHIAOet al., 1995; KIM et al., 1996). 7-phosphates yielding nonphosphorylated 1- These are formally all cyclitol derivatives, keto-2-deoxy-cyclitols. End products are, e.g., most probably formed via the Cb pathways. 6-deoxystreptamine and valienamine/valiol- An aliphatic m-C7N unit is found in valien-
404
I0 Aminoglycosides and Sugar Components in Other Secondary Metabolites

Enz
Enz
T
H o
T B-H*
H \w
OH
HO OH H
OH
AL 0-Ph
(5)
gH)
HO
AD*]
[i
__c
rOH
OH
HO
OH
AL 0-Ph
b
0 CH2
OH
OH
aminehalidamine-containing substances, such The common dehydroquinatehhikimate as validoxylamines and validamycins (cf. Sect. pathway of many aromatic compounds should 10, Fig. A21), in acarbose, and in related gly- be regarded as a Cb pathway and is involved cosidase inhibitors (see Fig. A22; TRUSCHEITin the formation of many aromatic m-C7N et al., 1984; MULLER,1989). units (cf. Fig. 4. These units occur in many )
405
4 Fig. 4. Routes to the formation of (amino-)cyclitols other molecule derived from the same pathvia myo-inositol-~-3-phosphatesynthase (Ca) and way (e.g., validamycin, acarbose, and other a dehydroquinate synthase-like enzyme mechanism related compounds; see Sect. 4.4). In extreme (Cb; cf. reactions 8-7). Substrates can be D-glucose- cases, such as in pactamycin (see Fig. A30), ) 6-phosphate (1 and 3, sedoheptulose-7-phosphate (5, or its 5-epimer), or 3-deoxyarabinoheptulosonic they can constitute the central building block acid (7). The products of Ca or Cb pathways are which is condensed with four or more side either cyclitolphosphates (2) which have to undergo groups of different origin and varying comdephosphorylation and further oxidation before plexity. In pactamycin they consist of two arotransamination or 1-keto-2-deoxycyclitols (4,6, and mates (one derived from the dehydroquinate 8) which can be transaminated directly, respective- pathway, a m-C7N unit, the other from the ly. The numbering of ring atoms in (2) and (4) is polyacetate/polyketide pool), a dimethylated according to the counting system used in strept- urea group, a hydroxyethyl group derived amine derivatives (streptidine, actinamine, 2-deoxy- from the methyl groups of methionine, and a streptamine); numbers in (6) and (8) are given ac- C-bound methyl group. In addition, four of cording to the nomenclature in valienamine and dehydroquinate, respectively. 0 (C-1 or C-2) and A the five C atoms of the pentitol moiety which (C-6 or C-7) mark the original carbohydrate atoms seems to be a myo-inositol derivative are subforming the new C-C bond in cyclitol products. stituted with nitrogenous group. However, it is doubtful whether some of The dehydroquinate pathway (7-8) can also be started with a 5-amino-3,5-deoxyarabinoheptulo- the representatives of a common chemical sonic acid or an intermediate becomes transamin- group of aminoglycosides share the same cyated after cyclization to yield an aromatic m-C7N clitol pathway. For instance, the destomycin unit as is found in many secondary metabolites (for group contains two compounds with aminocydetail, see FLOSS BEALE,1989; RINEHART and et clitols derived from streptamine, one of which al., 1992; YAMAUCHI and KAKINUMA, 1992c, is identical to 1-N-amidinostreptamine (see 1993).
Fig. A 1 6 IKEDAet al., 1985b), an intermediate in the streptomycin pathway (Ca pathway; see Sect. 4.1.1). The other members of well-known secondary metabolites, such as this family have 2-deoxy-scyllo-streptamine many ansamycins, e.g., rifamycins, streptovar- (2DOS) as cyclitol moieties (Cb pathway). icins, geldanamycin (not shown), in candicid- Similarly, the fortimicins (Ca) and the istamyins (not shown), and in pactamycin (cf. Figs. 5 cins (Cb) could be put into separate groups and A30), which is unusual in that it contains (see Fig. A10 and Sect. 4.2). This might sugcomponents seemingly derived from the two gest even two different initial pathways for alternate cyclitol pathways, one following the cyclitol formation resulting in very similar Ca and the other an aromatic Cb route differ- end products or, alternatively, a subsequent ent from that of the ansamycin family (RINE- modification by oxidoreductases (dehydroxylHART et al., 1981, 1992; FLOSSand BEALE, ation of Ca derivatives or hydroxylation of 1989). A detailed account of the aromatic Cb Cb products) as optional routes to produce pathway is beyond the scope of this chapter; the alternate category of building blocks. the reader is referred to the literature However, the latter possibility is less likely (CHIAOet al., 1995; FLOSS,in press; KIM et since in no case were both pathways found in the same producer. al., 1996). The cyclitol moieties generally may be regarded as aglyca in glycosylation reactions, e.g., in streptomycins (see Sect. 4.1), fortimi- 3.2 Sugar Components cins, and istamycins (see Sect. 4.2), most 2The sugar derivatives which can be at.deoxystreptamine-(2DOS-) containing aminoglycosides (see Sect. 4.3), and some other tached to the cyclitols of both the Ca and the products such as the nucleoside antibiotic ad- Cb type, may stem either from nucleotidyl-acenomycin (cf. Fig. A27 and Sect. 5.2). Alter- tivated monomeric precursors (glycosyltransnatively, they are (co-)substrates in other fer reactions) or from the pool of polymertypes of condensation reactions, e.g., with an- ized D-glucose directly (e.g., maltodextrins;
406
Sy
HO OH 0O H 6H HO OH o O 6H OH HO bSO3H
O OH H OH HO
OH
rnyeinositol
HN=C-NHz
D-chifeinositol (kasugamycin)
(adenomycin)
streptamine (SS-56C)
H3c00 H2N-zo
streptidine (streptomycin) fortamine (fortimicin A) rnyeinosamine (rninosaminomycin)
HzN-C=O H-CH, OH H~
-: !:
HO OH
HO OH OH (LL-BM 12%)
OH
OH
actinamine (spectinomycin)
HO-$H
H3C
0; :HoooH " 0
CH20H
(myomycins)
) formed via a Ca pathway (see Fig. 4 . Some examples of their occurrence are given in brackets. Known or assumed positions derived from the original C-1 (0)and C-6 (A) of the glucose precursor are labeled in the pentitols of trehazolin and allosamidin no suggestion
Fig. 5. Structures of (amino-)hexito1 and pentitol components in secondary metabolites known or assumed to be
OH ACH~OH
OH
NH2
(pactamycin)
(trehazolin)
(allosamidin)
can be given without any direct experimental evidence.
transglycosylations). For a systematic analysis of the pathways involved we define here the following general routes by which carbohydrate building blocks are preformed:
(1) the 6-deoxyhexose pathway (6DOH); (2) the pyranosidic or furanosidic pentose (P), hexose (H), heptose (Hep), octose (Oct), or the hexosamine (HA) pathway; (3) some glucosyl residues ((H),) are possibly incorporated via transglycosylation from di- or oligosaccharides such as maltodextrins, especially where a-l,4-bound glucose or maltose moieties are encountered
(4) spacing non-carbohydrate-derived residues are unspecified (X). Thus, for the purposes of the discussion here and for simplification a "pathway formula" is defined for each carbohydrate-containing compound considered. The postulated pathway for the derivation of the mono- and oligosaccharidic structures is given by connecting the above abbreviations for the individual precursor pathways where the figures in brackets indicate the points of glycosidic (or other) substitution. For example, streptomycin (Ca(4)-6DOH(2)-HA) or mannosylmannosidostreptomycin (Ca(4)-6DOH(2)-
407
2-deoxy-scylb inosose
P-deoxystreptamine
(istamycinA: R1,2 NHz, H) = (istamycinB: R1,2 H, NHz) =
CH2OH
CHzOH
CHeOH
C;H~OH
1-ketovaliol
OH
valienamine
OH
validamine valiolamine
OH
Fig. 6 Structures of (amino-)hexi. tols and pentitols or benzoic acid derivatives in secondary metabolites known or assumed to be formed via a Cb pathway (see
occurrence are given of their Fig. 4). Some examplesin brackets. Known or assumed positions derived from the original C-1 or C-2 (0)and C-6 or C-7 (A) of the glucose or heptulose (heptulosonic acid) precursors, respectively, are labeled.
OH
2- hydroxyvalidamine
6H
(oligostatins)
HH A$ad o$ d A Qn
no
OH
~~~~~~~
O U. -.
I
dehydroquinate
OHA
OH
OH
OH
OH
COOH
COOH
(aristeromycin)
(neplanocinA)
3-aminod-hydroxy3-aminobenzoic acid benzoic acid
HA(4)-(H)J would be classified together with spectinomycin (Ca(4,5)-6DOH) and set apart from neomycins (HA-(4)Cb(6)-P(3)HA), validamycins (Cb(1)-Cb), and amylostatins ((H),,-(4)Cb(l)-6DOH(l)-(H),,) (see Sect. 10 and Tab. 1 for a compilation).
3.2.1 6-Deoxy- and Other Deoxyhexoses

D-glucose, a precursor of most or all 6deoxyhexoses (6DOH) in prokaryote-formed (antibiotic-like) secondary metabolites, seems to be generally activated by deoxythymidine diphosphate (dTDP). Catalysis of the first
two steps in their pathway is accomplished by the enzymes dTDP-D-glucose synthetase and dTDP-D-glucose 4,6-dehydratatase yielding 4keto-6-deoxyhexose intermediates (Fig. 7). The 4-keto compounds can be used as a common precursor for branching the further pathways into the D- and L-series of hexose derivatives (Figs. 8 and 9). The ~ - 6 D 0 H are isos merized from the D-configurated precursors by a 3,5-epimerase (see Fig. 7). In many cases both biosynthetic routes are followed in the same producing cell. Frequently, by such branching routes both D- and L-6DOH derivatives are formed and are specifically incorporated into a particular complex end product, e.g., macrolides and some angucyclins. In aminoglycosides, 6DOH components are re-
408

hexose-1 -P nucleotidyC transferase NDP-hexose 4,g-dehydratase
CHzOH HO Qo-P OH
CH*oH
HO QO-NDP
OH
O o O - N D p
D-GDOH
OH
NDP4ketoMeoxyhexose 3,S-epimerase
1
L-6DOH
OH-D Q ON P
Fig. 7. General pathway for the biosynthesis of 6-deoxyhexoses. Hexose-1-phosphate precursors are either
D-glucose-1-phosphate or D-mannose-1-phosphate. P: phosphate; NDP dTDP, CDP, or GDP.
latively rare compared with the dominance and variability among modifying sugar side chains in other chemical classes of microbial natural products. Important and interesting types of further modification are (1) the deoxygenations (formally: dehydroxylations) at C-2, C-3, and C-4, (2) the transaminations (formally: exchange of hydroxyl for amino groups) at C-2, C-3, and C-4, and (3) the isomerization and epimerization steps. Other types of modifications, such as C, N, 0, and S methylations or transfer reactions for more complex side groups are also common in the 6DOH pathways. Here we briefly describe only the first mechanism; the transaminations and other types of isomerization (e.g., epimerization) reactions are discussed below. A mechanism for the deoxygenation steps was elucidated by the recent studies of the CDP-3,6-deoxyhexose pathway (REEVES, 1993; SHNAITMAN and et KLENA,1993; THORSON al. 1993; THORSON and LIU, 1993a, b; LIU and THORSON, 1994). This is an alternative pathway yielding 6DOHs which is abundantly used in gram-negative bacteria besides the dTDP-pathway, e.g., in the biosynthesis of lipopolysaccharide 0-chains. The enzyme system involved consists of two iron-sulfur proteins one of which (El) catalyzes the pyridoxamine phosphate (PMP)-dependent dehydration of the 4-keto-
6-deoxy hexose via a radical mechanism and by forming a covalently bound PMP-hexose intermediate (Fig. 10). The electrons for this process are delivered via a second enzyme (E2) which contains FAD in addition to the [2Fe-2S] cluster and uses NADH as an electron donor. This type of mechanism could easily be envisaged also to be involved in the 2-deoxygenation of many other 6DOHs in secondary metabolites, such as the sugar constituents in the daunorubicin-cytorhodinrhodomycin group of anthracyclines or some of the 6DOHs occurring in chromomycins (cf. Figs. 8 and 9).
3.2.2 Other Sugar Components

As outlined above biogenesis studies in early phases of antibiotic research have often misleadingly suggested that D-glucose, D-glucosamine, glycerol, or related carbon sources are directly incorporated into sugar constituents in secondary metabolites. This was in most cases interpreted as an indication that the regular routes of sugar activation and further processing are adopted from' primary metabolic routes, e.g., the UDP-hexosamine pathway in bacterial cell wall biosynthesis. However, our rapidly increasing knowledge of the extreme specificity of secondary meta-
3 Biogenesis and Basic Pathways of Cyclitoki and Sugar Components
409
bolic traits does not support this view. Unex- hexose-derived moieties in many chemical pected modes of sugar activation (e.g., nu- groups of heterogenously composed secondcleotidylation by CTP instead of UTP), 2-ami- ary metabolites. no-hexose derivation from precursors other than D-glucosamine, and unknown mechanisms of modification and condensation have been found (some examples are given below; see Sects. 4.1, 4.2, and 5.1). Some of the known or hypothetical general routes for the biosynthesis of hexosamines are outlined in Fig. 11; it should be emphasized that 6-amino6-deoxyhexoses and 2-amino-2-deoxyhexoses also could be formed in the same way as the 3- and 4-isomers and that the same principal route also should yield amino-group-containing pentoses, heptoses, and octoses. Further variation may come from the transamination Aminoglycosides are largely actinomycete step, before or after sugar activation by nucleotidylation or even after condensation into products though rare occurrences in other bacterial groups are known (e.g., butirosincomplex molecules. Several of the unusual C-5 to C-9 sugar producing Bacillus circulans, sorbistin in derivatives encountered in many groups of pseudomonads, and N-methyl-scyllo-inosmicrobial products are found in the aminogly- amine in rhizobia; see Fig. 1). Aminoglycoside coside-related and other microbial secondary production is mainly a property of members metabolites. Some examples comprise pyra- of the filamentous actinomycete genera Strepnosidic 2- or 3-pentosamines (e.g., in seldo- tomyces spp., Streptoverticillium spp., and mycins and gentamicins, respectively; cf. Figs. Saccharopolyspora spp., and of MicromonoA14 and A17), 2-deoxyhexoses (e.g., in cytos- spora spp., Amycolatopsis spp., and Actinoet aminomycins; see Fig. A27), 3-deoxyhexos- planes sp. (WILLIAMS al., 1989). The bioloamines (e.g., in lividomycins; Fig. A l l ) , 4- gy, biochemistry, mode of action, biotechnoldeoxyhexosamines (e.g., in seldomycins; see ogy, and clinical applications of the aminoglyFig. A17), L-hexoses (e.g., L-mannose in de- coside antibiotics in addition to aminoglycosertomycin; not shown), and L-hexosamines side resistance have been often reviewed et (e.g., in streptomycin; see Sect. 4.1 and Fig. (e.g., KORZYBSKI al., 1978; DAVIESand 1978; WALLACE al., 1979; PIEPERSet Al). These moieties are most probably all ac- SMITH, tivated and modified via nucleotidylated in- BERG et al., 1980 WALKER,1980 PEARCE 1981; UMEZAWA and Hootermediates. However, it is thought that sim- and RINEHART, ple glucosylations and mannosylations in- PER, 1982; DAVIESand YAGISAWA, 1983; 1983; UMEZAWA al., 1986; CUNDet volve UDP-D-glucose and GDP-D-mannose, FOSTER, respectively. Furanosidic pentoses, such as ri- LIFFE, 1989, 1990; GRAFE, 1992; PIEPERSbose (e.g., in neomycins; see Fig. A l l ) or ara- BERG,1995; HOTTA et al. 1995). Recent inbinose moieties (e.g., in nucleoside antibiot- vestigations have concentrated mainly on the ics), might be introduced from 5-phosphori- molecular biology of the clinically relevant bosyl-l-diphosphate (PRPP) as activated pre- aminoglycoside resistance and molecular ascursors or, again, via an unknown activation pects of aminocyclitol aminoglycoside biosynand transfer mechanism. The dehydroxyla- thesis, resistance, and regulation in the protion reactions of pyranosidic hexoses (other ducing organisms (mainly with respect to than 6DOH) at positions C-2, C-3, and C-4 streptomycins and fortimicins; see Sects. 4.1 are especially interesting from a mechanistic and 4.2). During the past decade, semisynpoint of view. Several examples are known in thetic modification, new screening methods, aminoglycosides (see, e.g., the discussion in and new fields of application have resulted in Sect. 4.2 on the fortimicin group) and other the discovery of several new aminoglycoside
4 Genetics and Biochemistry of the Biosynthesis and Functions of Aminoglycosides
410
HO D
O
OH
HZN D
03)(J -#
HZN
OH
6-deoxy-D-glucose (acarbose?)
0 OH actinospedose (spectinomyan)
NHC(NH)NHz glucocinnamoylspermidines (LL-BM123-complex)
CH3 H3C0 &OH

CH3 D-sibirosamine (sibiromyan)
0-methyl-D-rhamnose (antibiotic A201)
D-mycosamine (amphotherian B; nystatin)
D-perosamine (perimycin)
HO
i=)
OCH3
CHa
OH OH
D-mycaminose (tylosin)
H3C0
OH
Ddesosamine (erythromycin)
OH
Damosamine (amicetin)
D-mycinose (angolamyan)
D-everminose (eveminomicin)
N(CH3)2 (ravidomyan)
6H
D-kasugamine (kasugamycin) D-digitoxose (lipomycin)
"0
OH
OH D-fucose (benanomian A; chartreusin)
OH D-digitalose (chartreusin)
D-ossamine (ossamydn)
D-chromose A (chromomycins)
AHCHI
D-olivose (urdamydns; 0- & Cglycosidc) D-forosamine (forosamydn) (neocarzinostatin) D-isomycamine (spiramycin)
Fig. 8
411
"PoNH2HO NH2 D-baallosamine (LPS, 0-chains; Pseudomonas)
OH
H3CHN O
OH
HO O
D-3-epimycosamine (macrolactam)
D-thomosamine (pradimidns; R1.2 = H OT CH3)
" 0
OH
D-vicenisamine (vicenistatin)
D-2,3,6-trideoxyglucose (dutomycin)
GOH
OH D-fucose (benanomicin A; chartreusin)
OH 0-olivose (olivomycin)
D-oliose (olivomydn)
"0
R
OH Ow0 (avlamycin; R = acetyl, hydroxyethyl)
OH (avilamydn) Dnoviose (novobiodn) D-vicenisamine (vicenistatin)
"0HoooH "0OH
OCH3
OH
OH
Ddiginose (heart glycosides)
H3C0
bN<
HO
OH HO
D-sarmentose (heart glycosides)
D-boivinose (stroboside)
Dchalcose (chalcomydn)
>H (o I.
H3C0 O H3C0
OH
H3CHN
(staurosporine) 6deoxy-D-mannose (lienomycin)
(notonesomycin)
CH3
4-O-methyl-2,3,6-lriDdeoxyglucose (A204A, polyether)
Gr;)
OAc OH [(-)griseusin A]
Fig. 8. Structural variants of D-6-deoxyhexoses (D6DOH) found in secondary metabolites. Examples of their occurrence in particular natural products are given in brackets.
412
Fig. 9
HO OH
aG
CHO(or CH@H) OH OH OH
L-rhamnose (oxopropalines)
L-fucose (EPSa)
CHO (or CH@H)
HO
0 0 0
OH OH HO OH 2-deoxy-L-fume (cytorhodins)
413
L-rhodinose (cytorhodins)
OH
OH
NH2
L-(dihydm-)str*tose (streptomyans)
L-(dihydm-)-5h y & o ~ s ~ e p ~ ~ (streptomyans; LPS. 0-chains)
L-daunosamine (daunorubicins)
L-rhodosamine (cytorhodins)
OH
'
o OH
eOH
@O
OH
CH3
L-mycarose (macrolides)
Lchromose B (chromomycins)
HO
0
L-dneruloseA (cytorhodins)
OH
L-dnerulose B (cytorhodins)
@O
L-awlose (cytorhodins)
OH
NH-R
(glycopeptides: R = -H, -methyl, -CH&~tyl)
4-methyl-Lrhodosamine (saptomyans)
"
HO OH O D
Hou
OH H3C0 OH HQH oo
L-amicetose (cytorhodins)
CH3
CH3
Lcladinose (maadides)
L-axenose (axenomycins)
L-ristosamine (ristomycin)
(calicheamidn)
CH3 O
I
NH2
"VoH
OCH3 H3Cb OCH3
CH3
(balhimycin)
L-oleandrose (avermectin)
L-nogalose (anthracyclines)
Ldecilnitrose (anthracydines)
Fig. 9
414
HO QOH
OOH OOH
SUC-0
H
NH2 O
V OH
OH OH
( R ) L H (TAN-1120; R = cydic ether)
H3C0
CH3
6deoxy-L-talose (phenazoviridin)
(avidinorubicin; SUC= sucdnyl)
avidinosamine (avidinorubicin)
H~NOCO OH
L(or D)-noviose (novobiocin)
2-O-methylL-rhamnose (helvecardins)
(foraminosyl(+)griseusin A)
F g 9 Structural variants of L-6-deoxyhexoses (L-6DOH) found in secondary metabolites. Examples of i. .

their occurrence in particular natural products are given in brackets.
0
0
OH
Ph-0
PMP El
H N
\
~ ~ C )
0-NDP
' /
CH,
0 H'
OH OH
QNDP
Ph-0
0 HN+2 >No
Ph-0
1I
>
'0-NDP
OH
0-NDP
F g 1 . Mechanism of 3-dei. 0 hydroxylation in the 3,6-dideoxyhexose pathway of gram-negative bacteria. El: PMP-dependent, [2Fe2S]-containing dehydrase (AscCIRfbH-type); E2: NADH-dependent, FAD/ [2Fe-2S]-containing dehydrase (for details, see LIU and THORSON, 1994).
415
(R = Acor H)
D4HA
o w O : -D p
R
0-NDP
DIL3HA (3PA)
NH2
ONDP
L4HA
(R = CYOH or H )
Fig. 11. General scheme for the derivation of 2-, 3-, or 4-aminated hexoses and pentoses. Also, 2- and 6-aminohexoses (HA), 2-aminopentoses (PA), or aminoheptoses and aminooctoses could be formed in their furanosidic or pyranosidic forms via the oxidation and transamination pathway.
structures and, in at least one case, in the introduction into pharmaceutical use. This chapter will focus mainly on the latter aspects.
4.1 Streptomycins and Related Ca Aminoglycosides

The streptomycins and bluensomycin are a relatively homogenous group of basically pseudotrisaccharidic aminoglycosides (Fig. A l ) which can be elongated by one or two further glycosidic residues (mannosyl or et ashimosyl; IKEDA al., 1985a; TOHMA al., et
1989) or be cleaved into two hexose derivatives as in AC4437 (dihydrostreptosyl-streptidine; AWATAet al., 1986). Compounds of this group are produced by a broad range of species (KORZYBSKI al., 1978). They have et been found in the genera Streptomyces, Streptoverticillium, Amycolatopsis, and may be present in others. A survey of streptomycinproducing strains from collections and a taxonomic study of new isolates that was recently started (PHILLIPS al., 1992; MARSHand et WELLINGTON, 1994) indicated that they can mainly be grouped into three clusters within the family of Streptomycetaceae. These clus. ters are represented by S. griseus, S hygroscopicus, and Streptoverticillium mashuense.
416
Therefore, the presently accepted taxonomy have been clarified, it is known that the comof the family Streptomycetaceae (WILLIAMS pounds of the streptomycin family are syntheet al., 1989; EMBLEY STACKEBRANDT,sized via 25-30 enzyme-catalyzed steps and 1994) assigns those streptomycin-producing (WALKER, 1975a; RINEHART and STRO1976; GRISEBACH, 1978; RINEHART, species that have been investigated in detail SHANE, and ITO, 1982; PIEPERSBERG, to clearly separated species clusters. Recently, 1980; OKUDA this view was confirmed and extended by se- 1995). The currently hypothesized streptomyquencing the 16s rRNA genes from several cin and bluensomycin biosynthesis pathways streptomycin-producers (MEHLING et al., are summarized in Fig. 12. They involve the 1995). As with other secondary metabolites, formation of the activated precursors, strepthe reason for this taxonomically scattered, tidine (bluensidine)-6-phosphate (cf. Sect. but relatively stable distribution in indiviual 4.1.1.1), dTDP-dihydrostreptose (cf. Sect. 4.1.1.2), and NDP-N-(methyl)-L-glucosamine strains worldwide is not known. (cf. Sect. 4.1.1.3) which are made most likely from ~-glucose-6-phosphate.However, it has 4.1.1 Streptomycins been demonstrated that D-glucosamine can be incorporated without breakage of its C-C The biosynthesis of streptomycins, includ- bonds into the N-methyl-L-glucosamine subing bluensomycin, was intensively studied by unit of streptomycin (reviewed by OKUDA feeding labeled primary metabolic precursors and ITO, 1982). In a second phase, the precur(DEMAIN INAMINE, and 1970 MUNRO al., sors are condensed forming dihydro-streptoet 1975) and analyzing streptomycin biosynthet- mycind-phosphate (cf. Sect. 4.1.1.4) which is ic reactions in cell-free systems and with pu- secreted and probably oxidized while passing rified biosynthetic enzymes (WALKER, 1975a, through the cytoplasmic membrane to give b; GRISEBACH, 1978; RINEHART, 1980). Al- streptomycin-6-phosphate outside the cell (cf. though not all of the biosynthetic reactions Sect. 4.1.1.4). The biologically active strepto-
F(NH)-NH,
in CM out
dTDPG
dTDP-Dihydrostreptose
OH
NHCH,
NDPG(A) (R NHZ OT OH)
L-glucosamine
NDP-&Methyl-
Fig. 12. Gerneral outline of the streptomycin (SM) pathway. The three activated intermediates principally formed from glucose-6-phosphate (G-6-P) are condensed in the cytoplasm to dihydro-SM-6-phosphate (DHSM-6-P) which is oxidized and dephosphorylated to SM during or after transport through the cytoplasmic membrane (CM). mIP: myo-inositolphosphate; dTDPG: deoxythymidinediphosphate-glucose; NDPG(A): nucleosidediphosphate-glucose (or -glucosamine) (NDP = CDP or UDP).
417
mycins are liberated finally by a specific phosphatase (cf. Sect. 4.1.1.4), and after re-uptake they can be phosphorylated by a streptomycin 6-phosphotransferase representing the resistance mechanism which protects the producers from their own products (cf. Sect. 4.5). Genetic studies of streptomycin biosynthesis in the various producers started with the cloning of resistance genes (reviewed in CUNDLIFFE, 1989; PIEPERSBERG, 1995). Two genes were initially cloned, strA (aphD) and aphE, which encode two different streptomycin phosphotransferases, APH(6) and APH(3 ), respectively (WALKER, 1975b; WALKER and WALKER,1975; DISTLERand PIEPERSBERG, 1985; DISTLERet al., 1987a; HEINZELet al., 1988). Only the strA gene could be localized up to now in the production gene clusters of S. griseus and S. gfuucescens GLA 0. The aphE gene, however, occurs only in S. griseus strains and does not appear to be linked to the strlsts cluster. Subsequently, further streptomycin biosynthetic genes were identified by chromosome walking and genetic complementation of mutants from different S. griseus strains deficient in streptomycin biosynthesis (DISTLER et al., 1985; OHNUKI al., 1985a, b). About 30 et genes for (5 -hydroxy-)streptomycin (strlsts) and bluensomycin (bfu) production have been cloned and analyzed from various strains of S. griseus, from S. gfuucescens GLA.0 (ETH 22794), and from other Streptomyces spp., and all were found to be clustered in one region of about 3 0 4 0 kb of genomic DNA (Tab. 2; Fig. 13; MANSOURI al., 1989; et DISTLER al., 1990, 1992; RETZLAFF al., et et 1993; PIEPERSBERG, 1995). Comparison of the primary structures of the homologous strl sts genes and their gene products in S. griseus and S. gfuucescens revealed identity values varying between 58% and 86%. In contrast, the corresponding nonencoding intercistronic DNA sections have much less or no significant homology. The order of the strlsts genes identified so far within the respective operons is similar although the arrangement of the operons differs considerably among producing species. Usually, the genes encoding enzymes for the synthesis of a subunit of streptomycin (e.g., streptidine) are not arranged in subpathway-specific operons. Instead, mixed
operons are generally found which may reflect the need for a strictly coordinated regulation of strlsts gene expression in order to guarantee a coordinated supply of the activated precursors in streptomycin synthesis. The phenomena involved in genetic regulation of streptomycin biosynthesis are discussed below (cf. Sect. 4.6).
4.1.1.1 Biosynthesis of Streptidine and Bluensidine

The biochemistry of the eleven steps involved in streptidine synthesis was evaluated by WALKER al. (WALKER, et 1975a, 1990) by means of enzymology and led to the proposed pathway outlined in Fig. 14. The initial step is the formation of ~-myo-inositol-3-phosphate from D-glucose-1-phosphate via the Ca route (cf. Sect. 3.1) by an ATP-dependent myo-inositol phosphate synthase (WALKER,1975a; SIPOSand SZABO,1989). Up to now, no strlsts gene could be identified which encodes the myo-inositol synthetase. Also attempts to purify the enzyme failed because of its instability (SIPOSand SZABO,1989). The end product is streptidine (SD)d-phosphate, which also seems to be the first intermediate made from externally supplied SD in SD- mutants (OHNUKI et al., 1985a, b; DISTLERet al., 1985). SD-6-phosphate is synthesized by two sets of five parallel enzymatic reactions each presumably catalyzed by individual enzymes: two cyclitol phosphate phosphatases, two cyclitol dehydrogenases, two aminotransferases, two phosphotransferases, and two amidinotransferases in streptomycin producers (cf. Tab. 2, Fig. 14). In contrast to this the producer of bluensomycin S. hygroscopicus forma gfebosus lacks steps 8 to 11 (see Fig. 14; WALKER,1990) which are replaced by carbamoylation and phosphorylation reactions at positions 5 and 4, respectively, yielding bluensidined-phosphate (see Fig. 15). The product of the strO gene has significant similarity to eukaryotic inositolmonophosphate phosphatases (RETZLAFF et al., 1993). Therefore, this protein is assumed to be the enzyme which catalyzes step 2 in the streptidine pathway (RETZLAFF
Tab. 2. Gene Products and Enzymes Presumed to be Involved in Streptomycin Production Molecular Datab MW [kDa] 216 (native) 28 37 45 260 (256) 348 424 320 (316) 347 36 (35) 39 aa Enzymatic Function' (Preliminary Assignment) Remarksd References"
Gene Producta
Unkown
~-myo-Inositol-3-phosphate synthetase
ATP-dependent
StrO StrI StsC
StrN StrBl
Unknown 52 43 32 38 (35) 38 36 22 32 355 328 200 304 349 (319) 312 410 490
StsB
StsA
StsE
StrB2
StrD StrE StrM
StrL
SIPOSand SZABO (1989) A, RETZLAFF al. (1993) et (~-myo-Inositol-3-phosphate phosphatase) NAD(P)-dependent MANSOURI PIEPERSBERG and (1991) (scyllo-Inosose dehydrogenase) A, RETZLAFF al. (1993) et Rel. to StrS and scyllo-Inosose aminotransferase StsA; PLP-dependent PISSOWOTZKI al. (1991) et (scyllo-Inosamine 4-phosphotransferase) DISTLER al. (1987b) et scyllo-Inosamine-4-phospha te amidinotransferase (cf. WALKER, 1975a) N-Amidino-scyllo-inosamine4-phosphate phosphatase NAD(P)-dependent A (NAmidino-scyllo-inosamine dehydrogenase) A, RETZLAFF al. (1993) et (3-Keto-N-amidino-scyllo-inosamine Rel. to StsC and aminotransferase) StrS; PLP-dependent (N-Amidino-streptamine A 6-phosphotransferase) N-Amidino-streptamine-6-phosphate PISSOWOTZKI al. (1991) et amidinotransferase PISSOWOTZKI al. (1991) et dTDP-Glucose synthetase PISSOWOTZKI al. (1991) et NAD-dependent dTDP-Glucose 4,6-dehydratase dTDP-4-Keto-6-deoxyglucose PISSOWOTZKI al. (1991) et 3,5-epimerase PISSOWOTZKI al. (1991) et dTDP-4-Keto-~-rhamnose dehydrogenase NADP-dependent [ = dTDP-L-dihydrostreptosesynthase]
StrX StrU StrF StrG StsG StrS 33 24 26 (ca. 45-50) 63 46 33 38 (46) 29 272 350 (424) 307 449 (462) 592 300 213 236 (inc.)
20 46 32 23 27 40
182 428 281 199 246 378 (377)
(NDP-Hexose 3,5-epimerase) (NDP-Hexose oxidoreductase) (NDP-Hexose epimerase) (NDP-Hexose epimerase) (N-Methyltransferase) (Aminotransferase; unknown function)
StrT StsD StsF StrV
StrW
StrK
StrA (APhD) StrR
Unknown function Unknown function Unknown function (Exporter for streptomycin 6-(or 3 "-)phosphates) (Exporter for streptomycin 6-(or 3 "-)phosphates) Streptomycin 6-(or 3 "-)phosphate phosphatase Streptomycin 6-phosphotransferase
A, BEYERet al. (1996) Rel. to StrM et NAD(P)-dependent A, BEYER al. (1996) MANSOURI PIEPERSBERG and (1991) MANSOURI PIEPERSBERG and (1991) A A, RETZLAFF al. (1993) et Rel. to StsC and StsA; PLP-dependent A A A A, BEYERet al. (1996) ATP-dependent (ABC-transporter) ATP-dependent A, BEYERet al. (1996) (ABC-transporter) Extracellular MANSOURI PIEPERSBERG and (1991) protein ATP-dependent DISTLER al. (1987a) et
AphE
DNA-binding protein, activator of gene expression streptomycin 3 " -phosphotransferase
ATP-dependent
A, DISTLER al. (1987b); et RETZLAFF DISTLER and (1995) HEINZEL al. (1988) et
Cf. Fig. 13 Genetic data from Streptomyces griseus N2-3-11 andor S. gluucescem GLA.0 (from the latter strain given in brackets if different); inc.: incompl sequence Cf. Figs. 14-17 PLP: pyridoxalphosphate A J. AmERT, S. BEYER, DISTLER, MANSOURI, MAYER, J. K. G. and W. PIEPERSBERG, unpublished data
420
421
4 Fig. 13. Gene clusters for the production of streptomycins (SM) and the related Ca aminoglycosides bluensomycin (BM) and spectinomycin (SP). The streptomycete producers investigated most intensively are S. griseus (Sgr) strains N2-3-11 and DSM40236, S. glaucescens (Sgl) GLA.0 (ETH 22794), S. bluensis (Sbl) DSM40564, and S. flavopersicus (Sfl) NRRL 2820. Restriction maps for a few enzymes are given for orientation; the clusters are aligned according to their homologous amidinotransferase [strB( l)] genes.
et al., 1993; see Fig. 14). The StrI and StsB proteins are clearly members of the oxidoreductase class with an N-terminal dinucleotide and PIEcoenzyme binding site (MANSOURI PERSBERG, 1991). Therefore, they are good candidates for the step-3 and -8 cyclitol oxi-
doreductases which form the keto groups preceeding the two transamination steps (see Fig. 14). The StrI protein significantly resembles the myo-inositol-2-dehydrogenase enzyme from Bacillus subtilis (EUJITAet al., 1992) and is the enzyme which catalyzes the first cyclitol dehydrogenation step (J. AHLERT, W. unpublished data). PIEPERSBERG; The StsC protein is the scyllo-inosose aminotransferase and catalyzes the transfer of the a-amino group of glutamine to scyllo-inosose yielding scyllo-inosamine and a-ketoglutaramate, an unusual transamination reaction in bacterial cells (cf. step 4 in Fig. 14; WALKER, 1975a; LUCHER et al., 1989; J. AHLERT, J. DISTLER, W. PIEPERSBERG, and unpublished data). This enzyme, like the strlsts-gene products StrS and StsA, is a member of a new class
(StsE or SWN)
(3
(StrO)
(Strl or StSB)
stsc
SWBl
aKGN
1
(11
(2)
(3)
(StsB or SW
(4)
(5)
(6)
(3
(StsA or StrS)
(StrN or StSE)
StrB2
YH C-NH,
HO
HO
(6)
(I)
Ho (8)
HO (9)
lo
0-P (10)
0-P
'1
Streptidlne&phosphate
O=C-NH,
P)
ea
O=C-NH,
HO
Bluensldins &phosphate
Fig. 1 . The streptidine and bluensidine pathways. Enzymatic steps are numbered; where a particular gene 4 product is known or postulated to be involved this is given (cf. Tab. 2 and Fig. 13). Known or postulated (2) intermediates (numbered in brackets) are (1) ~-myo-inositol-3-phosphate, myo-inositol, (3) scylloinosose, (4) scyllo-inososamine, (5) scyllo-inososamine-4-phosphate, (6) N'-amidino-scyllo-inososamine4-phosphate, (7) N'-amidino-scyllo-inososamine,(8) 3-keto-N1-amidino-scyllo-inososamine, N'-amidi(9) (7a) no-streptamine, (10) N1-amidino-streptamine-6-phosphate, bluensidine; P: phosphate residues.
422

StrM
0-dTDP
13 (12) (13) (14) 14 (15)
OH
OH
dTDP-Dihydrosireptose
Fig. 1 . The dTDP-dihydrostreptose pathway. For the numbering and labeling system, see legend of Fig. 14; 5 intermediates are (12) D-glucose-1-phosphate,(13) dTDP-D-glucose,(14) dTDP-4-keto-6-deoxy-~-glucose, (15) dTDP-4-keto-~-rhamnose.
of pyridoxalphosphate(PLP)-dependent aminotransferases, the so-called secondary metabolic aminotransferases (SMAT). Several other antibiotic biosynthetic aminotransferases (see the compilation and discussion in PIEPERSBERG, 1994), the protein MosB of Rhizobium meliloti involved in the metabolism of L-3-0-methyl-scyllo-inosamine (MURPHY et al., 1993), and the PMP-dependent enzyme El catalyzing the 3-dehydroxylation during the formation of 3,6-dideoxyhexosesin gram-negative bacterial LPS biosynthesis (THORSON al., 1993; LIU and THORSON, et 1994) belong to the SMAT protein family. Therefore, it seems obvious that one of the genes, either sfsA or strS, encodes the N-amindino-scyllo-inosamine L-alanine aminotransferase, the second transaminase necessary for streptidine-6-phosphate formation (cf. Fig. 14, step 9). Alternatively, one of the remaining SMAT enzymes, StrS or StsA, may be involved as a biosynthetic enzyme in the synthesis of the NDP-N-methyl-L-glucosamine (NMLGA) subunit (cf. Sect. 4.1.1.3). The enzymes catalyzing the phosphotransfer (cf. Steps 5 and 10, Fig. 14) seem to be StrN or StsE proteins. Although this function has not yet been proven by in vitro assays of the individually expressed enzymes there is much evidence for this assumption. Both proteins contain in their C-terminal portion the characteristic signature motifs which are typical for aminoglycoside phosphotransferases and eukaryotic protein kinases (cf. Sect. 4.5; PIEPERSBERG et al., 1988; HEINZEL al., 1988). Reet cently, a gene homologous to strN was found on a DNA fragment from S. flavopersicus, which confers spectinomycin resistance to S. lividans (J. ALTENBUCHNER D. LYUTZand
KANOVA, unpublished data; see below Sect. 4.1.2). In addition, a spectinomycin phosphorylating activity was detected in this recombiJ. nant S. lividans strain (J. DISTLER, ALTENBUCHNER, and D. LYUTZKANOVA, unpublished data). These findings support the assumption that sfrN encodes a phosphotransferase. Two closely related genes, strBI and strB2, were identified which encode the amidinotransferase engaged in the biosynthesis of the streptidine moiety (OHNUKI al., 1985a, et b; DISTLERet al., 1987b; TOHYAMA al., et 1987; MAYERet al., 1988). Both amidinotransferases of S. griseus cloned in S. lividuns were active in a nonspecific assay which did not differentiate between the first and second transamidination steps (OHNUKI al., 1985a, b; DISTLERet et et al., 1987b; TOHYAMA al., 1987; S. EHRIGFRANTZKE, and W. PIEPERSBERG, unpublished data). Since the enzymes encoded by the cloned genes have not as yet been tested with their postulated substrates (WALKER, 1975a, see steps 6 and 11, Fig. 14) it remains uncertain whether the assumptions regarding their functions (cf. StrB1, step 6, StrB2, step 11, Fig. 14) are correct (PISSOWOTZKI al., et 1991) or whether StrBl carries out both steps (OHNUKIet al., 1985b). The latter is supported by the puzzling finding that extracts of S. bluensis seemed to contain both enzymatic activities although only StrBl is used in the bluesidine pathway (WALKER, 1990) and only the strB1 gene could be detected by hybridization in two bluensomycin producers, S. bluensis DSM 40564 and S. hygroscopicus ssp. glebosus DSM 40823 (cf. Fig. 13; G. MAYER, A. MEHLING, W. PIEPERSBERG, and unpublished data).
4 Genetics and Biochemistry of the Biosynthesis and Functions of Arninoglycosides
423
4.1.1.2 The Pathway
L-Dihydrostreptose
GRISEBACH (1978) postulated a route for dTDP-L-dihydrostreptose similar to that leading to activated L-rhamnose in gram-negative bacteria: activation of D-glucose in form of dTDP-D-glucose (step 12, Fig. 15), dehydratation to dTDP-4-keto-6-deoxyglucose (step 13), epimerization to dTDP-4-keto-~-rhamnose (step 14), and reduction coupled to a rearrangement of the hexose carbon chain yielding dTDP-L-dihydrostreptose (step 15). The 4 enzymes catalyzing the synthesis of dTDP-L-dihydrostreptose are encoded by the genes strD, strE, strM, and strL (see Tab. 2 and Fig. 14; DISTLERet al., 1987b; PISSOWOTZKI et al., 1991; PIEPERSBERG, 1994). Evidence in support of this is the high level of homology which they share with the respective genes encoding the L-rhamnose biosynthesis enzymes in salmonellae, rfbA, B, C, D , and their activity in Escherichiu coli (REEVES, and 1993; PIEPERSBERG, 1994; S. VERSECK W. PIEPERSBERG, unpublished data). StrD is related to NDP-hexose synthases (pyrophosphorylases) (DISTLERet al., 1987b). Genes similar to all or part of the strDELM cluster are present in many other gene clusters which encode enzymes for the production of streptomycete secondary metabolites containing a 6DOH sugar moiety (PIEPERSBERG, 1994; LIU and THORSON, 1994; VINING and STUTTARD, 1994). In a screening of streptomycete strains more than 50 of which produced 6deoxyhexose-containing secondary metabolites a majority seemed to possess genes which and hybridize to strD, E(L,M) (STOCKMANN PIEPERSBERG, 1992). Thus, this part of the str gene cluster appears to be widespread among antibiotic-producing streptomycetes and other bacterial groups (cf. Sects. 5.3 and 6).
4.1.1.3 Hexosamine Pathway

The synthesis of the third moiety of streptomycin, (NDP-activated) N-methyl-L-glucosamine (NMLGA), has been intensively studied. Several speculative pathways have been postulated, including nucleotidylation, deace-
tylation of one of the intermediates (if UDPN-acetyl-hexosamines are formed), and at least three epimerization steps, one of which could be divided into separate oxidation and reduction steps at C-4 of the hexoseamine, followed by N-methylation (RINEHART and STROSHANE, 1976; GRISEBACH, 1978; OKUDA and ITO, 1982; HIROSE-KUMAGAI al., et 1982; KUMADA al., 1986). The direct preet cursor still remains unknown but could perhaps be D-glucose-1-phosphate, D-glucosamhe-1-phosphate, N-acetylm-glucosamine-lphosphate, or N-methyl-D-glucosamine (S. BEYER and W. PIEPERSBERG, unpublished data). Also, the published structures, UDPactivated and phosphorylated hexosamines (HIROSE-KUMAGAI al., 1982; KUMADA et et al., 1986), of putative intermediates of the Vmethyl-L-glucosamine (NMLGA) pathway which were accumulated in the wild type and in a mutant blocked in the NMLGA pathway cannot be explained by the currently postulated pathway. Another unsolved problem is the formation of the N-methyl group in Nmethyl-L-glucosamine which might be introduced at the D-glucosamine level or later in the pathway, perhaps even after condensation of the three streptomycin moieties (OKUDA and ITO, 1982, op. lit.). There is indirect evidence that the genes strPQX, isolated and sequenced from S gluu. . cescens, strFG, and stsG, analyzed from S grisew, are involved in the N-methyl-L-glucosamine pathway (MANSOURIand PIEPERSBERG, 1991; J. AHLERT, G. MAYER, S. BAYER,and W. PIEPERSBERG, unpublished data). StrQ is a CDP-D-glucose pyrophosphorylase which could catalyze the activating step of the N-methyl-L-glucosamine pathway (Fig. 16, step 16; S. BAYERand W. PIEPERSBERG, unpublished data). The StrP protein shares a higher degree of similarity with UDP-glucose 4-epimerases than with NDP-hexose 4,6-dehydratases, which are distantly related, and it could, therefore, be a C-4 dehydrogenase (or epimerase; Fig. 16, steps 17,19,18a, or 20) for NDP-hexose derivatives (PISSOWOTZKI al., et 1991). Alternatively, StrP could be an oxidoreductase, introducing a keto group at C-2 of a CDP-activated hexose (Fig. 16, step 20). This could then be the substrate for an aminotransferase, encoded by either strS or stsA,
424

CHPH
A
HO
Cm a
CH,OH
(slrporu?)
CHPH
(strx?)
f&lCD?
OH (16) 16 OH (17) 17
0=QiO-mp QQICDP
OH (18) 18 OH OH (19)
NDP-N-MethylLglucosamine
/( W
(17a)
( W
( w
(2W
Fig. 1 . The NDP-N-methyl-L-glucosamine (NMLGA) pathway. For the numbering and labeling system, 6 see legend of Fig. 14. The exact route of formation of NMLGA is unknown and could either procede via CDP-glucose (A) or an unknown derivative of NDP-D-glucosamine (B) as precursors (for details, see text). Possible intermediates in (A) are (16) D-glucose-1-phosphate, (17) CDP-D-glucose,(18) CDP-4-keto(22) D-glucose, (19) CDP-4-keto-~-mannose, CDP-L-mannose, (21) CDP-2-keto-~-glucose, CDP-L(20) glucosamine; possible intermediates in (B) are (16a) NDP-D-glucosamine, (17a) NDP-N-methyl-D-glucos(20a) amine, (18a) NDP-N-methyl-D-mannosamine, (19a) NDP-4-keto-N-methyl-~-mannosamine, NDP-
4-keto-N-methyl-~-glucosamine.
which introduces the 2-amino group (Fig. 16, step 21). The strFG genes were mapped in a region which complemented a mutant blocked in the N-methyl-L-glucosamine pathway (KUMADA al., i986): Protein compariet sons suggest that both StrF and StrG could be members of the group of dinucleotide-independent (non-oxidoreductase type) sugar isomerases (or epimerases) and might even form a heterodimeric enzyme (MANSOURI and PIEPERSBERG, 1991). A candidate for a 35epimerase (Fig. 16, steps 18 and 19a) in the N-methyl-L-glucosamine pathway is StrX because of its significant similarity -to other 3 5 epimerases such as StrM (see Sect. 4.1.1.2; S. BEYER and W. PIEPERSBERG,unpublished data). The StsG protein has three conserved motifs which are generally found in methyltransferases (KAGAN and CLARKE, 1994). Therefore, stsG could encode the N-methyltransferase necessary for streptomycin formation (Fig. 16, steps 22 and 16a). Based on
these genetic data, the following routes for the synthesis of N-methyl-L-glucosamine can be proposed among several other possibilities:
(1) CDP activation of D-glucose-1-phosphate (StrQ), epimerization and oxidoreduction followed by a transamination and N-methylation yielding CDP-L-glucosamine (Fig. 16A); (2) activation and modification of D-glucosamine (or a derivative thereof) which would account for the earlier observations that D-glucosamine is preferentially incorporated into N-methyl-L-glucosamine (Fig. 16B) (OKUDAand ITO, 1982; KUMADA al., 1986). et
However, the sequence of enzymic steps or the still unknown reactions remain even more speculative. Corresponding to the cyclitol transaminase reactions of the cyclitol an addi-
425
tional phosphotransferase could also be involved in the N-methyl-L-glucosamine pathway forming the identified, additionally phosphorylated NDP-hexosamine (HIROSE-KuMAGAI et al., 1982; KUMADA al., 1986). et
(MAIER GRISEBACH, and 1979) that this oxidase is in the particulate (membrane) fraction and does not require the addition of an electron acceptor such as NAD(P)+. Nevertheless, these phenomena could be explained by a hypothetical oxidase/exporter complex formation between the StrU protein and the cytoplasmic domain of the StrU/W transmem4.1.1.4 Condensation of Subunits, brane complex and a strongly bound dinuProcessing, and Export cleotide coenzyme in StrU. This would also explain the coupling of oxidation and transThe condensation of the activated precur- port steps which have been interpreted as besors requires two glycosyltransferase steps ing a functional unit (MAIER and GRISEwhich are catalyzed by two enzymes localized BACH,1979). An attractive speculation could in the soluble cytoplasmic fraction resulting in be that the oxidation is coupled to both the dihydro-streptomycin-6-phosphate,which is ATP-driven export of a phosphorylated amithe last soluble intermediate detected inside noglycoside and a membrane-bound electron producing cells (KNIEP and GRISEBACH, transport. The final dephosphorylation to release the 1976, 1980). The dihydrostreptosyl transferase was partially purified and found to be a biologically active antibiotic is catalyzed by dimeric enzyme (KNIEP and GRISEBACH, StrK, a streptomycin-6-phosphate specific ex1980) with a likely subunit molecular weight tracellular phosphatase (step 27, Fig. 17; and PIEPERSof 35kDa. Evidence that the strH gene could WALKER,1975a; MANSOURI be one of the two genes needed for glycosyl- BERG,1991). The gene products of the strK transfers is weak (OHNUKI al., 1985a, b; genes of S. griseus and S. glaucescens (orfZ of et MANSOURI and PIEPERSBERG, 1991). The VOGTLI and HOTTER, 1987) are highly hocondensation product dihydro-streptomycin- mologous to the alkaline phosphatase (PhoA) and PIEPERSBERG, 6-phosphate is converted, probably coupled of E. coli (MANSOURI with the active transport, to streptomycin-6- 1991). The properties of the StrK phosphatphosphate by a membrane-associated dehy- ase of S. griseus when expressed in S. lividans drogenase (step 26, Fig. 17; MAIERand GRI- are similar to those reported earlier for the streptomycin phosphate phosphatase (WALSEBACH. 1979). The proteins encoded by 1973; WALKER, 1975a; strVW and strU recently detected in both S. KER and SKORVAGA, and PIEPERSBERG, 1991). Each glaucescens and S. griseus (PIEPERSBERG, MANSOURI 1994; BEYER al., 1996) are able to form this of the five steps (two glycosyltransfers, dehyet membrane-bound transport/dehydrogenase drogenation, phosphatase reaction, and transcomplex. The StrU protein has significant port) should have an equivalent in 5 '-hysimilarity to the alcoholic hydroxyl-group-ox- droxy-streptomycin producers and, except for idizing dehydrogenases, and the strV(W) gene the dehydrogenation, also in the dihydroproduct(s) represent a new member of the streptomycin and bluensomycin producers family of the so-called ABC transporters (cf. Fig. Al). The 5 '-hydroxylation reaction (BEYER et al., 1996) suggesting that they forming 5 '-hydroxy derivatives of streptomycould be engaged in the oxidation and export cins is still obscure. It could take place by a of dihydro-streptomycin-6-phosphate.How- hydroxylase reaction during the release at the ever, StrU clearly is a member of the dinu- cytoplasmic membrane or at the stage of the cleotide coenzyme-dependent dehydrogenase dTDP-hexose. However, it seems unlikely family and does not contain any transmem- that the original 6-hydroxy group of D-glubrane domains nor membrane-association cose remains in the intermediates since the sites. Therefore, the suggested involvement of following steps would then require enzymes StrU in the oxidation of dihydrostreptomy- with altered substrate specificity or even alcin-6-phosphate does not easily correlate to tered reaction mechanisms. D-Mannosylation group at position 4 of the N-methyl-L-glucosamine the earlier findings of H. GRISEBACH'S
in
(7)
out
y(NH)-NHp HN HN=C-NH,
Streptidinr6-phosphate dTDP-Dihydre streptose
C(NH)-NH,
26
H3C OH OH
{ I -0 @
+
H32: OH
(24)
,cj
0
OH
NDP-N-MethylL-glucosamine hHCH3 y(NH)-NH, y(NH)-NH, HN HN=C-NH? HN HN=CAphD (StrA
SM-6-P
(or SM-T-P)
28
OH
4 Genetics and Biochemistry of the Biosynthesis and Functions of Arninoglycosides
427
moiety occurs in some streptomycin-producing biovars of S. griseus. This appears to be a nonspecific peripheral reaction rather than an integral step in the biosynthetic pathway, since in the same strains a mannosidohydrolase is formed under carbon catabolite depletion (INAMINE and DEMAIN,1975). Later, also a dimannosylated product was detected in S. griseus strains (IKEDAet al., 1985a), in addition to N-methyl-L-glucosamine amino group modifications as in the ashimycins produced by S. griseus FT3-4 (TOHMA et al., 1989). Also, pseudodisaccharidic end products of the streptomycin family missing the Nmethyl-L-glucosamine moiety have antibiotic activity and occur as natural products in fermentations of the Streptomyces sp. strain AC4437 (AWATAet al., 1986).
4.1.2 Streptomycin-Related Ca Aminogl ycosides

Spectinomycins (Actinospectacin). The spectinomycin molecule (see Fig. A2) is composed of two double condensed hexose derivatives, a myo-inositol and a 6DOH-derivative (actinospectose, a 4,6-dideoxyhexose; cf. Fig. 8), and therefore is a Ca(4,5)-6DOH compound as is the dihydro-streptosylstreptidine (AC4437) produced by some streptomycetes (AWATA al., 1986). The incorporaet tion of radioactively and stable isotope-labeled glucose into both moieties clearly supports this interpretation (OTSUKA et al., 1980). However, no further biochemical studies on the pathway have been reported, except for the recent demonstration in a spectinomycin producer of an L-g1utamine:scyffoinosose aminotransferase similar to that found in streptomycin producers (cf. Sect. 4.1.1.1; WALKER, 1995). Recently, a spectinomycin-resistance-conferring DNA segment (3.65 kb) was cloned from the spectinomycin producer S. flawpersicus NRRL 2820 (J. ALTENBUCHNER and D. LYUTZKANOVA, personal communication). The DNA sequence suggested that it was derived from a rearranged and degenerated streptomycin gene cluster since it showed striking similarity with three genes, strBl, strR, and strN in the re-
spective clusters of S. griseus and S. gfaucescens (cf. Fig. 14). The strB2-related gene is deleted and its truncated protein product (80 aa) is certainly nonfunctional. The deletion was probably accompanied or preceded by an insertion of an IS112-like IS element part of which has been retained. Also, the similarity pattern among the reading frames is puzzling: the product of the strB2-related gene shows 83.8% and 92.5%, the product of the strR-related gene 45.9% and 47.5%, and the product of the strN-related gene 29.5% and 29.4% identity to the StrB1, StrR, and StrN proteins, respectively, of S. griseus and S. gfaucescens. It is not yet known whether there are other production genes for spectinomycin adjacent to this cluster. However, this finding strongly supports the hypothesis that an ancestral production gene cluster for streptomycin-like aminoglycosides could have developed into another variant by divergent evolution after degeneration, modification, and, later on, frequent recombination with an intact strlsts gene cluster thereby creating a new pathway which yields simpler but still effective end products. More recently, a new spectinomycin derivative, spenolimycin (Fig. A3), has been described (KARWOWSKI al., 1984). Its et 6DOH moiety is altered in that the 3' position contains an oxymethyl group and the C-3 '14'-bond is unsaturated, reminiscent of modifications which occur in other 6DOH residues, e.g., in anthracyclines (see Figs. 8 and 9 and Sect. 5.3). Kasugamycins. Not much work has been done to elucidate the biochemical pathway for the production of kasugamycin (Fig. A4) and related Ca(4)-6DOH compounds (minosaminomycin; Fig. A5). However, it is reasonable to postulate basically streptomycin-like pathways also for these aminoglycosides. The cyclitols are again clearly derived from myoinositol. However, the postulate by UMEZAWA et al. (1986) that the 6DOH moiety kasugamine could be derived from UDP-N-acetylD-glucosamine seems unlikely in view of what has been learned from the streptomycin pathway and from other 6DOH or hexosamine pathways in the recent past. Rather, it seems likely that a dTDP- or a CDP-hexose biosynthetic route is used, since the final product is
428
a 2,3,4,6-tetradeoxy-2,4-diaminohexose derivative (cf. Sects. 3.2.1 and 5.3).

Myomycin. Myomycin, besides being a possible Ca(4)-HA (Fig. A8) compound modified by a varying number of plysyl residues, is another streptomycin-related product in that it contains some distant structural similarities to bluensomycin, such as carbamoyl groups in the cyclitol moiety in addition to a guanidino group (in the hexosamine moiety) and apparently a mode of action and binding site which are identical to those of the streptomycins (DAVIES al., 1988). Since these aminoglyet cosides are produced by nocardioforms (or coryneforms; FRENCH al., 1973) it will be et interesting to see whether genes related to known strlsts genes in the streptomycin or bluensomycin producers are also present in the lower actinomycetes. Examples of such genes which could be found in the myomycin producer are those encoding enzymes of the following families (see Tab. 2): amidinotransferases (e.g., StrB1, StrB2), carbamoyltransferases (bluensomycin pathway), and those related to some of the enzymes involved in the synthesis of the cyclitol (e.g., StrO, StrN) and in other parts of the pathway, e.g., activation, synthesis, and transfer of the 3-aminohexose precursor (e.g., StrQ, StrS, or StsA), or involved in export and resistance phenomena (e.g., StrV, StrW, StrA). Genetic relationships to producers of antibiotics containing /3lysine tails (e.g., streptothricin; cf. Fig. A27; or viomycin) may also become apparent. The 3-guanidinomannose moiety in myomycin could be derived from an unusual NDP-glucose or NDP-mannose pathway similar to the NMLGA unit in streptomycin (cf. Sect. 4.1.1.3). Boholmycin. The last, structurally new class of aminoglycosides to be described is represented by the pseudotetrasaccharidic compound boholmycin (Fig. A7; SAITOHet al., 1988) and is produced by a strain of Streptomyces hygroscopicus. It clearly belongs to the Ca aminoglycosides according to our definitions. It is composed of a dicarbamoyl scylloinositol, two amino sugars condensed via glycosidic bonds to the 4- and 6-positions of the cyclitol, and a heptose (Ha-(4)Ca(6)-PA-
Hep). Its clear structural relationships with both myomycin (in the cyclitol(4)-3-aminohexose pseudodisaccharidic unit; see above) and seldomycins (in the cyclitol(6)-2-aminopentose unit; cf. Sect. 4.3) gives it a clear bridging role between the Ca and the (2DOS-)Cb aminoglycosides, and it will be interesting to see whether this is reflected by similarities at the genetidenzymic level. The incorporation of a D-mannoheptose makes it unique among the cyclitol-containing aminoglycosides and could be another indication for the hypothesis that the production of cell wall components of gram-negative bacteria (e.g., lipopolysaccharide) and many antibiotics in streptomycetes may be based on the same gene pool (PIEPERSBERG, 1992,1993).
Ca Aminoglycosides with Monoaminocyclitols. Three monoaminocyclitol aminoglycosidic antibiotics with no importance in pharmaceutical application but with interesting structures have been reported which should be mentioned here. (1) Minosaminomycin, a kasugamycin-related Ca(4)-6DOH product of Streptomyces sp., has already been mentioned above. Its unique structural feature is the presence of a 1-D-1-amino-1-deoxy-myo-inositolwhich a to histidinyl-valine dipeptide is bound via an amide bond (Fig. A5). The introduction of the amino group into the cyclitol could principally follow the same route as in the streptidine pathway, although with altered stereoselectivity in the steps catalyzed by the first-step dehydrogenase and transaminase. (2) Hygromycin A (Fig. A9) was detected prior to hygromycin B in the same strain of Streptomyces hygroscopicus in the early 1950s and was later also found as a product of other Streptomyces sp. and of Corynebacterium equi. It is also a Ca(l)-[X]-6DOH compound and is interesting in several respects: - it has an aminocyclitol unit, a 2-amino-neoinosamine, the derivation of which is totally obscure, but suggesting, however, that both a Ca and a Cb (for the biosynthesis of the 2DOS moiety in hygromycin B) pathway is functional in this strain, and that these two pathways can only coexist because of their completely different stereoselectivity of oxidationltransarnination steps;
'
429
two of the cis-hydroxyl groups are bridged by a methylene residue in the cyclitol moiety; - the furanosidic 5-ketod-deoxysugar could be another rare 6DOH unit derived from a dTDP-glucose (cf. Sects. 3.2.1 and 5.3).
-
(3) The aminoglycosides of the LL-BM123 series (Fig. A6), produced by Nocardia sp., are again compounds of biosynthetically mixed origin since they also contain amino acid residues such as minosaminomycin. Their unique pathway formula is Ca(4)-H(4)HA, where the disaccharide D-glucosaminyl(pl,4)-~-rnannoseis glycosidically linked to the 4 position of the 2-amino-2-deoxy-myoinositol moiety. The latter aminocyclitol should be derived from myo-inositol, which again can only be formed via an aminotransferase with different substrate selectivity relative to the StsC protein involved in the biosynthesis of streptidine (cf. Sect. 4.1.1.1).
4.2 Fortimicins, Istamycins

This group of very similar Ca(6)-HA or Cb(6)-HA compounds (Fig. A10) is widespread among filamentous actinomycete genera (cf. Fig. l), namely Micromonospora spp. (fortimicins: FTM; SF-2052), Dactylosporangium spp. (dactimicins), Streptomyces spp. (istamycins: ISM; sannamycins), and Saccharopolyspora spp. (sporaricins). In fact, the only major biosynthetic difference between the two groups containing either fortamine (FTM, dactimicins) or 2-deoxyfortamine (ISM, sporaricins, and sannamycins) as aminocyclitols seems to be the formation of the cyclitol via myo-inositol phosphate or scylloinosose, respectively (cf. Sect. 3.1). The FTM pathway has been most extensively studied in Micromonospora olivasterospora ATCC 21819 (FTM-A producer). It has been found to be almost congruent in substrate specificity with that of the producers of SF-2051 Micromonospora sp. SF-2089 ATCC 31580, dactimicin Dactylosporangium matsuzakiense ATCC 31570, sannamycin Streptomyces sunnanensis I F 0 14239, istamycin s. tenjimariensis ATCC 31603, and sporaricin Saccharopo-
lyspora hirsuta ATCC 20501 (Fig. 18 and Tab. 3; ITOH et al., 1984; ODAKURA al., 1984; et DAIRI and HASEGAWA, 1989; HASEGAWA, 1991, 1992; DAIRIet al., 1992a, b, c; OHTAet al., 1992a, b, 1993a, b; OHTA and HASEGAWA, 1993a, b; HOTTA et al., 1995). Of the roughly 20 steps of FTM-A biosynthesis, 14 have been identified by various methods, including induction and analysis of blocked mutants, gene cloning, and feeding of intermediates. Most of the production genes (fms) seem to be clustered on a DNA fragment of ca. 30 kb or more in M. olivasterospora. The order of identified genes in M. olivasterospora ATCC 21819 is fmsl0, 13, 3, 4, 5, 12, 8, 7, 14,1,11, (orf2), fmr0, (orf4). This DNA segment as a whole only hybridizes with the genomic DNA from Micromonospora sp. SF2089 and D. matsuzakiense (DAIRI et al., 1992b). However, the restriction pattern of the hybridizing bands were almost identical only in the strain Micromonospora sp. SF2089. In the DNA of the other three producers investigated no hybridization with the 30 kb fragment was observed, but when individual genes for conserved functions were taken as probes, e.g., the fmsl3 (smsl3; encoding the N-glycyltransferase) genes, significant hybridization was seen with the DNA from all six producers (OHTA et al., 1992b). Thus, it seems likely that all producers of FTM-like aminoglycosides contain highly related gene clusters originating from a common evolutionary source with some minor modifications, such as the use of a different pathway for the formation of the (2-deoxy)scyllo-inosose precursor. The difference between the two more distant groups (from the hybridization data; see above) is also reflected by the aquisition of two types of resistance genes: (l), f m r 0 (fmrM, fmrD, from M.olivasterospora ATCC 21819, Micromonospora sp. SF-2098, and Dactylosporangium matsuzakiense ATCC 31570, respectively); (2) fmrT f i r s , fmrH, from Streptomyces tenjimariensis ATCC 31603, S. sannanensis I F 0 14239, and Saccharopolyspora hirsuta ATCC 20501, respectively). Both encode members of the 16s rRNA methyltransferases but these enzymes methylate different residues (G-1405 and A-1408; CUNDLIFFE, 1989; HASEGAWA, 1991; OHTA
430

FTM-A0
myeinositd
scybinosamine D-glucosamine
1'2
(OR. TA)
(OR. MT) TA.
4.5
(GT)
3
OH
HjCHN
I 6
FTM-FU-10
OH
(OR. TA)
FTM-KL1
8.9
y 3
(3'-PT, DHslORs?)
H3CHN
FTM-AP
H3CHN
H3C
10
fH3
12
(MT)
OH
FTM-B
13
FTM-KR
FTM-KH
H3CHN fH3
(GLY)
14
T o 4y 3
(FIT)
OH
CHNh
H3C
HCH=NH
FTM-A
0%
OH
dactimicin
0'3%
Fig. 18. The fortimicin (FTM, astromicin) pathway. The same pathway starting from 2-deoxy-scyllo-inosamine seems to be established in istamycin/sannarnycin/sporaricin producers. The known intermediates (1992) and HO'ITA et al. and postulated enzymatic steps are given; for further details, see HASEGAWA (1995).
and HASEGAWA, 1993a, b; OHTAet al. 1993a, b; see Sect. 4.5). In each case the single resistance gene seems to reside in the production gene cluster. However, they seem to be differently organized in the gene clusters in both
groups (OHTAet al., 1993a). Thus, the set of genes used in the dactimicin producer could be a mixture of the two extreme evolutionary lines found in the other producers of the FTMlISM group aminoglycosides: (1) it con-
Tab. 3 Gene Products and Enzymes Known or Presumed to be Involved in the Production and Resistance of Forti .
Gene Producta Fmsl Fms2 Fms3 Fms4 Fms5 Fms7 Fms8 FmslO Fmsll Fmsl3 (Smsl3) Fmsl4 FmrO (FmrM, FmrD) FmrT KamC
Coding Capacity of Gene aab
Enzymatic Function or Step (Preliminary Assignment)
Remarksd
Organism
Mol, Msp, (Dma) Mol, Msp, (Dma) Mol, Msp, (Dma)
Mol, Mol, Mol, Mol,
Msp, Msp, Msp, Msp,
(Dma) (Dma) (Dma) (Dma) Mol, Msp, (Dma) Mol, Msp, (Dma)
(D-myo-Inositol 2-dehydrogenase) NAD(P)-dependent? (scyllo-Inosose aminotransferase) PLP-dependent? FTM-FU-10 synthesis (D-G~UCOSaminyltransferase) FTM-A0 synthesis FTM-A0 synthesis FTM-KKl synthesis FTM-AP synthesis (FTM-KK1 ATP-dependent, phosphotransferase) homologous to APH(3)-II FTM-KH synthesis FTM-KR synthesis (FTM-KH epimerase) FTM-B N-glycyltransferase FAD, 4-mer SAM-dependent SAM-dependent SAM-dependent (FTM-A synthesis?) Unknown (downstream f m r 0 ) (Epoxide hydrolase) Unknown (downstream f m r q
a
Mol, Msp, Dma, San, Shi Mol, Msp, (Dma)

Mol; Msp, Dma Ste, (San) Shi
Mol
N-Formimidoyl FTM-A synthase (oxidase) 16s rRNA (G-1405) methyltransferase 16s RRNA (A-1408) methyltransferase 16s rRNA (A-1408) methyltransferase Unknown (upstream f m r 0 )
(FTM-A synthesis?)
Mol
Ste Ste Cf. Fig. 18 In brackets: partial sequence data Cf. Fig. 18 P L P pyridoxalphosphate; SAM: S-adenosyl methionine
Mol: Micromonospora olivasterospora; M tylosporangium matsuzakiense; San: Strepto mariensis: Shi: SaCCharODOhSDOra hirsuta A HASEGAWA (199i) and DAIRI al. ( et
Unknown ORF (ORF-2) Unknown ORF (ORF-4) Unknown ORF (ORF-1) Unknown ORF (ORF-3)
(Sannamycin synthesis?) (Sarmamycin synthesis?)
432
10
Aminoglycosides and Sugar Components in Other Secondary Metabolites Fig. 10); (2) a 4'5'-dehydratase reaction as is suggested by the occurrence of 4'3'-dehydro-FTM-A; (3) a reductase/dehydrogenase) step reducing the 4',5 ' double bond.
tains probably a Cb pathway as the ISM (sannamycin) type producers; but (2) it has the resistance gene and profile as well as the stronger DNA sequence similarity to the FTM(SF2051) type producers. The last two steps in the formation of FTM-A (glycyltransfer) and FTM-C (N-formimidoylation of the glycyl amino group; cf. Fig. 18), catalyzed by the gene products Fmsl3 and Fmsl4, respectively, in M. olivusterosporu (Tab. 3) represent the biochemically best investigated phase of the biosynthetic pathway for FTM-like aminoglycosides. The genes for these two steps have been cloned, analyzed in part, and found to be present in all producers of FIM/ISM type aminoglycosides either by hybridization or by activity (DAIRI et al., 1992c; OHTA et al., 1992b). In a blocked mutant of the ISM producer S. tenjimuriensis FTM-B was converted into 1-epi-FIM-B, dactimicin, and l-epi-dactimicin; M. olivasterosporu in turn converted ISM-A. and ISM-Bo into ISM-A3 and ISMB3, respectively (cf. Fig. A1 0 HOTTAet al., 1989; DAIRI and HASEGAWA, 1989). The mechanism of the glycyl transfer and the putative activation of the glycyl residue (e.g., aminoacyl-AMP) has not yet been studied. The N-formimidoyl group was shown to be derived from glycine, the C-2 group of which is converted via an unusual oxidase mechanism to the formimidoyl group and probably COa in the presence of molecular oxygen only; this is catalyzed by the FAD-containing Fmsl4 enzyme (DAIRIet al., 1992~). Another interesting gene product is the Fms8 phosphotransferase, which probably catalyzes the 3 '-OH phosphorylation of the purpurosamine moiety in the FTM-KK1 intermediate. This enzyme is homologous to the APH(3') enzymes encoded by the nmrA and aph genes of neomycin-producing Micromonosporu sp. and Streptomyces fradiae, respectively, and can be replaced by the latter gene products (DAIRI et al., 1992a). However, its involvement in the interesting 3 ' ,4 '-dehydroxylation is unclear (cf. Fig. 18, steps 8, 9). For this phase of the pathway probably several steps are required: (1) dehydratation at C-3 'could occur via a mechanism similar to that operating in the 3,6-dideoxyhexose pathway in gram-negative bacteria (cf. Sect. 3.2.1 and
4.3 2-DeoxystreptamineContaining Aminoglycosides

The large and clinically important group of 2-deoxystreptamine(2DOS)-containing aminoglycosides was extensively studied with regard to its biogenesis in wild type and mutant strains using 14C-, I3C-, 3H-, and "N-labeled precursors, such as D-glucose, D-glucosamine, and 2DOS. The resulting data obtained mainly with the producers of neomycin, paromomycin, ribostamycin, butirosin, and the gentamicidsagamicin group antibiotics have been reviewed extensively (RINEHART and STROSHANE, 1976; PEARCEand RINEHART, 1981; KAKINUMA, 1982; KASE et al., 1982; OKUDAand ITOH, 1982; UMEZAWA al., et 1986; GRAFE, 1992). However, since about 1985 only very few new findings have been published. Therefore, only a brief summary of what is currently known is given here. The 2DOS moiety which is the basic building block in this family of compounds is made directly from glucose-6-phosphate via the Cb route (see Sect. 3.1; Fig. 19, cf. Fig. 4) as has been clearly demonstrated recently in the neomycin producer S. fradiae (YAMAUCHI and KAKINUMA, 1992b, c; 1993; 1995) and earlier postulated (KAKINUMA, 1982). Previously, the biosynthesis of the cyclitol moiety was obscure and was believed to occur either via myo-inositol or directly from glucose by an unknown mechanism (RINEHARTand STROSHANE, 1976). However, it was already known from the labeling pattern of the C1 and C, positions of glucose and from the incorporation of the first amino group, that 2deoxy-scyllo-inosamineis not derived from Dglucosamine and that the direction of the second-step transamination is opposite to that of the streptidine and actinamine pathways (PEARCEand RINEHART,1981; UMEZAWA et al., 1986). The first transamination step could be catalyzed by an aminotransferase very similar to the S. griseus StsC enzyme (see Sect. 4.1.1.1) since it was shown by WALKER
HO
boQ7
OH
OH
OH
D-Xyb-, (R = -NDP. -P, or -
Dglucosarnine NDP. -P, or -H)

I
NH2
OH
/
(5-0-rbosyl) pyranosyl ?)
paromamine
genlarnicin
NH2
HO
(y
b~
pseudotrisaccharide
gentamicin sagamicin, sisomicin, verdamicin G418
434
et al. (CHENand WALKER, 1977; LUCHER et al., 1989; WALKER,1995) that ketocyclitols were transaminated by the a-amino group of glutamine in extracts of the neomycin producer S. fradiae and the gentamicin producer Micromonospora purpurea as in streptomycin and spectinomycin producers. Therefore, it will be interesting to see whether stsC probes, e.g., from S. griseus, will detect a respective gene in the production gene clusters of 2DOS-producing actinomycetes. The later steps in the formation of 2DOS could also be closely related to the respective steps in the streptidine pathway (cf. Fig. 14), such as dehydrogenation (enzyme StsB- or StrI-like?) and second-step transamination (enzyme StsA-like?). It is unknown whether 2DOS precursors are phosphorylated at some stage or enter the further condensation and secretion steps in a form similar to streptomycin. The presence of aminoglycoside-3-phosphotransferases as resistance mechanisms in some of the 2DOS producers (see below, Sect. 4.5) would suggest, however, that the phosphorylation does not occur in the cyclitol moiety. All major 2DOS-containing aminoglycosides, except for the destomycin-hygromycin B group and perhaps also the apramycins, seem to be formed via a common pseudodisaccharidic intermediate, paromamine, which is formed as an early intermediate from 2DOS and a molecule of D-glucosamine, which is probably nucleotide-activated, via the hypothetical pathway outlined in Fig. 19. Whether the precursor formation and the attachment reactions of the D-glucosamine moiety somehow relate to that of the NMLGA unit of streptomycin (see Sect. 4.1) remains to be shown. Further on, the 2DOS pathways branch into various alternate routes resulting in the large variety of end products (see Figs. All-A17).
4,5-Disubstituted 2DOS Aminoglycosides. The compounds which are 4,5-substituted at the aminocyclitol, such as the neomycins, paromomycins, lividomycins (Fig. A1l ; basically Ha-(4)Cb(S)-P(3)-HA) or ribostamycins, and butirosins (Fig. A13; basically Ha-(4)Cb(5)P), are probably synthesized directly from paromamine or via a second pseudodisaccharide, neamine (Fig. 19), followed by the at-
tachment of a furanosidic pentose, xylose (forming xylostasins), or ribose (forming ribostamycins) which finally can be further glycosylated by a 2,6-diamino-2,6-dideoxyhexose (neosamine B or C) in the neomycins and paromomycins. The pseudodisaccharides are clearly intermediates and can be converted directly to the respective end products in mutants blocked in their formation, e.g., in the biosynthetic pathway of the aminocyclitol. Neamine (neomycin A) has measurable antibiotic activity. The pathway of the neosamines B and C is as obscure as that of the first hexosamine unit, though it is clear that D-glucosamine is preferentially incorporated into them. The type of activation, the glycosyltransferase(s), and the route of incorporation of a second amino-N into the C, position of the 2-aminohexose remain to be established enzymatically. A similarity to the situation in streptomycin-producing S. griseus could exist also here: UDP-(N-acetyl-)D-glucosamine might not be an immediate precursor as in the cell wall biosynthesis. Therefore, it will be necessary to screen also for a StrQ-related nucleotidylating enzyme among the gene products involved in biosynthesis of neomycin-like aminoglycosides. Important in this context are also the results of more recent biogenesis studies with l-13C- and 6-13C-labeled glucose, which suggest that all building blocks of neomycin seem to be formed from intermediates of the pentose phosphate cycle ,(Fig. 20; RINEHART al., 1992). If this et equilibration of all or most glucose molecules through the pentose phosphate cycle generally occurs for all hexose and other carbohydrate components in secondary metabolites within the production phase, the earlier results of isotope-labeling studies would have to be reinterpreted.
4,6-Disubstituted 2DOS Aminoglycosides. The kanamycins-tobramycin (Fig. A 1 2 HA(4)Cb(6)-HA), the gentamicins-sagamicin (Fig. A14; HA-(4)Cb(6)-PA), and probably also the seldomycins (Fig. A17; HA-(4)Cb(6)PA) groups of 2DOS-containing aminoglycosides are also synthesized from paromamine. However, their further substitution at the C6 position of the cyclitol and the nature and modification of the second glycosidic residue
435
Fig. 20. Biogenesis of the hexose-, pentose- and cyclitol-derived components in neomycin from 6-(CL3)-~glucose (according to RINEHART et al., 1992). The labeling patterns measured by (C)-NMR prove that all units are built up preferentially from C, (0), or C3 C2, units (thick lines) rearranged by transketolase- and transaldolase-catalyzed reactions in passages through the pentosephosphate cycle. The earlier labeling patterns obtained with 1- (*) or 1,6-labeled (UlA) D-glucoses are also given from which a differentiation between direct or indirect incorporation of glucose was not possible.
0
OH
OH
HO
OH
HO
distinguish the gentamicins from the seldomycins in which a pyranoid C-5 sugar moiety is replaced by a hexosamine. This sugar in the gentamicin-related compounds is D-xylose (the pseudodisaccharide unit formed with the 2DOS cyclitol is called garamine) and could be either D-xylose or 2-amino-2-deoxy-~-xylose in the seldomycins. Since the gentamicins are only produced in Micromonospora spp., and the seldomycins are only found in Srrepromyces spp. it will be of interest to study the mutual relationships between the respective sets of biosynthetic genedenzymes relative to those involved in the production of the other groups of 2DOS-containing aminoglycosides. The most intensive study on the design of an individual 2DOS pathway was carried out on that of the gentamicin(GM)-sisomicin-sagamicin group (KASEet al., 1982; cf. UMEZAWA et al., 1986) which is one of the most versatile with more than 20 end products identified. Some details should be mentioned for comparison (for formulae see Fig. A 14): (1) a minimum of 20 enzymes is probably involved in the biosynthesis of, e.g., GM-C,; (2) many of the steps, e.g., dehydrogenation and transamination in positions 3 and 6 of pyranoses,
or N-methylation and dehydration, resemble those of other aminoglycoside pathways; (3) the multiply branching pathway proceeds from the first trisaccharidic intermediate GMA2 to GM-X2 and then branches to yield the two unsaturated intermediates sisomicin (via JI-20A) and verdamicin (via G-418, a compound now frequently used for cloning vector selection in plant and animal cells). In some strains these two intermediates can already be the major end products which are released. This biosynthetic phase strongly resembles the 3,4-dehydroxylation steps in the fortimicin pathway (cf. Fig. 18; steps 8, 9); however, the occurrence of a 3,4-unsaturated 2,3,4,6deoxy-2,6-aminohexose indicates that dehydratation (probably of a 4-hydroxylated precursor after 3-dehydroxylation; cf. LIU and THORSON, 1994) is a step in this process (see also Sect. 4.2). The last phase of GM biosynthesis yielding the final products GM-C,, GM-C,,, GM-C2, and sagamicin, which are mainly used in therapeutics variably involves reduction, epimerization, and N-methylation steps, probably by action of the same enzymes on different but structurally related intermediates.
436
10
Aminoglycosides and Sugar Components in Other Secondary Metabolites
Apramycins and Destomycins. A more distant 1-N-amidinostreptamine-6-phosphate interposition relative to the other 2DOS aminogly- mediate (cf. Fig. 14) could be used. The streptcosides is taken by apramycin (Fig. A15; amine unit in SS-56-C could be formed directCB(4)-OctA(S)-HA) and destomycin-hygro- ly from scyllo-inosamine or via hydrolysis of mycin B (Fig. A 1 6 Cb/Ca(S)-H(2,3)-HepA) the l-N-amidino group from l-N-amidinogroup. In the apramycins monosubstituted in streptamine. This also could mean that 6the 4-position of the 2DOS moiety parom- phosphate intermediates of destomycins are amine could be an intermediate. The further formed inside the cells as in the case of the pathway would then require an elongation at streptomycin producers. In summary, compared with the streptomythe 6'-hydroxymethyl group of the hexosamine moiety by a C-2 unit which, however, cins and the fortimicins the 2DOS aminoglyseems rather unlikely. An alternative route cosides mentioned so far seem to take an inwould be the condensation of 2DOS with an termediate position with respect to the distrioctose derivative, which is reminiscent of the bution of modifying steps relative to the conunusual octose pathway in the lincosamides densation reactions. The modifications are (cf. Sect. 5.1; RINEHART, 1980; see also the practically all finished before condensation in discussion in OKUDAand ITO, 1982). Also, streptomycins, but in the 2DOS antibiotics moiety glyco- much more take place after condensation by the 4-amino-4-deoxy-~-glucose sidically linked to the 8' position is quite unu- glycosyltransfer reactions by which only the sual among the aminoglycosides; 4 transami- diaminocyclitol is completely preformed. nation also occurs in the kasugamycin-related compounds, but on a 6DOH unit (see Sect. 4.1). This unit could be synthesized by trans- 4.4 Other Aminoglycosides amination of a NDP-4-ketoglucose intermediate formed by an enzyme related to the Monomeric sugar derivatives. The number UDP-glucose 4-epimerase (cf. Sect. 3.2). In of bioactive and stable molecules directly dethe destomycins and hygromycin B the 2DOS rived from monosaccharidic units or sugar moiety is 5-substituted with an unusual hex- analogs described in the literature is increasose, D-talose. This in turn is fused via a ing steadily (Fig. A18; cyclitol-related comunique type of chemical bonding, an ortho-es- pounds are treated below; cf. Sect. 5.2). Some ter linkage between the 2'- and 3'-hydroxyls examples are mentioned here: (1) A group of glycosidase inhibitors, such and the 1"-position of a sugar acid derived from a 6-amino-6-deoxyheptose, destomic as the nojirimycins (inhibit glucosidases and acid. These structural details suggest that mannosidases), galactostatin (inhibits galactothere is very little resemblance between the sidases), and siastatin (inhibits sialidases), are pathways of hygromycin B and destomycin sugar analogs with a substituted pyridine ring production and those of other aminoglyco- and can be regarded as a group of bacterial sides, except for the biosynthesis of the ami- alkaloids derived from aminohexoses or aminocyclitol. Interestingly, strains of Strepto- nopentoses (GRAFE, 1992). The sugar-like 1myces eurocidicus and of Saccharopolyspora deoxynojirimycin (DNJ) and a similar struchirsuta produce the destomycin derivatives, tured plant alkaloid (castanospermine; not shown) can act as anti-HIV drugs by preventSS-56-C and l-N-amidino-l-N-demethyl-2hydroxydestomycin A (INOUYE al., 1973; ing the maturation of the gp120 envelope glyet IKEDAet al., 1985b; cf. Fig. A16), respective- coprotein. This finding motivated a recent ly, which are clearly Ca compounds and are study of the biogenesis of these unusual amirelated to streptidine in their diaminocyclitol no sugars by stable isotope labeling in Strepet moieties. Here, a DNA recombination event tomyces subrutilus (HARDICK al., 1992). It resulting in the fusion of two gene clusters, was found that DNJ is derived from a glucose those for streptomycin and destomycin pro- molecule converted first via fructose, 6-oxidaduction, could have created a new mixed tion and reductive 2- or 6-transamination pathway. From the streptomycin gene cluster steps to mannonojirimycin (Fig. 21). This inthose genes needed for the production of the termediate can then be dehydrated and re-
437
HO +OH
'YOH
tOH kHpOH
oxidation/
%H*OH
toH
1
transamination oxidation
[ '(1
cyclisation
transamination
/
Or
OCHpOH
OCHpOH
OH
" I O H OH
]
DCH,OH
OH
%H~-NH,
A - 0
II
- H
epimetisation
CHIOH
1. dehydration 2 . r e y *
OH
OH
NJ
DNJ
also been prepared, two of which, miglitol and emiglitate (Fig. AM), have reached the phase of clinical development (MULLER, 1989). (2) The recently described D-glucosamine derivative CV-1 (Fig. A18), which is produced by a Streptomyces spp. and is a weak antibiotic by itself, has a very interesting cooperative effect together with spiramycin on gram-negative bacteria (ICHIMURA al., et 1987). The inhibitory effect of CV-1 was demonstrated to be on LPS synthesis in E. coli thereby relieving the barrier effect of the outer membrane for spiramycin which by itself is ineffective on gram-negatives. The biosynthesis of CV-1 involves N-carbamoylation of Dglucosamine, probably from L-citrulline, which conceivably occurs either on a l-phosphate- or nucleotide-activated precursor or the free sugar. Subsequent reorganization of N-carbamoyl-D-glucosamine to the unique open ring hemiaminal was shown to proceed spontaneously (YASUZAWA al., 1987). et (3) Valiolamine (cf. Fig. A21) is a new monomeric aminocyclitol member of the validamycin group of aminoglycosides (KAMEDA et al., 1984; see below) and has inhibitory activity against a-glucosidases.
Trehalosamines and Other Aminodisaccharides. A larger group of nitrogen-containing and carbohydrate-related actinomycete products are compounds with structural analogy to the disaccharides trehalose or saccharose. Most of these have some biological activity either as antibiotics or as glycosidase inhibitors. The &,a-glycosidic trehalosamines (Fig. duced to 1-deoxymannonojirimycin or, alter- A19; HA(1)-H) are known for a long time natively, epimerized at C-2 to nojirimycin and and were isolated on account of their antibacsubsequently dehydroxylated to DNJ. Thus, terial activity which, however, is only weak et several enzymedgenes related to those used (cf. UMEZAWA al., 1986; ASANOet al., in the formation of other amino and/or deoxy 1989). The biosynthesis of 2-trehalosamine sugar components in known antibiotics could and mannosyl glucosaminide from a molecule be used. The same might hold for other each of D-glucosamine and either D-glucose monomeric amino sugars isolated from cul- or D-mannose, respectively, by enzymes retures of microorganisms such as 3-amino-3- lated to trehalose synthases can easily be endeoxy-D-glucose, N-carbamoyl-D-glucosamine, visaged. In contrast, the derivation of 3- and prumycin, and streptozotocin (Fig. A18; see 4-trehalosamines which also occur as streptoalso the compilations in UMEZAWA al., mycete products could follow at least two difet 1986; GRAFE, 1992). Semisynthetic deriva- ferent routes: (1) formation of an activated 3tives of DNJ with improved inhibitory activity or 4-aminohexose which is condensed with Don a-glucosidases and pharmacokinetics have glucose or (2) modification of preformed tre-
Fig. 21. Proposed biogenesis of nojirimycin and related monosaccharide analogs. (D)NJ: (1-deoxy-)nojirimycin; (D)MJ: (1-deoxy-)mannonojirimycin. The labeling patterns obtained from 1,6-labeled D-glucose was adopted from RINECHART al. (1992) (cf. Fig. 20). et
438
halose. It would be interesting to know whether these compounds have (auto-)regulatory functions in adaptive processes, such as osmoregulation, cell differentiation, or secondary metabolism; trehalose metabolism could play an important role in those physiological phenomena in streptomycetes (CHAMPNESS and CHATER, 1994). A new member of the trehalosamine family is the qP-glycosidic antibiotic 3,3'-neotrehalosadiamine (Fig. A19; HA(1)-HA) isolated from a Bacillus pumilus for its antibacterial activity on an aminoglycoside-hypersensitivestrain of Klebsiella pneumoniae (NUMATA al., 1986; et TSUNO al., 1986). et The 3-amino-3-deoxy analogs of trehalose and sucrose and their epimers, T-I, T-I1 (3trehalosamine), and T-I11 or S-I and S-I1 (3sucrosamine) should be regarded as members of this family of compounds (Fig. A20; ASANO et al., 1989). These have been obtained by
chemo-enzymatic synthesis from the 3-keto forms of trehalose and sucrose, which were prepared by treatment with D-glucoside 3-dehydrogenase from Flavobacterium saccharophilum, and by reductive chemical amination. Except for S-I, these compounds are weak antibiotics; S-I is an inhibitor of invertases. Validamycins, Acarbose, and Related Oligosaccharidic Aminoglycosides. The cyclitol moieties formally derived from valiolamine, such as valienamine or validamine (cf. Figs. 6 and 22), are most probably products formed via a Cb pathway and may be regarded as aliphatic m-C7N units, though they are not synthesized using erythrose-4-phosphate and phosphoenolpyruvate as the aromatic m-C7Nunits (cf. Sect. 3; FLOSSand BEALE,1989; RINEHART et al., 1992). Thus, the suggestion that valienamine is not derived from the shikimate pathway (DEGWERT et al., 1987)
13
0-Ph
(-HzOO)
(R-NHd
R=O)
HO
valienamine
HO
(-JoHoo 4
CH2OH HO HO
OH
OH
OH
validoxylamine A validamycin A
validoxylamine B validamycin B
validoxylamineG vaiidamycin G
Fig. 22. Proposed scheme of biogenesis of C7 sugar-derived cyclitols of the validamycin family. The labeling pattern from 1,6- Or 6-(C'3)-~-glucoses adopted was from RINEHART al. (1992) (cf. Fig. 20). et
439 PAPE,
probably hold true only for the origin of the C7 precursor (which is either sedoheptulose7-phosphate or an epimer thereof). However, at least the cyclization and dehydration steps are probably catalyzed by enzymes homologous to dehydroquinate synthase and dehydroquinate dehydrase, respectively. This type of cyclitol is found in validoxylamines and validamycins (Fig. A21; basically Cb(1)Cb(4)-H compounds), and in acarbose, amylostatin, and other structurally related glycosidase inhibitors (Fig. A22; general formula (H),-Cb( 1)-6DOH/H(l)-(H),; TRUSCHEIT et al., 1981; MULLER, 1989; YOKOSE et al., 1989). Valienamine and the related C7cyclitols are all condensed via an imino group with either another molecule of the same origin and similar structure (validamycin family) or at the 4 position with a 4,6-dideoxy-~-glucose or 4-deoxy-~-glucoseunit (acarviosine family of a-glucosidase and trehalase inhibitors). This suggests that the aminotransfer reaction takes place on a precursor of the C, cyclitol moiety and that valienamine might be an intermediate for all other derivatives (cf. Figs. 6,21 and A21, A22; cf. RINEHART al., et 1992). or The validamycins are a-D- P-D-glucosylated at various positions (Fig. A21) which could be achieved by extracellular or cellwall-associated glucosyltransferases. Also, the acarviosine family of glycosidase inhibitors, comprising acarbose, amylostatins, oligostatins, amino-oligosaccharides (epoxy derivatives of amylostatins), adiposins, trestatins, and AI-5662, are all variably glycosylated, mainly by condensation with di- or oligosaccharidic units such as maltose, oligomaltodextrins, trehalose, or an acarviosine units (TRUSCHEIT al., 1981; MULLER, 1989). et Again, the mechanism of addition of such blocks of oligosaccharidic sugars could be a membrane-associated extracellular process similar to the bactoprenol-dependent formation of extracellular heteropolysaccharides, e.g., lipopolysaccharide 0-chains (GOEKE, 1986; RAETZ, 1996). Alternatively, these compounds could be excreted as inactive precursors by special exporters similar to streptomycin (cf. Sect. 4.1.1.4). This view is supported by the recent identification of an acarbose 7-phosphotransferase in the producing
Actinoplanes sp. (DREPPER and 1996).
Trehazolin. A new type of aminoglycosidic and very specific trehalase inhibitor, trehazolin (or trehalostatin), a product of Micromonospora sp. and Amycolatopsis sp., and the chemical synthesis of this compound and its P-anomer has recently been reported (ANDOet al., 1991; KOBAYASHI SHIOZAand KI, 1994; Fig. A23; Ca(l,2)-HA). The aminocyclitol moiety in this molecule is unusual in that it could be formed in a Ca pathway (cf. Fig. 4) via myo-inositol and later ring contraction as is suggested for pactamycin (see Sect. 5.2) or by direct reductive ring closure from a ketohexose precursor, e.g., fructosed-phosphate. Also, the presence of a carbamoyl group is reminiscent of other groups of aminoglycosides; the incorporation of this group may involve enzymes similar to those used in pathways of other aminoglycosides. However, the trehazolin molecule exhibits the only known true aminoglycosidic linkage in its bonding to the am-glucose moiety. Sorbistins. The pseudodisaccharidic aminoglycosides of the sorbistin family (Fig. A24) are interesting in two respects: they are produced both by pseudomonads and higher actinomycetes, and they contain an unusual open-chained diaminohexitol, 1P-diaminosorbitol, of unknown biosynthetic origin (reviewed in UMEZAWA al., 1986). It is temptet ing to speculate that these compounds could be derived from a pathway similar to that of the fortimicins followed by a reductive ring cleavage between the C-1 and C-2 positions of an aminocyclitol similar to fortamine later in the pathway (cf. Sect. 4.2; cf. Fig. 18). In fact, a nonacylated and nonmethylated fortamine precursor could be the direct precursor of 1,4-diaminosorbitol with an additional epimerization in position c-5 (C-3 of fortarnine) creating the stereochemical configuration of sorbitol. The presence of a 4-amino-4-deoxyglucose can be explained in two ways, as for the biogenesis of 4-amino-4-deoxytrehalose (see the discussion on trehalosamines above). In accordance with the similarity to the fortimicin pathway a preformed (NDP-)Caminoglucose would be the preferred precursor.
440
producers (WALKER,1975b; COURVALIN et al., 1977) and clinical isolates of gram-negative and gram-positive bacteria, such as obligate pathogens or from opportunistic infections, where they are mostly plasmid-determined (BENVENISTEand DAVIES, 1973; UMEZAWA, 1974; DAVIESand SMITH,1978 FOSTER,1983), were the first among all antiAllosamidin. A new aminoglycosidic product biotic resistance phenomena to be identified of Streptomyces spp., allosamidin and its de- at the molecular level. In fact, the first resistmethyl and didemethyl derivatives (Fig. A25; ance gene cloned from an antibiotic producer Ca(4)-HA(4)-HA), is the first chitinase inhib- was the gene encoding butirosin-3 -phosphoitor isolated from actinomycetes (SAKUDA et transferase, APH(3 )-IV, from Bacillus circuet al., 1987; ZHOUet al., 1992, 1993). Allos- lans (COURVALIN al., 1977). Also, the coramidin is composed of unusual components, a relation between the phosphorylation of branched five-membered C6 aminocyclitol streptomycin and its occurrence in a strepto(allosamizoline) and two N-acetyl-D-allos- mycin producers, S. griseus (MILLER and 1969; NIMI et al., 1971) and in cliniamine units linked via p-1,4-glycosidic bonds WALKER which could be synthesized via a novel Ca cal strains of Escherichia coli and Pseudopathway and via 3-epimerizations from monas aeruginosa (UMEZAWA al., 1967a, et (N-acety1)-D-glucosamine, respectively. The b), for the first time led to the speculation exact route of the synthesis of the allosami- that transferable antibiotic resistance could in zoline cyclitol is unknown, though its C and N general have evolved in the producers of and atoms were shown to be directly derived from these natural products (BENVENISTE D-glucosamine. Therefore, it could either be DAVIES,1973). Later, ribosomal target site synthesized from myo-inosamine formed by modification via specific 16s rRNA methylaan enzyme analogous to myo-inositolphos- tion was also identified as a major aminoglyphate synthase via successive ring contrac- coside resistance mechanism in producers tion, oxidoreduction/epimerization, and (PIENDL et al., 1984; CUNDLIFFE, 1989, transamination steps. Alternatively, it could 1992a). For the first time, the transfer of a be directly cyclized from ~-glucosamined- gene between gram-positive and gram-negaphosphate via a Ca type (or other) enzyme tive bacteria in nature was demonstrated to forming pentacyclitols by activating C-5 and occur with an aminoglycoside resistance gene: linking it to C-1. The aminooxazoline ring in the gene aphA-3 (kanamycinheomycin 3 the allosamizoline moiety (Fig. A25) is proba- phosphotransferase) was transferred between bly introduced via N-amidinotransfer from L- enterococci (also streptococci and staphyloarginine, N-monomethylation, cyclization, cocci) and Campylobacter coli (reviewed in 1994). Some of the aminoglycoand a second N-methylation step (ZHOU et COURVALIN, al., 1993). Therefore, also in this pathway sev- side-resistance genes might have spread from eral steps could be catalyzed by enzymes re- aminoglycoside producers to other bacteria in earlier evolutionary periods, the mechanisms lated to strlsts gene products. of which are now becoming apparent (MAZODIER and DAVIES, 1991; COURVAIJN, 1994). In this context, it is also interesting that 4.5 Resistance in Aminoglycoside related multifactorial systems exist for the inProducers tercellular transport of DNA and protein molecules which function specifically between Interestingly, aminoglycoside and in parti- bacteria and other, unrelated types, such as cular streptomycin resistance mechanisms by cells of other bacterial genera, and plant and mutation (OZAKI et al., 1969; WALLACEet animal cells (POHLMAN al., 1994). et al., 1979; PIEPERSBERG al., 1980), and et Much progress has been made in the analythose encoded by specific resistance genes in sis of resistance mechanisms in bacterial proThis could be synthesized via reactions similar to the initial steps postulated for the biosynthesis of the NMLGA moiety of streptomycin (cf. Sect. 4.1; cf. Fig. 16), whereby a convenient 4-ketohexose intermediate is formed which could be used as an aminotransferase substrate.
441
ducers of aminoglycosides and other carbohydrate-containing self-toxic compounds (Tab. 4, Fig. 23). Our present knowledge can be summarized as follows: (1) There are two basic resistance-conferring biochemical phenomena: first, specific elimination of the inhibitory function inside the producing cell (e.g., inactivation by modification, modification of the target site, or the production of a new, insensitive version of the normal target complex) and second, active transport of the inhibitor out of the cells via energy-driven exporters. The first type could be regarded as a mere self-protection mechanism which employs protective chemical groups or bypass mechanisms, whereas the second has its main function in transporting the compounds out of the cells in order to enable the bioactive end products to reach their natural destinations, namely other cells. We only can speculate on the natural functions of these compounds and the nature of the target cells. In general these target cells could be other cells of the same organism (hormonelike functions: the target cells would be either nonproducing or otherwise differently differentiated stages of the life cycle) or cells of other, e.g., competitive, organisms (DAVIES et al., 1992; PIEPERSBERG, 1993). (2) Active exporters have been identified so far only for antibiotics such as macrolides, anthracyclines, tetracyclines (see Tab. 4), but not for aminoglycosides. Interestingly, two genes, strV and strW, have recently been detected in the streptomycin production gene clusters of both S. glaucescens and S. griseus; these genes encode a new type of ABC transporters (PIEPERSBERG, 1995; BEYER et al., 1996). Since streptomycin is secreted in an inactive, phosphorylated form (WALKER, 1975b; MANSOURI and PIEPERSBERG, 1991; PIEPERSBERG, 1995) these transmembrane exporters, if responsible for secretion, would not give rise to a resistance phenotype (RETZLAFF al., 1993). There is evidence in et support of this hypothesis: S. lividans 66 strains carrying a combination of the strA [APH(6)] and strVW transcription units on plasmids convert added streptomycin to an extracellularly accumulated streptomycindphosphate (unfortunately those clones turned out to be very instable; S. BEYERand W. PIE-
PERSBERG, unpublished data). Therefore, the above view would be additionally supported by the possible existence of an active export system, also for antibiotically inactive precursors of aminoglycosides from the producing cells. It will be of further interest to investigate the presence of similar transporter genes in other aminoglycoside production gene clusters. A different type of membrane-anchored protein has recently been identified as the product of the butB gene in Bacillus circulans NRRL B3312 (AUBERT-PIVERT and DAVIES, 1994; HOTTA et al., 1995). The ButB protein is related to the cell-wall-associated Slayer proteins in low-G C gram-positive bacteria and interruption of its gene blocks butirosin production, thus indicating that. it might also be involved in aminoglycoside export. (3) Some of the aminoglycoside-resistance mechanisms listed in Tab. 4, e.g., phosphorylation and acetylation, are also among those which could give us the biochemical basis for an understanding of the evolution of this type of resistance determinants. These are suspected to be derived from biosynthetic enzymes or serve both purposes at the same time, such as the pac and bar genes encoding puromycin and phosphinothricin acetyltransferases, respectively (WALKER, 1975a, b; PIEPERSBERG et al., 1988; CUNDLIFE, 1992a; et THOMPSON SETO,1995; TERCERO al., and 1996). Again, it was the biosynthetic pathway of streptidine which first supported this speculation since intermediates in this pathway are successively dephosphorylated and rephosphorylated twice, the last time at the same position (C-6 of the aminocyclitol) as is phosphorlyated by the resistance enzyme, StrA (AphD or APH(6); see below), in streptomycin producers (cf. Fig. 17; WALKER, 1975b). Since this enzyme also has a phosphorylating activity for streptidine and its immediate precursor, N-amidinostreptamine (WALKER,1975b; however, at much higher K M values; DISTLER PIEPERSBERG, and 1985) it was suggested to be also a biosynthetic enzyme. However, among the gene products for streptomycin production there are two, StrN and StsE, with peptide motifs similar to those of the catalytic centers of antibiotic and protein kinases, especially the H X D X ~ N X , - ~ ~ U D
Tab. 4. Resistance Mechanisms against Glycosidic Antibiotics in Producers (Examples)
Antibiotic
Resistance Mechanism"
Producing Organismb S. griseus, S. glaucescens S. fradiae M. chalcea S. rimosus forma paromomycinus S. ribosidificus S. kanamyceticus S. hygroscopicus S. tenebrarius M. purpurea M. olivasterospora S. tenjimariensis S. kasugaensis S. spectabilis S. pactum S. lincolnensis S. alboniger S. peucetius Saccharopolyspora erythraea S. fradiae S. sphaeroides
Reference A A, G B A A A A A C D A G G A E A F A A A
Streptomycin Neomycin Neomycin Paromomycin Ribostamycin Kanamycin Hygromycin B Nebramycin
Gentamicin Fortimicin (astromicin) 1stamycin Kasugamycin Spectinomycin Pactamycin Lincomycin Puromycin Daunorubicin Erythromycin Tylosin Novobiocin
SM-6-PhT; (SM-3 "-PhT); (export: ABC?) neomycin-3-AcT, ne0m.d '-ACT, neom.-3 '-PhT neomycin-3-ACT paromomycin-3-AcT, par.-3 '-PhT ribostamycin-3-AcT, rib.-3 '-PhT 16s rRNA MT (G-1405) hygromycin B-7-PhT nebramycin-3-AcT, par.-3 '-PhT 16s rRNA MT (A-1408, G-1405); SM-6-PhT 16s RRNA MT (?) (16s rRNA MT?) 16s RRNA MT (A-1408) kasugamycin-2'-ACT fortimicin/istamycin-2'( +2 ")-ACT spectinomycin-2" -ACT;spectinom-?-PhT 16s rRNA MT (A-964) 23s rRNA MT (?); export:ABC; export:ApH puromycin-ACT, export:ApH(?) export:ABC 23s rRNA MT (A-2058); export:ABC; export:ApH 23s rRNA MT (A-2058); export:ABC, export:ApH res. target enzyme: DNA gyrase
a ACT:acetyltransferase; M T methyltransferase; P h T phosphotransferase; A B C ATP-dependent transporter family; ApH: pH gradient-depende transporter family S.: Streptomyces; M.: Micromonospora ' A: CUNDLIFFE (1989); B: SALAUZE al. (1992); C: PIENDLet al. (1984); D: HASEGAWA et (1991); E: ZHANGet al. (1992; and unpublish and (1991); G: HOTTAet al. (1995) observations); F GUILFOILE HUTCHINSON
443
biosynthesis
r*
Y
Fig. 2 . Schematic representa3 tion of self-protecting resistance mechanisms in producers of antibiotically active secondary carbohydrates.
modification
=
3 active export (ATP-driven)

4
@,a activehnactive secondary metabolite
2 resistant target site (ts)
active export (ApH-driven)

+=
modifying group
(4) Resistance to aminoglycosides also oc(U = hydrophobic residue, e.g., I, V, L) motif probably involved in the phosphate group curs as a cryptic phenotypic property in strepet transfer (PISSOWOTZKI al., 1991; RETZ- tomycetes, such as a second streptomycin resistance enzyme (streptomycin-3 -phosphoLAFF et al., 1993; KNIGHTON al., 1991; cf. et also Sect. 6). These are more likely to be the transferase) and kanamycin resistance entwo phosphotransferases involved in the zyme (kanamycin-3-N-acetyltransferase) in S. et streptidine pathway (see Sect. 4.1.1.1). Re- griseus (HEINZEL, al. 1988; HOITA et al., cently, a spectinomycin resistance gene en- 1988). It can be used either as a basis for the coding a spectinomycin phosphotransferase, screening for new producers of aminoglycoet was cloned from S. flavopersicus NRRL 2820 side-like antibiotics (ETIENNE al., 1991) or which showed striking similarity to StrN of S. to stimulate aminoglycoside production, e.g., griseus and S. glaucescens (J. ALTENBUCH-the 6-acetyltransferase gene from S. kanaNER, D. LYUTZKANOVA, J. DISTLER, myceticus stimulates aminoglycoside producand personal communication). This finding tion when several copies are introduced into strongly supports the hypothesis that resist- kanamycin and neomycin producers (CRA1986). ance genes originate from biosynthetic genes. MERI and DAVIES, Here, this hypothesis also has to be extended to include the possibility that an ancestral production gene could become a resistance gene by divergent evolution or after degener- 4.6 Regulation in Streptomycetes ation and modification of a pre-existing pathThe synthesis of aminoglycosides like the way to yield simpler but still effective end products (this could be the case for spectino- production of other secondary metabolites in mycin; see Sect. 4.1). The kanamycin-6-ace- Streptomyces is regulated by a complicated tyltransferase in S. kanamyceticus could also network of regulator proteins encoded by the be involved primarily in the acetylation of ka- individual biosynthetic gene clusters and othnamycin precursors for several reasons (CRA- er pleiotropic regulatory genes. Besides these regulatory genes, hormone-like autoregulaMERI and DAVIES, 1986; see also below).
444
tors (e.g., A-factor), intracellular signal molecules (e.g., ppGpp), and nutrients (e.g., phosphate, glucose, N sources, etc.) control the production of aminoglycosides at the level of gene expression and/or enzyme activity. Mainly for the streptomycin producing S. griseus as a model system we will here describe some of the most important details known or postulated (summarized in Fig. 24). The depression of the production of aminoglycosides by glucose has been reported for streptomycin, kanamycin, istamycin, and neomycin (VININGand DOULL,1988; DEMAIN, 1989) and is caused by the repression of antibiotic synthases (e.g., N-acetylkanamycin aminohydrolase; DEMAIN, 1989) rather than by an influence on the formation of precursors of secondary metabolites. Carbon catabolite control mechanisms extensively studied in other bacterial systems such as E. coli and Bacillus subtilis (reviewed by FISCHER,1992) typically use cAMP/CAP-mediated gene regulation. Although cAMP relieves glucose repression of N-acetylkanamycin aminohydrolase in S. kanamyceticus (SATOHet al., 1976) there is only weak evidence for the involvement of cAMP in regulating aminoglycoside production in Streptomyces. The intracellular cAMP concentration falls sharply in the mid to late vegetative growth phase of S. griseus S104 before the onset of secondary metabolism (RAGAN and VINING,1978). Glucose increases cAMP levels in S. antibioticus while simultaneously repressing oleandomycin formation (LISHNEVSKAYA al., 1986). There et is no evidence that cAMP directly affects streptomycin production (NEUMANN al., et 1996) or controls secondary metabolism in general (MARTINand DEMAIN, 1980). It is more likely that the carbon source repression is mediated by glucose kinase. Glucose kinase-deficient mutants of S. coelicolor having normal intracellular levels of cAMP do not exhibit a glucose repression phenotype (ANGELL et al., 1992; KWAKMAN and POSTMA, 1994). In contrast to these results, glucose-insensitive mutants of S. kanamyceticus blocked in kanamycin production were isolated which have wild-type levels of glucose kinase (FLORES al., 1993). Thus, a phosphorylated et sugar could mediate carbon source repression of antibiotic production (DEMAIN, 1989).
There is conflicting evidence in the literature concerning the influence of the nitrogen source on the production of streptomycin and other aminoglycosides. Nevertheless, ammonium seems to impede the synthesis of streptomycin, neomycin, and kanamycin (SHAPIRO, 1989) whereas nitrate and some amino acids support the production of aminoglycosides, e.g., kanamycin (BASAK and MAJUMDAR, 1973) and streptomycin. The stimulation of streptomycin production by alanine, arginine, andlor glutamine can be explained in terms of their being direct donors of nitrogenous groups in enzyme-catalyzed steps during streptomycin biosynthesis (cf. Sect. 4.1). The mechanisms underlying the positive effect of asparagine and proline (SHAPIRO,1989) or the repression of streptomycin synthesis by valine (ENSIGN, 1988; NEUMANN al., 1996) et in S. griseus are presently not understood. These compounds could influence the expression of biosynthetic key enzymes at the genetic level. This influence could be mediated by a NtrB/NtrC-like system or by a modulation of the activities of primary metabolic enzymes and streptomycin biosynthetic enzymes at the physiological level via metabolite accumulation and excretion by, e.g., feedback repression or induction of gluconeogenetic pathways or metabolic conditions favoring special routes of intermediary metabolism, such as the pentosephosphate cycle used in the reverse direction. The biosynthesis of aminoglycosides (e.g., streptomycin, neomycin, kanamycin) is sensitive to a high concentration ( > 5 mM) of inorganic phosphate (MARTIN,1989). The extracellular aminoglycoside phosphate phosphatase which forms the biologically active antibiotic, in the case of streptomycin (see Sect. 4.1), from an inactive phosphorylated precursor is inhibited by phosphate. Streptomycin6-P is accumulated in cultures of S. griseus grown at a high concentration of phosphate or at low pH due to the inhibition of the streptomycin-6-P phosphatase (MANSOURI und PIEPERSBERG, 1991). In addition, there is evidence that some secondary metabolic genes are controlled by a PhoB/PhoR-like system; so-called pho boxes have been detected in phosphate-regulated promoter regions (cf. Fig. 24B; MARTIN,1989; MARTIN
445
10
signal 7
20
30
40
50
S.
Fig. 2.Model for the regulation of streptomycin production in Streptomyces griseus. A Growth and strep4 tomycin production of S. griseus. The kinetics of biomass accumulation (0)and of streptomycin production (0)are shown. The columns indicate the relative expression rates ol StrS, StsA, and StsC as determined by densitometry of the autoradiographed protein patterns from S. griseus pulse-labeled with (35S)methionine after separation on 2-dimensional protein gels. B Regulation of the expression of strlsts genes by pleiotropic (e.g., A-factor) and pathway-specific (e.g., StrR) factors. P: promoters; arrows indicate positive regulatory effects; tRNA-molecules are symbolized by cloverleaves. The dominance of the StrR-dependent regulation is indicated by thick lines, for further explanantions see the text.
446
et and LIRAS,1989; LIRASet al., 1990). Similar DNA-binding protein (ONAKA al., 1995). structures were also found in the promoter In a proposed model the A-factor receptor regions of the aphDp2 gene of S. griseus and protein acts in the absence of A-factor as a of the strK (streptomycin-6-P phosphatase) repressor of streptomycin biosynthesis and et gene of S. glaucescens (LIRASet al., 1990; differentiation (MIYAKE al., 1990). In addition, A-factor stimulates membrane-bound DISTLER al., 1990). et The possible role of guanosine tetraphos- GTPases in S. griseus in vivo and in vitro (PEphate or guanosine pentaphosphate (ppGpp NYIGE et al., 1992). A-factor induces the exor pppGpp) and GTP pools in controlling pression of StrR, the activator protein of secondary metabolism in streptomycin-pro- some of the transcription units in the gene ducing S. griseus and other antibiotic produc- cluster for streptomycin biosynthesis (cf. Sect. et ers was extensively studied by OCHI (1987, 4.1; DISTLER al., 1987b; RETZLAFFet al., and 1995). The A1988,1990). Although relC mutants showed a 1993; RETZLAFF DISTLER, marked reduction in antibiotic production on factor-dependent transcription of sfrR seems complete medium it seemed difficult to assess to be controlled via regulation by the A-facwhether the effect on streptomycin produc- tor receptor of at least one further DNAtion was a direct consequence of the defect in binding protein (pX; cf. Fig. 24) which binds ppGpp formation or an indirect effect of the to an enhancer-like element located upstream and relC mutation (OCHI,1990). In S. clavuligerus of the strR promoter (HORINOUCHI BEPet there is no relationship between ppGpp and PU, 1992; VUJAKLIJA al., 1991). Three ad~; antibiotic production (BASCARANet al., ditional DNA-binding proteins ( P Y ~ - cf. Fig. 1991). The same observation was made for 24B), not dependent on A-factor induction in streptomycin synthesis in S. griseus on mini- S. griseus, interact with neighboring sites in mal medium; no ppGpp formation could be the same strR promoter region, suggesting an detected at any growth phase (NEUMANN even more complex regulatory network govet erning StrR expression (VUJAKLIJA al., et al., 1996). A-factor (2-(6-methylheptanoyl)-3R-hy- 1993). droxymethyl-4-butanolide), an extracellular The StrR proteins encoded by the streptodiffusible autoregulatory molecule, triggers mycin biosynthesis gene cluster of S. griseus both streptomycin biosynthesis differentia- and S. glaucescens share an identity of 62.8%. tion and streptomycin biosynthesis in S. gri- The strR gene encodes an activator of streptoseus. This factor, which was discovered by mycin and OH-streptomycin production. The KHOKHLOV al. (1967, 1988), belongs to a existence of the StrR activator protein was et group of chemically similar y-butyrolactone first postulated in mutants of S. griseus and its autoregulators which are synthesized by a va- role was suggested from the mutant phenoand riety of Streptomyces (HORINOUCHI BEP- type and its suppression by complementation et PU, 1992, 1994). A-factor induces streptomy- with wild type DNA (OHNUKI al., 1985a, cin production and sporulation in A-factor- b). Analysis of the mode of action of StrR in negative mutants of S. griseus at concentra- S. griseus showed that StrR is a DNA-binding tions as low as 1 nM. The induction by A-fac- protein which activates the expression of the tor exhibits a strict growth phase dependence, sfrlsts genes strBl and sfsC at the level of in that A-factor has to be present during the transcription by binding to upstream promotfirst hours of growth (decision phase moder sequences (hatched boxes in Fig. 24B; et and DISTel, see below; PIEPERSBERG, NEUMANN RETZLAFF al., 1993; RETZLAFF 1995; et al. 1996). A-factor-induced regulation in S. LER, 1995; BEYERet al., 1996). These and a griseus depends on an A-factor-binding pro- third identified StrR-binding site within the tein, which is present in very low amounts (ca. sfrR gene are palindromic sequences with the 37 molecules per genome) and has a binding consensus sequence: GTTCGAnnGn(11)Cnnand 1995). constant (Kd) of 0.7 nM (MIYAKE al., 1989, CTCAACG (RETZLAFF DISTLER, et 1990). Recently, the gene of the A-factor- The function of the StR-binding site within binding protein, arpA, was cloned from S. gri- the sfrR gene, e.g., negative feedback regulaseus and found to encode a repressor type tion of StrR expression or activation of the
447
promoter aphDp2 located approximately thetic genes could also be dependent on such 400bp downstream of this element, is un- sigma specific for secondary metabolic genes. clear. In the OH-streptomycin gene cluster of In S. griseus the similarity of the strlsts proS. glaucescens an StrR-binding site was also moters (strRp, aphDP, and strBlp) supports identified upstream of the StrBl and strX this assumption (DISTLER et al., 1987b). genes and within the strR gene. Recently, a There are several reports of the possible ingene homologous to strR was identified on a volvement of other factors in the regulation DNA fragment of S. spectabilis conferring of streptomycin production in S. griseus spectinomycin resistance in S lividuns (D. strains: the DNA-binding protein ORF1590 . unnecessary for sporulation and probably strepLYUTZKANOVA J. ALTENBUCHNER, and published data). Therefore, it can be specu- tomycin production in S. griseus (MCCUE et lated that StrR homologous proteins are al., 1992), ADP-ribosylation of proteins (PEwidely distributed activators of the synthesis NYIGE et al., 1992) possibly via influencing of aminoglycoside antibiotics, especially for membrane-bound G proteins (GTPases; PEthose using the Ca cyclitol pathway, which NYIGE et al., 1992), and C-factor, a cytodiffercould have evolved from a common ancestral entiation protein, excreted into the medium gene cluster. et by S. griseus 45H (SZESZAK al., 1991). ReThe expression of aminoglycoside biosyn- cently, it has been found that serinelthreonine thetic genes in Streptomyces is regulated by and tyrosine protein kinases are present in andlor is dependent on additional gene prod- most or all streptomycetes and, therefore, ucts having pleiotropic effects and influencing could be part of the complicated regulation both secondary metabolism and differentia- network necessary for the induction of antition on a more general level. The bldA gene biotic synthesis and differentiation, also in s. encodes a leucine-specific t RNA which recog- griseus. Proteins phosphorylated at a tyrosine nizes the rare codon UUA (cf. Fig. 24; CHAT- residue were detected in S. griseus and other et ER,1989,1992; LESKIW al., 1991a, b). BldA Streptomyces spp. (WATERS al., 1994). Speet mutants of S. griseus are deficient in aerial cific inhibitors of eukaryotic protein kinases, mycelium formation and streptomycin pro- such as staurosporin (cf. Fig. A29), inhibit duction (MCCUE et al., 1992). Recently, it sporulation (HONGet al., 1993) and streptoet was demonstrated that tRNAUUA-dependent mycin production (NEUMANN al., 1996) of control could be a key switching process in S. griseus without affecting vegetative growth. the onset of differentiation and secondary These results suggest that in Streptomyces a metabolism (LESKIW al., 1991a, b). In S. signal transduction pathway similar to that in et griseus, S. glaucescens, and S. spectabilis the eukaryotic organisms may control cell differstrR genes contain a respective TTA triplet in entiation and secondary metabolism. conserved positions in the 5-section of the S. griseus, like other streptomycetes (HOLT respective reading frames (DISTLERet al., et al., 1992), undergoes a decision-making 1987b; PISSOWOTZKI al., 1991; MAYER, process during a rather short period of 1-2 h et 1994; D. LUTZKANOVA J. ALTENBUCH- the mid logarithmic growth phase (Fig. 24A; and NER, unpublished). In the strN genes of S. griNEUMANN al., 1996). This decision phase et seus and S. glaucescens additional TTA co- depends on pre-existing factors (such as Adons were detected. The strA (sph) gene of S. factor) and can be suppressed by certain meglaucescens possesses also a TTA codon, tabolites (e.g., valine) or inhibitors of proteinwhich, however, is absent from the counter- modifying processes (e.g., 3-aminobenzampart of this gene, strA (uphD), in S. griseus ide). Various changes in physiology, e.g., a (VOGTLIand HOTTER,1987; DISTLER al., temporal increase in intracellular CAMP levet 1987a). In S. coelicolor seven different RNA els and gene expression, can be observed durpolymerase sigma factors were identified and ing and after this period (DISTLERet al., et the role of special sigma factors (e.g., WhiG) 1990 NEUMANN al., 1996), which altogethin controlling differentiation and antibiotic er seem to be an absolute prerequisite for latproduction was elucidated (BUTTNER, 1989). er cell differentiation and streptomycin proThe transcription of aminoglycoside biosyn- duction. It is notable that the A-factor-de-
448
pendent expression of some strlsts genes (stsC, stsA, stsS) occurs 2-4 h after this decision phase, but 10 h before streptomycin is detectable in the culture medium for the first time (cf. Fig. 24A). This concept of the induction of secondary metabolism during a very early growth phase temporally separated from the production phase demands a global signal transduction network which might include autoregulator molecules, receptors, transducers, protein phosphorylation/dephosphorylation systems, and second messengers. Although the direct regulation of the strlsts genes and the superimposed regulatory cascades in S. griseus are still incompletely understood, a model for the regulation of streptomycin production which includes current hypotheses can be proposed (Fig. 24).
4.7 Overview of the Aminoglycoside Pathways

When we summarize what we have learned from the genetics and biochemistry of aminoglycoside synthesis in bacteria and when we compare the respective pathways with each other, several similarities, but also quite divergent traits become apparent. The initial pathways of aminocyclitol formation starting at the (deoxy-)scyllo-inosose level, whether via the Ca or Cb routes, and some of the intermediate steps, e.g., the involvement of a Dglucosamine moiety in the formation of a pseudodisaccharide in the 2DOS and fortimicidistamycin groups, could use very similar enzymes (genes). Also, the resistance mechanisms may be very similar and based on the evolution of common ancestral genes (see also Sect. 4.5 and 6). Whether the Ca or Cb routes for aminohexitol formation are used might depend on the availability of the respective enzymedgenes. However, other more specific factors might force cells to use either one or the other. The Cb pathway is more economical because it uses fewer steps (cf. Sect. 3.1). However, the initial step enzyme (~-myo-inosotol-3-phosphatesynthase) for the Ca route might be more wide-spread in streptomycetes because it is probably used also for other, more generally distributed bio-
synthetic pathways (e.g., for the formation of the recently detected compound mycothiol, which is a major thiol in most actinomycetes; NEWTONet al., 1996). Also, part of the enzymes for the next steps (e.g., an StrI-like oxidoreductase, cf. Sect. 4.1.1.1) could be used in myo-inosotol degradation, a frequent property of Streptomyces spp. other than S. griseus. For these reasons it is an interesting speculation that both the Ca and Cb pathways are used alternatively when a metabolic sorting mechanism is needed for the alternative use of different anabolic and/or catabolic pathways in the same cell. The various NDP activation mechanisms of monosaccharides before modification and glycosyltransfers might have a similar metabolic sorting effect. However, another explanation might be that they are designed in order to facilitate the horizontal genetic transfer of whole biosynthetic branches between bacteria and maybe even higher taxa. The 6DOH pathways are used less in aminoglycoside production than are the pathways leading to 6hydroxylated hexosamines; however, they are preferred over other carbohydrate pathways in the modification of nonaminoglycosidic secondary metabolites (cf. Sect. 3.2 and 5.2). In aminocyclitol aminoglycosides the 6DOH routes are practically exclusively coupled to the Ca type, streptomycin-like compounds (cf. Sect. 4.1). This could reflect a common evolutionary link between both pathways or a tight genetic coupling of both sets of genes required (or both). This coupling also implies that the sets of functionally identical enzymedgenes involved should show a higher degree of relatedness in producers of compounds such as streptomycins, spectinomycins, or kasugamycins than in Cb (e.g., 2DOS) producers. In contrast to this, amino sugars can either be derived from preformed aminated precursors (e.g., D-glucosamine), NDPactivated hexoses, or formed only after condensation of a carbohydrate precursor into (pseudo-)oligosaccharidiccompounds and occur in both Ca and Cb aminoglycosides. Therefore, it will be interesting to determine by which alternative route most of the aminoglycoside hexosamine pathways proceed. In particular, the biosynthesis route of the Nmethyl-L-glucosamine precursor of strepto-
449
mycin (mode (2), Sect. 4.1.1.3), the details of which are still unknown, could be an example for a wider distribution of the HA pathway. The streptomycins and probably some of the more related substances (e.g., spectinomycin, kasugamycin) are made via the condensation of highly modified precursors in the final stage of the pathway. In contrast, the larger group of 2DOS-containing aminoglycosides and the compounds of the fortimicin/istamycin family are condensed first from relatively simple precursors, and the intermediates are strongly modified in the later pathway. Some substances, such as the acarboselike a-glucosidase inhibitors might be condensed with additional subunits only after export to the cell surface (outer side of the cytoplasmic membrane) with the involvement of membrane anchors and carriers, e.g., undecaprenyl phosphate residues, for the activated intermediates. This could also indicate a more general principle in the formation of secondary metabolites where frequently mixtures of very similar end products formed by a given strain of the producing microorganism are found. Rather than being side products of inefficient or nonspecific biosynthetic enzymes, complex product mixtures frequently observed in fermentations might be the result of the individual specificities of the export systems which are responsible for the intermediate/product patterns released from the cells. Thus, the strategies and routes used by the various producers for the production of the chemically relatively homogenous class of aminocyclitol-containing aminoglycosides described so far do not seem to be of similar homogeneity. Instead, a strong selective advantage could exist in the long-term diversification by means of the modulation of the pathway design (e.g., by natural genetic engineering). Hence, in the long-term possibly from time to time a new pathway for an altered end product develops which uniformly serves very similar functions of which the individual producers take advantage and at the same time escape the defense mechanisms of related producers or nonproducers in a competing situation in natural environments. The resistance mechanisms themselves in the aminoglycoside producers do not seem to be similar,
although evidence is emerging that some of them are based on biosynthetic functions (e.g., phospho- and acetyltransferases). However, a connection between individual aminoglycoside pathways might exist at a higher level of intracellular regulation or extracellular cell-cell communication between the producing cells and/or other cell types or organ1993). isms (see discussion in PIEPERSBERG,
4.8 Aminoglycoside-Target Site Interactions and General Effects on Bacterial and Eukaryotic Cells
The interaction of the classic aminoglycosides with the small (30s) subunits of eubacteria-type ribosomes (bacterial, mitochondrial, and plastidal ribosomes) has been well investigated and reviewed several times 1974; SCHLESSINGER et in the past (GORINI, et al. 1975; WALLACE al., 1979; VAZQUEZ, 1979; PIEPERSBERG al., 1980; GALEet al., et 1993). 1981; HILL et al., 1990; NIERHAUS, Two functional groups can be distinguished among the aminoglycosidic translational inhibitors: (1) translational misreading-enhancing and bactericidal compounds, e.g., the streptomycins and all 2DOS aminoglycosides; (2) bacteriostatic compounds which do not affect translational accuracy, e.g., spectinomycin and kasugamycin. This difference is also reflected by the alterations in overall translation patterns after aminoglycoside addition to cultures of E. cofi and B. subtifis, where group 1 compounds induce mistranslation and a rapid loss of overall translation ability and group 2 aminoglycosides a quasi-relaxed (Re1 - ) phenotype similar to that induced by chloramphenicol with an uncoupled residual translation of a special group of proteins for a longer period of time (PIEPERSBERG, 1985). An exceptional position is taken by hygromycin B which exhibits functional traits of both groups but does not seem to induce misreading of the genetic code (BAKKER, 1992). Originally, it was believed that certain small subunit ribosomal proteins mediate aminoglycoside interaction with the ribosome (OZAKI et al., 1969; PIEPERSBERG al., et 1980). In contrast, an early suggestion of Go-
450
RINI (1974) that 16s rRNA could be the primary target site of aminoglycosides was proven more recently; evidence is also accumulating that 16s and 23s rRNAs are the catalytically active components in both decoding and peptidyltransfer in the eubacterial ribosome (MOAZEDand NOLLER,1987; DE STASIO et al., 1989; CUNDLIFFE, 1990; NOLLER,1993). Mistranslation is also observed with some aminoglycosides in some but not all Archaea tested (LONDEIet al., 1988). The decoding process involves an interaction of a short RNA duplex (paired codon-anticodon) with the P and A sites of the 30s subunits, a process similar to that of the self-splicing group I introns (Fig. 25; VON AHSENand NOLLER, 1993). Therefore, it is not surprising that the RNA molecules of group I introns specifically bind aminoglycosides and that the splicing reaction is inhibited by these antibiotics (VON AHSENet al., 1991; SCHROEDER al., 1993). et It has also been demonstrated that short RNA analogs of the 16s rRNA decoding site specifically interact with both aminoglycosides and its RNA ligands, tRNA and mRNA, in the absence of ribosomal proteins (PUROHIT STERN,1994). These findings and initiated speculations that aminoglycosides are molecular fossils echoing their possible functions as ligands of catalytically active RNAs in a precellular RNA world (DAVIES et al., 1992). An argument in favor of this hypothesis is the finding that all known translational inhibitors among the aminoglycosides,
although they bind at different positions, seem to bind within a common domain in the f catalytic center o the 16s rRNA of bacterial ribosomes (CUNDLIFFE, 1990; NOLLER, 1993; BRINKet al., 1994). Effects other than the direct interaction of the group 1 bactericidal aminoglycosides with the bacterial ribosome have been suggested to be responsible for causing cell death in eubacteria (DAVIS, 1987). Besides mistranslation at actively elongating ribosomes (GoRINI,1974; WALLACE al., 1979; PIEPERSet BERG et al., 1980), two phenomena were observed to be coupled with lethality, mainly in studies with streptomycin and gentamicin: (1) a two-step uptake kinetics of aminoglycosides, the first phase of which is dependent on the APcomponent of the proton motive force or driven by ATP (HANCOCK,1981a, b; et BRYANand KWAN,1983; FRAIMOW al., 1991), and (2) membrane damage, probably via the induction of membrane channels (DAVISet al., 1986; BUSSEet al., 1992). Two models could explain the pleiotropy of action of these compounds: (1) incorporation of misread proteins into the cytoplasmic membrane, and (2) mistranslation of proteins with a short half-life involved in DNA replication and/or cell division and made at very few copies in a distinct phase of the cell cycle. The first model would explain also the observed drastic increase of aminoglycoside uptake in the second, killing phase, where the antibiotics are irreversibly accumulated inside the cells in a
Decoding
3
5
Splice-site selection
5ss
16s rRNA-A site
Fig. 25. Similarities between interaction of aminoglycosides in the decoding site of bacterial ribosomes
(left) and self-splicing type I introns (modified according to SCHROEDER a]., 1993). et
5 Related Sugar Components in Other Secondary Metabolites
451
large excess over the number of ribosomes. Aminoglycosides accumulate by binding electrostatically to anionic groups of cytoplasmic macromolecules or by being caged into degradation products of mistranslated proteins (DAVIS, 1987; BUSSEet al., 1992). Also, passage through the outer membrane of gramnegative bacteria could be affected by direct interaction of aminoglycosides with porins and/or lipopolysaccharide components (HANCOCK et al., 1991). In patients treated with aminoglycosides the most prominent adverse side effects encountered are oto- and nephrotoxicity (PRATTand FEKETY, 1986). Aminoglycoside hypersensitivity in humans is a maternally inherited trait, suggesting a mitochondrion-associated genetic defect, and a molecular mechanism has been proposed recently (HUTCHIN and CORTOPASSI, 1994). Hypersensitive persons have been shown to carry a 1 5 S Gmutation, i.e., a replacement of the A residue which is normally in this position of the small (12s) mitochondrial rRNA by a G. This area of small, 16s type rRNAs is known to interact with aminoglycosides and to be involved in the control of codon-anticodonpairing (see above; cf. Fig. 25). This defect leads to frequent death of hair cells in the ear, which is proposed to be due to enhanced production of superoxide by a mistranslated mitochondrial complex I, the proteins of which account for more than half of the coding capacity of the mitochondrial genome. Nephrotoxicity is primarily accompanied by membrane damage in the renal tubular cells. The biochemical basis of this effect in man is still not fully understood and could be caused either by an effect on lipid metabolism and structure or by mistranslation of proteins inducing malfunctions of membrane traffic (KOHLHEPP al., 1994; and op. cit.). The et nephrotoxic effect is also seen in neonates whose mother was treated with aminoglycosides during gestation (SMAOUI al., 1993) et and can be suppressed by polyaspartic acid (SWAN al., 1991). et

5.1 Lincosamides
Characteristic of the lincosamides is an interesting C-8 aminosugar component which is represented by the methylthiolincosaminide (MTL) moiety of lincomycin A (Fig. A26; WRIGHT,1983). Lincomycins (LM) A and B and celesticetin are members of the lincosamide group of antibiotics produced by S. lincolnensis and several other streptomycete species. Intensive biogenesis studies on LM-A involving measurement of stable isotope labeling patterns led to the proposal of a biosynthetic pathway for this antibiotic from an octulose and L-tyrosine. The former which is presumably derived from intermediates of the pentose phosphate cycle is a precursor of the MTL moiety and the latter is a precursor of the propylproline (PPL) subunit of LM-A (Fig. 26; BRAHME al., 1984a, b). However, et for the MTL subunit the genetic record seems to suggest a participation of a nucleotide (probably dTDP) activation step and a series of modification steps, including dehydratation, on NDP-activated sugar intermediates (PESCHKE al., 1995). Therefore, either a toet tally different initial pathway starting from Dglucose with the familiar first steps of the 6deoxyhexose pathways (see Sect. 3.2.1 and below) or an NDP-activation and modification of a C8 sugar intermediate are the routes for the formation of the MTL subunit which currently can be postulated (cf. Fig.26; PESCHKE al., 1995). Also, it was found that et ~-3,4-dihydroxyphenylalanine (L-DOPA) and 3-propylidene-A -pyrroline-5-carboxylic acid are probable intermediates in the PPL biosynthetic branch of the pathway (BRAHME et al., 1984a; K u o et al., 1992). In addition to a specific set of biosynthetic enzymes a special coenzyme, the so-called co-synthetic factor, which is structurally identical to the ribodeazaflavin moiety of the F420 coenzyme of methanogenic bacteria is needed for the PPL branch of the LM production pathway in S.
452
A
Sedoheptulose-7- (P) Desoxythymidintri- (P) (dl-rP) Erythrose - 4 - (P) Xylose-5- (P) Glucose-1- (P)
4 .
L- Tyrosine
dTDP-Glucose
(LmbM?) (LmbS?)
H2N
(LmbO?)
i b
CH20-P
OH
bH
OH
n steps (?)
(LmbO,M,S?)
3.
HNL
MethylthioPropylproline (PPL)
COOH
bH
Fig. 26. Hypothetical pathway of lincomycin A. For the biosynthesis of the Cs sugar moiety, methylthiolincosaminide (MTL), two alternative pathways are proposed (A) from intermediates of the pentose phosphate cycle via formation of a C8sugar precursor; (B) from dTDP-glucose via a 6DOH pathway an a later extension of the carbon chain. The possible involvement of three of the gene products of the putative MTL genes (cf. Fig. 27) is indicated.
453
Streptomyces lincolnensis 78-1 1 (Lincornycin A)
N Z
P O S R 0
Q
(ImbL) lmbM (ImbN
K I
KS
KS
lmbZ
ImbP)
lmb0
lmbS
lmbR
lmbQ
dTDP-hexose synthase dTDP-hexose 4,6-dehydratase
sugar arninotransferase(SMAT) other sugar-modifying enzymes
Fig. 27. The, sugar subcluster of the biosynthetic gene (lmb) cluster for lincomycin A. Genes marked by lmr encode resistance (or export) proteins.
lincolnensis (Kuo et al., 1989, 1992). In contrast, no intermediates of the MTL subpathway could be identified so far. The biosynthesis of the related antibiotic celesticetin (Fig. A26) probably proceeds via a very similar route. The LM-A production gene clusters of two overproducing industrial strains derived from Streptomyces lincolnensis NRRL2936 were cloned and analyzed by mutagenesis and hybridization: one of them (strain 78-11) has been sequenced (CHUNG and CROSE, 1990; ZHANG al., 1992: PESCHKEet al., 1995). et The lmbllmr gene cluster is composed of 27 open reading frames with putative production functions (biosynthetic or regulatory: lmb genes) and three resistance (or export; lmr) genes, and is flanked by two of them, the lmrA and 1mrC genes (Fig. 27). Compared with the respective genome segment of other lincomycin producers, the Imbllmr clusters seem to have a very similar overall organization in the other LM producers, S. pseudogriseolus NRRL3985, Streptomyces sp. NRRL3890, and S. vellosus NRRL8037. However, they are embedded in nonhomologous genomic environments and exhibit polymorphic restriction patterns. In the wild-type strain (S. lincolnensis NRRL2936) the lmbllmr cluster is apparently
present in a single copy. However, in the industrial strain S. lincolnensis 78-11 the gene clusters for the production of LM and melanin (melc) are duplicated on a large (450500 kb) DNA segment by transposition to another genomic region accompanied by deletion events (PESCHKE al., 1995). This fact et indicates that enhanced gene dosage is one of the factors underlying overproduction in the developed industrial strains. Only a minority of the putative Lmb proteins belong to known protein families which among the supposed PPL biosynthetic enzymes include members of the y-glutamyl transferases (LmbA), an L-tyrosine oxidase (LmbB2), an L-DOPA oxidase (LmbBl), amino acid acyladenylate synthetases (LmbC), aromatic amino acid aminotransferases (LmbF), and imidazoleglycerolphosphate dehydrataseslhistidinolphosphate phosphatases (LmbK) (PESCHKEct al., 1995; U. PESCHKE, D. NEUSSER, KASCHABECK, W. PIEPERSS. and BERG, unpublished data). However, except for LmbB1,2, even these proteins do not easily fit into any hypothetical route from L-Tyr to PPL, suggesting that the whole pathway is largely unknown. Similarly, in the right hand part of the lmbl lmr cluster a set of eight genes, lmbLMNZPOSQ, encode proteins which to var-
454
ious extents are related to enzymes involved in the sugar converting or modifying metabolism. The existence of these putative enzymes, dTDP-glucose synthase (LmbO), dTDP-glucose 4,6-dehydratase (LmbM), and (NDP-)ketohexose aminotransferase (LmbS), suggests that in contrast to the earlier proposals of the hypothetical biosynthetic pathway resulting in the MTL moiety (BRAHME al., et 1984b) this branch of the LM-pathway seems to be based on nucleotide-activated sugar intermediates. The stable isotope-labeling pattern found by these authors, suggesting an octulose - phosphate intermediate to be formed first, could also be explained by the same type of equilibration of externally applied glucose through the pentose phosphate cycle as was found for the biosyntheses of neomycin and validamycin (RINEHART al., 1992; cf. et Fig. 20). However, this would probably lead to another labeling pattern as that found in the case of the MTL subunit (BRAHME al., et 1984b). At first, for the reaction sequence synthesis/4,6-dehydratation NDP-pyranose (cf. Fig. 7) only a hexose seemed likely as a precursor. Therefore, the alternative pathway given in Fig. 26 (route B) was proposed (PESCHKE al., 1995; PIEPERSBERG, et 1994). The later detection of a gene ZmbR encoding a transaldolase-like enzyme in the sugar subcluster of the lmb cluster (cf. Fig. 27) could support an alternative route via a pyranosidic octose intermediate which is NDP-activated and further modified in this form. Finally, a thiomethyl unit to the C1 position of the postulated (NDP-)6-amino-6,8-deoxyoctose intermediate would be added which could be transferred from 5 -thiomethyladenosine, a side product of polyamine biosynthesis from S-adenosyl-methionine.
5.2 Cyclitols
Other natural products containing cyclitol moieties or their (e.g., aromatic, pentitol, and cyclohexane carboxylic acid) derivatives are more widespread than is at first obvious. For instance, cyclitol derivatives formed via the Ca (cf. Sect. 3.1) pathway are components in nucleoside type antibiotics, e.g., adenomycin
which in addition to an inositol unit contains an L-hexosamine derivative (Fig. A27). Therefore, their formation could have several steps in common with the streptomycin pathway. In this case equivalent enzymes could be used beyond the cyclitol formation, via Dmyo-inositol-3-phosphate synthase and phosphatase (StrO-like enzyme?). The initial nucleotidylation, e.g., part of the oxidoreduction and epimerization, or transamination steps, which might be catalyzed by the gene products StrQ, StrP, StrX, and StsA (or StrS), respectively, during the formation of the N-methyl-L-glucosamine subunit of streptomycin (see Sect. 4.1.1.3), could also be involved in the biosynthesis of the hexosamine moiety. An example is the cyclopentane ring in pactamycin (Fig. A30) which according to earlier biogenesis studies should also be derived from myo-inositol, i.e., via a 1,243diaminocyclitol (RINEHART al., 1981, 1992; et GRAFE,1992). Similarly the pentitols in allosamidin and trehazolin (Figs. A23 and A25) are possible products of the Ca route. The ring contraction of a hexitol to a pentitol could be accomplished by a mechanism similar to that operating in the formation of a furanoid ring from a pyranose derivative during dihydrostreptose biosynthesis (ses Sect. 4.1.1.2; cf. Fig. 15). In such a mechanism a dehydrogenase is believed to reduce a keto group and concomitantly introduces a rearrangement in the C chain resulting in a C, branch. Other possible routes for the synthesis of pentitols as in pactamycin or in allosamidin could be a new and so far unknown Cb mechanism acting on a ketohexose phosphate (e.g., fructose-6-phosphate) as a precursor and subsequent hydroxylation of the 2-deoxypentitol via 2,3-dehydratation and isomerizing rehydratation such as in the formation of 2-hydroxyvalidamines (cf. Fig. 22). In contrast to this, as a second example the pentitol moieties of some nucleoside homologs, aristeromycin, neplanocin A (inhibitors of S-adenosylcysteine hydrolase; Fig. A27), and adecypenol, which are inhibitors of various enzymes of nucleoside metabolism, are though to be synthesized from a fructose derivative and via pathways either similar to the Ca or analogous to the Cb routes (PARRY et al., 1989; PARRY,1992). Stable and radioac-
Related Sugar Components in Other Secondary Metabolites
455
tive isotope labeling experiments have shown that the stereochemistry of the C-6 atom in the cyclopentane ring formation between C-6 and C-2 of the hexose precursor is opposite to that in enzymatic myo-inositolphosphate synthesis. Also, the inversion of stereochemistry is unexplained by a mechanism of the Ca type. Therefore, it is more likely that the aristeromycin pentitol is formed by a Cb type enzyme cyclizing fructose-6-phosphate first to a 2-deoxyketopentitol. Successive reduction of the keto group to hydroxyl, and phosphorylation and pyrophosphorylation in the 5'- and 1 '-positions, respectively, could result in an analog of 5 '-phosphoribosyl-1 '-pyrophosphate (PRPP). The addition of the adenine moiety could then be catalyzed by either a salvage enzyme, e.g., purine-RRPP transferase, or a set of enzymes equivalent to the purine biosynthetic enzymes. Indirect evidence has been presented that both could occur in the organism producing aristeromycin and neplanocin, i.e., Streptomyces citricolor (PARRY, 1992). An interesting monoaminocyclitol, N-methyl-scyllo-inosamine, is found as a so-called rhizopine in rhizobia (MURPHY al., 1987, et 1993). Its anabolic metabolism is largely unknown so far. However, it will be interesting to see whether enzyme-catalyzed steps equivalent to those in the streptidine pathway are used (cf. Sect. 4.1.1.1). Also, nonglycosylated and neutral cyclitol derivatives, with unknown biological activities and biosynthetic origins, such as 2,3,4-trihydroxy-6-methylcyclohexanone (Fig. A28; MULLERet al., 1986), seem to be common and stable actinomycete products. Similar compounds of this group have been found in the chemical screening of actinomycete -secpersonal ondary metabolites (S. GRABLEY, communication). Cyclophellitol, produced by the mushroom Phellinus sp. and active as a specific inhibitor of almond p-glucosidase (ATSUMI al., 1990a, b), is another example et for this type of compound. It is plausible that this is synthesized by a basic pathway and cyclization mechanism (Cb) similar to that of the valienamine-related C, cyclitols (cf. Sect. 4.4). The exact biosynthetic routes and their relationships to the known cyclitol pathways of all these compounds await further clarifica-
tion, and can now be studied in a more direct way.
5.3 6-Deoxyhexoses
The 6DOH-derivatives, both of D- and Lconfiguration (cf. Sect. 3.2.1; Figs. 8 and 9), are the most variable group of sugar components in low molecular weight natural products. They are widely distributed especially in the actinomycete antibiotic groups of the aromatic (type I; anthracyclines, angucyclines, granaticin, etc.) and macrolide type polyketides (type 11; macrolides, avermectins, POlyenemacrolides) and in the glycopeptides (vancomycin family, thiopeptides) (PIEPERS1994). They BERG, 1994; LIU and THORSON, also are building blocks in highly modified and variable extracellular polysaccharides, such as the lipopolysaccharides of gram-negative bacteria (LIUand THORSON, 1994). Their biosynthetic pathways have not been studied extensively in most cases. However, it was shown that DNA probes taken from the genes encoding enzymes involved in the basic steps of the 6DOH pathway (Sects. 3.2.1 and 4.1.1) detect the gene clusters relevant for 6DOH biosynthesis in many streptomycetes and other actinomycetes (PIEPERSBERG et al., 1991b; STOCKMANN PIEPERSBERG, and 1992). Hybridization signals obtained with these cloned fragments were localized in the predicted production gene clusters. From such experiments it was found that the dTDPglucose 4,6-dehydratases, StrE, and related enzymes are the most highly conserved ones. In contrast, the genedenzymes for the first step related to strDIStrD from S. griseus (dTDP-glucose synthetase) were more divergent and, interestingly, some of them did not hybridize at all, such as strD or tylAl from tylosin-producing S. fradiae. The tylAl gene is more closely related to gram-negative bacterial genes which encode the same enzyme (such as rfbA from salmonellae). Again, this suggested that these genes can be recruited from a common gene pool independent of any taxonomic distances by all eubacteria whenever a particular secondary carbohydrate metabolism is required.
456
The basic enzyme complement and the cin as a single bond only the N-glycosidic gene and protein families involved in 6DOH linkage exists between the 4 -0-methylglubiosynthesis is becoming increasingly appar- cose and the indolocarbazole moieties (LAM ent (see reviews by SHNAITMAN KLENA, et al., 1989; Fig. A29). The transferring glycoand 1993; PIEPERSBERG, 1994; LIU and THOR- syltransferases, so far postulated to be inSON,1994; cf. Sect. 3.2). Enzymatic synthesis volved in 6DOH transfer (LIU and THORSON, et et of the activated sugar components dTDP-L- 1994; OTTEN al., 1995; PIEPERSBERG al., et oleandrose (MACNEIL,1995; see Fig. 9) and 1995; DICKENS al., 1996) are related to the dTDP-L-rhamnose (REEVES, 1993; see Fig. 9) macrolide glucosyltransferases (MgtA, MgtB) which are naturally derived from 6DOH causing resistance to 14- and 16-membered 1992b; VILCHES et pathways has been achieved by complete in macrolides (CUNDLIFFE, vitro enzymic synthesis from dTDP-D-glu- al., 1992). cose. Also, total chemical synthesis and introduction into the natural aglyca have been reported for some 6DOH derivatives, e.g., the 5.4 Other Pentose, Hexose, and 6DOHs D-mycosamine (BEAU, 1990) which Heptose Derivatives occurs in polyene macrolides (see Fig. 8) and Nitrogen-containing derivatives of pentoses L-daunosamine (THOMAS, 1990) which occurs and hexoses (other than 6DOHs) are found in anthracyclines (see Fig. 9). The types of chemical bonds involved in in many microbial secondary metabolites, esthe linkage between 6DOH moieties and pecially in nucleoside type antibiotics, either their aglyca also vary widely and are worth as (deoxy-)ribose analogs or as additionally some attention. Besides the usual O-glyco- modifying components in products originatsidic bonding C- and N-glycosidic or more ing from pathways with chemically heterogecomplex linkages are also observed. C-glyco- nous precursors. sidic bonds occur with both aromatic and aliphatic C atoms, e.g., in altromycins (BRILLet Pentosamines. Examples for aminofuranoses al., 1990). Examples are the 2,6-dideoxy-~- are found in some well-studied aminonucleoside glucose and 3-amino-3-N-methyl-4-O-methyl- antibiotics, e.g., puromycin and the re2,3,6-trideoxy-~-hexosemoieties in granaticin lated antibiotic A201A (cf. Fig. A27). In the and staurosporine, respectively (Fig. A29). In puromycin-related aminonucleosides the biogranaticin the incorporation of the double- synthetic origin of the 3-amino group of the (1 ,4-)C-bound 6DOH into the benzoiso- ribose derivative, for which an adenine nuchromanequinone polyketide aglycone could cleotide is the precursor, is not yet known. be achieved via two different routes from a However, intensive investigations of the gedTDP-4-keto-2,6-~-hexose precursor accord- netics and the enzymology of these coming to earlier proposals (FLOSSand BEALE, pounds have been initiated recently (LA1989): via the formation of a C-glycosidic CALLE et al., 1992). Interestingly, the putative linkage after dTDP elimination or via a reac- products of the two genes pur3 and prgl in tion of the aromatic C-10 and the 4-keto the puromycin biosynthetic gene cluster of group of the sugar. The latter route is consid- Streptomyces alboniger (TERCERO et al., ered more likely since other configurations in 1996) are related to the StrO (possible cyclithe pyrane ring do not alter the regioselectivi- tol-phosphate phosphatase) and StrSIStsAl ty of the condensation reaction. In staurospo- StsC (SMATs) proteins, respectively, from S. rin and in the staurosporin-related compound griseus (cf. Sect. 4.1). Besides the 3-amino-3K-252a (Fig. A29; KASE et al., 1986) an N- deoxyribose moiety A201A contains two adglycosidic linkage is probably formed first be- ditional rare sugar units: an unusual furanotween the 1-position of the aminodDOH sidic hexose with an unsaturated C-C bond and one of the indole N atoms of the trypto- between C-1 and C-2 branching off directly phan-derived indolocarbazole heterocyclic from the furane ring, and a 6DOH, an O-mesystem. This may also be the case for other thyl-D-rhamnose (cf. Fig. A27). Therefore, a very similar metabolites since in rebeccamy- strE probe was tested for hybridization and
6 Evolutionary Aspects
457
and 1992; PIEPERSBERG,1993; EMBLEY STRACKEBRANDT, 1994). Although there are no data at present which could unequivocally prove this hypothesis it nevertheless has a high likelihood. Evidence is accumulating Hexosamines. Hexosamines of various struc- that the genes for these pathways have been tures and positions of amino-N substitution transmitted horizontally between the actinooccur in many secondary metabolites (see ex- mycetes and have also spread to other microamples in Figs. A7, A8, All-15, A17-19, bial groups. Hence, the butirosin (Fig. A13) A24, and A27) and many of them may share biosynthetic genes are not likely to have common biosynthetic traits with aminoglyco- evolved independently in the Bacillaceae. sides. Most of them have retained a pyrano- Rather, it seems that they have been derived sidic ring structure, but incorporation into from an actinomycete gene cluster involved in other heterocyclic ring systems via linearized the biosynthesis of a ribostamycin-like aminointermediates can also occur, e.g., the D-glu- glycoside (cf. Fig. A13) which was laterally cosamine-derived building block in mitomy- transferred to Bacillus circulans or an ancescins (Fig. A31; HORNEMANN al., 1974; tor thereof. The elucidation of the structures et OKADA al., 1988). Examples of aminopyra- of the gene clusters which encode enzymes inet noses are encountered in some nucleoside volved in the formation of ribostamycin and type antibiotics, e.g., streptothricins and blas- butirosin and of other 2DOS aminoglycosides ticidin S (cf. Fig. A27). The origin of the 2- will clarify these evolutionary aspects. The amino-D-glucose derivative in streptothricins horizontal dissemination of aminoglycoside is still unknown. resistance genes among all major eubacterial The biogenesis of the 4-amino-2,3,4-deoxy- groups is particularly well documented (Fosglucuronic acid moiety in blasticidin S (cf. TER, 1983; PIEPERSBERG et al., 1988; CUNDFig. A27), which is produced by Streptomyces LIFFE, 1989; SHAWet al., 1993). The primary griseochromogenes and used against rice blast structure similarities, in particular, between disease, was studied by isotope labeling the aminoglycoside phosphotransferase en(GOULD,1992). The fully retained position- zymes (APH) suggest a common origin of anspecific labels of fed D-glucose (or even with tibiotic phosphotransferases in bacteria and higher yield with D-galactose) in this com- eukaryotic protein kinases (DISTLERet al., et et pound indicated that it is directly derived 1987a; HEINZEL al., 1988; PIEPERSBERG 1990; RETZLAFF al., et from a UDP-D-glucose or UDP-D-galactose al., 1988,1991; KIRBY, precursor with UDP-D-glucuronic acid and 1993). This common evolutionary origin is cytosylglucuronic acid as intermediates. The also supported by site-directed mutation and latter intermediates were confirmed by direct gene fusion experiments which yield hybrid measurements in cell-free extracts of the en- APH enzymes. These experiments have zymes UDP-D-glucose epimerase, UDP-D- shown that the essential amino acid residues glucose oxidase, and cytosylglucuronic acid and the two-domain structure of the catalytic subunit in the CAMP-dependent protein kinsynthase. ase of eukaryotes are conserved in the APHs (TAYLOR al., 1990 BLAQUEZ al., 1991; et et KNIGHTON al., 1991; PIEPERSBERG al., et et 1991a; RETZLAFF al., 1993). et The above-mentioned versatile enzyme families, which are involved in secondary suMost of the aminoglycosides and related gar metabolism, such as aminocyclitol biosyncompounds considered here are actinomycete thesis and hexose activation and modification, products. Therefore, the question of whether are obviously products of a modularly used the evolution of the respective pathways also gene pool, the products of which are mainly occurred in this molecularly defined lineage involved in the production of highly variable of bacterial taxonomy arises (DAVIESet al., and mostly secreted biomolecules not used in
found to give a signal in the genomic DNA of the A201A producer Streptomyces capreolus (A. JIMENEZ,personal communication; cf. Sect. 5.3).
6 Evolutionary Aspects
458
primary cell functions. Instead, they seem to be mainly involved in the extracellular communication of the producing cell with other biological systems (PIEPERSBERG, 1992, 1993). The secreted low molecular weight molecules, e.g., antibiotics, enzyme inhibitors, and autoregulators, could be used for the defense against agonistic or competing organisms or as hormone-like signal transmitters in differentiation processes. The extracellular polysaccharides and other cell surface-bound compounds (LPS, capsular polymers, etc.) are synthesized on the basis of very similar pathways and gene clusters; they may be cell surface individualizing material for cell-specific recognition and attachment or protection, i.e., these compounds could serve similar functions and be regarded as cell surfacebound secondary metabolites. In accordance with this, both the genes for secondary metabolites and the genes for LPS biosynthesis have frequently been suggested to be transferred horizontally (PIEPERSBERG,1993, 1994; REEVES, 1993; SHNAITMANand KLENA, 1993; LIU and THORSON, 1994). Therefore, the rfa genes and the secondary carbohydrate biosynthetic genes in actinomycetes represent an interesting basis for the study of the evolutionary links and dynamics in secondary, highly mobile gene pools. The biosynthesis pathway of streptidine represents another interesting topic of evolutionary studies. The bluensidine pathway, e.g., was suggested to be ancestral to that of streptidine (WALKER,1990). Amidinotransferase activities catalyzing both reactions 6 and 11 of the streptidine pathway (cf. Sect. 4.1.1.1 and Fig. 14) were found in the bluensomycin producer S. hygroscopicus ssp. glebosus ATCC 14607 (WALKER, 1990). However, no strB2 gene is present in the two bluensomycin producers, S. bluensk DSM 40564 and S. hygroscopicus ssp. glebosus DSM 40823 (MAYER,1994; G. MAYER, MEHLING, A. and W. PIEPERSBERG, unpublished data.). In these strains only one strB gene is conserved which clearly belongs to the sfrBl group, since the adjacent gene downstream is strF in both cases (see Fig. 13) and these genes encode proteins with a much higher amino acid sequences identity (ca. 85%) relative to the StrBl than to StrB2 proteins of streptomycin
producers. Therefore, in streptomycin producers gene duplication could have resulted during the evolution of two amidinotransferase with district substrate specificities from one ancestral enzyme exhibiting both activities. Alternatively, the bluensidine pathway could be a degenerated streptidine pathway after loss of the sfrB2 gene and changes in the substrate specificity of the StrBl protein. Also, an open question is whether or not myo-inositol is a specific precursor and, therefore, its synthase is an intrinsic enzyme of the streptomycin pathway, since measurements of this enzyme under various culture conditions and in various strains suggested a rather non-specific distribution (SIPOS and SZABO,1989). The recent finding that probably all actinomycetes contain a major thiol compound called mycothiol (NEWTONet al., 1996) which also could be regarded as an aminoglycoside indicates that the anabolic myoinositol pathway is generally present in this taxonomic group, in contrast to other bacteria. Therefore, the evolution of Ca type pathways could be based on a preferred metabolic route in actinomycetes. The protein similarities between StrS, StsA, and StsC are in the range of 25% identity and, therefore, too low to suggest a gene duplication event during the evolution of the streptomycin pathways which might explain the origin of these proteins. The occurrence of a third possible aminotransferase is puzzling since it was thought that the third amino group introduced during the biosynthesis of streptomycin into NMLGA was derived via the primary metabolic pathway from the Dglucosamine pool (GRISEBACH, 1978). If this precursor is not formed de novo under conditions of streptomycin production another explanation for the need of an additional transamination step would be the regeneration of glutamine from a-ketoglutamine, the unusual by-product of step 4 in S D biogenesis. This aminotransferase reaction is not normally observed in prokaryotes (WALKER, 1975a). The N-terminal sequence of the StrS protein is also identical to that of one of the proteins expressed at high concentrations in the streptomycin production phase in S. griseus but not in the mutant M881 (see Tab. 2) (DISTLER et al., 1992).
8 Pathway Engineering and Other Types of Application in Biotechnology
459
7 Involvement of Primary Metabolism in the Delivery of Carbohydrate Precursors

The involvement of primary metabolic traits in the activation of precursors, the dynamics and alterations of precursor pool sizes, and the use of nutritional sources for precursor formation during the production phase in the producers of aminoglycosides and other secondary carbohydrates has not been studied intensively (SHAPIRO, 1989). The regulation of streptomycin production and its growth phase dependence has already been discussed above (see Sect. 4.6). These aspects will become more accessible for investigation in the future when the genetic programs for the biogenesis of representative members of this group of compounds are analyzed and available for manipulation in different organisms. Biotechnological development and use of production strains in fermentation could then become more efficient; mathematical modeling of the producers physiology would also greatly facilitate the production processes. It is especially important to gain specific knowledge about the sources and the routes of delivery of the carbohydrate precursors and the nitrogenous groups in the case of the aminoglycosides. The finding that compounds such as neomycins and validamycins are synthesized from a precursor pool related to the pentosephosphate (PP) cycle intermediates (RINEHART al., 1992), and that the et NDP-hexose forming enzymes are encoded by genes in the clusters for aminoglycosides and other sugar-based secondary metabolites (see Sect. 4) in streptomycetes could help to clarify these aspects. If hexose intermediates are derived from the PP cycle mainly, this raises the question of whether compartmentalization or a difference in the efficiencies of the enzymes involved in the conversion of the primary hexosephosphate are responsible for this phenomenon. Alternatively, the inversion of the major route(s) of intermediary C metabolism to the predominant use of the gluconeogenetic pathway could occur during
the production phase. In this context it is interesting to note that amino acids seem to be preferred or even essential nutrients in streptomycin or neomycin producers (SHAPIRO, 1989; PIEPERSBERG, 1995).

The availability of the genes which encode enzymes necessary for the biosynthesis of secondary metabolites recently has allowed the initiation of experiments designed for the production of new hybrid molecules (HOPWOOD et al., 1990; KATZ and DONADIO, 1993; several articles in VINING and STUTTARD, 1995; PIEPERSBERG, 1994). For the array of aminoglycosides and other carbohydrate-derived natural products this phase of biotechnology has not yet been as successful as it was for the polyketides. The reasons for this are as follows: (1) the pathways involved are composed of multiple steps catalyzed by highly specific and in general monofunctional enzymes; (2) the biochemistry and chemistry of carbohydrates is complicated, and the intermediates of biosynthetic pathways are mostly quite instable and inaccessible to synthetic chemistry; and (3) only relatively few research groups are engaged in the investigation of secondary carbohydrate metabolism. Nevertheless, several fields of biotechnological application of our increasing knowledge of the biochemical and genetic components involved in the formation of activated and modified sugar or cyclitol molecules can be envisaged (cf. also PIEPERSBERG, 1994): (1) genetic engineering of pathways for improved fermentations, with respect to nutrient control and incorporation, optimization of yield, and restriction of product patterns, or for the production of new hybrid end products in the producers themselves (e.g., the hybrid, glycosylated tetracenomycins; DECKER et al.,
460
1995); (2) development of biotransformation aminobutyryl-dibekacin; cf. Fig. 28B); for systems for the in vivo glycosylation of fed most of those modifications, natural counteraglycones; (3) in vitro enzymatic production parts exist as models. (2) Designed mixing of subpathways could of activated sugar derivatives and/or glycosyltransfer to synthetic aglycones; or (4) transfer be envisaged. For example, first the Cb pathof the genetic complement for complete pathways could be exchanged for Ca routes and ways or branching subpathways to new host vice versa. In another line of experimental systems and/or their redesign for improved models the exchange of the sugar modifying production characteristics or completely new and transferring subpathways between prometabolites. However, most of the prerequi- ducers of related groups of aminoglycosides sites for the planned redesign of the pathways could be attempted, e.g., the PA pathways befor aminocyclitol aminoglycosides and related tween the producers of gentamicins (cf. Fig. secondary carbohydrates are not yet availa- A14) and seldomycins (cf. Fig. A17) or the ble. These prequisites include the complete 6DOH or H A pathways between the producgenetic and biochemical analyses of several ers of streptomycins (cf. Fig. Al), spectinokey pathways and knowledge of substrate mycins (cf. Figs. A2 and A3), kasugamycins specificities of key enzymes (especially the (cf. Figs. A4 and A5), and boholmycin (cf. glycosyltransferases and other condensing en- Fig. A7). (3) Some aminoglycosides are produced in zymes) and of the bottleneck steps or the flux rates of individual precursors and interme- genera or strains which are not easily accessidiates through the pathway. ble for physiological or genetic manipulation Some interesting goals for pathway engi- or which produce unwanted side products. neering in aminocyclitol aminoglycoside pro- The ability to produce could, therefore, be ducers are as follows: transferred to hosts which can be more easily (1) Transfer of the production and transfer manipulated. For instance, the aminoglycogenes for the Nl-a-hydroxy-yaminobutyryl side production genes from Micromospora moiety of butirosin from Bacillus circulans to spp. could be transferred to Streptomyces spp. a kanamycin-producing Streptomyces kana- or even to more distantly related GRAYmyceticus in order to produce in vivo the organisms (GRAS = generally regarded as semisynthetic compound amikacin ( =N I - a - safe) such as corynebacteria (cf. PIEPERShydroxy-yaminobutyryl-kanamycin A), as al- BERG,1993). ready suggested much earlier (Fig. 28A; (4) The regulation of production genes and DAVIESand YAGISAWA, 1983). This would the nutrient flow in aminoglycoside producers require first that the N1-a-hydroxy-yamino- could be further targets of pathway engineerbutyryltransferase also recognizes the kana- ing. Because of their complicated physiology mycin A molecule as a substrate and in its and regulation in complex cell differentiation correct amino acceptor group and, second, cycles (cf. Sect. 4.6) this would be of particuthat the possible kanamycin exporter in S. ka- lar importance when continuous culture technamyceticus transports the new end product niques are used in production. with equivalent efficiency. Similarly, it might Another application of the new genetic and be possible to find enzymic reactions for the biochemical data could be the development designed modification of other aminoglyco- of genetic screening systems for the search for sides, e.g., group transfers or dehydroxyla- a particular production ability. For this purtions such as those used in the synthesis of pose, we have to find more genes for key well-established semisynthetic chemothera- functions which are diagnostically relevant in peutics with activity against bacteria with clin- the detection of a specific pathway. An examically important resistance patterns (cf. SHAW ple is the family of strE-related genes encodet al., 1993) and with reduced ototoxicity. Ex- ing the dTDP-glucose 4,6-dehydratases charamples are (1) netilmicin ( =l-N-ethylsisomi- acteristic for the 6DOH pathways (STOCKcin; cf. Fig. A14), (2) isepamicin (cf. Fig. 28B), MA and PIEPERSBERG, 1992). This type Of dibekacin ( =3 ,4 -dideoxykanamycin B; cf. diagnostic material, together with that typical Fig. A12), and arbekacin ( =N,-a-hydroxy- y- of other chemical classes of secondary meta-
461
Butirosin A
Amikacin
(= AHB)
2
OH
genetic engineering ?
(= AHP)
CH9NH9
,:uH3
OH
OH
CH2NH2
NH-AC
OH
lsepamicin (N1-AHP-3*-N-acetyIgentarnicin B)
Arbekacin (Habekacin) (N'-AHB-3',4'-dideoxykanamycin6)
Fig. 28. Semisynthetic aminoglycosides as models for the production of modified aminoglycosides by pathway engineering. A Hypothetical production of the semisynthetic amikacin in a genetically engineered kanamycin A producer; B structures of two other semisynthetic aminoglycosides, derivatives of gentamicin B and kanamycin B, with good pharmacological properties and activity against multiple-resistant pathogens. The chemical groups that are introduced by semisynthetic additions (given in bold face) or deleted (arrows) mostly mimic known modifications of related natural components in other molecules.
bolites, could be applied together with classical screening methods for the detection of new compounds. This could be of interest if, e.g., a leading structure is known and new derivatives related to that special target group are searched for, or if in particular microbial
groups (e.g., actinomycetes) an unwanted subgroup of producers has to be excluded. In the future, when sufficient and highly predictive pathway-specific gene probes become available this method can also be used for a more rapid prescreening method.
462
9 Conclusions and Perspectives

The knowledge of all the variants of secondary metabolites composed of (amino-)sugars or their derivatives (secondary carbohydrates) and their biosyntheses is still sparse. Evidence is accumulating, however, that the various pathways of aminocyclitol aminoglycoside production share several common features and that a common gene pool is used for the modular design of these pathways. It can be predicted that the application of molecular genetics and biochemistry on several production systems for secondary carbohydrates will bring some unification to several aspects of this field of research. Concomitantly, also an intensification of applied research on secondary carbohydrates will arise, since these are components in many applied natural products where they are essential for bioactivity. Thus, the data reported in this chapter will influence other fields of research and development, especially of other chemical groups of low-molecular weight bioactive molecules or of polysaccharides and glycoconjugates. Also, the research and development in the field of chemo-enzymatic synthesis will be fertilized by providing the biochemical tools such as enzymes and their substrates (e.g., NDP-activated sugars). The basis for pathway engineering for the designed production of new variants of secondary metabolites will also be available very soon. With
respect to the more restricted group of aminocyclitol aminoglycosides in particular, there still seems to be a potential for finding new structural classes as is indicated by the recent detection of new groups via target-directed screenings for glycosidase inhibitors (e.g., acarbose, trestatins, trehazolin, and allosamidins). Other fields of application and new structures will also become available by continuing research efforts and the input of intelligence and scientific skills. Acknowledgements We thank all our collaborators for their enthusiastic participation in aminoglycoside research. The work on secondary carbohydrate genetics and metabolism in the laboratory of the authors was generously supported by the Deutsche Forschungsgemeinschaft, the Bundesministerium fur Forschung und Technologie, the European Commission, and the pharmaceutical companies Bayer AG and Hoechst AG.
10 Appendix (Chemical Structures)

The figures compiled in the Appendix (Figs. Al-A31) summarize the chemical families of secondary carbohydrates mentioned in the text; their pathway formulae are indicated in the legend (cf. Sect. 3.2).
463
NH
II
OH
6
streptidine (bluensidine)
0
streptose
N-methyl-L-glucosamine (demethyl-, 4"-mannosido-, 4"-dimannosido-,4"-ahimosido-, N-glycolyI-)

OH
R'
streptomycin dihydrostreplomyan 5'-hydroxystreptomycin Ndemelhylstreptomycin 5'-hydroxy-N-demethyIdihydroslreptomycin mannosido5'-hydroxyslreptomycin dimannosidcstreptom ycin Muensomycin ashmycin A NH-CNH-NH2 NH-CNH-NH2 NH-CNH-NH, NH-CNH-NH2 NH-CNH-NH2 NH-CNH-NH2 NH-CNH-NH2 0-CO-NH2 NH-CNH-NH2
R2
CHO CHP-OH CHO CHO CH2-OH CHO CHO CH2-OH CHO
R3
Ft
CH3 CH3 CH3 H H CH3 CH3 CH3 CH3 C, H
~~~ ~
I +
H H H H H
Ff
CH3 CI H CH2-OH CH3 CH2-OH CHp-OH CH3 CH3 CH3 CH3

~
H
a-Dmannose (= DM) a-DM-1.6-a-DM H 2" '-carboxy-xylofuranose (ashimose) a-DM-1,Ga-DM
H H
H H
ashmyan 8 NH-CNH-NH2 CHO AC4437 = 5'-hydroxystreplomycin lacking NMLGA
CO-CH2-OH
Fig. Al. Streptomycins; basic pathway formula Ca(4)-6DOH(2)-HA.
464
u
F g A2. Spectinomycin;Ca(4,5)-6DOH. i.
iH H3C-YH OH
CH3
F g A7. Boholmycin; HA-(4)Ca(6)-PA(4)-Hep. i.
Fig. A3. Spenolimycin; Ca(4,5)-6DOH.
HN
I HN=C I COOH
boo
E H
OH HN=C AH2
OH
F g A4. Kasugamycin;Ca(4)-6DOH. i.
NH
F g AS. Myomycins; Ca(4)-HA. i.
@2
2 NH-CO-CH-NH-CO-NH
HZN
OH
HOOC
OH
F g AS. Minosanimomycin; Ca(4)-6DOH. i.
F g A9. Hygromycin A; Ca(l)-[X]-6DOH. i.
bH
L-Arg
Fig. A6. LL-BM132a; Ca(4)-H(4)-HA.
465
H3C0
d6
OH CH3 CH3 CH3 CH3 H CH3 CH3 CH3 CH3 H H H H H H H H
H H H H H H H H H CH3 CH3
H3CN
I
OH
H3CHN
Lysinomicin(Ca(G)-HA)
NHp H fortimicin A NHp H fortimicin B 1-epkfortimicinB H NHp N, H H fortimicin C NHp H fortimicin D NHp H dactimicin 1-epi.dactimicin H NHp sporaricin A H NHp sporaricin B H NH2 NHp H istamycinA , NHp H istamycinA (= sannamycin) NHp H istamycin A3 istamycinBo H NHp istamycin B H NHp istamycin B3 H N, H istamycin C NHp H istamycin A2 NHp H fortimicin KG3 = 4',5'-dehydrofortimicin A
fortimicin KH fortimicin KR istamycinYo istamycinXo
OH OH OH OH OH OH OH
H H H H
COCH2NHp H H COCHpNHCONHp COCHpNH2 COCHpNHCH=NH COCHzNHCH=NH COCHpNHp H H COCHpNHp COCHpNHCH=NH H COCHpNHp COCHpNHCHsNH COCHpNHp COCHpNHCONHp CH3 CH3 H H
H H H H H H OH OH H H
CH3 CH3 CH3 CH3 CHpCH3 CH3
OCH3 H OCH3 H
~ ~~~~
H OCH3 H OCH3
Fig. A10. Fortimicinhstamycin family; Ca(6)-HA or Cb(6)-HA.
466
HzN
R2 NH2 NH2 OH OH OH OH OH
R3
R4
R5
neomycin B neornycin C paromornycin I parornomycin II mannosylparornomycin lividomycin A lividomycin B
OH OH OH OH OH H H
H H H H a-D-Man a-D-Man H
~
CH2NH2 H CH2NHz H CHzNHz CHzNHz CH2NHz
H CH2NH2 H CHzNHz H H
Fig. All. Neomycin family; HA-(4)Cb(S)-P(3)-HA.
I0 Appendix (Chemical Structures)
467
HO
Q
CH2R3 RO I
R2
OH OH OH OH OH OH H OH H
R3 NHp NHp OH NH2 NH2 NHp NHp NH2 NHp
R4
NH2 NHp NHp NHp OH OH NHp NHp NHp
R5 OH OH OH OH OH OH OH OCONHp OCONHp
R6
Fig. Al2. Kanamycin family; HA-(4)Cb(6)-HA.
kanamycin kanamydn B kanamydn C amikacin NK-1001 NK-1012-1 tobramycin nebramYCin nebramycin5
OH NHp NH? OH OH NH2 NH2

NH2
NH2
H H H ahb H H H H H
R4
OH
R2
R3 NH2 NH2 OH OH NH2 NH2

W 2
R4 H OH H OH H OH OH H OH
R5
OH H OH H OH H H OH
Fig. A13. Butirosinlribostamycin family; HA(4)Cb(S)-P.
butirosin A butirosin B butirosin El butirosin E2 butirosin C1 butirosin Cp ribstamycin xylostatin LL-BM 408a
ahb ahb ahb ahb ahb ahb H H H
OH OH OH OH H H
OH
OH OH
NH2 OH
468
gentamicinA gentamicinAl gentamkin4 gentamicinA3 gentamkin& gentamkinB gentamicinB, gentamicinXz G-418 JI-20A JI-206
0
NH2 NHz NHz OH N, H OH OH NH2 NHz NHz N, H
H H H H H H CH3 H CH3 H CH3

R'z
NHCH3 NHCH3 OH N H ~ NHCH, OH NCH3CH0 N, H NHCH3 N, H NHCH3 OH NHCH3 OH NHCH3 NH2 NHCHj NHZ NHCH3
R'3
R'4
OH OH OH
OH H OH H OH CH3 CH3 CH3 CH3 CH3 CH3

R'5
H OH H OH H OH OH OH OH OH OH
R'6
R'I
CH3 H CH3 H H H CH3 H H H
gentamicin C1 gentamicin C , 1 gentamicin C2 gentamicin CZa sagamicin sisomicin verdamicin G-52 66-408 66-40D
H H H CH3 H H H H H H
CH3 H H H CH3 H H CH3 H H
CH3 CH3 CH3 CH3 CH3 CH3 CH3 CH3 OH H
OH OH OH OH OH OH OH OH H OH
+ +
+ +
Fig. A14. Gentamicin family; HA-(4)Cb(6)-PA.
469
R1
R2
apramycin oxyapramycin saccharwin
H OH
H
NH2 NH2 OH
H2N
Rl
R2
R3 OH OH NH2
R4
seldomycin 1 seldomycin3 seldomycin 5
OH OH H
OH NH2 NH2
OH OH OCH3
Fig. A15 Apramycin family; Cb(4)-OctA(8)-HA.
Fig. A17. Seldomycin family; HA-(4)Cb(6)-PA.
OH
hygromycin B destomycin A destomycin B destomycin C A-396-1 A-163164
Fig. A16. Hygromycin Bldestromycin family; Cb(S)-H(2,3)-HepA or Ca(S)-H(2,3)-HepA.
ss-56c
1-N-amidino-I -Ndemethyl-2-hydroxydestomycin A
CH3 H CH3 CH3 H CH3 H H
H H H H OH OH
OH O H H OH O H H OH OH
H H OH H H
OH
H H
OH OH H OH OH OH OH
OH
H H OH H H H H
H
470
NH-Ac siastatin
Ri
RP
R3
R4
nojirimycin A mannonojirimycin ldeoxynojirimycin (DNJ) 1-deoxymannonojirimycin miglitol emiglitate
H OH OH H OH H OH H H OH H H H H OH H H OH H ( H)O C ,, H H OH H (CH2)20R (R = -gbenzoic acid ethylester)

y- 42
HGoH
OH galactostatin CH2OH HQH oo NHCOyCH3
CHgH HO 0
HmH2
0
OH
Hb;;
OH
NHCOCHCH3
QOH
3-amino-3-deoxyPglucose
cv-1
do
NH2
No
streptozotocin
prumycin
Fig. A18. Monomeric aminoglycosides which are products of HA or PA pathways.
trehalosamine mannosylglucosaminide 4-amino-4-deoxytrehalose
H OH H
OH H OH
NH2 NH2 OH
OH
OH
NH2
RP
3,3'-neotrehalosadiamine (BMY-28251)
OH
Fig. A19. Trehalosamine family; HA(1)-H or HA(1)-HA.

CH2OH CH,OH
471
Ri
T-l T-ll T-Ill H NH2 NH2
R2
R3
R4
Ri S-l S-ll H NH2
R2 NH2 H
NH2 H H H H OH
OH OH H
Fig. A 0 Trehalosamine-related synthetic disaccharides. 2.
validoxylamine A validoxylamine B validoxy\amine G validamycin A validamycin B validamycin C validamycin D validamycin E validamycin F validamycin G validamycin H
H H OH H H H H H
H OH H
H H H H H H a-D-Gk H
H H H
H OH H H OH H H H
H H H
H H H P-D-GlC P-D-GlC P-D-GlC H a-D-Glc(l,4) -P-D-Glc P-D-GlC P-D-GlC a-D-Glc(1.6) -&D-Glc

CHpOH
H H H H H a-D-Glc H H H H H
H H H H H H H H a-D-Glc H H
CHpOH
CH20H
HOCH2 OH
valienamine
validamine
valiolamine
2-hydroxyvalidamine
Fig. A21. Validamycin family; Cb(l)-Cb(4)-H. Valiolamine (Cb) was also found as a separate end product.
412

CH20H
R3-O OH
NH
0
OH
0-R2
acarbose H (amylostatins) OH adiposins trestatin H
a-l.4-maltosyl a-l,4-maltosyl a-1,Gmahosyl-
H, or (a-D-glucopyranosyl), H. or (a-D-glucopyranosyl), H. or (core pseudotrisaccharide),
2.3-epoxyderivatives of acarbose oligostatins= 2,3-dihydro-2-hydroxyderivativesof acarbose
a-1,rl-trehalosyI
Fig. A22. A carbose/amylostatin family of glycosidase inhibitors; (H),,-Cb(1)6DOH(l)-(H). or (H).-Cb(1)-H(1)-
HO
bI=
HO NH
HO ~
o
OH NH
I
0 0 :
OH
OH CH20H R
Ethyl Propyl Methyl H
OH NH
I
Ac
Ac
I
'';J-Rl
R2
allosamizoline (RI = R2 = CH3)
R1
R2
Fig. A23. Trehazolin; Ca(l,2)-HA.
allosamidin demethylallosamidin didemethylallosamidin
CH3 CH3 H
CH3 H H
Fig. A25 Allosamidins; Ca(4)-HA(4)-HA.
co
R
I
OH
sorbistin Al sorbistin A2 sorbistin B sorbistin D
Fig. A24. Sorbistins; Ca(2)-HA.

p5
473
OH
Fig. A26. Lincosamides; HA or OctA.
lincomycin A lincomycin B clindamycin N-demethyllincomycin A celesticetin desalicetin celesticetin B celesticetin C celesticetin D
CH3 CH3 CH3 CH3 (CHz)z-O-salicylyl (CHz)z-OH bu (CH2)2-0-iso ryryl (CH2)z-O-ant hranilyl (CH,),-O-acetyl
CH2-CHz-CH3 CH2-CH3 CHz-CHz-CH3 CH2-CHz-CH3 H H H H H
414
b HOH2C HO
OH Ph-0 OH Adenophostin A (Ad = Adenin; Ph = P03H2) (P(3FH) Cytosaminomycinr (R = acyl residues) (6DOH(4)6DOH)
Adenomycin (P-(l)Ca(3)HA)
ro\
CH~
OH
(3-h ydroxyDamicetose)
(D-amosamine)
" O H 2 Y Y
0
mOecH2
OCH3 CH20H
H2NOC9~0\~=
A201C (Ad = Adenin) (PA(3)-[Xl-H(6)-6DOH)

I
HOH2C
Aristeromycin (Ad = Adenin) (Ca or Cb)
a
Ad OH OH
streptothricins (R = deazapurin) (HA)
Ad - H 2 C f j
J - - ( NH - R
COOH
OH
OH Blasticidin S (Cyl = cytosin; R = y-N-methyl-P-arginine) (HA)
Neplanocin A (Ad = Adenin) (Ca or Cb)
Fig. A27. Nucleoside-type secondary metabolites with unusual sugar components or cyclitols.
475
H0 Q
HO
$ (
OH
CH3
2,3,4-trihydroxy-6methylcyclohexanone
cyclophellitol
F g A28. Natural neutral cyclitols; Ca or Cb? i.

OH
F g AM. Pactamycin; Ca? i.
Granaticin
mitomycinA mitomycinB mitornycin C porfiromycin
CH3 OW3 H
OCH3
NHz NHz
CH3
CH3 H
CH3 CH3
(thick line, bold face = D-glucosamineunit)
F g A31. Mitomycins. i.
K-252a
Rebeccamycin
Fig. A29. Granaticin, staurosporine, and staurosporine-related compounds. Unusual bonds in the linkage of 6DOH components in microbial secondary metabolites. C-glycosidic binding in the polyketide granaticin and N-glycosidic binding in the indolocarbazole alkaloids staurosporine, K-252a (same aglycone as staurosporine), and rebeccamycin which are all produced by actinomycetes.
476
10 Aminoglycosides and Sugar Components in Other Secondary Metabolites of antibiotic resistant bacteria, Proc. Natl. Acad. Sci. USA 70,2276-2280. BEYER,S., DISTLER, PIEPERSBERG, (1996), J., W. The str gene cluster for the biosynthesis of 5'hydroxystreptomycin in Streptomyces glaucescens GLA.0 (ETH 22794): new operons and evidence for pathway-specific regulation by StrR, Mol. Gen. Genet. 250,775-784. BILLINGTON, C. (1993), The Inosotol PhosD. phates. Weinheim: VCH. BLAZQUEZ, DAVIES,J., MORENO,F. (1991), J., Mutations in the aphA-2 gene of transposon Tn5 mapping with the regions highly conserved in aminoglycoside-phosphotransferases strongly reduce aminoglycoside resistance, Mol. Microbiol. 5, 1511-1518. BRAHME, M., GONZALEZ, E., ROLLS,J. P., N. J. HESSLER, J., MIZSAK,S., HURLEY, H. E. L. (1984a), Biosynthesis of the Lincomycins. 1. Studies using stable isotopes on the biosynthesis of the propyl- and ethyl-L-hygric acid moieties of lincomycins A and B, J. Am. Chem. SOC. 106, 7873-7878. BRAHME, M., GONZALEZ, E., MIZSAK,S., N. J. ROLLS, J. R., HESSLER, J., HURLEY, H. E. L. (1984b), Biosynthesis of the Lincomycins. 2. Studies using stable isotopes on the biosynthesis of the methylthiolincosaminide moiety of Lincomycin A, J. Am. Chem. SOC. 106,7878-7883. BRILL,G. M., MCALPINE, B., WHITTERN, N., J. D. BUKO,A. M. (1990), Altromycins, novel pluramycin-like antibiotics. 11. Isolation and elucidation of structure, J. Antibiot. 43, 229-237. BRINK, F., BRINK, VERBEET, P., DEM. G., M. BOER, H. A. (1994), Spectinomycin interacts specifically with the residues G1064 and C1192 in 16s rRNA, thereby potentially freezing this molecule into an inactive conformation, Nucleic Acids Res. 22, 325-331. BRYAN, E., KWAN, (1983), Roles of ribosomal L. S. binding, membrane potential, and electron transport in bacterial uptake of streptomycin and gentamicin, Antimicrob. Agents Chemother. 23,835-845. BUSSE, H.-J., WOSTMANN, BAKKER, P. K., E. (1992), The bactericidal action of streptomycin: membrane permeabilization caused by the insertion of mistranslated proteins into the cytoplasmic membrane of Escherichia coli and subsequent caging of the antibiotic inside the cells due to degradation of the protein, J. Gen Microbiol. 138,551-556. BUTTNER, J. (1989), RNA polymerase heteroM. geneity in Streptomyces coelicolor A3(2), Mol. Microbiol. 3, 1653-1659. J. CHADWICK, J., WHELAN, (Eds.) (1992), SecD. ondary Metabolites: Their Function and Evolu-
11 References
ANDO,O., SATAKE, ITOI,K., SATOA., NAKAH., JIMA, TAKAHASHI, HARUYAMA, OHM., S., H., KUMA, KINOSHITA, ENOKITA, (1991), Y., T., R. Trehazolin, a new trehalase inhibitor, J. Antibiot. 44,1165-1168. ANGELL, SCHWARZ, BIBB,M. (1992), The S., E., glucose kinase gene of Streptomyces coelicolor A3(2): its nucleotide sequence, transcriptional analysis and role in glucose repression, Mol. Microbiol. 6, 2833-2844. ASANO,N., KATAYAMA, TAKEUCHI, FuK., M., RUMOTO, T., KAMEDA, Y., MATSUI, (1989), K. Preparation of 3-amino-3-deoxy derivatives of trehalose and sucrose and their activities, J. Antibiot. 42, 585-590. ATSUMI, IINUMA, NOSAKA, UMEZAWA, S., H., C., K. (1990a), Biological activities of cyclophellitol, J. Antibiot. 43, 1579-1585. S., K., H., ATSUMI, UMEZAWA, IINUMA, NAGANAH., Y., WA,H., NAKAMURA, IITAKA, TAKEUCHI, T. (1990b), Production, isolation and structure determination of a novel Pglucosidase inhibitor, cyclophellitol, from Phellinus sp., J. Antibiot. 43, 49-53. AUBERT-PIVERT, DAVIES, (1994), BiosyntheE., J. sis of butirosin in Bacillus circulans NRRL B3312 identification by sequence analysis and insertional mutagenesis of the butB gene involved in antibiotic production, Gene 147, 1-11. AWATA,M., MUTO,N., HAYASHI, YAGINUM., MA, S. (1986), A new aminoglycoside antibiotic, substance AC4437, J. Antibiot. 39, 724-726. BAKKER, P. (1992), Aminoglycoside and aminoE. cyclitol antibiotics: hygromycin B is an atypical bactericidal compound that exerts effects on cells of Escherichia coli characteristic for bacteriostatic aminocyclitols, J. Gen Microbiol. 138, 563-569. BASAK,K., MAJUMDAR, K. (1973), Utilization S. of carbon and nitrogen sources by Streptomyces kanamyceticus for kanamycin production, Antimicrob. Agents Chemother. 4, 6-11. BASCARAN, SANCHEZ, HARDISSON, V., L., C., BRANA, F. (1991), Stringent response and inA. itiation of secondary metabolism in Streptomyces clavuligerus, J. Gen. Microbiol. 137, 1625-1634. BEAU, J.-M. (1990), Polyene macrolides: stereostructural elucidation and synthetic studies of a few members, in: Recent Progress in the ChemiG., cal Synthesis of Antibiotics (LUKACS, OHNO, M., Eds.), pp. 135-182. Berlin: Springer-Verlag. J. BENVENISTE, DAVIES, (1973), AminoglycoR., side antibiotic inactivating enzymes in actinomycetes similar to those present in clinical isolates
I1 References
tion, Ciba Foundation Symposium 171. Chichester: John Wiley & Sons. CHAMPNESS, C., CHATER, F. (1994), RegulaW. K. tion and interaction of antibiotic production and morphological differentiation in Streptomyces spp., in: Regulation of Bacterial Differentiation (PIGGOT, P., MORAN c. P., YOUNGMAN, P., JR., Eds.), pp. 61-94. Washington, DC: American Society for Microbiology. CHATER,K. F. (1989), Multilevel regulation of Streptomyces differentiation, Trends Genet. 5, 372-377. K. CHATER, F. (1992), Genetic regulation of secondary metabolic pathways in Streptomyces, in: Secondary Metabolites: Their Function and EvoJ., D., J., lution (DAVIES, CHADWICK, WHELAN, K., WIDDOWS, Eds.), pp. 144-156. Chichester: John Wiley & Sons. CHEN,Y.-M., WALKER, B. (1977), TransaminaJ. tions involving keto- and amino-inositols and glutamine in actinomycetes which produce gentamicin and neomycin, Biochem. Biophys. Res. Commun. n, 688-692. CHIAO, S., XIA,T. H., MEI,B. G., JIN,Z. K., Gu, J. W. L. (1995), Rifamycin SV and related ansamycins, in: Biochemistry and Genetics of Antibiotic L., C., Biosynthesis (VINING, S T U ~ A R D , Eds.), pp. 477498. Boston, MA: Butterworth-Heinemann. CHUNG, T., CROSE,L. L. (1990), Transposon S. Tn4556-mediated DNA insertion and site-directed mutagenesis; in: Proc. GIM 1990, Vol. I (HESLOT,H., DAVIES,J., FLORENT, BOBIJ., CHON, DURAND, PENASSE, Eds.), pp. L., G., L., 207-218. Paris: SociCtC FranGaise de Microbiologie. COLEMAN, S., FRASER, R. (1993), Acylketene R. J. [4 +2]cycloadditions: divergent de novo synthesis of 2,6-dideoxy sugars, J. Org. Chem. 58, 385392. COOK,G. R., BEHOLZ, G., STILLE, R. (1994), L. J. Construction of hydroxylated alkaloids ( + / -)mannonolactam, ( + )-deoxymannojirimycin, and ( +/-)-prosopinine through aza-annulation, J. Org. Chem. 59,3575-3584. COURVALIN, (1994). Transfer o antibiotic reP. f sistance genes between gram-positive and gramnegative bacteria, Antimicrob. Agents Chemother. 38, 1447-1451. COURVALIN, WEISBLUM, DAVIES, (1977), P., B., J. Aminoglycoside-modifying enzyme of an antibiotic-producing bacterium acts as a determinant of antibiotic resistance in Escherichia coli, Proc. Natl. Acad. Sci. USA 74, 999-1003. CRAMERI, DAVIES, E. (1986), Increased proR., J. duction of aminoglycosides associated with am-
477
plified antibiotic resistance genes, J. Antibiot. 39, 128-135. CUNDLIFFE, (1989), How antibiotic-producing E. organisms avoid suicide, Ann. Rev. Microbiol. 43,207-233. CUNDLIFFE, (1990), Recognition sites for antiE. biotics within rRNA, in: The Ribosome. Structure, Function, and Evolution (HILL, W. E., DAHLBERG, GAR RE^, R. A,, MOORE,P. A., B., SCHLESSINGER, WARNER, R., Eds.), D., J. pp. 479490. Washington, DC: American Society for Microbiology. CUNDLIFFE, (1992a), Self-protection mechaE. nisms in antibiotic producers, in: Secondary Metabolites: Their Function and Evolution, Ciba Foundation Symposium I71 (CHADWICK, D, WHELAN,J., Eds.), pp. 199-214. Chichester: John Wiley & Sons. CUNDLIFFE, (1992b), Glycosylation of macrolide E. antibiotics in extracts of Streptomyces lividans, Antimicrob. Agents Chemother. 36,348-352. DAIRI,T., HASEGAWA, (1989), Common bioM. synthetic feature of fortimicin-group antibiotics, J. Antibiot. 42, 934-943. DAIRI, OHTA,T., HASHIMOTO, HASEGAT., E., WA,M. (1992a), Self cloning in Micromonospora olivasterospora of fms genes for fortimicin A (astromicin) biosynthesis, Mol. Gen. Genet. 232, 262-270. DAIRI,T., OHTA,T., HASHIMOTO, HASEGAE., WA,M. (1992b), Organization and nature of fortimicin A (astromicin) biosynthetic genes studied using a cosmid library of Micromonospora olivasterospora DNA, Mol. Gen. Genet. 236, 3948. DAIRI, T., YAMAGUCHI, HASEGAWA, K., M. (1992c), N-formimidoyl fortimicin A synthase, a unique oxidase involved in fortimicin A biosynthesis: purification, characterization and gene cloning, Mol. Gen. Genet. 236, 49-59. DAVIES, (1992), Another look at antibiotic resistJ. ance, J. Gen. Microbiol. 138, 1553-1559. DAVIES, SMITH,D. I. (1978), Plasmid-deterJ., mined resistance to antimicrobial agents, Ann. Rev. Microbiol. 32, 469-518. DAVIES, YAGISAWA, (1983), The aminocycliJ., M. to1 glycosides (aminoglycosides), in: Biochernistry and Genetic Regulation of Commercially Important Antibiotics (VINING, C., Ed.), pp. 311L. 354. London: Addison-Wesley Publishing Company. DAVIES, CANNON, MAUER, B. (1988), J., M., M. Myomycin: mode of action and mechanism of resistance, J. Antibiot. 41, 366-372. B. DAVIS, D. (1987), Mechanism of bactericidal action of aminoglycosides,Microbiol. Rev. 51,341350.
478
DAVIS, D., CHEN, TAI,P. C. (1986), Misread B. L., J., FLORENT, BOBICHON, DURAND, J., L., G., protein creates membrane channels: an essential PENASSE, Eds.), pp. 379-392. Paris: Societ6 L., FranGaise de Microbiologie. step in the bactericidal action of aminoglycosides, Proc. Natl. Acad. Sci. USA 83, 6164- DISTLER, MANSOURI, MAYER,G., STOCKJ., K., 6168. MANN, PIEPERSBERG, (1992), StreptomyM., W. DE STASIO, A., MOAZED, NOLLER, F., E. D., H. cin production and its regulation, Gene 115,105DAHLBERG, E. (1989), Mutations in 16s riA. 111. bosomal RNA disrupt antibiotic-RNA interac- DREPPER, PAPE,H. (1996), Acarbose 7-phosA., tions, EMBO J. 8, 1213-1216. photransferase from Actinoplanes sp.: purificaDECKER, HAAG, UDVARNORKI, ROHR, H., G., G., tion, properties, and possible physiological funcJ. (1995), Novel genetically engineered tetracetion, J. Antibiot. 49, 664-668. nomycins, Angew. Chem. (Int. Ed Engl.) 34, EMBLEY, M., STACKEBRANDT, (1994), The T. E. 1107-1 110. molecular phylogeny and systematics of the actiDEGWERT, VAN HULST, R.,PAPE, H., HERU., nomycetes, Ann. Rev. Microbiol. 48,257-289. J. P. ROLD, E., BEALE, M., KELLER, J., LEE, ENGLUND, T. (1993), The structure and biosynR. P. J. P., FLOSS, G. (1987), Studies on the syntheH. thesis of glycosyl phosphatidylinositiol protein sis of the a-glucosidase inhibitor acarbose: valanchors, Annu. Rev. Biochem. 62, 112-138. ienamine, a m-C7N unit not derived from the ENSIGN, C. (1988), Physiological regulation of J. shikimate pathway, J. Antibiot. 40,855-861. sporulation of Streptomyces griseus, in: Biology DEMAIN L. (1989), Carbon source regulation of A. of Actinomycetes 88 (OKAMI, BEPPU,T., Y., idiolite biosynthesis in actinomycetes, in: RegulOGAWARA, Eds.), pp. 309-315. Tokyo: JaH., ation of Secondary Metabolism in Actinomycetes pan Scientific Societies Press. (SHAPIRO, Ed.), pp. 127-134. Boca Raton, S., ETIENNE, ARMAU, DASSIN, TIRABY, G., E., M., G. FL: CRC Press. (1991), A screening method to identify antibiotDEMAIN, L., INAMINE, (1970), Biochemistry A. E. ics of the aminoglycoside family, Antibiot. 44, and regulation of streptomycin and mannosido1357-1366. streptomycinase (a-D-mannosidase) formation, FISHER, H. (1992), Glutamine synthesis in StrepS. Bacteriol. Rev. 34, 1-19. tomyces - a review, Gene 115, 13-17. W. DICKENS, L., YE, J., STROHL, R. (1996), M. M., M. E., Cloning, sequencing and analysis of aklavike- FLORES, E., PONCE, RUBIO, HUITRON, C. (1993), Glucose and glycerol repression of atone reductase from Streptomyces sp. strain C5, amylase in Streptomyces kanamyceticus and isoJ. Bacteriol. 178, 33843388. lation of deregulated mutants, Biotechnol. Lett. DISTLER, PIEPERSBERG, (1985), Cloning and J., W. 15,595-600. characterization of a gene from Streptomyces griseus coding for a streptomycin-phosphorylating FLOSS,H. G. (in press), Biosynthesis of Antibiotics, in: Proc. 6th Con$ Genetics and Molecular Bioactivity, FEMS Microbiol, Lett. 28, 113-117. J., K., W., DISTLER, KLIER, PIENDL, WERBITZKI, logy of Industrial Microorganisms, Bloomington, IN, Oct. 1996 (HEGEMANN, D; Baltz, R. H., G. 0. . BOCK, A., KRESZE, G., PIEPERSBERG, W. BALTZ, H., Skatrund, P. L., Eds.). R. (1985), Streptomycin biosynthesis in StreptoH. J. myces griseus. I. Characterization of streptomy- FLOSS, G., BEALE, M. (1989), Investigation of the biosynthesis of antibiotics, Angew. Chem. cin-idiotrophic mutants, FEMS Microbiol. Lett. (Int. Edn.) 28, 146-177. 30,145-150. DISTLER, BRAUN, EBERT, PIEPERSBERG, FOSTER,T. J. (1983), Plasmid-determined resistJ., C., A., ance to antimicrobial drugs and toxic metal ions W. (1987a), Gene cluster for streptomycin bioin bacteria, Microbiol. Rev. 41, 361-409. synthesis in Streptomyces griseus: Analysis of a H. J. I. central region including the major resistance FRAIMOW, S., GREENMAN, B., LEVITON, M. M., DOUGHERTY, J., MILLER, H. (1991), T. gene, Mol. Gen. Genet. 208,204210. Tobramycin uptake in Escherichia coli is driven J., K., DISTLER, EBERT,A., MANSOURI, PISSOWOTZKI, STOCKMANN, PIEPERSBERG, by either electrical potential or ATP, J. BacteriK., M., 01. 173,2800-2808. W. (1987b), Gene cluster for streptomycin bioJ. H. synthesis in Streptomyces griseus: nucleotide se- FRENCH, C., BARTZ,Q. R., DION, W. (1973), Myomycin, a new antibiotic, J. Antibiot. 26,272quence of three genes and analysis of transcrip283. tional activity, Nucleic Acids Res. 15, 80418056. FUJITA, SHINDO, MIWA, YOSHIDA, I. Y., K., Y., K. DISTLER, MAYER, PIEPERSBERG, (1990), J., G., W. (1992), Bacillus subtilis inositol dehydrogenaseRegulation of biosynthesis of streptomycin, in: encoding gene (idh): sequencing and expression Proc. GIM 1990, Vol. 1 (HESLOT, DAVIES, H., in Escherichia coli, Gene 108, 121-125.
I1 References
479
GALE, E. F., CUNDLIFFE, REYNOLDS, E., E., P. tion of biosynthetic genes of fortimicin A (astromicin), Actinomycetol. 5, 126131. M. M. RICHMOND, H., WARING, J. (1981), The Molecular Basis of Antibiotic Action, 2nd Edn. HASEGAWA, (1992), A novel, highly efficient M. London: Wiley Interscience. gene-cloning system in Micromonospora applied GODA,S. K., AKHTAR, (1992), Neomycin bioM. to the genetic analysis of fortimicin biosynthesis, Gene 115,85-91. synthesis: the incorporation of D-6-deoxyglucose derivatives and variously labelled glucose into HEINZEL, WERBITZKY, DISTLER,J., PIEO., P., the 2-deoxystreptamine ring. Postulated involvePERSBERG, w. (1988), A second streptomycin ment of 2-deoxyinosose synthase in the biosynresistance gene from Streptomyces griseus codes thesis, J. Antibiot. 45, 984-994. for streptomycin-3 -phosphotransferase. Relationship between antibiotic and protein kinases, K. GOEKE, (1986), Enzymatische Untersuchungen Arch. Microbiol. 150, 184-192. zum Zuckerstoffwechsel und zur Biosynthese des a-Glucosidase-Inhibitors Acarbose bei Acti- HILL, W. E., DAHLBERG, GARRETT, A., A., R. noplanes spec., Dissertation, University of MunJ. MOORE, B., SCHLESSINGER, WARNER, P. D., ster. R. (1990), The Ribosome. Structure, Function, GORINI, (1974), Streptomycin and misreading of L. and Evolution. Washington, DC: American SoM., ciety for Microbiology. the genetic code, in: Ribosomes (NOMURA, TISSIERES, LENGYEL, Eds.), pp. 791-803. HIROSE-KUMAGAI, YAGITA, AKAMATSU, A., P., A., A., New York Cold Spring Harbor Laboratory. N. (1982), UDP-N-methyl-D-glucoseamine-phosGOULD, J. (1992), Exploring the intricate details S. phate. A possible intermediate of N-methyl-Lof antibiotic biosynthesis, in: Secondary-Metabglucosamine moiety of streptomycin, A . Antibiot 35, 1571-1577. olite Biosynthesis and Metabolism (PETROSKI, R. J., MCCORMICK, P., Eds.), pp. 11-25. New HOLMES, J., DROCOURT, TIRABY, GUNS. D. D., G., York: Plenum Press. DIFFE, E. (1991), Cloning of an aminoglycoside GRAFE,U. (1992), Biochemie der Antibiotika. Heiresistance-endodinggene, kame, from Saccharodelberg: Spektrum Akademischer Verlag. polispora hirsuta: Comparison with kamB from Streptomyces tenebrarius, Gene 102, 19-26. GRISEBACH, (1978), Biosynthesis of sugar comH. C., C., ponents of antibiotic substances, Adv. Carbo- HOLT, T. G., CHANG, LAURENTWINTER, MURAKAMI, GARRELS, I., DAVIES, E., T., J. J. hydr. Chem. Biochem. 3581-126. THOMPSON, J. (1992), Global change in gene C. GUILFOILE, G., HUTCHINSON, R. (1991), A P. C. expression related to antibiotic synthesis in bacterial analogue of the mdr gene of mammalStreptomyces hygroscopicus, Mol. Microbiol. 6, ian tumor cells is present in Streptomyces peuce969-980. tius, producer of daunorubicin and doxorubicin, Proc. Natl. Acad. Sci. USA 88, 8553-8557. A., S., HONG,S. K., MATSUMOTO, HORINOUCHI, BEPPU, (1993), Effects of protein kinase inhiT. ANCOCK, R. E. W. (1981a), Aminoglycoside upbitors on in vitro protein phosphorylation and take and mode of action with special reference cellular differentiation of Streptomyces griseus, to streptomycin and gentamicin. I. Antagonists Mol. Gen. Genet. 236, 347-354. and mutants, J. Antimicrob. Chemother. 8, 249276. HOPWOOD, A., SHERMAN, H., KHOSLA, C., D. D. T. BIBB, J., SIMPSON, J., FERNANDEZ-MOREM. ANCOCK, E. W. (1981b), Aminoglycoside upR. NO, M. A., MARTINEZ, MALPARTIDA, E., F. take and mode of action with special reference (1990), Hybrid pathways for the production of to streptomycin and gentamicin. 11. Effects of secondary metabolites, in: Proc. 6th Int. Symp. aminoglycosides on cells, J. Antimicrob. CheH., J., Genet. Ind. Microorg. (HESLOT, DAVIES, mother. 8, 429-445. FLORENT, BOBICHON, DURAND, PENJ., L., G., S. HANCOCK, E. W., FARMER, W., LI, Z., R. EASSE, L., Eds.), pp. 259-270. Paris: SocietC POOLE,K. (1991), Interaction of aminoglycoFransaise de Microbiologie. sides with outer membranes and purified lipopoS., lysaccharide and OmpF porin of Escherichia HORINOUCHI, BEPPU,T. (1992), Regulation of secondary metabolism and cell differentiation in coli, Antimicrob. Agents Chemother. 35, 1309Streptomyces: A-factor as a microbial hormone 1314. and the AfsR protein as a component of a twoHARDICK, J., HUTCHINSON, W., TREW, J., D. D. S. component regulator system, Gene 115, 167WELLINGTON, M. H. (1992), Glucose is a preE. 172. cursor of 1-deoxynojirimycin and l-deoxymanS., nonojirimycin in Streptomyces subrutilus, Tetra- HORINOUCHI, BEPPU,T. (1994). A-factor as a microbial hormone that controls cellular differhedron 48,6285-6296. entiation and secondary metabolism in StreptoM. HASEGAWA, (1991), A gene cloning system in myces griseus, Mol. Microbiol. 12, 859-864. Micromonospora which revealed the organiza-
480

ITOH,S., ODAKURA, KASE,H., SATHO, TAY., S., KAHASHI, IIDA, T., SHIRAHATA, NAK., K., KAYAMA, (1984), Biosynthesis of astromicin K. and related antibiotics. I. Biosynthetic studies by bioconversion experiments, J. Antibiot. 37,16641669. KAGAN, M., CLARKE, (1994), Widespread ocR. S. currence of three sequence motifs in diverse Sadenosylmethionine-dependent methyltransferases suggests a common structure for these enzymes, Arch. Biochem. Biophys. 310,417-427. KAKINUMA, (1982), Biosynthesis of ribostamyK. cin. Application of the deuterium label, in: Trends in Antibiotic Research. Genetics, Biosyntheses, Actions, and New Substances (UMEZAWA, H., DEMAIN,A. L., HATA, T., HUTCHINSON, C. R., Eds.), pp. 185-194. Tokyo: Japan Antibiotics Research Association. KAMEDA, ASANO,N., YOSHIKAWA, TAY., M., KEUCHI, YAMAGUCHI, MATSUI, SAM., T., K., TOSHI,H., FUKASE,H. (1984), Valiolamine, a new 0-glucosidase inhibiting aminocyclitol produced by Streptomyces hygroscopicus, J. Antibiot. 37, 1301-1307. KARWOWSKI, P., JACKSON, BOBIK, A., J. M., T. R. PROKOP, F., THERIAULT,J. (1984), SpenolJ. imycin, a new spectinomycin-type antibiotic. I. Discovery, taxonomy and fermentation, J. Antibiot. 37, 1513-1518. KASE,H., ODAKURA, TAKAZAWA, KITAY., Y., MURA, NAKAYAMA, (1982), Biosynthesis s., K. of sagamicin and related aminoglycosides, in: Trends in Antibiotic Research. Genetics, Biosyntheses, Actions, and New Substances (UMEZAWA,H., DEMAIN, L., HATA, T., HUTCHINA. SON,C. R., Eds.), pp. 195-212. Tokyo: Japan Antibiotics Research Association. KASE,H., IWAHASHI, MATSUDA, (1986), KK., Y. 252a, a potent inhibitor of protein kinase C of microbial origin, J. Antibiot. 39, 1059-1065. KATZ,L., DONADIO, (1993), Polyketide syntheS. sis: prospects for hybrid antibiotics, Ann. Rev. Microbiol. 47, 875-912. KENNEDY, F. (1988), Carbohydrate Chemistry. J. Oxford Clarendon Press. KHOKHLOV, S. (1988), Results and perspectives A. of actinomycete autoregulators studies, in: BiolY., ogy of Actinomycetes '88 (OKAMI, BEPPU,T., OGAWARA, Eds.), pp. 338-345. Tokyo: JaH., pan Scientific Societies Press. I. L. KHOKHLOV, S., TOVAROVA, I., BORISOVA, A. N., PLINER,S. A., SHEVCHENKO, A., KORL. I. NITSKAYA, Y., IVKINA, S., RAPOPORT, E. N. A. (1967), A-factor responsible for the biosynthesis of streptomycin by a mutant strain of Actinomyces streptomycini, Doklady Akad. Nauk SSSR 177,232-235.
HORNEMANN, KEHRER,J. P., NUNEZ,C. S., U., RANIERI, L. (1974), D-Glucosamine and L-ciA. trulline, precursors in mitomycin biosynthesis in Streptomyces verticillatus, J. Am. Chem. Soc. 96, 320-322. D., L. G. HORTON, HAWKINS, D., MCGARVEY, J. (1989), Trends in synthetic carbohydrate chemistry, ACS Symp. Ser. 386. Washington, D C American Chemical Society. HOTTA, K., ISHIKAWA, ICHIHARA, NAGAJ., M., NAWA, MIZUNO, (1988), Mechanism of inH., S. creased kanamycin-resistance generated by protoplast regeneration of Streptomyces griseus. I. cloning of a gene segment directing a high level of an aminoglycoside 3-N-acetyltransferase activity, J. Antibiot. 41, 94-103. HO-ITA,K., MORIOKA, OKAMI, (1989), BioM., Y. synthetic similarity between Streptomyces tenjimariensis and Micromonospora olivasterospora which produce fortimicin-group antibiotics, J. Antibiot. 42, 745-751. HO-ITA, K., DAVIES, YAGISAWA, (1995), J., M. Aminoglycosides and aminocyclitols (other than streptomycin), in: Biochemistry and Genetics of Antibiotic Biosynthesis (VINING, STU-ITARD, L., Eds.), pp. 571-595. Stoneham: Butterworth-Heinemann. T., M. HUDLICKY, CEBULAK, (1993), Cyclitols and Their Derivatives. Weinheim: VCH. HUTCHIN, CORTOPASSI, (1994), Proposed T., G. molecular and cellular mechanism for aminoglycoside ototoxicity, Antimicrob. Agents Chemother. 38,2517-2520. ICHIMURA, M., KOGUCHI,T., YASUZAWA, T., TOMITA, (1987), CV-1, a new antibiotic proF. duced by a strain of Streptomyces sp. I. Fermentation, isolation and biological properties of the antibiotic, J. Antibiot. 40,723-726. IKEDA,Y., GOMI,S., YOKOSE,K., NAGANAWA, H., IKEDA, T., MANABE,M., HAMADA,M., KONDO, S., UMEZAWA,H. (1985a), A new streptomycin group antibiotic produced by Streptomyces sioyaensis, J. Antibiot. 38, 18031805. IKEDA, KONDO,S., KANAI, SAWA, HAY., F., T., MADA, M., TAKEUCHI, T., UMEZAWA,H. (1985b), A new destomycin-family antibiotic produced by Saccharopolyspora hirsuta, J. Antibiot. 38, 436-438. INAMINE, DEMAIN, L. (1975), MannosidoE., A. streptomycin hydrolase, Methods Enzymol. 43, 637-640. INOUYE, SHOMURA, WATANABE, TOTS., T., H., SUGAWA,K., NIIDA,T. (1973), Isolation and gross structure of a new antibiotic SS-56C and related compounds, J. Antibiot. 26, 374-385.
I1 References
481
KIM,C.-G., KIRSCHNING, BERGON, ZHOU, A., P., ase has a regulatory role in carbon catabolite reP., Su, E., SAUERBREI, NING,S., AHN, Y., B., pression in Streptomyces coelicolor, J. Bacteriol. BREUER, M., LEISTNER,E., FLOSS, H. G. 176,2694-2698. (1996), Biosynthesis of 3-amino-5-hydroxyben- LACALLE, A., TERCERO, A., JIMENEZ, R. J. A. zoic acid, the precursor of mC7N units in ansa(1992). Cloning of the complete biosynthetic mycin antibiotics, J. Am. Chem. SOC. 118, 7486gene cluster for an aminonucleoside antibiotic, 7491. puromycin, and its regulated expression in heteroKIRBY, (1990). Evolutionary origin of aminoglyR. logous host, EMBO J. 11,785-792. coside phosphotransferase resistance genes, J. LAM, K. S., FLORENZA, SCHROEDER, R., S., D. Mol. Evol. 30,489-492. DOYLE, W., PEARCE, J. (1989). BiosyntheT. C. KNIEP,B., GRISEBACH, (1976), Enzymatic synH. sis of rebeccamycin, a novel antitumor agent, in: thesis of streptomycin. Transfer of L-dihydroNovel Microbial Products for Medicine and Agstreptose from dTDP-L-dihydrostreptose to riculture (DEMAIN,A. L., SOMKUTI, A., G. streptidine-6-phosphate, FEBS Lett. 65, 44-46. HUNTER-CREVA, C., ROSSMOORE, W., J. H. KNIEP,B., GRISEBACH, (1980), Biosynthesis of H. Eds.), pp. 63-66. Amsterdam: Elsevier Science Publishers. streptomycin. Purification and properties of a dTDP-~-dihydrostreptose:streptidine-6-phos- LESKIW, K., BIBB,M. J., CHATER, F. (1991a), B. K. phate dihydrostreptosyltransferasefrom StreptoThe use of a rare codon specifically during demyces griseus, Eur. J. Biochem. 105, 139-144. velopment, Mol. Microbiol. 5, 2861-2867. KNIGHTON, R., ZENG, TENEYCK, F., ASH-. LESKIW, K., LAWLOR, J., FERNANDEZ-ABAD. J., L. B. E. FORD, v. A., XUONG, N. H., TAYLOR, s. s. LOS, J. M., CHATER, F. (1991b), TTA codons K. (1991), Crystal structure of the catalytic subunit in some genes prevent their expression in a class of cyclic adenosine monophosphate-dependent of developmental, antibiotic-negative Streptoprotein kinase, Science 253, 407-414. myces mutants, Proc. Natl. Acad. Sci. USA 88, KOBAYASHI, SHIOZAKI, (1994), Synthesis of Y., M. 2461-2465. trehazolin panomer, J. Antibiot. 47, 243-246. LIRAS, ASTURIAS, A., MARTIN, F. (1990), P., J. J. KOGA, Y., NISHIHARA, MORII,H., AKAGAM., Phosphate control sequences involved in tranWA-MATSUSHITA, (1993), Ether lipids of meM. scriptional regulation of antibiotic biosythesis, TIBTECH 8,184-189. thanogenic bacteria: structures, comperative aspects, and biosynthesis, Microbiol. Rev. 57, 164- LISHNEVSKAYA,B., KUZINA, A., ASINOOVSE. Z. 182. KAYA, K., BELOUSOVA, I., MALKOV, N. I. M. KOHLHEPP, J., Hou, L., GILBERT, N. (1994), S. D. A., RAVINSKAYA, YA. (1986), Cyclic adenoA. Pig kidney (LLC-PKl) cell membrane fluidity sine-3',5 '-monophosphoric acid in Streptomyces antibioticus and its possible role in the reguladuring exposure to gentamicin or tobramycin, Antimicrob. Agents Chemother. 38,2169-2171. tion of oleandomycin biosynthesis and culture growth, Mikrobiologiya 55, 350. KORZYBSKI, KOWSZYK-GINDIFER, KURYT., Z., LOWICZ, (1978), Antibiotics. Origin, Nature, LIU, H.-W., THORSON, S. (1994), Pathways and W. J. and Properties. Washington, D C American Somechanisms in the biogenesis of novel deoxysuciety for Microbiology. gars by bacteria, Ann. Rev. Microbiol. 48, 223256. KROHN, KIRST,H. A., MAAG,H. (1993), AntiK., biotics and Antiviral Compounds. Chemical Syn- LONDEI, ALTAMURA, SANZ, L., AMILS, P., S., J. R. thesis and Modification. Weinheim: VCH. (1988), Aminoglycoside-induced mistranslation in thermophilic archaebacteria, Mol. Gen. Genet. KUMADA, HORINOUCHI, UOZUMI, BEPY., S., T., 214,48-54. PU, T. (1986), Cloning of a streptomycin-production gene directing synthesis of N-methyl-L-glu- LOOK,G. C., FOTSCH, H., WONG,C.-H. (1993), C. cosamine, Gene 43, 221-224. Enzyme-catalysed organic synthesis: practical J. routes to aza sugars and their analogs for use as Kuo, M. S., YUREK, A., COATS, H., LI, G. P. D. (1989), Isolation and identification of 7,8-dideglycoprocessing inhibitors, Acc. Chem. Res. 26, 182-190. methyl-8-hydroxy-5-deazariboflavin, unusual an cosynthetic factor in streptomycetes from Strep- LORIAN, (Ed.) (1980), Antibiotics in Laboratory V. tomyces lincolnensis, J. Antibiot. 42, 475-478. Medicine. Baltimore, M A Williams & Wilkins. Kuo, M. S., YUREK, A., COATS,J. H., CHUNG, LUCHER,L. A., CHEN, Y.-M., WALKER,J. B. D. S. T., LI, G. P. (1992), Isolation and identifica(1989), Reactions catalysed by purified L-gluttion of 3-propylidene-A'-pyrroline-5-carboxylic amine:keto-scyllo-inositol aminotransferase, an acid, a biosynthetic precursor of lincomycin, J. enzyme required for biosynthesis of aminocycliAntibiot. 45,1773-1777. to1 antibiotics, Antimicrob. Agents Chemother. KWAKMAN, POSTMA, (1994), Glucose kinM., W. 33,452-459.
482
LUKACS, OHNO,M. (1990), Recent Progress in G., the Chemical Synthesis of Antibiotics. Berlin: Springer-Verlag. MACNEIL, J. (1995), Avermectins, in: BiochemD. istry and Genetics of Antibiotic Biosynthesis (VINING, STUTTARD, Eds.), pp. 421442. L., C., Boston, MA: Butterworth-Heinemann. H. MAIER,S., GRISEBACH, (1979), Biosynthesis of streptomycin. Enzymic oxidation of dihydrostreptomycin (6-phosphate) to streptomycin (6phosphate) with a particulate fraction of Streptomyces griseus, Biochim. Biophys. Acta 586, 231241. MALLAMS, K. (1988), The carbohydrate-conA. taining antibiotics, in: Carbohydrate Chemistry (KENNEDY, F., Ed.), pp. 73-133. Oxford: J. Claendron Press. MANSOURI, PIEPERSBERG, (1991), Genetics K., W. of streptomycin production in Streptomyces griseus: nucleotide sequence of five genes, strFGHIK, including a phosphatase gene, Mol. Gen. Genet. 228, 459469. MANSOURI, PISSOWOTZKI, DISTLER,J., K., K., MAYER,G., HEINZEL, BRAUN, EBERT, P., C., A., PIEPERSBERG, (1989), Genetics of strepW. tomycin production, in: Genetics and Molecular Biology of Industrial Microorganisms (HERSHBERGER, c. L., QUEENER, w., HEGEMAN, s. G., Eds.), pp. 61-67. Washington, D C American Society for Microbiology. MARSH, WELLINGTON, M. H. (1994), MolecP., E. ular ecology of filamentous actinomycetes in soil, in: Molecular Ecology of Rhizosphere Microorganisms. Biotechnology and the Release of D. GMOs (O'GARA, F., DOWLING, N., BOESTEN,B., Eds.), pp. 133-149. Weinheim: VCH. MARTIN, F. (1989), Molecular mechanisms for J. the control by phosphate of the biosynthesis of antibiotics and other metabolites, in: Regulation of Secondary Metabolism in Actinomycetes (SHAPIRO, Ed.), pp. 213-237. Boca Raton, S., FL: CRC Press. J. A. MARTIN, F., DEMAIN, L. (1980), Control of antibiotic biosynthesis, Microbiol. Rev. 44, 230251. MARTIN, F., LIRAS,P. (1989), Organization and J. expression of genes involved in the biosynthesis of antibiotics and other secondary metabolites, Ann. Rev. Microbiol. 43,173-206. MAYER, (1994), Molekulare Analyse und EvoG. lution von 5 '-Hydroxystreptomycin- und Bluensomycin-Biosynthesegenen aus Streptomyces glaucescens GLA.0 und Streptomyces bluensis ISP 5564, Dissertation, Bergische Universitat GH, Wuppertal. MAYER, VOGTLI,M., PISSOWOTZKI, HUTG., K., TER, R., PIEPERSBERG, (1988), Colinearity W.
of streptomycin production genes in two species of Streptomyces. Evidence for occurrence of a second amidino-transferase gene, Mol. Gen. (Life Sci. Adv.) 7, 83-87. MAZODIER, DAVIES,J. (1991), Gene transfer P., between distantly related bacteria, Annu. Rev. Genet. 25, 147-171. MCCUE,L. A., KWAK,J., BABCOCK, J., KENM. DRICK, F. (1992), Molecular analysis of spoK. rulation in Streptomyces griseus, Gene 115, 173179. MEHLING, WEHMEIER, F., PIEPERSBERG, A., U. W. (1995), Nucleotide sequence of Streptomyces 16s ribosomal DNA: towards a specific identification system for streptomycetes using PCR, Microbiology 141, 2139-2147. MILLER,A. L., WALKER, B. (1969), Enzymatic J. phosphorylation of streptomycin by extracts of streptomycin-producing strains of Streptomyces, J. Bacteriol. 99, 401-405. MIYAKE, HORINOUCHI, YOSHIDA, CHIK., S., M., BA, N., MORI,K., NOGAWA, N., MORIKAWA, N., BEPPU,T. (1989), Detection and properties of A-factor-binding protein from Streptomyces griseus, J. Bacteriol. 171, 4298-4302. MIYAKE,K., KUZUYAMA, HORINOUCHI, T., S., BEPPU,T. (1990), The A-factor-binding protein of Streptomyces griseus negatively controls streptomycin production and sporulation, J. Bacteriol. 172,3003-3008. MOAZED,D., NOLLER,H. F. (1987), Interactions of antibiotics with 16s rRNA, Nature 327, 389394. MULLER,L. (1989), Chemistry, biochemistry and therapeutic potential of microbial a-glucosidase inhibitors, in: Novel Microbial Products for Medicine and Agriculture (DEMAIN, L., SOMKUTI, A. G. A., HUNTER-CREVA, C., ROSSMOORE, J. H. W., Eds.), pp. 109-116. Amsterdam: Elsevier Science Publishers. MULLER,A., KELLER-SCHIERLEIN, BIELECKI, W., J., RAK,G., STUMPFEL, ZAHNER, (1986), J., H. (2S,3R,4R,6R) - 2,3,4,- Trihydroxy - 6 - methylhexanon aus zwei Actinomyceten-Stammen, Helv. Chim. Acta 69, 1829-1832. MUNRO, H. G., TANIGUCHI, RINEHART, M. M., K. L., GOTTLIEB, STOUDT, H., ROGERS,T. D., T. 0. (1975), Carbon-13 evidence for the stereochemistry of streptomycin biosynthesis from glucose, J. Am. Chem. SOC.91,4782-4783. MURPHY, J., HEYCKE, BANFALVI, TATE, P. N., Z., M. E., DE BRUJIN, KONDOROSI, TEMPE, F., A., J., SCHELL, (1987), Genes for the catabolism J. and synthesis of an opine-like compound in Rhizobium meliloti are closely linked and on the Sym plasmid, Proc. Natl. Acad. Sci. USA 84, 493-497.
I 1 References
483
MURPHY J., TRENZ, P., GRZEMSKI, DE P. S. W., F. J. BRUIJN, J., SCHELL, (1993), The Rhizobium meliloti rhizopine mos locus is a mosaic structure facilitating its symbiotic regulation, J. Bacteriol. 175,5193-5204. NEUMANN, T., PIEPERSBERG, DISTLER,J. W., (1996), The decision phase model of the growth phase-dependent regulation of biosynthesis of streptomycin in Streptomyces griseus, Microbiology 142, 1953-1963. NEWTON,G. L., ARNOLD, PRICE, S., SHERK., M. s. RILL, c., DELCARDAYRE, B., AHARONOJ., R. WITZ, COHEN, DAVIES, FAHEY, C., Y., G., DAVIS, (1996), Distribution of thiols in microC. organisms: mycothiol is a major thiol in most actinomycetes, J. Bacteriol. 178, 1990-1995. NIERHAUS, H. (Ed.) (1993). The Translational K. Apparatus. New York: Plenum Press. NIMI, O., ITO, G., SUEDA,S., NOMI, R. (1971), Phosphorylation of streptomycin at C6-OH of streptidine moiety by an intracellular enzyme of Streptomyces griseus, Agr. Biol. Chem. 35, 848855. NOLLER, F. (1993), On the origin of the riboH. some: Coevolution of subdomains of tRNA and rRNA, in: The RNA World (GESTLAND, F., R. ATKINS, F., Eds.), pp. 137-184. New York: J. Cold Spring Harbor Laboratory Press. NUMATA, K., YAMAMOTO, H., HATORI, M., MIYAKI, KAWAGUCHI, (1986), Isolation T., H. of an aminoglycoside hypersensitive mutant and its application in screening, J. Antibiot. 39, 9941O. OO &HI, K. (1987), Metabolic initiation of differentiation and secondary metabolism by Streptomyces griseus: significance of the stringent response (ppGpp) and GTP content in relation to A-factor, J. Bacteriol. 169, 3608-3616. &HI, K. (1988), Nucleotide pools and stringent response in regulation of Streptomyces differentiation, in: Biology of Actinomycetes '88 (OKAMI, Y., BEPPU,T., OGAWARA, Eds.), pp. 330H., 337. Tokyo: Japan Scientific Societies Press. &HI, K. (1990), Streptomyces griseus, as an excellent object for studing microbial differentiation, Actinomycetol. 4, 23-30. S., ODAKURA, KASE,H., ITOH,S., SATHO, TAY., KASAWA, S., TAKAHASHI, SHIRAHATA, K., K., NAKAYAMA, (1984), Biosynthesis of astromiK. cin and related antibiotics. 11. Biosynthetic studies with blocked mutants of Micromonospora olivasteropora, J. Antibiot. 37, 1670-1680. OGURA,H., HASEGAWA, SUAMI, (1992), A., T. Carbohydrates - Synthetic Methods and Applications in Medicinal Chemistry. Tokyo: Kodansha. OHNUKI, IMANAKA, AIBAS. (1985a), SelfT., T., cloning in Streptomyces griseus of an str gene
cluster for streptomycin biosynthesis and streptomycin resistance, J. Bacteriol. 164, 85-94. OHNUKI, IMANAKA, AIBA, (1985b), IsolaT., T., S. tion of streptomycin non-producing mutants deficient in biosynthesis of the streptidine moiety or linkage between streptidine-6-phosphate and dihydrostreptose, Antimicrob. Agents Chemother. 27, 367-374. OHTA,T., HASEGAWA, (1993a), Analysis of the M. self-defense gene ( f m r 0 ) of a fortimicin A (astromicin) producer, Micromonospora olivasterospora: comparison with other aminoglycosideresistance-encoding genes, Gene 127, 63-69. OHTA,T., HASEGAWA, (1993b), Analysis of the M. nucleotide sequence of fmrT encoding the selfdefense gene of the istamycin producer, Streptomyces tenjimariensis ATCC 31602; comparison with the sequences of kamB of Streptomyces tenebrarius NCIB 11028 and kamC of Saccharopolyspora hirsuta CL102, J. Antibiot. 46, 511-517. OHTA, T., HASHIMOTO, E., HASEGAWA, M. (1992a), Characterization of sannamycin A-nonproducing mutants of Streptomyces sannanensis, J. Antibiot. 45, 289-291. M. OHTA, T., HASHIMOTO, E., HASEGAWA, (1992b), Cloning and analysis of a gene (smsl3) encoding sannamycin B-glycyltransferase from Streptomyces sannanensis and its distribution among actinomycetes, J. Antibiot. 45, 11671175. OHTA,T., DAIRI,T., HASHIMOTO, HASEGAE., WA,M. (1993a), Use of a heterologous gene for molecular breeding of actinomycetes producing structurally related antibiotics: self-defense genes of producers of fortimicin-A (astromicin)group antibiotics, Actinomycetol. 7, 145-155. OHTA, T., DAIRI,T., HASEGAWA, (1993b), M. Characterization of two different types of resistance genes among producers of fortimicin-group antibiotics, J. Gen. Microbiol. 139, 591-599. OKADA, YAMAMOTO, TSUTANO, NAKAH., K., S., MURA, s. (1988), A new group of antibiotics, hydroxamic acid antimycotic antibiotics. I. Precursor-initiated changes in productivity and biosynthesis of neoenactins NL, and NL,, J. Antibiot. 41,869-874. OKUDA, ITO,Y. (1982), Biosynthesis and mutaT., synthesis of aminoglycoside antibiotics, in: AmiH., noglycoside Antibiotics (UMEZAWA, HOOPER, I. R., Eds.), pp. 111-203. Berlin: SpringerVerlag. ONAKA, ANDO,N., NIHIRA, YAMADA, H., T., Y., S. BEPPU,T., HORINOUCHI, (1995), Cloning and characterization of the A-factor receptor gene from Streptomyces griseus, J. Bacteriol. 177, 6083-6092.
484
W. OTSUKA, MASKARETTI, A., HURLEY, H., PIEPERSBERG, (1993), Streptomycetes and coH., 0. L. rynebacteria, in: Biotechology, 2nd Edn., Vol. 1 FLOSS,H. G. (1980), Stereochemical aspects of (REHM, H.-J., REED, G., Eds.), pp. 433468. the biosynthesis of spectinomycin, J. Am. Chem. Weinheim: VCH. SOC. 102,6817-6820. O ~ E NS., L., LIU, X., FERGUSON, HUTCHIN- PIEPERSBERG, (1994), Pathway engineering in J., W. SON,C. R. (1995), Cloning and characterization secondary metabolite-producing actinomycetes, of the Streptomyces peuceticus dnrQS genes enCrit. Rev. Biotechnol. 14,251-285. coding a daunosamine biosynthesis enzyme and PIEPERSBERG, (1995), Streptomycin and related W. a glycosyltransferase involved in daunorubicin aminoglycosides, in: Biochemistry and Genetics biosynthesis, J. Bacteriol. 177, 6688-6692. of Antibiotic Biosynthesis (VINING, STUTL., OZAKI, MIZUSHIMA, NOMURA, (1969), M., S., M. TARD, Eds.), pp. 531-570. Boston, M A ButC., Identification and functional characterization of terworth-Heinemann. the protein controlled by the streptomycin-reH., PIEPERSBERG, GEYL,D., HUMMEL, BOCK, W., sistant locus in E. coli, Nature 222, 333-339. A. (1980), Physiology and biochemistry of bactePARRY, J. (1992), Investigation of the biosyntheR. rial ribosomal mutants, in: Genetics and Evolusis of aristeromycin, in: Secondary-Metabolite tion of RNA Polymerase, tRNA and Ribosomes R. Biosynthesis and Metabolism (PETROSKI, J., (OSAWA, OZEKI, UCHIDA, YURA, S., H., H., T, MCCORMICK, P., Eds.), pp. 89-104. New S. Eds.), pp. 359-377. Tokyo: University of Tokyo York: Plenum Press. PARRY,R. J., BORNEMANN, SUBRAMANIAN, Press. V., W., J., P., R. (1989), Biosynthesis of the nucleoside anti- PIEPERSBERG, DISTLER, HEINZEL, PEREZ-GONZALEZ, J.-A. (1988), Antibiotic resistbiotic aristeromycin, J. Am. Chem. SOC. 111, ance by modification: many resistance genes 5819-5824. could be derived from cellular control genes in PEARCE, J., RINEHART, L. (1981), BiosyntheC. K. actinomycetes - A hypothesis, Actinomycetol. 2, sis of aminocyclitol antibiotics, in: Antibiotics, 83-98. Vol. 4, Biosynthesis (CORCORAN, W., Ed.), J. pp. 74-100. Berlin: Springer-Verlag. PIEPERSBERG, HEINZEL, MANSOURI, W., P., K., MONINGHOFF, PISSOWOTZKI, (1991a), U., K. PENYIGE, VARGHA, G., ENSIGN, C., BARAA., J. Evolution of antibiotic resistance and producBAS,G. (1992), The possible role of ADP ribotion genes in streptomycetes, in: Genetics and sylation in physiological regulation of sporulaProduct Formation in Streptomyces (BAUMtion in Streptomyces griseus, Gene 115, 181-185. H., D., BERG, s., KRUGEL, NOVACK, Eds.), pp. PESCHKE, SCHMIDT, ZHANG, U., H., H.-Z., PIE161-170. New York Plenum Press. PERSBERG, (1999, Molecular characterizaW. tion of the lincomycin production gene cluster of PIEPERSBERG, STOCKMANN, MANSOURI, W., M., Streptomyces lincolnensis 78-11, Mol. Microbiol. K., DISTLER,J., GRABLEY,S., SICHEL, P., 16,1137-1156. BRAu, B. (1991b), German Patent Application PHILLIPS, WELLINGTON, M. H., REES,S. B., L., E. P4130 967 - HOE 91/F 300. JUN,L. S., KING,G. P. (1992), The distribution PIEPERSBERG, JIMENEZ, CUNDLIFFE, W., A., E., of DNA sequences hybridizing with antibiotic GRABLEY,S., BRAu, B., MARQUARDT, R. production and resistance gene probes within (1994), Glycosyltransferases from streptomytype strains and wild isolates of Streptomyces cetes as tools in biotransformations, BRIDGE species, J. Antibiot. 45, 1481-1491. A., Progress Report 1994 (CEC) (VASSAROTTI, PIENDL, BOCK,A., CUNDLIFFE, (1984), InW., E. Ed.), pp. 128-133. Brussels: Printeclair. volvement of 16s ribosomal RNA in resistance K., K., of the aminoglycoside-producers Streptomyces PISSOWOTZKI, MANSOURI, PIEPERSBERG, W. (1991), Genetics of streptomycin production tenjimariensis, Streptomyces tenebrarius, and Miin Streptomyces griseus. Molecular structure and cromonospora purpurea, Mol. Gen. Genet. 197, putative function of genes strELMB2N, Mol. 24-29. Gen. Genet. 231,113-123. PIEPERSBERG, (1985), Aminoglycoside AntiW. POHLMAN, F., GENETTI, D., WINANS, C. R. H. S. biotika: Wichtige Therapeutika und Objekte der (1994), Common ancestry between IncN conjuGrundlagenforschung, Forum Mikrobiol. 8/85, gal transfer genes and macromolecular export 153-161. systems of plant and animal pathogens, Mol. MiPIEPERSBERG, (1992), Metabolism and cell indiW. crobiol. 14,655-668. vidualization, in: Secondary Metabolites: Their R. Function and Evolution, Ciba Foundation Sym- PRATT,W. B., FEKETY, (1986), The AntimicroD., J., bial Drugs. New York: Oxford University posium 171 (CHADWICK, WHELAN, Eds.), pp. 294-299. Chichester: John Wiley & Sons. Press.
I1 References
PUROHIT,P., STERN,S. (1994), Interactions of small RNA with antibiotic and RNA ligands of the 30s subunit, Nature 370, 659-662. RAETZ,C. R. H. (1996), Bacterial lipopolysaccharides: a remarkable family of bioactive macroamphiphiles, in: Escherichia coli and Salmonella. F. Cellular and Molecular Biology (NEIDHARDT, C., Ed.), pp. 1035-1063. Washington DC: American Society for Microbiology. RAGAN, M., VINING, C. (1978), Intracellular C. L. cyclic adenosine 3 3-monophosphate levels and streptomycin production in cultures of Streptomyces griseus, Can. J. Microbiol. 24, 1012-1015. REEVES, (1993), Evolution of Salmonella 0 anP. tigen variation by interspecific gene transfer on a large scale, TIG 9, 17-22. RETZLAFF, DISTLER, (1995), The regulator of L., J. streptomycin gene expression, StrR, of Streptomyces griseus is a DNA binding activator protein with multiple recognition sites, Mol. Microbiol. 18, 151-162. RETZLAFF, MAYER, BEYER, AHLERT, L., G., S., J., VERSECK, DISTLER,J., PIEPERSBERG, S., W. (1993), Streptomycin production in streptomycetes: a progress report, in: Industrial Microorganisms: Basic and Applied Molecular Genetics (HEGEMAN, D., BALTZ,R. H., SKATRUD, G. P. L., Eds.), pp. 183-194. Washington: American Society for Microbiology. RINEHART, L., JR. (1980), Biosynthesis and muK. tasynthesis of aminocyclitol antibiotics, in: Aminocyclitol Antibiotics, ACS Symp. Ser. 125 (RINEHART, L., SUAMI, Eds.). WashingK. T., ton, D C American Chemical Society. RINEHART, L., JR., STROSHANE, M. (1976), K. R. Biosynthesis of aminocyclitol antibiotics, J. Antibiot. 29, 319-353. RINEHARD, L., POTGIETER, DELAWARE, K. M., D. L. (1981), Direct evidence from multiple 13C labeling and homonuclear decoupling for the labeling pattern by glucose of the m-aminobenzoyl (C7N) unit in pactamycin, J. Am. Chem. SOC. 103,2099-2101. RINEHART, L., JR., SNYDER, C., STALEY, K. W. A. L., LAU, R. C. M. (1992), Biosynthestic studies on antibiotics, in: Secondary-Metabolite Biosynthesis and Metabolism (PETROSKI, R. J., S. MCCORMICK, P., Eds.), pp. 41-60. New York: Plenum Press. SAITOH, TSUNAKAWA, TOMITA, MIAKI, K., M., K., H. T., KONISHI, KAWAGUCHI, (1988), BoM., holmycin, a new aminoglycoside antibiotic. I. Production, isolation and properties, J. Antibiot. 41,855-861. SAKUDA, ISOGAI, MATSUMOTO, SUZUKI, S., A., s., A. (1987), Search for microbial insect growth
485
regulators. 11. Allosamidin, a novel insect chitinase inhibitor, J. Antibiot. 40,296-300. SALAUZE, PEREZ-GONZALEZ,A., PIEPERSD., J. BERG, w., DAVIES, (1991), Characterisation J. of aminoglycoside acetyltransferase-encoding genes of neomycin-producing Micromonospora chalcea and Streptomycesfradiae, Gene 101,143148. SATOH,A., OGAWA, SATOMURA, (1976), H., Y. Regulation of N-acetylkanamycin amidohydrolase in the idiophase in kanamycin fermentation, Agric. Biol. Chem. 40,191. SCHLESSINGER, MEDOFF,G. (1975), StreptoD., mycin, dihydrostreptomycin, and the gentamiJ. cins, in: Antibiotics, Vol. 111 (CORCORAN, W., F. HAHN, E., Eds.), pp. 535-550. Berlin: Springer-Verlag. SCHROEDER, R., STREICHER,B., WANK, H. (1993), Splice-site selection and decoding: are they related? Science 260, 1443-1444. SHAPIRO, (1989), Nitrogen assimilation in actiS. nomycetes and the influence of nitrogen nutrition on actinomycete secondary metabolism, in: Regulation of Secondary Metabolism in ActinoS., mycetes (SHAPIRO, Ed.), pp. 135-211. Boca Raton, FL: CRC Press, Inc. SHAW, J., RATHER, N., HARE,R. S., MILLER, K. P. G. H. (1993), Molecular genetics of aminoglycoside resistance genes and familial relationships of the aminoglycoside-modifying enzymes, Microbiol. Rev. 57, 138-163. SHNAITMAN, A., KLENA, D. (1993), Genetics C. J. of lipopolysaccharide biosynthesis in enteric bacteria, Microbiol. Rev. 57, 655-682. SIPOS,L., SZABO,G. (1989), Myo-inositol-l-phosphate synthase in different Streptomyces griseus variants, FEMS Microbiol. Lett. 65, 339-364. SMAOUI, MALLIE,J.-P., SCHAEVERBEKE, H., M., ROBERT, SCHAEVERBEKE,(1993), GenA., J. tamicin administered during gestation alters glomerular basement membrane development, Antimicrob. Agents Chemother. 37, 1510-1517. STOCKMANN, PIEPERSBERG, (1992), Gene M., W. probes for the detection of 6-deoxyhexose metabolism in secondary metabolite-producing streptomycetes, FEMS Microbiol. Lett. 90, 185190. SWAN, K., KOHLHEPP, J., KOHNEN, W., S. S. P. GILBERT, N., BENNETT, M. (1991), LongD. W. term protection of polyaspartic acid in experimental gentamicin nephrotoxicity, Antimicrob. Agents Chemother. 35,2591-2595. F., s., SZESZAK, VITALIS, BBKEsI, I., SZABb, G. (1991), Presence of factor C in streptomycetes and other bacteria, in: Genetics and Product ForS., mation in Streptomyces (BAUMBERG, KRU-
486
Neotrehalosadiamine (BMY-28251), a new amiGEL, H., NOVACK, D., Eds.), pp. 11-18. New nosugar antibiotic, J. Antibiot. 39, 1001-1003. York: Plenum Press. H. S. J. W. UMEZAWA, (1974), Biochemical mechanisms of TAYLOR, S., BUECHLER, A., YONEMOTO, resistance to aminoglycoside antibiotics, Adv. (1990), CAMP-dependent protein kinase: FrameCarbohydr. Chem. Biochem. 30,183-225. work for a diverse family of regulatory enzymes, UMEZAWA, HOOPER, R. (Eds.) (1982), AmiH., I. Ann. Rev. Biochem. 59,971-1005. noglycoside Antibiotics. Berlin: Springer-Verlag. TERCERO, A., ESPINOSA, C., LACALLE, A., J. J. R. JIMBNEZ, (1996), The biosynthetic pathway UMEZAWA, KONDO, ITO, Y. (1986), AminoA. s., s., of the aminonucleoside antibiotic puromycin, as glycoside antibiotics, in: Biotechnology, 2nd deduced from the molecular analysis of the pur Edn., Vol. 4 (REHM,H.-J., REED,G., Eds.), pp. cluster of Streptomyces alboniger, J. Biol. Chem. 309-357. Weinheim: VCH. 271, 1579-1590. UMEZAWA, H., OKANISHI, M., UTAHARA, R., TESTA,B., KYBURZ, FUHRER, GIGER,R. E., W., MAEDA,K., KONDO,S. (1967a), Isolation and (1993), Perspectives in Medicinal Chemistry. structure of kanamycin inactivated by a cell-free Weinheim: VCH. system of kanamycin-resistant Escherichia coli, J. Antibiot. A20,136-141. THIEM, (Ed.) (1990), Carbohydrate Chemistry. J. Topics in Current Chemistry 154. Berlin: Sprin- UMEZAWA, OKANISHI, KONDO, HAMAH., M., S., ger-Verlag. NA, K., UTAHARA, MAEDA, R., K., MITSUHASHI, S. (1967b), Adenylylstreptomycin, a product THOMAS, J. (1990), Synthetis of anthracyclines G. of streptomycin inactivated by E. coli carrying related to daunomycin, in: Recent Progress in the R-factor, Science 157, 1559-1561. Chemical Synthesis of Antibiotics (LUKAS,G., OHNO,M., Eds.), pp. 467-496. Berlin: Springer- VAZQUEZ, (1979), Inhibitors of protein syntheD. Verlag. sis, in: Molecular Biology, Biochemistry, and A., Biophysics, Vol. 30 (KLEINZELLER, SPRINTHOMPSON, J., SETO,H. (1995), Bialaphos, in: C. GER, G. F., WITTMANN, G., Eds.). Berlin: H. Biochemistry and Genetics of Antibiotic BiosynSpringer-Verlag. C., thesis (VINING,L., STUTTARD, Eds.), pp. 197-222. Boston, MA: Butterworth-Heinemann. VILCHES, HERNANDEZ, MENDEZ, SALC., C., C., AS,J. A. (1992), Role of glycosylation in biosynTHORSON, S., LIU,H.-W. (1993a), CharacterizaJ. thesis of and resistance to oleandomycin in the tion of the first PMP-dependent iron-sulfur-conproducer organism, Streptomyces antibioticus, J. taining enzyme which is essential for the biosynBacteriol. 174, 161-165. thesis of 3,6-dideoxyhexoses,J. Am. Chem. SOC. 115,7539-7540. VINING, C., DOULL,J. L. (1988), Catabolite reL. pression of secondary metabolism in actinomyTHORSON, S., LIU, H.-W. (1993b), Coenzyme B6 J. cetes, in: Biology of Actinomycetes '88 (OKAMI, as a redox cofactor: a new role for an old coenY., BEPPU,T., OGAWARA, Eds.), pp. 406H., zyme? J. Am. Chem. SOC. 115, 12177-12178. 411. Tokyo: Japan Scientific Societies Press. THORSON, S., Lo, S. F., LIU, H.-W., HUTCHINJ. VINING, STUTTARD, (1995), Biochemistry L., C. SON, C. R. (1993), Biosynthesis of 3,6-dideoxyand Genetics of Antibiotic Biosynthesis. Boston, hexoses: New mechanistic reflections upon 2,6M A Butterworth-Heinemann. dideoxy, 4,6-dideoxy, and amino sugar construction, J. Am. Chem. SOC. 115,69936994. VOGTLI, HUTTER, (1987), Characterization M., R. of the hydroxystreptomycin phosphotransferase TOHMA, KONDO, H., YOKOTSUGA, IWAMOS., J., gene (sph) of Streptomyces glaucescens: nucleoTO, J., MATSUHASHI, ITO, T. (1989), AshiG., tide sequencing and promoter analysis, Mol. mycins A and B, new streptomycin analogues, J. Gen. Genet. 208, 195-203. Antibiot. 42, 1205-1212. U., H. TOHYAMA, OKAMI, UMEZAWA, (1987), VON AHSEN, NOLLER, F. (1993), FootprintH., Y., H. ing the sites of interaction of antibiotics with caNucleotide sequence of the streptomycin phostalytic group I intron RNA, Science 260, 1500photransferase and amidinotransferase of Strep1503. tomyces griseus, Nucleic Acids Res. 15, 18191834. VON AHSEN, U., DAVIES,J., SCHROEDER, R. (1991), Antibiotic inhibition of group I ribozyme TRUSCHEIT, FROMMER, JUNGE, MULE., W., B., function, Nature 353, 368-370. LER, L., SCHMIDT, D. D., WINGEDER, W. (1981), Chemistry and biochemistry of bacterial VUJAKLIJA, UEDA, K., HONG,S., BEPPU,T., D., alpha-glucosidase inhibitors, Angew. Chem. (Int. HORINOUCHI, (1991), Identification of an AS. Edn.) 20, 744-761. factor-dependent promotor in the streptomycin biosynthetic gene cluster of Streptomyces griseus, TSUNO, IKEDA,C., NUMATA, T., K.-I., TOMITA, Mol. Gen. Genet. 229, 119-128. K., KONISHI, KAWAGUCHI, (1986), 3,3'M., H.
I 1 References
487
VUJAKLIJA,D., HORINOUCHI, BEPPU, T. S., (1993), Detection of an A-factor-responsive protein that binds to the upstream activation sequence of strR, a regulatory gene for streptomycin biosynthesis in Streptomyces griseus, J. Bacteriol. 175,2652-2661. WALKER,J. B. (1975a), Pathways of the guanidinated inositol moieties of streptomycin and bluensomycin, Methods Enzymol. 43,429-470. WALKER, B. (1975b), ATP:streptomycin 6-phosJ. photransferase, Methods Enzymol. 43, 628-632. WALKER, B. (1980), Biosynthesis of aminoglycoJ. side antibiotics, Dev. Ind. Microbiol. 21, 105113. WALKER, B. (1990), Possible evolutionary relaJ. tionships between streptomycin and bluensomycin biosynthetic pathways: Detection of novel inositol kinase and 0-carbamoyltransferase activities, J. Bacteriol. 172,58445851. J. WALKER, B. (1995), Enzymatic synthesis of aminocyclitol moieties of aminoglycoside antibiotics from inositol by Streptomyces spp.: detection of glutamine-aminocyclitol aminotransferase and diaminocyclitol aminotransferase activities in a spectinomycin producer, J. Bacteriol. 172,58445851. J. M. WALKER, B., SKORVAGA, (1973), Phosphorylation of streptomycin and dihydrostreptomycin, J. Biol. Chem. 248, 2435-2440. J. M. WALKER, B., WALKER, S. (1975), ATP:streptomycin 3 " -phosphotransferase, Methods Enzymol. 43, 632-634. WALLACE, J., TAI, P.-C., DAVIS,B. D. (1979), B. Streptomycin and related antibiotics, in: Antibiotics, Vol. V-1, Mechanism of Action of AntiJ. bacterial and Antitumor Agents (CORCORAN, W., HAHN,F. E., Eds.), pp. 272-303. Berlin: Springer-Verlag. WATERS,B., VUJAKLIJA, GOLD, M. R., DAD., VIES,J. (1994), Protein tyrosine phosphorylation in streptomycetes, FEMS Microbiol. Lett. 120, 187-190. WELLINGTON, M. H., MARSH, P., TOTH, I., E. CRESSWELL, HUDDELSTON, SCHILHAN., L., BEL, M. B. (1993), The selective effects of antibiotics in soil, in: Trends in Microbial Ecology (GUERRERO,R.,PEDR6S-AL6, C., Eds.), pp. 331-336. Madrid: Spanish Society for Microbiology. WIDLANSKI, BENDER,S. L., KNOWLES,J. R. T., (1989), Dehydroquinate synthase: A sheep in wolf's clothing? J. Am. Chem. SOC. 111, 22992300. WILLIAMS,S. T., SHARPE,M. E., HOLT, J. G. (1989), Bergey ' Manual of Systematic Bacterios fogy, Vol. 4. Baltimore, ML: Williams and Wilkins.
WONG,Y.-H. H., SHERMAN, R. (1985), AnomW. eric and other substrate specificity studies with myo-inositol-1-P synthase, J. Biol. Chem. 260, 11083-1 1090. WRIGHT, L. C. (1983), The lincomycin4elesticeJ. tin-anthramycin group, in: Biochemistry and Genetic Regulation of Commercially Important AnL. tibiotics (VINING, C., Ed.), pp. 311-328. London: Addison-Wesley. N., K. YAMAUCHI, KAKINUMA, (1992a). Biochemical studies on 2-deoxy-scyllo-inosose an early intermediate in the biosynthesis of 2-deoxystreptamine. I. Chemical synthesis of 2-deoxy-scylloJ. inosose and [2,2-ZHz]-2-deoxy-scyllo-inosose, Antibiot. 45,756-766. N., K. YAMAUCHI, KAKINUMA, (1992b), Biochemical studies on 2-deoxy-scyllo-inosose an early intermediate in the biosynthesis of 2-deoxystreptamine. 11. Quantitative analysis of 2-deoxy-scyllo-inosose, J. Antibiot. 45, 767-773. YAMAUCHI, KAKINUMA, (1992c), ConfirmaN., K. tion of in vitro synthesis of 2-deoxy-scyllo-inosose, the earliest intermediate in the biosynthesis of 2-deoxystreptamine, using cell free preparations of Streptomyces fradiae, J. Antibiot. 45, 774780. N., K. YAMAUCHI, KAKINUMA, (1993), Biochemical studies on 2-deoxy-scyllo-inosose an early intermediate in the biosynthesis of 2-deoxystreptamine. VI. A clue to the similarity of 2-deoxyscyllo-inosose synthase to dehydroquinate synthase, J. Antibiot. 46, 1916-1918. N., K. YAMAUCHI, KAKINUMA, (1995), Enzymatic carbocycle formation in microbial secondary metabolism. The mechanism of the 2-deoxy-scylloinosose synthase reaction as a crucial step in the 2-deoxystreptamine biosynthesis in Streptomyces fradiae, J. Antibiot. 60, 5614-5619. YASUZAWA, YOSHIDA,M., ICHIMURA,M., T., SHIRAHATA, SANO,H. (1987), CV-1, a new K., antibiotic produced by a strain of Streptomyces sp. 11. Structure determination, J. Antibiot. 40, 727-731. YOKOSE,K., FURUMAI, SUHARA, PIRSON, T., Y., W. (1989), Trestatin: alpha-amylase inhibitor, in: Novel Microbial Products for Medicine and Agriculture (DEMAIN, A. L., SOMKUTI,G. A., J. HUNTER-CREVA, C., ROSSMOORE,H. W., Eds.), pp. 117-126. Amsterdam: Elsevier Science Publishers. ZHANG,H., SCHMIDT, H., PIEPERSBERG, W. (1992), Molecular cloning and characterization of two lincomycin-resistance genes, lmrA and ImrB, from Streptomyces lincolnensis 78-11, Mol. Microbiol. 6, 2147-2157.
488

ZHOU, Z.-Y., SAKUDA, S., KINOSHITA, M., YAMADA, Y. (1993), Biosynthetic studies of allos-
ZHOU, Z.-Y., SAKUDA,
S., YAMADA, Y . (1992), Biosynthetic studies on the chitinase inhibitor, allosamidin. Origin of the carbon and nitrogen atoms, J. Chem. SOC. Perkin Trans. I 1992,16491652.
amidin. 2. Isolation of didemethylallosamidin, and conversion experiments of 14C-labeled demethylallosamidin, didemethylallosamidin and their related compounds, J. Antibiot. 46, 15821588.
Biotechnology
11 Products from Basidiomycetes
GERHARD ERKEL TIMM ANKE

Kaiserslautern, Germany
1 Introduction 490 2 Cultivation of Basidiomycetes 490 3 Primary and Secondary Metabolites from Basidiomycetes - Biosyntheses and Possible Functions 491 4 Screening Methods Used for the Detection of Potentially Useful Metabolites 496 5 Bioactive Metabolites from Basidiomycetes 496 5.1 Pleuromutilin (Tiamulin) 496 5.2 The Strobilurins and Oudemansins 496 5.2.1 Mode of Action - Selective Toxicity 497 5.2.2 Structure-Activity Relationships - Development of Agricultural Fungicides 500 5.2.3 Biosynthesis 500 5.2.4 Possible Functions of Strobilurins and Oudemansins in the Producing Fungi 501 5.3 Other Antibacterial and Antifungal Metabolites 501 5.4 Cytotoxic and Antitumor Metabolites 503 5.4.1 Antitumor Polysaccharides 508 5.4.2 Immunosuppressive Metabolites 509 5.5 Antiviral Compounds and Inhibitors of Reverse Transcriptases 509 5.6 Inhibitors of Platelet Aggregation 512 5.7 Herbicidal Compounds 513 5.8 Insecticidal and Nematicidal Metabolites 515 5.9 Inhibitors of Cholesterol Biosynthesis 518 5.10 Inhibitors of Aminopeptidases 518 5.11 Inhibitors of Phospholipases C and A2 519 5.12 Inhibitors of (Na+-K+)-ATPases 520 5.13 Addendum 521 5.13.1 Inhibitors of Leukotriene Biosynthesis 522 5.13.2 Inducers of Differentiation of Promyelocytic Leukemia Cells and Inhibitors of Signal Transduction in Tumor Cells 524 6 Future Perspectives 525 7 References 526
490
1 Introduction
The basidiomycetes (mushrooms) constitute a large class of fungi and are estimated to consist of 30000 species (MOLLER and LOEFFLER, 1982) which is approximately one third of all fungi known. Since ancient times many of them were used as food (e.g., boletuses, chantarelles, Agaricus spp.) or for cultural purposes (hallucinogenic mushrooms). One of the first to describe pharmacological and toxic activities of fruiting bodies was (A.D. 23-79). Although PLINIUS SECUNDUS a direct use of the fruiting bodies was common practice all over the world, a detailed study of their contents and the metabolites produced by cultured mycelia started only in this century with FLEMINGS discovery of the imperfect fungus Penicillium notatum as the producer of penicillin, the first antibiotic metabolite for the treatment of bacterial infections in humans. This led to an intensive search for new antibiotics produced by other microorganisms, especially easily available soil-living forms. The pioneers in the search for antibiotics from basidiomycetes are ANCHEL,HERVEY, WILKINS and their coworkers who investigated extracts from fruiting bodies and mycelial cultures of approximately 2000 species (for a review, see FLOREY al., et 1949). Their outstanding work offered a first glance at the basidiomycete chemistry and succeeded in the isolation of pleuromutilin (KAVANAGH al., 1951), the first basidiomyet cete metabolite to serve later as a leading structure for the development of a commercial antibiotic. The work on antibiotic producing basidiomycetes was almost completely discontinued following WAKSMANS discovery of the streptomycetes as most promising antibiotic producers. These bacteria can be obtained from soil samples and grow easily in a variety of technical media. Up to now, the worldwide investigation of Streptomyces and related genera resulted in more than 6000 metabolites, many of them being used as antibiotics or for other pharmacological purposes. In spite of the incredible wealth of structures and activities which can to be found in streptomycetes and imperfect fungi it now has become increasingly difficult to find novel
chemical entities from these organisms. Because of that and of recent progress in fermentation technology, product recovery, and spectroscopy for structural analysis new investigations of other organisms seem to be attractive again. Among these are rare Actinomycetales, gliding bacteria, marine organisms, and some taxa of higher fungi including basidiomycetes. In the following chapter an overview of bioactive metabolites from mycelial cultures of basidiomycetes with special emphasis on antibiotics, cytotoxic, and antitumor compounds is given. In some cases secondary metabolites were obtained from fruiting bodies. Normally, basidiomycetes do not form fruiting bodies under laboratory conditions. Therefore, only the larger mushrooms can be used in chemical or biological investigations, and they have been studied intensively for the occurrence of toxins (reviewed by BRESINSKY and BESL, 1985), hallucinogens (reviewed by SCHULTESand HOFMANN, 1980), and pigments (reviewed by GILL, 1994; GILL and STEGLICH, 1987).
2 Cultivation of Basidiomycetes
The life cycle of a typical basidiomycete (e.g., Agaricus campestris, which does not form conidia and is devoid of a yeast phase) starts with haploid basidiospores germinating to form haploid mycelia, which - if compatible - fuse to give rise to dikaryotic mycelia from which fruiting bodies are derived. Karyogamy and meiosis take place in the basidia located in the hymenium (e.g., lamellae or pores) of the fruiting body. Usually, 4 haploid basidiospores are formed. Basidiomycetes are usually collected as fruiting bodies from their natural substrate: dead or living plants, soil or dung. Cultures can be derived either from spores (haploid or dikaryotic mycelia) or tissue plugs (dikaryotic mycelia) which can germinate and grow on
3 Primary and Secondary Metabolites of Basidiomycetes
491
complex media, typically containing yeast extract or peptone as a nitrogen source and glucose, maltose, or malt extract as a carbon source. A medium commonly used in the authors laboratory consists of 4 g glucose, 4 g yeast extract, 10 g malt extract, water to 1 L, pH adjusted to 5.5. The same media usually are suitable for submerged cultivation either in Erlenmeyer flasks or in fermenters. Metabolite diversity and production are mainly dependent on the biosynthetic capabilities of the strain and on the fermentation conditions. From the authors experience, these conditions can only be varied to a limited extent since many strains are highly sensitive to shear stress imposed by the impellers, and media have to be chosen so as to permit sufficiently fast growth. These problems were addressed by CHENINA al. (1993), GERMERet DONK et al. (1993), and BRAUER and KORN (1993). The modeling of a basidiomycete fermentation was achieved by HAS and MuNACK (1993). An example for a detailed description of a technical-scale process is the production of the antibiotic pleuromutilin (KNAUSEDER and BRANDL,1975; SCHNEIDER and MOSER,1987). Several peculiarities in the cultivation of basidiomycetes are commonly encountered: spores of many species, e.g., from the genera Znocybe and Russula, do not germinate, and no growth from tissue plugs can be observed. Many mycelial cultures grow very slowly on solid media or in submerged cultures. Fermentation times range from one to several weeks. Suitable methods for the preservation of cultures are keeping agar slants at 4C with periodical transfers (e.g., once a year) or storage in liquid nitrogen. Most cultures lose viability after freeze-drying.
3 Primary and Secondary Metabolites from Basidiomycetes Biosyntheses and Possible Functions
Secondary metabolites show diverse chemical structures that are often quite different from the primary metabolites (such as amino acids, acetyl coenzyme A, sugars, mevalonic acid, and intermediates from the shikimic acid pathway) from which they are synthesized. The starting point from primary metabolism is the basis of classification according to their biosynthetic precursors (TURNER, 1971; TURNER ALDRIDGE, and 1983). As shown in Fig. 1 the main branching points leading to secondary products are (HERBERT, 1989):
- acetyl coenzyme A, leading to polyketides,
polyins, terpenoids, steroids, or carotenoids; - shikimate, from which aromatic compounds can be derived; - amino acids, which serve as precursors for peptides and alkaloids; - glucose, for the biosynthesis of glycosides and aminoglycosides. Due to the widespread distribution of the biosynthetic pathways mentioned above, related secondary metabolites, like polyketides, steroids, and terpenoids have been isolated from bacteria, plants, fungi, and animals (BEALE, 1990 JANSEN and DE GROOT, 1991). The secondary metabolism of basidiomycetes is rich in terpenoids, especially sesquiterpenoids (AYERand BROWNE,1981) and polyacetylenes (JONESand THALLER, 1973). Many of these possess structures which up to now have only been detected in this class of fungi, whereas others closely resemble plant metabolites (FRAGA, 1990). Illudin M and illudin S (1 and 2, Fig. 2) were two of the first highly antimicrobial and cytotoxic metabolites of basidiomycetes detected by the screenings of ANCHELet al.
492
I1 Products from Basidiomycetes

Secondarv metabolism
Rm and inteamdim metabolism i m
.................................................................
Glycosides (e. g. Schizonellins) Kojic acid etc.
1Polysaccharides
1
Trio%-P
E l
+
Teeixe
,
Shikimate
I\
Pyruvate
1 1
Aromatic amino acids
\:
Aminoacids
1
,
_1
Fatty acids
Aromatic compounds(e. g. Strobilurins)
Alkaloids (e. g. Chalciporon)

Oligopeptides (e. g. Amanitins)
Acetyl-CoA
I /
Amino acids
wonylcoL
......................................
Fyesy-PP Geraayl-PP Steroids MewIonic acid
Depsipeptides(e. g. Beauvericin) Modified Amino acids (e. g. metabolites of Clitocybe acromelalga) Acetylenes (e. g. Siccayne) Polyketides (e. g. Merulinic acids)
h \? i
Citric acid cycle
- -
Sesquitemnoids (e. g. A h c o b ) Diterpenoids (e. g. Striatins) + Sesterterpenoids(e. g. Fasciculols)
F g 1 Interrelationship between primary and secondary metabolism. i. .
pR = H IlludinM(1) R = OH Illudin S (2)
F g 2 Biosynthesis o illudins. i. . f
493
(1950). The producing mushrooms, Clitocybe illudens and Lampteromyces japonicus, are highly toxic to humans. Illudin S is considered to be the toxic principle. The first information on stages in sesquiterpenoid biosynthesis were obtained by incorporation of radiolabeled mevalonic acid into illudins S and M from a head-to-tail condensation of isopentenyl pyrophosphate and dimethylallyl pyrophosphate by stationary cultures of Clitocybe illudens with a humulene-type precursor as the first cyclic intermediate (MCMORRIS and ANCHEL, 1965; HANSON and MARTEN, 1973). PRICE and HEINSTEIN(1978) confirmed this biosynthetic pathway (Fig. 2) by using a cell-free homogenate of Clitocybe illudens which incorporated labeled pyrophosphorylated isoprenyl alcohols into illudins. Another example of metabolites apparently occurring in basidiomycetes only are the diterpenoids - striatins A, B, C, D - and the corresponding striatals (Fig. 3) which are antibiotic and cytotoxic products of Cyathus striatus, C. poeppigii, C, limbatus, C. montagnei, and Gerronema fibula (ANKE et al., 1977a;HECHTet al., 1978). Striatins and striatals consist of a cyathan moiety with an attached pentose unit. In a resting cell system I4C- and I3C-labeled precursors were incorporated into striatins and striatals (RABE, 1989). Feeding experiments with l-I3C-glucose and 2-I3C-glucose and subsequent analysis by NMR spectroscopy revealed that the pentose unit was formed by decarboxylation of C-6 of glucose (70%) and to a smaller extent (30%) via the pentose phosphate cycle (Fig. 3). The labels observed in the cyathan skeleton were consistent with the formation of acetate via glycolysis and subsequent synthesis of mevalonate. Further cyclization of geranyl-geranyl pyrophosphate to the cyathan skeleton occurred according to the mechanism of the cyathine formation as described by AYERet al. (1979). Herical, an antibiotic metabolite of Hericium ramosum consists of a cyathan moiety and D-xylose attached to it by a glycosidic bond. When 14Cherical was prepared and fed to resting cells of C. striatus it was readily incorporated into the striatals A and B. Herical is thus considered as a direct precursor of striatals. The formation of the C-C bond between aglycone
and xylose involves a loss of protons, intramolecular cyclization, and acylation. Similar mechanisms may be assumed for the formation of the related metabolites dihydrostriatal C and hericin by H. ramosum. It has been proposed that molecules related to secondary metabolites played important roles in biochemical evolution as modulators or effectors, enhancing or controlling biological activities of primitive macromolecules. A number of antibiotics which inhibit translation, such as aminoglycosides, interact with ribosomes by directly binding to specific RNA conformations. It has been suggested that these secondary metabolites served as effectors of translation and other ribozyme-catalyzed reactions in early stages of evolution (DAVIS al., 1992). From another viewpoint et secondary metabolism might be an evolutionary playground from which new and useful biogenetic pathways could evolve (ZAHNER, 1982). The biological role of secondary metabolites is still a matter of debate. Secondary metabolites might be beneficial to producing organisms several ways: improving their ability to grow, reproduce, or disperse under appropriate conditions, or affording protection against competitors or predators. The majority of these compounds fit into these categories (VINING,1990). Among the growth-supporting substances are the sideramines (ferric ion-chelating compounds) and related metabolites which, in association with specific receptors, play an essential role in solubilization and uptake of iron. Fungal sideramines have been isolated from cultures of Penicillium sp., Neurospora crassa, some Fusarium strains, Ustilago sp., and the basidiomycetous yeast Rhodotorula pilimanae (WINKELMANN, 1986). Their biosynthesis is regulated by the concentration of soluble iron in the substrate. As has been shown for rhodotorulic acid synthetase (ANKE and DIEKMANN, 1972) and fusigen synthetase (ANKE al., 1973) these key et enzymes could not be detected as long as sufficient iron was present in the culture media. The characteristics of fungal sideramine biosynthesis is similar to the non-ribosomal biosynthesis of other peptide antibiotics (KLEINKAUF and VON DOHREN,1987). Recently, it has been shown that wood decaying basidio-
494
PP
Acctyl-CoA
4
Patose phosphotc Cycle
30 perccnt 70
Oxid. and Ueeorh. C-6
Cyathus striatus
Cyathie cyathus spp.
D-Xylose
Striatine A,
B, C, Cyathus spp.
OH
Herical, Hericium ramosum
8327-540,
H.ramoaum
Striatal D, G m n e m a fibula
8327-503, H. m o s u m
Fig. 3. Biosynthesis of striatals.
495
mycetes (brown and white rots) produce siderophores of the phenolate type (JELLISON et al., 1990). The enzymatic transformation of sesquiterpenes in various species of Russulaceae is an example for a proposed chemical defense system that preserves the fruiting bodies from attack by parasites and microorganisms (STERNER et al., 1985). The fruiting bodies of Laccarius vellereus contain large amounts of stearoylvelutinal (3, Fig. 4) which is transformed to the unsaturated dialdehyde isovelleral a few seconds after injury (4, Fig. 4). While stearoylvelutinal appears to have weak biological activity, isovelleral, like other unsaturated dialdehydes, has strong antimicrobial as et well as mutagenic properties (STERNER al.,
1987a). In addition, isovelIeral is a potent antifeedant for mammals that normally feed on mushrooms. Further reduction of an aldehyde group converts isovelleral to isovellerol (5, Fig. 4) the mutagenicity, pungency, and antimicrobial activities of which are diminished or lost upon reduction. It seems probable that injured fruiting bodies reduce the unsaturated dialdehydes in order to avoid prolonged contact with their own defense chemicals. Another example of a chemical defense mechanism in basidiomycetes are azepin derivatives occurring in fruiting bodies of Chalciporus piperatus. The main component chalciporon (6, Fig. 4) exhibits antibacterial and antifungal activity. Due to its strong pungency chalciporon is considered to be responsible
Stearoylvelutid(3)
Isovelleral(4)
Isovellerol(5)
Chalciporon (6)
Isochalciporon(7)
Laccarius vellerus andChalciporus piperatus.
Pleuromutiln (8)
Tiarnulin (9)
496
for the antifeedant acitivity. In solution chalciporon is converted to isochalciporon (7, Fig. 4) which still shows antibiotic activity but has lost pungency (STERNER al., 1987b). et
4 Screening Methods Used for the Detection of Potentially Useful Metabolites

Culture fluids or mycelial extracts of basidiomycetes are amenable to all screening methods applied to other microorganisms. Routine test systems comprise bacteria, filamentous fungi or yeasts, human and rodent cell lines, viruses, plants, and several enzyme assays.
5 Bioactive Metabolites from Basidiomycetes

5.1 Pleuromutilin (Tiamulin)
So far, the only commercial antibiotic produced by a basidiomycete is the diterpene pleuromutilin (8, Fig. 4). Pleuromutilin was first isolated from Pleurotus mutilus and Pleurotus passeckerianus in a screening for antibacterial compounds (KAVANAGH al., et 1951). The structural formula was elucidated by ARIGONI (1962) and BIRCHet al. (1963, 1966). In 1963, BRANDL al. (KNAUSEDER et and BRANDL,1976) isolated an antibiotic from Clitopilus passeckerianus which was identical with pleuromutilin. Pleuromutilin is active against gram-positive bacteria, but the most interesting biological activity is its high effectiveness against various forms of mycoplasms. The preparation of more than 66 derivatives of pleuromutilin by RIEDL (1976) and EGGERand REINSHAGEN (1976a, b) resulted
in the development of tiamulin (9, Fig. 4) which exceeds the activity of the parent compound against gram-positive bacteria and mycoplasms by a factor of 10-50. The minimal inhibitory concentrations (MIC) for different strains of Mycoplasma were in the range of 0.0039-6.25 p.g/mL-' (DREWSet al., 1975). Studies on the mode of action revealed that pleuromutilin and its derivatives act as inhibitors of prokaryotic protein synthesis by interfering with the activities of the 70 S ribosomal subunit. The ribosome-bound antibiotics lead to the formation of inactive initiation complexes which are unable to enter the peptide chain elongation cycle (HOGENAUER, 1979). In various bacteria drug resistance is developed stepwise. In some E. coli mutants the ribosome has lost its binding ability for tiamulin. Because of its outstanding properties tiamulin is currently used for the treatment of mycoplasma infections in animals. Pleuromutilin can be produced by fermentation in a medium composed of 50 g glucose, 50 g autolyzed brewer's yeast, 50 g KH2P04, 0.5 g MgS04x 7 H20, 0.5 g Ca(N03), 0.1 g NaCl, 0.5 g FeS04 x 7 H 2 0 , water to 1 L, pH 6.0. The yield after 6 d of growth in a 1000 L fermenter was reported to be 2.2gL-'. It could be demonstrated that during fermentation of pleuromutilin derivatives differing in the acetyl portions attached to the 14-OH group of mutilin were formed. The biosynthesis of these derivatives was strongly stimulated by addition of corn oil as a carbon source during fermentation (KNAUSEDER and BRANDL,1976). Pleuromutilin overproducers were obtained by conventional mutagenesis and selection programs as well as by protoplast fusion and genetic studies (STEWART, 1986).
5.2 The Strobilurins and Oudemansins

Initially, the strobilurins A (10, Fig. 5) and B (11, Fig. 5) were isolated from cultures of Strobilurus tenacellus, a small and very common edible mushroom growing on buried pine cones in early spring (ANKE et al., 1977b). Both compounds showed a remarka-
497
ble activity against a variety of filamentous fungi and yeasts but no antibacterial effects. The structure elucidation by W. STEGLICH'S group (SCHRAMM al., 1978; ANKEet al., et 1984) revealed that both compounds belonged to a new class of antifungal antibiotics. Close similarities between strobilurin A and mucidin, an antifungal antibiotic previously isolated from cultures of the wood-inhabiting basidiomycete Oudemansiella mucida, were recognized. Mucidin, however, had been described as a dextrarotatory crystalline compound (MUSILEK,1969). Fermentations of the Oudemansiella mucida strains used in the author's laboratory yielded in addition to strobilurin A a new antifungal antibiotic, oudemansin A (23, Fig. 5) (ANKEet al., 1979). Strobilurin A and mucidin were claimed to be identical by SEDMERA al. in 1982 and this et was finally proven by VON JAGOW et al. (1986) in a direct comparison of both compounds. In the meantime, numerous strobilurins and oudemansins were isolated from many genera of basidiomycetes (Fig. 5 ) , from tropical as well as from temperate regions; among them were many Mycena species (BAUERLE ANKE,1980; BAUERLE, and 1981) (Tab. 1). Surprisingly, several strobilurins were also isolated from an ascomycete, Bolinea lutea (FREDENHAGEN al., 1990a, b). Strobilurins et seem to be of worldwide occurrence. Most of their producers grow on wood or decaying plant material. Strobilurins and oudemansins inhibit fungal growth at very low concentrations (lo-'lo-") (ANKE et al., 1977b, 1979, 1983; BACKENS al., 1988 WEBERet al., 1990a, b; et et ANKEet al., 1990; ZAPF al., 1994) without any significant antibacterial activity. Weak phytotoxic activity of several strobilurin and oudemansin derivatives have been demonstrated (SAUTER al., 1995). Insecticidal acet tivity of strobilurin A against adults and larval stages of Epilachna varivestis (Mexican bean beetle), Aphis fabae (aphid), and Tetranychus urticae (mite) were also found at concentrations of 10-4-10-5 M (HOLST,University of GieSen, personal communication, 1978). Reversible cytostatic activity has been described for all strobilurins, with strobilurin E
(19, Fig. 5) being the most active. In HeLa S3 cells (human) these are accompanied by a 30% drop of the cellular ATP content and a change in morphology. The observed antiviral effects of strobilurin E (vesicular stomatitis virus in baby hamster kidney cells) are probably due to an inhibition of host cell growth (WEBERet al., 1990a).
5.2.1 Mode of Action - Selective Toxicity

Respiration in fungi and other eukaryotes is completely blocked by strobilurin A and oudemansin A. In Ehrlich ascitic carcinoma cells (ECA, mouse) syntheses of macromolecules (proteins, RNA, DNA) are inhibited due to a depletion of their ATP pool caused by the inhibition of oxidative phosphorylation. Upon addition of glucose this effect is completely reversed since ATP supply by glycolysis seems to be sufficient in these cells. In rat liver mitochondrial oxygen uptake and ATP synthesis were blocked by both a-ketoglutaric acid and succinate as substrates which gave the first evidence of a target within the respiratory chain (ANKEet al., 1979). This molecular target was precisely identified by VON JAGOW and coworkers (BRANDT al., et 1988, 1993). Strobilurins and oudemansins specifically inhibit the ubiquinol oxidation (Qp center) of the mitochondrial bcl complex (Fig. 6). Like other specific inhibitors strobilurins and oudemansins have become valuable tools for the development of a more detailed model of the structure and function of their target. In heterocysts of cyanobacteria the cytochrome b/f complex, which utilizes light energy to generate the proton gradient used for ATP synthesis and transfers electrons to nitrogenase, the key enzyme of nitrogen fixation, plays a central role. Using heterocysts of and an Anabaena sp. HOUCHINS HIND(1983) found that strobilurin A inhibited the electron flow from reduced plastoquinone to the cytochrome b/f complex in a similar way as in the mitochondrial bcl complex. The similarity of the b/f complexes in cyanobacteria and in higher plants, where the b/f complex plays a
498
C H = ; R2 3 "
H3CW
' " p
n H3COOC ; C
StrobilurinC (15)
Rl R2
Me0 Cl, Strobilurin B (1 1)

HO H
H, StrobilurinA(10)
L o n ; H3COOC C H 3
Me0 H, StrobilurinH(13)
H, StrobilurinFl (12)
Me0, Strobilurin X (14)
StrobilurinF2 (1 6)
StrobilurinD* (17)
A
OCH3
HydroxystrobilurinD* (18)
StrobilurinE (19)
StrobilurinG (20)
0ch3
A 9-MdhO~StrObild (2 1)
9-MethoxystrobilurinK* (22)
Oudemansin B (24)
QCH3
H3C0 m
;H3COOC C
Oudemansin X (25)
Fig. 5. The strobilurins and oudemansins. * The structures of the side chains are currently under investigation.
499
Fig. 5
BAS 490 P ( . ) 26
ICW504(27)
Tab. 1 Fungi Producing Strobilurins and Oudernansins .

~~
Producer Basidiomycetes
Compound
References
Agaricus sp. 89139 Crepidotus fulvomentosus Cyphellopsis anomala Favolaschia sp. 87129 Filoboletus sp. 9054 Hydropus scabripes Mycena aetites M. alkalina M. atromarginata M. avenacea M. cf. capillaripes M. crocata M. fagetorurn M. galapoda M. galopoda var. alba M. oregonensis M. polygramma M. purpureofima M. rosella M. sanguinolenta M. vitilis M. zephirus Oudemansiella mucida 0. radicata Strobilurus conigenoides S. esculentus S. tephanocystis S tentacellus . Xerula longipes X. melanotricha Ascomycete Bolina lutea
10, 12, 17 19 10, 12, 17 10, 12, 17, 19, 21, 22, 23 19 10 10 11 10 11 10 11 10 10 10 10 23 10 10 18 11 10 10,23 10,25 10 10 10 10,ll 11, 15 10, 11, 24 10, 11, 13, 16, 20
ZAPF(1994) WEBER al. (1990a) et WEBER al. (1990b) et ZAPFet al. (1995) SIMON (1994) BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) SCHRAMM al. (1978) et BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) BAUERLE (1981) BACKENS al. (1988) et BAUERLE (1981) et SCHRAMM al. (1978) ANKE al. (1979) et ANKE al. (1990) et unpublished data ANKE, ANKE and STEGLICH (1981) unpublished data ANKE, ANKE al. (1977b) et ANKE al. (1983) et et ANKE al. (1983)
FREDENHAGEN (1990a, b) et al.
500
Fig. 6 The Q-cycle mechanism of the bcl. complex and the mode of action of strobilet urins and oudemansis (BRANDT al., 1993). Q-Pool: ubichinone pool; Q, Q : QH,: , ubiquinone, ubisemiquinone, ubihydroquinone; c, cl: cytochrome cl; b566: low potential heme b; b562:high potential heme b FeS: iron-sulfur protein.
central role in cyclic photophosphorylation and in coupling photosystems 1 and 2, might contribute, together with an inhibition of mitochondrial respiration, to the phytotoxic activity observed for some strobilurin derivatives (SAUTER al., 1995). et The mitochondria1 bcl complex, the target of strobilurins and oudemansins, is common to many eukaryotic taxa. Mitochondria1 preparations of rat liver, beef heart, house fly, and corn are all sensitive to strobilurins (ANKE et al., 1979; BRANDTet al., 1993; SAUTER al., 1995). Surprisingly, strobilurin et A and other synthetic mimics exhibited no NCI, USA, toxicity to rodents (J. DOUROS, unpublished SAUTERet al., 1995). In fact, mucidermin, a preparation which apparently contains strobilurin A (mucidin) was marketed by Spofa, CSFR, for the treatment of dermatomycoses in humans. This lack of toxicity is probably due to enzymic degradation of the shobilurins by, e.g., mammalian esterases before reaching their target.
5.2.2 Structure-Activity Relationships - Development of Agricultural Fungicides

Extensive synthetic efforts lead to simple mimics and revealed that the E-p-methoxyacrylate unit is a prerequisite for the antifungal and respiration inhibiting properties of strobilurins and oudemansins (SCHRAMM, 1980; SCHRAMM al., 1982; T. ANKEet al., et 1988). Continuous efforts by STEGLICH and coworkers (ANKEand STEGLICH, 1989) and by BASF and ICI resulted in compounds with improved activity and light stability and lead to the development of BAS 490 F (26, Fig. 5) (SAUTER et al., 1995) and ICIA5504 (27, Fig.5) (CLOUGH,1993), which will be commercialized in the near future by ICI and BASF.
5.2.3 Biosynthesis
The biosynthesis of mucidin (28, Fig. 7) (the E, E, E geometry was revised to E, 2, E of strobulin A by VON JAGOWet al., 1986) was investigated by NERUDet al. (1982) by feeding isotopically labeled phenylalanine,
501
Mucidin (28)
V
C%C02Na
7
CH.,C02Na
Fig. 7. Biosynthesis of mucidin (NERUDet al., 1982). E, E, E, geometry has been revised to E, Z,
E, of strobilurin A; isotopically labeled carbon atoms in the precursors acetate, benzoic acid, and methionine are marked V,V,0, and
*.
benzoic acid, acetate, and methionine. The aromatic part of the molecule and the benzylic carbon atom are derived from the shikimate pathway. The side chain consists of acetate units, and all three methyl groups are derived from methionine (Fig. 7).
5.2.4 Possible Functions of Strobilurins and Oudemansins in the Producing Fungi

The biological activities in the producers of strobilurins and oudemansins suggest a possible role in the defence of habitats and substrates against competing fungi or predatory insects. The use of antifungal antibiotics against competing fungi would require the producing fungus to be resistant to its own product. This was clearly demonstrated for Strobilurus tenacellus and Mycena galapoda by VON JAGOW and coworkers (BRANDTet al., 1993). In the case of S. tenacellus, binding of strobilurin A and oudemansin A to the bcl complex was reduced by several orders of magnitude, whereas in the case of M.galopo-
da a markedly increased rate of respiration is likely to confer a greater resistance. Genetic characterization of the exon-intron organization, the deduced amino acid sequence of the cytochromes b from S. tenacellus, M . galopoda, and M. viriginata (which does not produce strobilurin A), and a comparative sequence analysis of two regions of cytochrome b contributing to the formation of the Qp center as demonstrated by VON JACOW and coworkers (KRAICZY al., 1996) et revealed, that the generally lower sensitivity of all three basidiomycetes was due to the replacement of a small amino acid residue in position 127 by isoleucine. For M. galopoda replacement of glycine-143 by alanine and glycine-153 by serine, and for S. tenacellus replacement of a small residue in position 254 by glutamine and asparagine-261 by aspartate were assumed to cause resistance to E-p-methoxyacrylates. The latter exchange is also found in Schizosaccharomyces pombe which shows a natural resistance to E-p-methoxyacrylates. On the other hand, it was demonstrated that Oudemansiella mucida produces oudemansin A together with strobilurin A on sterilized beech wood which is its natural sub1992). These findings strate (SCHWITZGEBEL, and the observation that many basidiomycetes belonging to different taxa (Tab. 1) use the same and obviously quite effective principle to secure their habitat in very different climates and locations suggest that strobilurins and oudemansis play an important role in the producing fungi.
5.3 Other Antibacterial and Antifungal Metabolites

Lentinellic acid (29, Fig. 8) from Lentinellus omphalodes and Lentinellus ursinus is a new protoilludine derivative. Interestingly, strains from Europe, USA, and Canada all produce the same antibiotic. Lentinellic acid shows strong antibacterial activity with minimal inhibitory concentrations of 1-5 pg mL-' for Bacillus brevis, Aerobacter aerogenes, and Corynebacterium insidiosum. Compared to lentinellic acid its methyl ester exhibits much higher antifungal activity. In
502

CHO
Lentidlic acid (29)
Aleurodiscal(30)
Scorodonin (31)
l-Hydroxy-2-mnyn-3-one (32)
3.4, I3-Trihydroxy-tetradeca-5,7,9,ll-tetra~c acid-Y-lactone (33)
Ficolon(34) R = OH Hydroxyfimicolon (35)

R =H
Hemimycin (36)
Fig. 8. Other antibacterial and antifungal metabolites.
503
Illudins isolated from Clitocybe illudens and Lampteromyces japonicus were two of the first known antitumor metabolites. The lifetime of Ehrlich ascites tumor mice was prolonged by illudin S (2, Fig. 2) when given at a dose of 166 pg kg-' i.p. Enclosure of illudin S into liposomes markedly enhanced this effect, apparently by decreasing the side effects observed under standard experimental conditions (SHINOZAWA al., 1979). 6et Deoxyilludin M (37, Fig. 9) was isolated from cultures of Pleurotus japonicus. This compound is closely related to illudin M and was effective against murine leukemia P388, showing a 24 % increase of life span at a daily dose of 5 mg kg-' i.p. (HARAet al., 1987). Several metabolites of basidiomycetes with strong cytotoxic and antifungal activitiy belong to the sesquiterpenoids with a marasmane or isolactarane skeleton. Marasmic acid (38, Fig. 9) was isolated from cultures of Marasmius conigenus in the course of the first extensive screenings for antibacterial comet 3,4,13,-trihydroxy-tetradeca-5,7,9,1l,-tetraynic pounds conducted by KAVANAGH al. acid-y-lactone (33, Fig. 8) from cultures of (1949). The sesquiterpenoid structure of maMycena viridimarginata (BAUERLE et al., rasmic acid was elucidated by DUGAN al. in et 1966. GREENLEE and WOODWARD achieved 1982). Fimicolon (34, Fig. 8) and hydroxyfimico- the first total synthesis in 1976 and several lon (35, Fig. 8) of Panaeolus fimicola and Psa- new synthetic approaches have been pubthyrella orbitarium are antibiotic and cyto- lished since then (MORISAKI al., 1980). Maet toxic guaianes (ANKE al., 1985a). Guaianes rasmic acid was also isolated from cultures of et are typical metabolites of higher plants. The Lachnella sp. and Peniophora laeta and exhistructures responsible for the biological activ- bits pronounced inhibitory action on nucleic ity of the Pleurotellus metabolites and fimico- acid syntheses in whole mammalian cells and lons are similar and consist of a five-mem- on some enzymes of nucleoside metabolism. bered ring with an exomethylene group adja- In isolated rat liver nuclei the guanylation of mRNA was strongly inhibited by 10 pg mL -' cent to an oxirane ring. et The antibiotic and cytotoxic compound marasmic acid (KUPKA al., 1983). The life hemimycin (36, Fig. 8) obtained from Hemi- span of P388 lymphocytic leukemia mice was mycena cucullata and H. candida is another prolonged by marasmic acid when given at a example for the occurrence of the same car- total dose of 3.5 mg kg-' i.p. The LDS0 for bon skeleton in basidiomycetes and higher tumor bearing mice was determined to be National Cancer plants (e.g., Acorus calamus). Hemimycin is 28 mg kg-' i.p. (J. DOUROS, highly oxygenated and contains a double Institute, USA, personal communication). It bond which easily reacts with nucleophiles was proposed that due to the reactive a$-un-
ECA cells DNA, RNA, and protein syntheses are inhibited by 50% using 20 pg mL-' lentinellic acid (STARKet al., 1988). Several related protoilludane orsellinate esters were isolated from cultures of Armillaria mellea. These compounds exhibit weak antibacterial and antifungal activity (OBOUCHI al., 1990; et YANGet al., 1991). Screening for antifungal compounds resulted in the isolation of aleurodiscal (30, Fig. 8) from mycelial cultures of Aleurodiscus mirabilis (LAUERet al., 1989). Aleurodiscal, a hydroxysesterterpene aldehyde P-D-xyloside with a novel carbon skeleton, is related to retigeranic acid A which was isolated from lichens (KANEDA al., 1972). Aleurodiscal poset sesses weak antibacterial activity and strongly inhibits the growth of several fungi in the agar diffusion assay at concentrations of 2-10 pg per disc. In addition, it causes abnormal branching of apical hyphae of Mucor miehei at a concentration of 1 pg mL-'. Acetylenes are strong antifungal metabolites commonly found in basidiomycetes. They also exhibit antibacterial and cytotoxic activity (TURNER, 1983). Examples are scorodonin (31, Fig. 8) from cultures of Marasmius scorodonius (ANKE al., 1980), l-hydroxy-2-nonyn-4-one et (32, Fig. 8) from fermentations of Zschnoderma benzoinum (ANKE et al., 1982), and
(BAUERLEet al., 1986). Acoranes have not yet been reported from microbial sources.
5.4 Cytotoxic and Antitumor Metabolites
504
I 1 Products from Basidiomycetes
4YH
% ,
CHO
Marasmic acid (38)
CHO
Pilatin (39)
Merulidial(40)
R1
R2 COCH3 COCH3
striatals
H
OH OH
HO
nc15H31c00
Schizonellin A ( 4 4)
Schizonellin B (45)
Fig. 9. Cytotoxic and antitumor metabolites.
505
Pleurotellol(47)
Pleurotellic acid (48)
Phellodonic acid (49)
Alliicol A (50)
Alliacol B (51)
Fig. 9
Fulvofemginin (52)
saturated aldehyde function marasmic acid covalently binds to nucleophilic (e.g., amino) groups of enzymes or to nucleic acids. The hydroxylated derivative of marasmic acid, 9Phydroxymarasmic acid, was isolated by H. ANKEet al. (1988). Introduction of a hydroxyl function reduces the biological activity of marasmic acid, but increases mutagenic activity in the Ames test. Pilatin (39, Fig. 9), a new marasmane derivative, was isolated from fermentations of Flagelloscypha pilatii, a cyphelloid fungus
(HEIM al., 1988). Pilatin inhibits the growth et of bacteria and fungi at concentrations of 550 pg mL-'. It strongly interferes with the DNA and RNA syntheses of ECA cells and both normal and Rous Sarcoma Virus (RSV)transformed chicken embryo fibroblasts (CEF). In addition, pilatin causes frameshift mutations in Salmonella typhimurium TA 98. In vivo no significant antitumor activity on P388 lymphocytic leukemia mice was observed for pilatin. The LDSofor tumor bearing mice was determined to be 125 mg kg-'
506
@+ Ho -

( ) I l11111< l
ok
Crinipellin B (54)
CrinipellinA (53)
OWwn3
11111<
IIII,
oii
0-aoetylcrinipellinA (55)
DhydrocrinipellinB (56)
11111<
Tetrahydrwrinipeh A (57)
11111
GWH3 qgH3
Ii
0
11111
Ii
Nematolin (58)
Nematolon (59)
Leianafulvene(60)
Fig. 9
507
i.p. (J. DOUROS,National Cancer Institute, USA, personal communication). The isolactarane merulidial (40, Fig. 9) was isolated from submerged cultures of Merulius et tremellosus (QUACK al., 1978). The crystalline sesquiterpene dialdehyde very strongly inhibits DNA synthesis in Ehrlich ascitic carcinoma (ECA) cells at 1 pg ml-'. In the assay of AMESet al. (1975) merulidial exhibits mutagenic activity. Comparative studies of merulidial and several hydroxylated and acetylated derivatives revealed that the molecular mechanism responsible for the mutagenicity of merulidial is different from the mechanism resulting in antimicrobial and cytotoxic activity. Acetylation of merulidial to S-acetylmerulidial, e.g., increases antifungal activity but diminishes mutagenic activity. Hydroxylation of merulidial to 9-a-hydroxymerulidia1 and 9-phydroxymerulidial as well as of acetylmerulidial to 9-a-hydroxyacetylmerulidial does not strongly affect mutagenic activity but dramatically reduces antimicrobial, cytotoxic, and phytotoxic activity (ANKEet al., 1989). The striatins A, B, and C (4143, Fig. 9) and the corresponding striatals were isolated from submerged cultures of Cyathus striatus, C. poeppigii, C. limbatus, and C. montagnei. They were also detected in the fruiting bodies (ANKE et al., 1977a). In ECA cells DNA, RNA, and protein syntheses are completely inhibited by 2 pg mL-' striatins. RSV-transformed CEF were found to be inhibited at lower concentrations as compared to their normal counterparts. Studies on the mode of action revealed that interference with the transport of essential precursors was mainly responsible for their cytotoxic activity (LEE and ANKE,1979). Striatins and striatals were found to prolong the life span of P388 lymphocytic leukemia mice and to be inhibitory in the system colon xenograft-athymic mouse. The LDso for tumor bearing mice was determined to be 150 mg kg-' i. p. In greenhouse experiments striatins exhibited good fungicidal activity against Plasmopara viticola on grape vine, Phytophtora infestans on potatoes, Botrytis cinerea on green pepper, and Septoria nodorum on wheat (ANKE et al., 1986). The schizonellins A and B (44 and 45, Fig. 9) are glycolipids produced by sub-
merged cultures of the smut fungus Schizonella melanogramma (DEMLet al., 1980). Like ustilagic acids which are glycolipids of a different type obtained from Ustilago maidis (LEMIEUX al., 1951), schizonellins exhibit et weak antibacterial and antifungal activity. In ECA cells the incorporation of leucine, uridine, and thymidine into protein, RNA, and DNA is completely inhibited by 25 pg mL-' of schizonellin A or B. Concomitant lysis of the cells suggests a detergent-like mode of action. Hypnophilin, pleurotellol, and pleurotellic acid (46-48, Fig. 9) were isolated from fermentations of Pleurotellus hypnophilus (KUPKA et a]., 1981a). While pleurotellol and pleurotellic acid belong to a new group of sesquiterpenoids, hypnophilin is a new member of the hirsutane family to which a number of typical basidiomycete metabolites belong. All three antibiotics exhibit antimicrobial and very high cytotoxic activity. However, in comparison to normal cells no selective toxicity for RSV-transformed chicken embryo fibroblasts (CEF) could be detected. Hypnophilin and pleurotellol also act as plant growth inhibitors. In the Avena coleoptile bioassay they strongly inhibit indole-3-acetic acid-induced growth of coleoptile sections. Like other exomethylene ketones and lactones Pleurotellus antibiotics very readily form adducts with cysteine or other thiols and they are mutagenic. Phellodonic acid (49, Fig. 9), a new hirsutane derivative closely related to hypnophilin, has recently been isolated from cultures of Phellodon melaleucus (STADLER et al., 1993b). Like hypnophilin, phellodonic acid exhibits antimicrobial and strong cytotoxic activity. The incorporation of radiolabeled precursors into DNA, RNA, and protein of L 1210 cells is almost completely inhibited at a concentration of 5 pg mL-'. The alliacols A and B (50 and 51, Fig. 9) from Marasmius alliaceus are a,punsaturated sesquiterpene lactones which exhibit rather low antimicrobial but highly cytotoxic properties (ANKE al., 1981). In ECA cells nucleic et acid biosyntheses are almost completely inhibited at concentrations of 2-10 pg mL-'. Like other a,punsaturated lactones, alliacols readily form adducts with nucleophilic thiols. It is assumed that a rapid reaction with SH
508
groups in enzymes or other proteins is responsible for most of the biological activity of these compounds. Deduced alliacolide shows no antibiotic or cytotoxic properties. Fulvoferruginin (52, Fig. 9), a sesquiterpenoid carotane derivative has been isolated from Marasmius fulvoferrugineus (KLEINet al., 1990). It is closely related to hercynolactone which was isolated from liverworths (HUNECK al., 1982). Several carotane seset quiterpenes were isolated from Ferula species. Fulvoferruginin exhibits modest antibacterial activity and inhibits the growth of several fungi at 5-50 pg mL-'. In ECA cells the incorporation of leucine, uridine, and thymidine into protein, RNA, and DNA was inhibited by 50% at a concentration of 10-50 pg mL-'. Crinipellins obtained from fermentations of Crinipellis stipitaria are the first known natural tetraquinanes (KUPKA al., 1979; ANKE et et al., 1985b). The crinipellins A and B and 0-acetylcrinipellin A (53-55, Fig. 9) containing an exomethylene ketone moiety are strong antibacterials and highly cytotoxic metabolites. The reduced compounds dihydrocrinipellin B and tetrahydrocrinipellin A (56 and 57, Fig. 9) are inactive. Like striatins and striatals crinipellins exert their cytotoxic activity mainly by interfering with the transport of essential nutrients and precursors. The cytotoxic activity on RSV-transformed CEF seems to be higher than on normal CEF (KUPKA al., 1980). et The caryophyllanes nematolin and nematolon (58 and 59, Fig. 9) were isolated from cultures of Naematoloma capnoides, N. sublateritium, N. fasciculare, and N. elongatipes (BACKENS al., 1984). Comparison to the et caryophyllanes of higher plants these basidiomycete metabolites contain more oxygen functions and one or two a,punsaturated carbony1 groups. Nematolon and nematolin are weakly antimicrobial, the cytotoxic activity of nematolon is 5-fold higher than that of nematolin. In ECA cells the incorporation of thymidine into DNA is inhibited by 50% at a nematolon concentration of 2 pg mL - I . In vivo no significant antitumor activity (B-16 melanocarcinoma, Lewis lung carcinoma, P-388 lymphocytic leukemia) was found for nematolon. The LD5,, for tumor bearing mice was de-
termined to be >225 mg kg-' (J. DOUROS, National Cancer Institute, USA, personal communication). Leaianafulvene (60, Fig. 9), an orange-yellow pigment, was isolated from mycelial culet tures of Mycena leaiana ( H A R ~ I G al., 1990). The compound is closely related to the illudins and represents the first example of a natural "isoilludane" derivative which may be formed from an illidane precursor by 1,2-migration of a methyl group. Leaianafulvene exhibits weak antibacterial activity, whereas its cytotoxic activity is quite pronounced. A 50 % lysis of ECA cells is observed at 2.5 pg mL-'. DNA and RNA syntheses are inhibited by 50% at a leaianafulvene concentration of 10 pg ml-'. In addition, mutagenic acitvity was also observed.
5.4.1 Antitumor Polysaccharides

Antitumor polysaccharides have been obtained from various kinds of preparations. They include lentinan (CHIHARA al., 1970), et a high-molecular weight p-1,3 glucan isolated from fruiting bodies of Lentinus edodes (SASAKI and TAKASUKA, 1976), and schizophylIan (KOMATSU al., 1969), a high-molecular et weight p 1 ,3 1,6 glucan obtained from cultured mycelia of Schizophyllum commune. These compounds inhibit the growth of various transplantable tumors in experimental animals, they increase the survival rate and are considered to exert their antitumor activity by the potentiation of the host animals' defense mechanisms rather than by direct inhibition of tumor cell growth (SUGA et al., 1984). Lentinan in its sulfated form was also used in conjunction with AZT to suppress HIV (DE CLERCQ,1990). Like several other sulfated polysaccharides, lentinan interferes with syncytium formation resulting from fusion of HIV-infected and uninfected cells. KS-2, a peptide containing a-linked mannose was extracted from the mycelia of Lentinus edodes (FUJI et al., 1978). KS-2 suppressed the growth of both Ehrlich tumors and Sarcoma 180 tumors at dose levels of 1 mg kg-' and 100 mg kg-' when administered intraperitoneally or orally. It was also capable of inducing interferone in mice.
J Bioactive Metabolites from Basidiomycetes 7
509
PSK (krestin), a polysaccharide preparation isolated from Coriolus versicolor predominantly consists of glucan and of ca. 25 % tightly bound protein (TSUGAGOSHI al., et 1984). Oral administration of PSK increased the survival rate in several animal cancer models, and PSK is now clinically used in Japan for the treatment of postoperative cancer patients. PSK was also reported to exhibit immunomodulating acitvity by regulating cytokine production and effector cell functions (reviewed by KOBAYASHI, 1993).
5.4.2 Immunosuppressive Metabolites

Only few immunosuppressive compounds from basidiomycetes have been reported. In the course of a screening for metabolites suppressing the proliferation of mouse lymphocytes stimulated with mitogens, three geranylphenols, flavidulol A, B, and C (61-63, Fig. 10) have been isolated from fruiting bodet ies of Lucturius fluvidulus (FUJIMOTO al., 1993). The ICso values for flavidulols A, B, and C were found to be 8.9 pg mL-', 4.9 pg mL-', and 36.3 pg mL-', respectively, in an assay measuring concanavalin A-induced proliferation of mouse lymphocytes, and 6.7 p,g mL-', 3.9 pg mL-', and 28.3 pg mL-', respectively, when the cells were stimulated with lipopolysaccharide.
5.5 Antiviral Compounds and

Inhibitors of Reverse Transcriptases
The nucleosides 6-methylpurine, 6-methyl9-/3-~-ribofuranosylpurine, and 6-hydroxy-
methyl-9-/3-~-ribofuranosylpurine (64-66, Fig. 11) were isolated from mycelial cultures of Collybiu maculutu in a screening for inhibitors of vesicular stomatitis virus (VSV) multiplication in baby hamster kidney (BHK) cells (LEONHARDT al., 1987). 6-methylpurine et had and 6-methyl-9-/3-~-ribofuranosylpurine been obtained before by chemical synthesis. All three nucleosides exhibit modest antifungal and cytotoxic activity; the effect on VSV multiplication in BHK cells is high and compares very favorably with that of aruA. Besides their antiviral activity 6-methyl-9-&~-ribofuranosylpurine and 6-hydroxymethyl-9-PD-ribofuranosylpurhe are inhibitors of adenosine desaminase and, therefore, interfere with the nucleoside metabolism. Other nucleosides that have been reported as secondary metabolites from basidiomycetes are nebularine, described as an antibacterial antibiotic from Clitocybe nebuluris (LOFGREN,1954), lentinacin, a hypercholesterolemic compound from Lentinus edodes (CHIBATA al., 1969), and the insecticidal et compound clitocine (67, Fig. 11) from Clitocybe inversu (KUBOet al., 1986). A compilation of nucleoside antibiotics from microbial sources and their biological activities was published by ISONO(1988) and by ISAACet al. (1991). A screening for inhibitors of avian myeloblastosis virus (AMV) reverse transcriptase resulted in the isolation of clavicoronic acid (68, Fig. 11) from fermentations of Cluvicoronu pyxidutu (ERKELet al., 1992). Clavicoronic acid is a non-competitive inhibitor of AMV (Ki: 130 pM) and Moloney murine leukemia (MMuLV) virus (Ki: 68 pM) reverse transcriptases. In permeabilized cells and isolated nuclei DNA and RNA synthesis are not
Fig. 10. Immunosuppressive metabolites.
Flavidulol A ( 1) 6
Flavidulol B (62)
Flavidulol C (63)
510
1
6-Melhylpudne (64)
6-Methyl-9-B-D-nboftuamsyIpurine (65)
H dH b
Clavicoronic acid (68) Clitocine (67)
Podoscyphic acid (69)
Fig. 1 . Antiviral com1 pounds and inhibitors of reverse transcriptases.
affected. Clavicoronic acid markedly inhibits the multiplication of VSV in BHK cells by interfering with the RNA-directed RNA polymerase of the virus. Clavicoronic acid exhibits no cytotoxic and very weak antimicrobial activity. Podoscyphic acid, (E)-4,5-dioxo-2-hexadecenoic acid (69, Fig. ll), isolated from fermentations of Podoscypha petalodes, is a noncompetitive inhibitor of AMV and MMuLV reverse transcriptase. The ICso values for the inhibition of AMV reverse transcriptase were 100 pg mL-' and for the MMuLV reverse DNA and transcriptase 10-20 pg mL-'.
RNA syntheses in whole cells and isolated nuclei are not affected by 100 pg mL-' podoscyphic acid. Comparison of the ethyl ester and the mono-oxo derivative of podoscyphic acid revealed the importance of the free y-oxoacrylate moiety for its biological activity (ERKEL al., 1991). et Several drimane sesquiterpenoids from basidiomycetes have been reported as inhibitors of reverse transcriptases. Drimanes had been isolated from a number of other natural sources including higher plants, ascomycetes, mollusks, and sponges (reviewed by JANSEN and DE GROOT,1991). The mniopetals A, 3,
511
Mniopetal B ( 1 7)
Mniopetal c (72)
Mniopetal D (73)
Mniopetal E (74)
Mniopetal F (75)
Kuehneromycin A (76)
Fig. 11
Kuehneromycin B (77)
Hyphodontal(78)
512
C, D, E, and F (70-75, Fig. 11)have been isolated from fermentations of a Canadian Mniopetulum species (KUSCHELet al., 1994). They most strongly inhibit the MMuLV reverse transcriptase at concentrations of 1.7-50 pM, with mniopetal B being the most active compound (ICso: 1.7 pM). The Icso values for the AMV reverse transcriptase are much higher. Inhibition of HIV-1 reverse transcriptase by mniopetals depends on the template primer used. With a natural heteropolymeric template inhibition of HIV-1 reverse transcriptase is most pronounced at concentrations of 30-190 pM. In addition, mniopetals exhibit cytotoxic properties which may be at least partly due to a lyctic action on the cytoplasma membrane. The kuehneromycins A and B (76 and 77, Fig. 11) have been isolated from cultures of a Tasmanian Kuehneromyces species (ERKEL al., 1995; ANKE et et al., 1993). Like mniopetals the kuehneromycins A and B preferentially inhibit the MMuLV reverse transcriptase with an ICsOof 36 pM, while inhibition of AMV reverse transcriptase is much less pronounced. The activity of HIV-1 reverse transcriptase with the natural heteropolymeric template is reduced to 50 % at a concentration of 64 pM. In addition, both compounds exhibit cytotoxic and antimicrobial activity.
0
Hyphodontal (78, Fig. l l ) , a new isolactarane sesquiterpenoid, has been isolated from fermentations of a Hyphodontiu species (ERKEL et al., 1994). Hyphodontal strongly inhibits the growth of several yeasts and is a noncompetitive inhibitor of AMV (Ki: 349 pM) and MMuLV (Ki:112 (LM)reverse transcriptase. The ICso for the HIV-1 reverse transcriptase with the natural heteropolymeric template was determined to be 77 pM (20 pg mL - I ) . The cytotoxic activity of hyphodontal is mainly due to the interference with DNA and RNA syntheses in whole cells and isolated nuclei. Up to now, none of the inhibitors described above has been shown to inhibit the multiplication of HIV-1 or HIV-2 viruses in cellular systems.
5.6 Inhibitors of Platelet Aggregation

A screening for antithrombotic compounds using platelet rich plasma from bovine slaughter blood resulted in the isolation 2-methoxy-5-methyl-l,4-benzochinone of (MMBC 79, Fig. 12) from mycelial cultures of Lentinus udhaerens (LAUERet al., 1991).
v
0
Lagopodin B (80)
Ompbalone (81)
Panudial (82)
Fig. U.Inhibitors of platelet aggregation.
513
MMBC inhibits aggregation of human blood platelets induced by U46619 a prostaglandine analog and thromboxane mimic with an ICs0 of 2.5 pg mL-' (16.45 pM). This effect is completely reversible by the addition of 1.8-2.4 pM U46619. Hence, it MMBC is proposed to act as a competitive thromboxane A2 receptor antagonist. Similar effects on platelet aggregation were observed with the benzoquinones lagopodin B (80, Fig. 12) obtained from Coprinus cinereus and omphalon (81, Fig. 12), an antimicrobial and cytotoxic metabolite isolated from cultures of Lentinellus omphalodes (STARKet al., 1991). The drimane panudial (82, Fig. 12) from cultures of a Panus species was detected as an inhibitor of platelet aggregation (LORENZEN et al., 1993). Panudial strongly interferes with the ADP-, collagen-, U46619-, ristocetin-, arachidonic acid-, and thrombine-induced aggregation of human and bovine platelets. The IC,, values of panudial for all inducers except for thrombine varied between 5 pM and 35 pM. In addition, panudial showed inhibitory activity against type I phospholipase A2 (ICs0=23 pg mL - ') from Naja mosambique. In HL-60 cells the incorporation of leucine and uridine into protein and RNA is markedly inhibited by 4-8 pg mL-' panudial.
5.7 Herbicidal Compounds

A screening for inhibitors of the key enzymes of the glyoxylate cycle, which are potential targets for herbicides, resulted in the isolation of mycenon (83, Fig. 13), a novel chlorinated benzoquinone derivative from et cultures of a Mycena species (HAUTZEL al., 1990). Mycenon inhibits isocitrate lyase preparations from plants, bacteria, and fungi. The Ki values were determined to be 5.2 pM, 11 pM,and 7.4 pM for the enzymes from Rhicinus communis, Acinetobacter calcoaceticus, and Neurospora crassa, respectively. Malate synthase, the second key enzyme of the glyoxylate cycle, was not affected. In addition to its phytotoxic activities mycenon exhibits antimicrobial and cytotoxic properties. Antibiotically active products which are structurally related to mycenon are siccayne (84, Fig. 13) from cultures of the marine ba-
sidiomycetes Halocyphina villosa (KUPKAet al., 1981b) and Helminthosporium siccans (ISHIBASHI al., 1968), frustulosinol, and et frustulosin (85 and 86, Fig. 13) isolated from cultures of Stereurn frustolosum (NAIR and ANCHEL, 1977). These compounds are hydrochinone derivatives with an isopentenyne side chain but without chlorine substitutions. Other acetylenic substances containing an aromatic ring are the antibacterial and antifungal antibiotics peniophorin B and A (87 and 88, Fig. 13) (GERBERet al., 1980). The pereniporins A and B (89 and 90, Fig. 13) were isolated from cultures of Perenniporia medullaepanis. Pereniporin A inhibits the root elongation of lettuce at 100 pg mL-' and is active against gram-positive bacteria. Both compounds show cytotoxic activity against Friend leukemia cells at 130 pg mL-' and 3.91 pg mL-', respectively (KIDAet al., 1986). Fomannosin (91, Fig. 13) was isolated from cultures of the wood-rotting basidiomycete Fomes annosus (Heterobasidion annosum). F. annosus is one of the relatively few basidiomycetes that cause the death of host cells in living trees and an extensive decay of heartwood of contaminated trees. Fomannosin exhibits phytotoxic activity in assays with Chlorella pyrenoidosa and against Pinus taeda seedlings when applied to the stem base or a lateral root at a concentration of 88 pg per seedling (BASSETT al., 1967). The absolute et configuration and the biosynthesis of fomannosin was elucidated by CAINEand NACHBAR (1978). Fomannoxin (92, Fig. 13), a dihydrobenzofuran which is 100 times more toxic to Chlorella pyrenoidosa, was isolated from the same fungus by HIROTANI al. (1977). et Several plant growth inhibitors were isolated from fruiting bodies of Naematoloma fasciculare. The fasciculols A, B, and C (9395, Fig. 13) are tetracyclic triterpenes with a hydroxylated lanostane skeleton. The fascicu101s D, E, and F (96-98, Fig. 13) are the corresponding esters with a novel depsipeptide group consisting of 3-hydroxy-3-methylglutaric acid and glycine (IKEDA et al., 1977). The fasciculols and their depsipeptides inhibited the growth of Chinese cabbage seedlings at concentrations of 100-300 pg mL-'. In addition, fasciculol D exhibited weak antibac-
514
Mycenon (83)
Siccayne (84)
Fmtulosinol(85)
Fmtulosin (86)
Peniophorin B (87) Pereniporin A (89)
Pereniporin B (90)
d==
0
cH2 --C %H
OCH3
--C -%OH
0
II
Peniophorin A (88)
9
HO
I t
Fomannoxin (92)
Fomannosin (91)
Fig. 13. Herbicidal compounds.
515
II cnjo -C -m2
"ii - --c N
-CH~-C
on 0 I -cnz -c II-
Fig. 13
terial activity against Staphylococcus aureus and Klebsiella pneumoniae. Several new related fasciculol esters, the fasciculic acids A, B, and C, were isolated from fruiting bodies of Naematoloma fasciculure as inhibitors of a calmodulin-dependent phosphodiesterase (PDE) (TAKAHASHI al., et 1989). The IC50values for the PDE from bovine heart were 6 pM for fasciculic acid B and 10 pM for fasciculic acid A. No data on phytotoxic activity have been published. Related tetracyclic triterpenoids with a lanostane skeleton have also been isolated from Gandoderma lucidum, G . applanatum, Piolithus tinctorius, P. arrhizus, and Fomes fastuosus (reviewed by CONNOLLY al., 1994). et
5.8 Insecticidal and Nematicidal Metabolites

Insecticidal activity against houseflies was described for ibotenic acid (99, Fig. 14) and its decarboxylation product muscimol (100,
Fig. 14). Ibotenic acid was isolated from fruiting bodies of Amanita muscaria, A . strobiliformis, and A . pantherina (for a review, see BRESINSKY BESL,1985). The cyclodepsiand peptide beauvericin (101, Fig. 14) which was isolated from the basidiomycete Polyporus sulphureus (DEOL et al., 1978) and from the entomopathogenic fungi Beauveria bassiana and Paecilomyces fumoso-roseus, exhibits insecticidal activity against mosquito larvae, brine shrimp, houseflies, and cockroach car1981). Beauveridiac cells in vitro (ROBERTS, cin is a ionophore forming complexes with alkali metals. Recently, inhibitory activity on acyl CoA-cholesterol acyltransferase (ACAT) in isolated rat liver microsomes has been described for beauvericin with an IC50 value of 3.0 pM (TOMODA al., 1992). In a et cell assay using 5774 macrophages the formation of cholesteryl esters was inhibited by 0.17 pM beauvericin. More than 150 species of nematophagous fungi belonging to Zygomycetes, Ascomycetes, Deuteromycetes, and Basidiomycetes are known to be capable of capturing nematodes, and hence the production of nematicidal toxins has been proposed (BARRON, 1977; SAYRE and WALTER,1991). 5-Pentyl-2-furaldehyde, 5-(4-pentenyl)-2furaldehyde, and methyl-3-p-anisoloxypropionate (102-104, Fig. 14), were isolated from cultures of Zrpex lacteus (HAYASHI al., et 1981). All three compounds caused 50% mortality of the nematode Aphelencoides besseyi at 25-50 pg mL-'. 2-Pecenedioic acid (105, Fig. 14), a fatty acid toxic to Panagrellus redivius, was isolated from a Pleurotus ostreatus strain. The compound immobilized 95% of the test nematode at a concentration of 300 pg mL-' within 1 h (KWOKet al., 1992). A screening using the saprophytic nematode Caenorhabditis elegans as a test organism resulted in the isolation of S-coriolic acid, linoleic acid, p-anisaldehyde, p-anisyl alcohol, l-(4-methoxyphenyl)-1,2-propanediol, 2and hydroxy-(4-methoxy)-propiophenone (106111, Fig. 14) from cultures of Pleurotus pulmonaris (STADLERet al., 1993b). The most active metabolites were S-coriolic acid and linoleic acid with LD50 values between 5 and 10 pg mL -'. Interestingly, the nematicidal
516
+
Ibotenic acid (99)
CHa
0.
6%
5-Pentyl-2-furaldehyde (102)
+
rrum-2-Decenoic acid (105)
5-(4-PentenyI)-2-1imldehyde(103)
IP
Beawericin (101)
Methyl-3-pankloxypropionate(104)
S-Coriolic acid (106)
Linoleic acid (107)
Fig. 14. Insecticidal and nematicidal metabolites.
517
p-Anisaldeyde (108)
p-Anisyl alcohol (109)
I -(CMethoxyphenyl)-l,2-propanediol(llO) 2-Hydroxy-(4'-methoxy)-propriophenone (11 1)
Cheimonophyllon A (1 12)
CheimonophyllonB (113)
Cheiinophyllon C (1 14)
CheimonophyllonD (1 15)
Fig. 14
cheimonophyllon E (116)
Cheimonophyllal(l17)
518
activity of different fatty acids depends on the chain length and the number of double bonds in the molecule. Furthermore, different species of nematodes varied in their sensitivity to various fatty acids. Six different bisabolane sesquiterpenes, the cheimonophyllons A-E (112-116, Fig. 14), and cheimonophyllal (117, Fig. 14) from cultures of Cheimonophyllum candidissimum have recently been described (STADLER et al., 1994). Cheimonophyllons A, B, D, and cheimonophyllal exhibit nematicidal, cytotoxic, and antimicrobial activity, whereas cheimonophyllons C and E show weaker effects. The LDso for Caenorhabditis eleguns was 10 Fg mL-' for cheimonophyllon A and D and 25 pg mL-' for cheimonophyllon B and cheimonophyllal.
5.10 Inhibitors of Aminopeptidases
Among the enzymes bound to the outer surface of mammalian cells aminopeptidases have been reported to be potential targets for immunomodulating drugs. A prominent example is bestatin (ubenimex), a strong inhibitor of aminopeptidase B and leucine aminopeptidase, which was isolated from cultures of Streptomyces olivoreticuli (UMEZAWA al., et 1976). Clinical studies of this drug were published by MATHB(1987). Tyromycin A (121, Fig. 16) from fermentations of Tyromyces lucteus is the first known naturally occurring citraconic anhydride derivative with two 3-methyl maleic anhydride units in its molecule. Tyromycin A strongly inhibits both leucine aminopeptidase and cys5.9 Inhibitors of Cholesterol teine aminopeptidase bound to the outer surface of HeLa S3 cells. The Ki values were deBiosynthesis termined to be 4.10-5 M for leucine aminoDrugs interfering with the biosynthesis of peptidase and 1.3.10-5 M for cysteine aminocholesterol are of potential value in the treat- peptidase. Tyromycin A also inhibits cytoment of hypercholesterolemia which is one of solic and microsomal leucine aminopeptidase the primary causes of arteriosclerosis and cor- of porcine kidney and carboxypeptidase onary heart disease. Mevinolin (monacolin of bovine kidney at concentrations of K), a specific inhibitor of eukaryotic 3-hy- 25-60 Fg mL-'. The inhibitory activity of droxy-3-methylglutaryl coenzyme A (HMG tyromycin A is due to the two maleic acid CoA) reductase isolated from cultures of anhydride moieties. Tyromycin amide (122, Monascus ruber (ENDO,1979) has been intro- Fig. 16) is devoid of any inhibitory activity on the cell-bound aminopeptidase of HeLa cells duced into clinical practice. Search for inhibitors of cholesterol biosyn- (WEBERet al., 1992). thesis in HeLa S3 cells resulted in the isolation of dihydroxerulin, xerulin, and xerulinic acid (118-120, Fig. 15), from surface cultures of Xerulu melanotricha (KUHNT et al., 1990). 5.11 Inhibitors of Phospholipases Dihydroxerulin strongly interferes with the C and A2 incorporation of 14C acetate into cholesterol, Caloporoside (123, Fig. 17), an inhibitor of while the incorporation of 14C mevalonate is hardly affected. It was shown that the inhibi- phospholipase C, has been isolated from fertory effect on cholesterol biosynthesis is due mentations of Caloporus dichrous (WEBERet to a strong inhibition of HMG CoA synthase al., 1994). Caloporoside is a new glycosylated salicylic acid derivative which exhibits weak at concentrations starting from 0.1 pg mL-' of dihydroxerulin. Similar results have been antibacterial and antifungal activity. Caloporoside shows a strong, selective inhibitory acobtained with xerulin. tivity on phospholipase C from pig brain with a Ki value of 12.3 pM. Phospholipases C from Clostridium welchii and Bacillus cereus, which act on other substrates are inhibited to a much lesser extent. Phospholipases A2 and D,
519
Dhydroxedi (118)
Xedin (119)
Xedinic acid (120)
Fig. 15. Inhibitors of cholesterol biosynthesis.
triglyceride lipases, and acetylcholin esterase are not affected. Several closely related derivatives of the aglycone part of caloporoside were isolated from fruiting bodies of Merulius tremellosus and Phlebia rudiafu. The merulinic acids A, B, and C (124-126, Fig. 17) exhibit antimicrobial and hemolytic activity which may be due to their lytic action on the cytoplasma membrane (GIANNETTI al., 1978). Merulinic et acids closely resemble the skin irritants of Anacardiaceae and of Ginkgo bilobu, anacardic acid 111, pelandjauic acid, and ginkgolic acid. Interestingly, mycelial cultures of M . trernellosus do not produce merulinic acids but the antibiotic sesquiterpenoid merulidial. 5-Hydroxy-3-vinyl-2(5H)-furanone (127, Fig. 17) has been isolated from cultures of Calypfella sp. (LORENZEN al., 1995). With reet
spect to the hydroxy-butenolide ring it resembles to the manoalides and luffarielloides which were isolated from Luffuriella variabilis and related sponges ( P o n s and FAULKNER, 1992). 5-Hydroxy-3-vinyl-2(5H)-furanone specifically inhibits the human synovial phospholipase AZwith ICs0 values of 100 nM. The compound also inhibits the aggregation of human and bovine platelets stimulated with different inducers.
5.12 Inhibitors of (Na +-K )-ATPases

+
Tyromycin A
(121) X = 0
Tyromycin amide (122) X = NH
Fig. 16. Inhibitors of aminopeptidases.
The tricyclic sesquiterpenoids coriolin and coriolin B (128 and 129, Fig. 18) were isolated from cultures of Coriolus consors (TAKEUCHI et al., 1969). While coriolin B shows neither antitumor nor antimicrobial activity, its oxidation product diketocoriolin B (130, Fig. 18) does. Studies on the mode of action of diketocoriolin B revealed that the antitumor activity is due to the inhibition of (Na+-K +)-ATPase localized in the cell membrane of tumor cells which causes a cessation of growth (KUNIMOTO et al., 1973). By chemical modification of coriolin NISHIMURA al. (1977) showed that et the keto group at C-5 and the two epoxy
520
I 1 Products from Basidiomycetes
OH
Caloporoside (123)
--C =C -(C!H2)6
i 7
R2
--C
H
R1
Merulioic acid A (124) OH H MedicacidB(125) H MedicacidC(126) H
R1 R2
OH H
P
Fig. 17. Inhibitors of phospholipases C and AZ
S-Hydroxy-3-vinyl-2(5~-tiua~ne (127)
groups greatly contribute to the antitumor and antibacterial activity. Diketocoriolin B augments antibody formation against sheep red blood cells (SRBC) in vivo at a concentration of 0.1 Fg per mouse or in in vifro using spleen cell cultures at 0.01 ng per culture (ISHIZUKA al., 1981). et
5.13 Addendum
This section describes compounds isolated from 1995 until November 1996. They are presented under the appropriate section number of the main text. Section 5.5: A screening for inhibitors of multiplication of vesicular stomatitis virus
(VSV) in baby hamster kidney cells (BHK21) resulted in the isolation of collybial (131, Fig. 19) from fermentations of Collybiu confluens (SIMONet al., 1995). The propagation of VSV in BHK-21 cells was reduced by a factor of lo3 at 21.5 pM of collybial with cytotoxic effects at 5-fold higher concentrations. Incorporation of labeled precursors into DNA, RNA, and proteins revealed that the antiviral effects of collybial are probably due to an interference with the macromolecular syntheses of the host. In addition, antibacterial activities against gram-positive bacteria were observed. Section 5.8 In addition to the recently described bisabolanes cheimonophyllons A-E (112-116, Fig. 14) and cheimonophyllal (117,
5 Bioactive Metabolitesfrom Basidiomycetes OH
521
Coriolin (128)
OH
Coriolim B (129)
0
Fig. 1 . Inhibitors of (Na+-K+)-ATPases. 8
Diketomriolin B (130)
Fig. 14), the p-menthane 1,Zdihydroxyminthlactone (132, Fig. 19) was isolated as a minor nematicidal component from fermentations of Cheimonophyllum candidissimum (STADLER et al., 1995). The LD50 against the nematode Caenorrhabditis elegans was determined to 25 pg mL- without any additional antimicrobial and cytotoxic activity. The structurally related minthlactone has been previously reported as a constituent of peppermint (Menet tha piperita) oil (TAKAHASHI al., 1980). Omphalotin (133, Fig. 19), a new cyclic dodecapeptide possessing strong and selective nematicidal activity against the plant pathogenic nematode Meloidogyne incognita was isolated from mycelial cultures of the basidiomycete Omphalotus olearius (MAYERet al., in press; STERNER al., in press). The LDgO et against M. incognita was determined to be 0.76 pg mL-, whereas the saprophytic nematode C. elegans was approximately 50 times less sensitive. Omphalotin exhibits no phytotoxic, antibacterial, or antifungal activities and is only weakly cytotoxic at high concentrations (100 pg mL-).
5.13.1 Inhibitors of Leukotriene Biosynthesis

Leukotrienes are potent biological mediators derived from arachidonic acid metabolism and are generated via the 5-lipoxygenase pathway. Leukotriene B4, a dihydroxy derivative, causes adhesion and chemotactic movement of leukocytes, enzyme release, and generation of superoxide in neutrophiles. The sulfopeptide leukotrienes C4,D4, and E4, are known as slow reacting substances of anaphylaxis and induce bronchoconstriction, stimulate mucus production, and increase vascular permeability (SAMUELSSONet al., 1987). Due to these effects the leukotrienes have been implicated as important mediators of inflammation and hypersensitivity reactions (FORD-HUTCHINSON al., 1994; SALet MON and GARLAND, 1991). A screening for inhibitors of leukotriene C4 biosynthesis resulted in the isolation of ( +)-10a-hydroxy-4-muurolen-3-one (134, Fig. 20) from fermentations of an Ethiopian Favolaschia species (ZAPFet al., 1996). The IC50 value for the inhibition of leukotriene C4 biosynthesis in rat basophilic leukemia (RBL-
522
OH
1,2-Dihydroxyminthlactone 32) (1
Collybial(l3 1)
Omphalotin (133)
Fig. 19. Antiviral and nematicidal compounds.
1) cells was determined to be 5-10 kg mL-'. Related cadinane sesquiterpenes, like ( + )-Tcadinol and ( -)-3-oxo-T-cadinol, have been et reported before (CLAERSON al., 1991) as constituents of scented myrrh (the resin of the plant Commiphoru guidotti), and similar effects on leukotriene C4 biosynthesis have been observed (ZAPFet al., 1996). Three known lacterane type sesquiterpenoids, blennin A (135, Fig. 20), blennin C (136, Fig. 20), and deoxylactarorufin A (137, Fig. 20) were obtained from fermentations of Lentinellus cochleutus (WUNDER al., 1996) et together with the new metabolites (2)-2-chlo-
ro-3-(4-methoxypheny1)-2-propen-l-o1 (138, Fig. 20) and lentinellone (139, Fig. 20), a protoilludane derivative. Blennin A (VIDARI et al., 1976), blennin C (DE BERNARDI al., et 1976), and deoxylactarorufin A (DANIEWSKI et al., 1977; DANIEWSKI KROL, 1981) are and strong inhibitors of leukotriene C, biosynthesis in RBL-1 cells with ICso values of 5 pgmL-' for blennin A, 4 pgmL-' for blennin C, and 2 Fg mL-' for deoxylactarorufin A. (Z)-2-chloro-3-(4-methoxyphenyl)-2propen-1-01 inhibited the leukotriene C4 biosynthesis with an IC5"of 15 kg mL -',whereas lentinellone was inactive.
5 Bioactive Metabolitesfrom Basidiomycetes
523
5.13.2 Inducers of Differentiation of Promyelocytic Leukemia Cells and Inhibitors of Signal Transduction in Tumor Cells
Among the phenotypic abnormalities in acute leukemia is a lack of granulocytes, macrophages, and platelets caused by the inability of the neoplastic leucocytes to undergo terminal differentiation and eventually apoptosis. The human HL-60 leukemia cell line is an excellent model for a study of functional and morphological differentiation in vitro, because the cells can be induced to differentiate into granulocytes or monocytes/macrophages. Differentiation may be followed by apoptosis, a process of active DNA fragmentation. The induction of differentiation and apoptosis are regulated by a network of signal transduction pathways and transcription factors which are
possible targets for a rational antitumor ther1994; THOMPapy (HASS, 1992; LEVITZKI, SON, 1995; MANNING, 1996). Pinicoloform (140, Fig. 21) an antibiotic from Resiniciurn pinicolu was detected in a screening for metabolites inducing the differentiation of HL-60 cells to monocytes and macrophages (BECKER et al., 1994) at concentrations between 0.5-1 yg mL -' (1.8-3.5 yM). Cytotoxic as well as antimicrobial activities have also been described at concentrations between 10-66 yM. The diterpenes lepistal (141, Fig. 21) and lepistol (142, Fig. 21) were isolated from cultures of Lepistu sordidu (MAZURet al., 1996). At a concentration of 0.2 yg mL lepistal induces the differentiation of 20 % of the HL-60 cells into granulocyte/monocyte-likecells and of 18% of the human histiocytic lymphoma (U-937) cells into monocyte-like cells. The related alcohol lepistol is 50-100 times less active. Cytotoxic activities of lepistal and lepis-
-'
(+)-l~-hydroxy-4-muurolen-3-one (1 34)
Blennin A (135)
Blennin C (136)
DeoxylactarorufinA (1 37)
Fig. 20. Inhibitors of leukotriene biosynthesis.
(Z)-2-chloro-3-(4-mehoxyphenyl)-2-prop~-~-ol(138)
Lmtindlone (1 39)
524
to1 were observed at 1 pg mL-' and 50 pg mL-', respectively. Lepistal exhibits pronounced antimicrobial activity whereas lepistol shows no antibacterial and only weak antifungal activities. Therefore, the aldehyde function is considered to substantially contribute to the biological activity of lepistal. Recently, two new bisabolane sesquiterpenes, nidulal (143, Fig. 21) and niduloic acid (144, Fig. 21), have been isolated from feret mentations of Nidulu cundidu (ERKEL al., in press). Nidulal and niduloic acid induce differentiation of 15-25% of HL-60 cells at concentrations of 72 p M and 36 pM, respectively. It has been shown that induction of dif-
ferentiation of HL-60 cells by nidulal is followed by apoptosis. In COS-7 cells (African green monkey) nidulal selectively activates the AP-1 dependent signal transduction pathways in a manner similar to the phorbol ester TPA, an activator of protein kinase C. In gel shift assays with extracts of nidulal-treated HL-60 cells a change of binding activities of the AP-1 transcription factor is observed, which may be the result of an altered composition of the AP-1 protein complex. The novel norilludane puraquinonic acid (145, Fig. 21) was isolated from mycelial culet tures of Mycenu puru (BECKER al., in press) as an inducer of morphological and
pinicolofom (140)
Lepistal(141) R==O Lepistol(142) R= -OH
Nidulal(143)
Niduloic acid (144)
OH F'uraquinonic acid (145)

Panepoxydone (146)
Fig. 21. Inducers of differentiation and inhibitors of signal transduction in tumor cells.
6 Future Perspectives
525
physiological differentiation of mammalian cells. It induces differentiation of 3 0 4 0 % of HL-60 cells into granulocyte- or monocyte/ It is evident that basidiomycetes provide a macrophage-like cells at 380 p,M. At the same rich, yet quite untapped source of compounds concentration U-937 cells, which are blocked at a later stage of development are affected to with novel structures and in some cases very interesting biological activity. Many of the a much lesser extent. NF-KB is an inducible, ubiquitous tran- compounds isolated so far seem to be proscription factor, which regulates the expres- duced exclusively by this class of fungi. From sion of various cellular genes involved in im- the number of estimated basidiomycetes spemune response, inflammation, acute phase re- cies (>30OOO) it may be assumed that there sponse, and several viral genes and inhibitors will be many more exciting discoveries made of NF-KB activation may, therefore, find in the future. broad application as novel therapeutics (BAEUERLE and HENKEL,1994; BALDWIN, 1996; MANNING ANDERSON, and 1994). In a search for new inhibitors of NF-KBmediated signal transduction in COS-7 cells using the secreted alkaline phosphatase J., E. (SEAP) as reporter gene, panepoxydone AMES,B. N., MCCANN, YAMASAKI, (1975), Methods for detecting carcinogens and muta(146, Fig. 21) was isolated from fermentations gens with the Salmonellalmammalian mutagenicof the basidiomycete Lentinus crinitus (ERity test, Mut. Res. 31, 347-364. KEL et al., 1996). Panepoxydone and several ANCHEL, HERVEY, ROBBINS, J. (1950), M., A., W. related derivatives have been previously reAntibiotic substances from basidiomycetes. VII. Clitocybe illudens, Proc. Natl. Acad. Sci. USA 36, ported as secondary metabolites from Punus 300-305. rudis, P. conchatus (KIS et al., 1970), and H. Penicilliurn urricue (SEGIGUCHI and GAU- ANKE,T., DIEKMANN, (1972), Metabolic products of microorganisms. 112. Biosynthesis of siCHER, 1979).Panepoxydone inhibits the NF-KB deramines in fungi. Rhodotorulic acid synthetactivated expression of the SEAP with an ase from extracts of Rhodotorula glutinis, FEBS ICso of 1.5-2 p,g mL-' (7.15-9.52 p,M). No inLett. 27, 259-262. hibition of AP-1-mediated expression of the ANKE,T., STEGLICH, (1981). Screening of baW. reporter gene could be observed at a concensidiomycetes for the production of new antibiotration up to 5 p,g mL-' panepoxydone. Pantics, in: Advances in Biotechnology, Vol. 1 (MOO-YOUNG, M., Ed.), pp 35-40. Pergamon epoxydone strongly reduces the TPA-, TNF-a-, Press, Canada. and ocadaic acid-mediated binding of NF-KB W. to the high affinity consensus sequence in ANKE,T., STEGLICH, (1989), P-Methoxyacrylate antibiotics: from biological activity to synCOS-7 and HeLa S3 cells as confirmed by thetic analogues, in: Biologically Active Moleelectrophoretic mobility shift assays. Pancules - Identification, Characterization, and Synepoxydone inhibits the phosphorylation of the U. thesis (SCHLUNEGGER, P., Ed.), pp. 9-25. inhibitory protein IKB and, therefore, sequesBerlin-Heidelberg: Springer-Verlag. ters the NF-KB complex in an inactive form. ANKE,H., ANKE,T., DIEKMANN (1973), BioH. Recently, the related cycloepoxydon, an inhisynthesis of sideramines in fungi. Fusigen synthetase from extracts of Fusarium cubense, bitor of AP-1- and NF-KB-mediated gene exFEBS Lett. 36,323-325. pression, has been isolated from fermentaF., W., T., tions of a deuteromycete strain (GEHRTet al., ANKE, OBERWINKLER,STEGLICH, HOFLE, G. (1977a), The striatins - new antibiotics in press). from the basidiomycete Cyathus striatus (Huds. ex Pers.) Willd., J. Antibiot. 30,221-225.
6 Future Perspectives
7 References
ANKE, T., OBERWINKLER, STEGLICH,W., F., SCHRAMM, (1977b), The strobilurins - new G. antifungal antibiotics from the basidiomycete Strobilurus tenacellus (Pers. ex. Fr.) Sing., J. Antibiot. 30,806-810.
526

(Relhan ex Fr.) Sing., J. Antibiot. 43, 10101011. ANKE,T., ERKEL,G., KOKSCH, KUSCHEL, G., A., GIMENEZ, A., VELTEN, R., STEGLICH,W. (1993), Inhibitoren der HIV-1 Reversen Transcriptase und Induktoren der Differenzierung menschlicher Zellinien aus Pilzen, in: Wege zu neuen Produkten und Verfahren in der BiotechT., U., nologie (ANKE, ONKEN, Eds.). DECHEMA Monographien 129, 15-25. ARIGONI, (1962), La struttura di un terpene di D. nuovo genere, Gazz. Chim. Itl. 22, 884-901. AYER,W. A., BROWNE, M. (1981), Terpenoid L. metabolites of mushrooms and related basidiomycetes, Tetrahedron Lett. 17, 2199-2248. T. AYER, W. A., LEE, S. P., NAKASHIMA, T. (1979), Metabolites of birds nest fungi. Part 12. Studies on the biosynthesis of cyathins, Can. J. Chem. 17,3338-3343. BACKENS, STEFFAN, STEGLICH, ZECHS., B., W., LIN,L., ANKE, (1984), Antibiotika aus BasidT. iomyceten. XIX. Naematolin und Naematolon, zwei Caryophyllan-Derivate aus Kulturen von Hypholoma-Arten (Agaricales), Liebigs Ann. Chem. 1984, 1332-1342. S., W., J., BACKENS, STEGLICH, BAUERLE, ANKE, T. (1988), Antibiotika aus Basidimyceten, 28. Hydroxystrobilurin D, ein antifungisches Antibiotikum aus Kulturen von Mycena sanguinolenta (Agaricales), Liebigs Ann. Chem. 1988, 405409. BALDWIN, S. (1996), The NF-KB and IKB proA. teins: new discoveries and insights, Annu. Rev. Immunol. 14,649-681. BARRON, L. (1977), The nematode-destroying G. fungi, in: Topics in Mycobiology, Vol. 1. Guelph, Ontario: Canadian Biological Publications. BASSETT, SHERWOOD, T., KEPLER,J. A., C., R. HARMILTON, B. (1967), Production and bioP. logical activity of famannosin, a toxic sesquiterpene metabolite of Fomes annosum, Phytopathology 57, 1046-1052. BAUERLE, (1981), Antibiotika aus BasidiomyceJ. ten der Gattungen Clitopilus, Hohenbuehelia, Hemimycena, und Mycena (Agaricales). Thesis, Department of Biology, University of Tubingen, FRG. BAUERLE,J., ANKE,T. (1980), Antibiotics from the genus Mycena and Hydropus scabripes, Planta Med. 39, 195-196. T. BAUERLE, A., HENKEL, (1994), Function and P. activation of NF-KB in the immune system, Annu. Rev. Immunol. 12,141-179. BAUERLE,J., ANKE,T., JENTE, R., BOSOLD,F. (1982), Antibiotics from basidiomycetes. XVI. Antimicrobial and cytotoxic polyines from M y cena viridimarginata, Arch. Microbiol. 132, 194196.
ANKE, HECHT,H. J., SCHRAMM, STEGLICH, T., G., W. (1979), Antibiotics from basidiomycetes. IX. Oudemansin, an antifungal antibiotic from Oudemansiella mucida (Schrader ex Fr.) Hoehnel (Agaricales), J. Antibiot. 32, 1112-1117. G., ANKE,T., KUPKA,J., SCHRAMM, STEGLICH, W. (1980), Antibiotics from basidiomycetes. X. Scorodonin, a new antibacterial and antifungal metabolite from Marasmius scorodonius (Fr.) Fr., J. Antibiot. 33, 4 6 3 4 7 . ANKE,T., GIANETTI, B.M., STEGLICH (1982), W. Antibiotika aus Basidiomyceten. XV. 1-Hydroxy-2-nonin-4-on,ein antifungischer und cytotoxischer Metabolit aus Ischnoderma benzoinum (Wahl.) Karst., Z. Naturforsch. 37c, 14. ANKE,T., BESL, H., MOCEK,U., STEGLICH, W. (1983), Antibiotics from basidiomycetes. XVIII. Strobilurin C and oudemansin B, two new antifungal metabolites from Xerula species (Agaricales), J. Antibiot. 36,661-666. ANKE,T., SCHRAMM, SCHWALGE, STEFFAN, G. B., B., STEGLICH, (1984), Antibiotika aus BasiW. diomyceten. XX. Synthese von Strobilurin A und Revision der Stereochemie der natilrlichen Strobilurine, Liebigs Ann. Chem. 1984 16161625. W. ANKE,T., BACKENS, STEGLICH, (1985a), A S., new antibiotic from Paneolus and Psathyrella species. Abstract Annu. Meeting ASM. Washington, D.C.: ASM Press. T., F., U., ANKE, HEIM,J., KNOCH, MOCEK, STEFFAN,B., STEGLICH, (1985b), Crinipelline, die W. ersten Naturstoffe mit einem Terquinan-Gerust, Angew. Chem. 27, 714-716. ANKE,T., STEGLICH, POMMER, H. (1986), W., E. The antifungal activities of striatals and striatins. Abstracts 6th Int. Congr. Pesticide Chem. (IUPAC), Ottawa. G., W., ANKE,T., SCHRAMM, STEGLICH, VON JACOW,G. (1988), Structure-activity relationships of natural and synthetic E-pmethoxyacrylates of the strobilurin and oudemansin series, in: The Roots of Modern Biochemistry (KLEINKAUF, H., VON DOHREN, H., JANICKE, Eds.), pp. 657L., 662. Berlin: de Gruyter. ANKE, H., HILLEN-MASKE, STEGLICH,W. E., (1988), 9-phydroxymarasmic acid and other sesquiterpenoids from submerged cultures of a basidiomycete, Z. Naturforsch. 44c,1-6. ANKE, H., STERNER, STEGLICH, (1989), O., W. Structure-activity relationship for unsaturated dialdehydes. 3. Mutagenic, antimicrobial, cytotoxic and phytotoxic activities of merulidial derivatives, J. Antibiot. 42, 738-744. ANKE,T., WERLE,A., BROSS,M., STEGLICH, W. (1990), Antibiotics from basidiomycetes. XXXIII. Oudemansin X, a new antifungal E-Pmethoxyacrylate from Oudemansiella radicata
7 References
527
BAUERLE, ANKE, HILLEN-MASKE, STEGJ., T., E., LICH, W. (1986), Hemimycin, a new antibiotic from two Hemimycena species (Basidiomycetes), Planta Med. 5, 418. BEALE,M. H. (1990), The biosynthesis of C5-Cz0 terpenoid compounds, Nat. Prod. Rep. 7, 25-39, 387-407. BECKER, ANKE,T., STERNER, (1994), A U., 0. novel halogenated compound possessing antibiotic and cytotoxic activities isolated from the fungus Resinicium pinicola (J. Erikss.) Erikss. & Hjortst., 2. Naturforsch. 49c, 772-774. BECKER, ERKEL, ANKE, STERNER, U., G., T., 0. (in press), Puraquinonic acid, a novel inducer of differentiation of human HL-60 promyelocytic leukemia cells from Mycena pura (Pers. ex Fr.), Nat. Prod. Lett. BIRCH, J., CAMERON, W., HOLZAPFEL, A. D. C. W., RICKARDS, W. (1963). The diterpenoid R. nature of pleuromutilin, Chem Ind. (London) 374-375. BIRCH, J., HOLZAPFEL, W., RICKARDS, A. C. R. W. (1966), The structure and some aspects of the biosynthesis of pleuromutilin, Tetrahedron (Suppl.) 8 Part 11, 359-387. BRANDT,U., SCHLAGGER, VON JAGOW, H., G. (1988), Characterisation of binding of the methoxyacrylate inhibitors to mitochondria1 cytochrome c reductase, Eur. J. Biochem. 173, 499506. BRANDT, HAASE,U., SCHLAGGER, VON U., H., JAGOW, (1993), Speziesspezifitat und WirkG. mechansimus der Strobilurine, in: Wege zu neuen Produkten und Verfahren in der Biotechnologie (ANKE, ONKEN, Eds). DECHET., U., MA Monographien 129,27-38. BRAUER, KORN, A. (1993), Untersuchung der H., Umsatzleistung im Hubstrahl-Bioreaktor bei der biotechnologischen Produktion mit Pilzen, in: Wege zu neuen Produkten und Verfahren in der Biotechnologie (ANKE,T., ONKEN, Eds.). U., DECHEMA Monographien 129, 147-157. BRESINSKY, BESL, H. (1985), Giftpilze. StuttA., gart: Wissenschaftliche Verlagsgesellschaft. CANE, E., NACHBAR, B. (1978), StereochemD. R. ical studies of isoprenoid biosynthesis. Biosynthesis of fomannosin from [1,2-"CC2]acetate, J. Am. Chem. SOC. 100,3208-3212. CHENINA, DOCHNAHL, HUFF,T., KUBALL, R., A., H. G., LANGE, ANKE,H., ANKE,T. (1993), B., Fluoreszenzspektrometer zur On-line-Verfolgung von Fermentationsprozessen, in: Wege zu neuen Produkten und Verfahren in der Biotechnologie (ANKE, ONKEN, Eds.). DECHET., U., MA Monographien 129, 171-182. CHIBATA, OKUMURA, TAKEYAMA, KoI., K., S., TERA, K. (1969), A new hypercholesterolemic
substance in Lentinus edodes, Experientia 25, 1237-1238. Y. CHIHARA, HAMURO, MADEA, Y., ARAI, G., J., Y., FUKUOKA, (1970), Fractionation and puF. rification of the polysaccharides with marked antitumor activity, especially lentinan from Lentinus edodes (Berk.) Sing. (an edible mushroom), Cancer Res. 30,2776-2781. CLAERSON, ANDERSON,R, SAMUELSSON, P., G. (1991), T-cadinol: a pharmacologically active constituent of scented myrrh: introductory pharmacological characterization and high field 'H and I3C NMR data, Planta Med. 57, 352-356. CLOUGH, M. (1993), The strobilurins, oudemanJ. sins, and myxothiazols, fungicidal derivatives of Pmethoxyacrylic acid, Nat. Prod. Rep. 10, 565574. CONNOLLY, D., HILL, R. A., NGADJUI, T. J. B. (1994), Triterpenoids, Nat. Prod. Rep. 1 , 911 117. DANIEWSKI, M., KROL, J. (1981), Constituents W. of higher fungi. Part XII, Pol. J. Chem. (Rocz. Chem.) 55, 1247-1252. DANIEWSKI, M., KOCOR,M., KROL,J. (1977), W. Constituents of higher fungi, Pol. J. Chem. (Rocz. Chem.) 51, 1395-1398. H., DAVIES, VON AHSEN,U., WANK, SCHROEJ., DER, R. (1992), Evolution of secondary metabolite production: potential role for antibiotics as prebiotic effectors of catalytic RNA reactions, in: Secondary Metabolites: Their Function and Evolution (CADWICK, J., WHELAN, Eds.), D. J., pp. 2444. Ciba Foundation Symposium 171. DE BERNARD], M., FRONZA,G., VIDARI,G., VITA-FINZI, (1976), Fungal metabolites 11: P. new sesquiterpenes from Lactarius scobiculatus Scop. (Russulaceae), Chim. Ind. 58, 177-178. DE CLERCQ,E. (1990), Perspectives for chemotherapy of the HIV infection: An introduction, Pharmacochem. Libr. 14 (Des. Anti-AIDS Drugs), 1-24. DEML,G., ANKE,T., OBERWINCKLER, GIANF., NETTI, B. M., STEGLICH, (1980), Schizonellin W. A and B, new glycolipids from Schizonella melanogramma (Ustilaginales), Phytochemistry 19, 83-87. D. P. DEOL,B. S., RIDLEY D., SINGH, (1978), Isolation of cyclodepsipetides from plant pathogenic fungi, Aust. J. Chem. 31, 1397-1399. DREWS, J., GEORGOPOULOS, LABER, G., A., SCHUTZE, UNGER, (1975), Antimicrobial E., J. activities of 81.723 hfu, a new pleuromutilin derivative, Antimicrob. Agents Chemother. 7, 507516. DUGAN, J., DEMAYO, NISBET, M., ROBBINJ. P., SON, J. R., ANCHEL, (1966), Terpenoids XIV. M. The constitution and biogenesis of marasmic acid, J. Am. Chem SOC. 88,2838-2844.
528
I I Products from Basidiomycetes

antitumor polysaccharide, KS-2, extracted from culture mycelia of Lentinus edodes, J. Antibiot. 31,1079-1090. FUJIMOTO, NAKAYAMA, YAMAZAKI, H., Y., M. (1993), Identification of immunosuppressive components of a mushroom, Lactarius flavidulus, Chem. Pharm. Bull. 41,654-658. GEHRT, A., ERKEL, G., ANKE, H., ANKE, T., STERNER, (in press), New hexaketide inhibi0. tors of eukaryotic signal transduction, Nat. Prod. Lett. GERBER, N., SHAW,S. A., LECHEVALIER, N. H. (1980), Structures and antimicrobial activities of peniophorin A and B, two polyacetylenic antibiotics from Peniophora afjinins Burt., Antimicrob. Agents Chemother. 17, 636-641. I. GERMERDONK, BECKER, GEHRIG, (1993), R., P., Verfahrenstechnische Methoden zur Verbesserung des Produktionsverhaltens bei der Fermentation des Pilzes Cyathus striatus, in: Wege zu neuen Produkten und Verfahren in der BiotechT., U., nologie (ANKE, ONKEN, Eds.). DECHEMA Monographien 129, 159-169. GIANNETTI, M., STEGLICH, B. W., QUACK,W., ANKE,T., OBERWINCKLER,(1978), MerulinF. sauren A, B und C, neue Antibiotika aus Merulius tremellosus Fr. und Phlebia radiata Fr., Z. Naturforsch. 33c, 807-816. GILL, M. (1994), Pigments of fungi (Macromycetes), Nat. Prod. Rep. 11, 67-90. GILL,M., STEGLICH, (1987), Pigments of fungi W. (Macromycetes), in: Progress in the Chemistry of Organic Natural Products, Vol. 51 (ZECHMEISTER, L., Ed.), pp. 1-317. Wien, New York: Springer-Verlag. W. R. GREENLEE, J., WOODWARD, B. (1976), Total synthesis of marasmic acid, J. Am. Chem. SOC. 98,6075476. HANSON, R., MARTEN, (1973), Incorporation J. T. of [2-H2]- and [4(R)-4-H]-rnevalonoid hydrogen atoms into the sesquiterpenoid illudin M., J. Chem SOC. Chem. Comm. 171-172. HARA,M., YOSHIDA, MORIMOTO, NAKAM., M., NO, H. (1987), 6-deoxyilludin M, a new antitumor antibiotic: Fermentation, isolation and structural identification, J. Antibiot. 40, 16731646. U., A., HARTTIG, ANKE,T., SCHERER, STEGLICH, W. (1990), Leianafulvene, a sesquiterpenoid fulvene derivative from cultures of Mycena leaiana, Phytochemistry 29, 3942-3944. HASS,R. (1992), Retrodifferentiation - an alternative biological pathway in human leukemia cells, Eur. J. Cell Biol. 58, 1-11. HASS, V. C., MUNACK, (1993), Modellierung A. und Regelung der Kultivierung von Cyathius striatus, in: Wege zu neuen Produkten und Verfahren in der Biotechnologie (ANKE,T., ON-
EGGER,H., REINSHAGEN, (1976a), New pleuroH. mutilin derivatives with enhanced antimicrobial activity. I. Synthesis, J. Antibiot. 29, 915-922. H. EGGER, REINSHAGEN, (1976b), New pleuroH., mutilin derivatives with enhanced antimicrobial activity. 11. Structure-activity correlations, J. Antibiot. 29, 923-927. ENDO, (1979), Monacolin K, a new hypercholesA. terolemic agent produced by a Monascus species, J. Antibiot. 32, 852-854. R., W. ERKEL,G., ANKE,T., VELTEN, STEGLICH, (1991), Podoscyphic acid, a new inhibitor of Avian myeloblastosis virus and Moloney murine leukemia virus reverse transcriptase from a Podoscypha species, Z. Naturforsch. &, 442-450. A., ERKEL,G., ANKE,T., GIMENENZ, STEGLICH, W. (1992), Antibiotics from basidiomycetes. XLI. Clavicornic acid, a novel inhibitor of reverse transcriptases from Clavicorona pyxidata (Pers. ex Fr.) Doty, J. Antibiot. 45, 29-37. R., A,, ERKEL,G., ANKE,T., VELTEN, GIMENEZ, STEGLICH, (1994), Hyphodontal, a new antiW. fungal inhibitor of reverse transcriptases from Hyphodontia sp. (Corticaceae, Basidiomycetes), Z. Naturforsch. 49c, 561-570. T., R., ERKEL, LORENZEN, ANKE, VELTEN, G., K., GIMENEZ, STEGLICH, (1995), KuehneroA., W. mycins A and B, two new biological active compounds from a tasmanian Kuehneromyces sp. (Strophariaceae, Basidiomycetes), 2. Naturforsch. 5oe, 1-10. ERKEL,G., ANKE, STERNER, (1996), InhibiT., 0. tion of NF-KBactivation by panepoxydone, Biochem. Biophys. Res. Commun. 224214221. 0. ERKEL,G., BECKER,U., ANKE,T., STERNER, (in press), Nidulal, a novel inducer of differentiation of human promyelocytic leukemia cells from Nidula candida, J. Antibiot. FLOREY,H. W., CHAIN,W., HEATLEY,A., JENNINGS, G., ABRAHAM, P., FLOREY, E. N. E. M. (1949), Antibiotics. London: Oxford University Press. FORD-HUTCHINSON, A. W., GRESSER, M., YOUNG,R. N. (1994), 5-Lipoxygenase, Annu. Rev. Biochem. 63, 383-417. FRAGA,B. M. (1990), Natural sesquiterpenoids, Nat. Prod. Rep. 7 , 61-84. FREDENHAGEN, KUHN, A., PETER, H. H. A., (1990a), Strobilurins F, G, and H, three new antifungal metabolites from Bolinea lutea. I. Fermentation, isolation and biological activity, J. Antibiot. 43, 655-660. FREDENHAGEN, HUG, P., PETER, H. H. A., (1990b), Strobilurins F, G and H, three new antifungal metabolites from Bolinea lutea. 11. Structure determination, J. Antibiot. 43,661-667. FUJI, T., MAEDA, H., SUZUKI,F., ISHIDA, N. (1978), Isolation and characterization of a new
7 References
529
KEN,U., Eds.). DECHEMA Monographien 129, 135-145. HAUTZEL, R., ANKE, H., SHELDRICK, W. S. (1990), Mycenon, a new metabolite from a Mycena species T A 87202 (Basidiomycetes) as an inhibitor of isocitrate lyase, J. Antibiot. 43, 12401244. HAYASHI, WADA,K., MUNAKATA, (1981), M., K. New nematicidal metabolites from a fungus, Irpex lacteus, Agric. Biol. Chem. 45, 1527-1529. HECHT, H. J., HOFLE, G., STEGLICH, ANKE, W., T., OBERWINKLER,(1978), Striatin A, B, and F. C: Novel diterpenoid antibiotics from Cyathus striatus; X-Ray crystal structure of striatin A, J. Chem. SOC. Chem. Commun. 665-666. HEIM, J., ANKE, T., MOECK, U., STEFFAN, B., STEGLICH, (1988), Antibiotics from basidioW. mycetes. XXIX. Pilatin, a new biologically active marasmane derivative from cultures of Flagelloscypha pilatii Agerer, J. Antibiot. 41, 17521757. HERBERT,R. B. (1989), The Biosynthesis of Secondary Metabolites. London, New York: Chapman and Hall. HIROTANI, O'REILLY,J., DONELLY, M. X. M., D. (1977), Fomannoxin - A toxic metabolite from Fomes annosum, Tertahedron Lett. 7, 651-652. HOGENAUER, (1979), Tiamulin and pleuromuG. tilin, in: Antibiotics, Vol. V-1 (HAHN, E., Ed.), F. pp. 340-360. Berlin, Heidelberg, New York: Springer-Verlag. HOUCHINS, HIND,G. (1983), Flash spectroscopJ., ic characterization of photosynthetic electron transport in isolated heterocysts, Arch. Biochem. Biophys. 224,272-282. HUNECK, CAMERON, F., CONOLLY, D., S., A. J. MCLAREN, RYCROFT, D. S. (1982), HercyM., nolactone, a new carotene sesquiterpenoid from the liverworts Barbilophozia lycopodioides and B. hatcheri. Crystal structure analysis, Tetrahedron Lett. 23,3959-3962. IKEDA, NIWA,G., TOHAYAMA, SASSA,T., M., K., MIURA, (1977). Structures of fasciculol C and Y. its depsipeptides, new biologically active substances from Nematoloma fasciculare, Agric. Biol. Chem. 41, 1803-1805. ISAAC, B. G., AYER, S. W., LETENDRE,L. J., STONARD, J. (1991), Herbicidal nucleosides R. from microbial sources, J. Antibiot. 44, 729-732. ISHIBASHI, NOSE, K., SHINDO, ARAI, M., K., T., MISHIMA, (1968), Siccayne: a novel acetylenH. ic metabolite of Helminthosporium siccans, Ann. Sankyo Res. Lab. 20,76-79. ISHIZUKA,M., TAKEUCHI, T., UMEZAWA,H. (1981), Studies on the mechanism of action of deketocoriolin B to enhance the antibody formation, J. Antibiot. 34, 95-102.
ISONO, (1988), Nucleoside antibiotics: structure, K. biological activity, and biosynthesis, J. Antibiot. 41, 1711-1739. JANSEN, J. M., DE GROOT,A. (1991), The occurB. rence and biological activity of drimane sesquiterpenoids, Nut. Prod. Rep. 8, 309-318. JELLISON, GOODELL,B., FEKETE,F., CHANDJ., HOKE, V. (1990), Fungal siderophores and their role in wood biodegradation. The International Research Group On Wood Preservation. Paper prepared for the 21st Annual Meeting, Rotorua, New Zealand, 13-18 May, 1990. JONES,R. H., THALLER, (1973), Microbial polyV. ynes, in: Handbook of Microbiology, Vol. 111, Microbial Products (LASKIN, I., LECHEVALa. IER, H. A., Eds.), pp. 63-74. Cleveland, OH: CRC Press. KANEDA, TAKAHASHI, IITAKA, SHIBAM., R., Y., TA, S. (1972), Retigeranic acid, a novel sesterpene isolated from the lichens of Lobaria retigera group, Tetrahedron Lett. 1972, 46094611. KAVANAGH, HERVEY, F., A., ROBBINS,W. J. (1949), Antibiotic substances from basidiomycetes. IV. Marasmius conigenus, Proc. Natl. Acad. Sci. USA 35, 343-349. KAVANAGH, HERVEY, A., ROBBINS,W. J. F., (1951), Antibiotic substances from basidiomycetes. VIII. Pleurotus mutilus (Fr.) Sacc. and Pleurotus passeckerianus Pilat, Proc. Natl. Acad. Sci. USA 37, 570-574. KIDA,T., SHIBAI, SETO,H. (1986), Structure of H., new antibiotics, pereniporins A and B, from a basidiomycete, J. Antibiot. 39, 613-615. KIS, Z., GLOSSE, A., SIGG, H. P., HRUBAN, L., SNATZKE,G. (1970), Die Struktur von Panepoxydon und verwandten Pilzmetaboliten, Helv. Chem. Acta 53, 185-186. KLEIN,K., ANKE,T., SHELDRICK, S., BROSS, W. M., STEFFAN,B., STEGLICH, (1990), FulvoW. ferruginin, a carotane antibiotic from Marasmius fulvoferrugineus Gilliam., Z. Naturforsch. 45c, 845-850. KLEINKAUF, VON DOHREN, (1987), BiosynH., H. thesis of peptide antibiotics, Annu. Rev. Microbiol. 41, 259-289. KNAUSEDER, BRANDL, (1976), PleuromutilF., E. ins: Fermentation, structure and biosynthesis, J. Antibiot. 29, 125-131. KOBAYASHI H., MATSUNAGA,K., FUJII, M. (1993), PSK as chemopreventive agent, Cancer Epidemiol. Biomarkers Prev. 2, 271-276. KOMATSU, OKUBO,S., KIKUMOTO, KIMUN., S., RA, K., SAITO,G., SAKAI, (1969), Host meS. diated antitumor action of schizophyllan, a glucan produced by Schizophyllum commune, Gann 60,137-144. KRAICZY, HAASE,U., GENCIC, FLINDT, P., S., S., ANKE, BRANDT, VON JAGOW,G. (1996), T., U.,
530
U., F. The molecular basis for the natural resistance of LAUER, ANKE,T., HANSSKE, (1991), Antithe cytochrome bc, complex from strobilurin biotics from basidiomycetes. XXXVIII. 2-methoxy-5-methyl-1,4-benzochinone, thromboxane a producing basidiomycetes to center Q, inhibiA2 receptor antagonist from Lentinus adherens, tors, Eur. J. Biochem. 235, 54-63. J. Antibiot. 44,59-65. KUBO, I., KIM, M., WOOD, W. F., NAOKI,H. (1986), Clitocine, a new insecticidal nucleoside LEE, S. G., ANKE,T. (1979), Die Wirkung von from the mushroom Clitocybe inversa, TetraheStriatin A, B, und C auf den Glucose-Transport dron Lett. 27,4277-4280. in normalen und Rous-Sarkom-Virus-transforKUHNT, D., ANKE, T., BESL, H., BROSS, M., mierten Huhner-Fibroblasten, Hoppe-Seylers Z. HERRMANN, MOCEK, U., STEFFAN,B., R., Physiol. Chem. 3 0 1170. 6, W. STEGLICH, (1990), Antibiotics from basidio- LEMIEUX, U., THORN, A., BRICE,C., HASR. J. mycetes. XXXVII. New inhibitors of cholesterol R. KINS, H. (1951), Biochemistry of the Ustilagibiosynthesis from cultures of Xerula melanotrinales. 11. Isolation and partial characterization of cha Dorfelt, J. Antibiot. 43, 1413-1420. ustilagic acid, Can. J. Chem. 29, 409-414. KUNIMOTO, HORI,M., UMEZAWA, (1973), LEONHARDT, ANKE,T., HILLEN-MASKE, T., H. K., E., Mechanism of action of diketocoriolin B, BioSTEGLICH, (1987), 6-Methylpurine,6-methylW. chim. Biophys. 298,513-525. 9-P-~-ribofuranosylpurine, and 6-hydroxyF., KUPKA, J., ANKE, T., OBERWINCKLER, methyl-9-P-~-ribofuranosylpurineas antiviral SCHRAMM, STEGLICH, (1979), AntibiotG., W. metabolites of Collybia maculata (Basidiomyics from basidiomycetes. VII. Crinipellin, a new cetes), 2.Naturforsch. 42c, 420-424. antibiotic from the basidiomyceteous fungus LEVITZKI, (1994), Signal-transduction therapy. A. Crinipellis stipitaria (Fr.) Pat., J. Antibiot. 32, A novel approach to disease managment, Eur. J. 130-135. Biochem. 226, 1-13. KUPKA, ANKE,T., LIPMANN, (1980), Com- LOFGREN, LUNING, HEDSTROM, (1954), J., F. N., B., H. parison of the cytotoxic effect of three antibiotThe isolation of nebularin and the determination ics from basidiomycetes in different cell systems, of its structure, Acta Chem. Scand. 8, 670-680. p. 17. Abstract Annu. Meeting ASM. WashingLORENZEN, ANKE,T., ANDERS, HINDERK., U., ton, D.C.: ASM Press. F. MAYR, HANSSKE, (1993), Two inhibitors H., KUPKA, ANKE, GIANETTI, J., T., B.M., STEGLICH, of platelet aggregation from a Panus species W. (1981a), Antibiotics from basidiomycetes. (Basidiomycetes), 2.Naturforsch. 49c, 132-138. XIV. Isolation and biological characterization of K., T., S., hypnophilin, pleurotellol, and pleurotellic acid LORENZEN, ANKE, KONETSCHNY-RAPP, SCHEUER, (1995), 5-Hydroxy-3-vinyl-2(5H)W. from Pleurotus hypnophilus (Berk.) Sacc., Arch. furanone - a new inhibitor of human synovial Microbiol. 130, 223-227. phospholipase A2 and platelet aggregation from KUPKA, ANKE, STEGLICH, ZECHLIN, J., T., W., L. fermentations of a Calyptella species (Basidio(1981b), Antibiotics from basidiomycetes XI. mycetes), Z. Naturforsch. 5Oc, 403409. The biological activity of siccayne, isolated from A. the marine fungus Halocyphina villosa, J. Anti- MANNING, M. (1996), Transcription factors: a new frontier for drug discovery, Drug Discovery biot. 34, 298-304. J., T., K., KUPKA, ANKE, MIZUMOTO, GIANETTI, Today 1,151-160. A. D. B. M., STEGLICH, (1983), Antibiotics from MANNING, M., ANDERSON, C. (1994), TranW. scription factor NF-KB:an emerging regulator of basidiomycetes. XVII. The effect of marasmic inflammation, in: Annu. Reports Med. Chem. 29 acid on nucleic acid metabolism, J. Antibiot. 36, (BRISTOL, A., Ed.), 235-244. J. 155-160. KUSCHEL, ANKE,T., VELTEN, KLOSTER- MATHB, G. (1987), Bestatin compared to other A., R., pharmacologic immunoregulator or modulating D., W. MEYER, STEGLICH, (1994), The mniopetagents, in: Horizons on Antibiotic Research als, new inhibitors of reverse transcriptases from (DAVIS,B. D., ICHIKAWA, MAEDA,K., T., a Mniopetalum species (Basidiomycetes), J. AnMITSCHER, E., Eds.), pp. 44-65. Tokyo: Japan L. tibiot. 47, 733-739. Antibiotics Research Association. 0. D., KWOK, C. H., PLATTNER,R., WEISLEDER, 0. WICKLOW, T. (1992), A nematicidal toxin MAYER,A., ANKE,H., STERNER, (in press), D. Omphalotin, a new cyclic peptide with potent from Pleurotus ostreatus NRRL 3526, J. Chem. nematicidal activity from Omphalotus olearius. Ecol. 18, 127-136. I. Fermentation and biological activity, Nut. LAUER, ANKE,T., SHELDRICK, S., SCHEU., W. Prod. Lett. RER, A., STEGLICH, (1989), Antibiotics from W. X., U., T., 0. basidiomyctes. XXXI. Aleurodiscal: an antifun- MAZUR, BECKER, ANKE, STERNER, (1996), Two new bioactive diterpenes from Legal sesterpenoid from Aleurodiscus mirabilis (Berk. & Curt.) Hohn, J. Antibiot. 42, 875-882. pista sordida, Phytochemistry 43, 405-407.
7 References
531
T. M. MCMORRIS, C., ANCHEL, (1965), Fungal metabolites. The structures of novel sesquiterpenoids Illudin S and M, J. Am. Chem. SOC.87, 1594-1600. MORISAKI, FURUKAWA, NOZOE, S., ITAI, N., J., A., ITAKA,Y. (1980), Synthetic studies on marasmane and isomarasmane derivatives, Chem. Pharm. Bull. 28, 500-507. MULLER, E., LOEFFLER,W. (1982), Mykologie. Stuttgart: Thieme Verlag. MUSILEK,V. (1969), Czech. Patent CS 136492, Chem. Abstr. 1969, 70, 18900~. M. NAIR,M. S. R., ANCHEL, (1977), Frustulosinol, an antibiotic metabolite of Stereum frustulosum: revised structure of frustulosin, Phytochemistry 16, 390-392. NERUD,F., SEDMERA, ZOUCHOVA, MUSIP., Z., LEK,V., VONDRACEK, (1982), Biosynthesis M. of mucidin, an antifungal antibiotic from basidiomycete Oudemansiella mucida. 2H-, I3C- and ''C-labelling study, Coll. Czech. Chem. Commun. 47, 1020-1025. Y., Y., S., NISHIMURA, KOYAMA, UMEZAWA, TAKEUCHI,T., ISHIZUKA,M., UMEZAWA,H. (1977), Chemical modification of coriolin B, J. Antibiot. 30,5945. N., OBUCHI,T., KONDOH,H., WATANABE, TAMAI, M., O M U R A , s., JUN-SHAN,Y., XIAOTIAN, (1990), Armillaric acid, a new antibiotic L. produced by Armillaria mellea, Planta Med. 56, 198-201. D. Po-r-rs, B. C. M., FAULKNER, J. (1992), Phospholipase A2 inhibitors from marine organisms, J. Nut. Prod. 55, 1701-1717. PRICE,M., HEINSTEIN, (1978), Cell-free biosynP. thesis of illudins, Lloydia 41, 574-577. F., QUACK,W., ANKE,T., OBERWINKLER, GIANN E ~ I , M., STEGLICH, (1978), Antibiotics B. W. from basidiomycetes. V. Merulidial, a new antibiotic from the basidiomycete Merulius tremellosus Fr., J. Antibiot. 31, 737-741. RABE,U. (1989), Fermentation von Cyathus strialus (Huds. ex Pers.) Willd (Basidiomycetes) und Biosynthese der Striatine. Dissertation, University of Kaiserslautern, FRG. RIEDL, K. (1976), Studies on pleuromutilin and some of its derivatives, J. Antibiot. 29, 132-139. ROBERTS,D. W. (1981), Toxins of entomopathogenic fungi. In: Microbial Control of Pests and Plant Diseases 1970-1980 (BURGES, D., Ed.), H. pp. 441-464. London, New York: Academic Press. SALMON, A., GARLAND, G. (1991), LeukoJ. L. trienes, antagonists and inhibitors of leukotriene biosynthesis as potential therapeutic agents, in: Progress in Drug Research (JUCKER,E., Ed.), pp. 9-90. Basel: Birkhauser.
SAMUELSSON, DAHLEN,S.-E., LINDGREN, B., J. A., ROUZER,C., SERHAN, N. (1987), LeukoC. trienes and lipoxins: structure, biosynthesis, and biological effects, Science 237, 1171-1176. SASAKI, TAKASUKA, (1976), Further study T., N. of the structure of lentinan, an antitumor polysaccharide from Lentinus edodes, Carbohydr. Res. 47, 99-104. SAUTER, H., AMMERMANN, BENOIT, R., E., BRAND, S., GOLD, R. E., GRAMMENOS, W., KGHLE, H., LORENZ,G., MULLER,B., ROHL, F., SCHIRMER, SPEAKMAN, B., WENDEU., J. ROTH, B., WINGERT, (1995), Mitochondria1 H. respiration as a target for antifungals: lessons from research on strobilurins, in: Antifungal Agents; Discovery and Mode of Action (DIXON, G. K., COPPING,L. G., HOLLOMON, W., D. Eds.), pp. 173-191. Oxford: BIOS Scientific Publishers. R. SAYRE, M., WALTER,D. E. (1991), Factors affecting the efficacy of natural enemies of nematodes, Annu. Rev. Phytopathol. 29, 149-166. SCHNEIDER, MOSER,A. (1987), Process kinetic H., analysis of pleuromutilin fermentation, Bioproc. Eng. 2, 129-135. SCHRAMM, (1980), Neue Antibiotika aus HOG. heren Pilzen (Basidiomyceten). PhD Thesis, Department of Chemistry, University of Bonn, FRG. SCHRAMM, STEGLICH,W., ANKE,T., OBERG., WINKLER, (1978), Antibiotika aus BasidiomyF. ceten, 111. Strobilurin A und B, antifungische Stoffwechselprodukte aus Strobilurus tenacellus, Chem. Ber. 1 1 2779-2784. 1, SCHRAMM, STEGLICH, G. W., ANKE, T. (1982), Structure-activity relationship of strobilurins, oudemansin, and synthetic analogues, Abstract 457, in: Abstracts 13th Int. Congr. Microbiol., August 8-13, Boston. SCHULTES, E., HOFMANN, (1980), Pflanzen R. A. der Cotter. Bern, Stuttgart: Hallwag. SCHWITZGEBEL, (1992), Untersuchungen an K. holzabbauenden und holzverfarbenden Pilzen, M . Sc. Thesis, Department of Biology, University of Kaiserslautern, FRG. SEDMERA,P., MUSILEK,V., NERUD,F. (1981), Mucidin: Its identity with strobilurin A, J. Anribiot. 34, 1069. J., SEGIGUCHI, GAUCHER,M. (1979), Isoepoxydon, a new metabolite of the patulin pathway in Penicillium urticae, Biochem. J. 182, 445-453. SHINOZAWA, TSUTSUI, ODA, T. (1979), EnS., K., hancement of the antitumor effect of illudin S by including it into liposomes, Experientia 35, 1102-1103. SIMON, B. (1994), Antivirale und cytotoxische Wirkstoffe aus Basidiomyceten. Ph.D. Thesis, University of Kaiserslautern, FRG.
532
SIMON, ANKE, ANDERS, NEUHAUS, B., T., U., M., HANSSKE, (1995), Collybial, a new antibiotic F. sesquiterpenoid from Collybia confluens (Basidiomycetes), 2. Naturforsch. 5Oc, 173-180. STADLER, ANKE,T., DASENBROCK, STEGM., J., LICH, (1993a), Phellodonic acid, a new bioW. logically active hirsutane derivative from Phellodon melaleucus (Thelephoraceae, Basidiomycetes), Z. Naturforsch. 48c,545-549. M., STADLER, MAYER,A., ANKE,H., STERNER, 0. (1993b), Fatty acids and other compounds with nematicidal activity from cultures of basidiomyocetes, Planta Med. 60, 128-132. STADLER, ANKE STERNER, (1994), New M., H., 0. nematicidal and antimicrobial compounds from the basidiomycete Cheimonophyllum candidissimum, J. Antibiot. 47, 1284-1289. O., STADLER, FOURON, M., J.-F., STERNER, ANKE, H. (1995), 1,2-dihydroxyminthlactone, a new nematicidal monoterpene isolated from the basidiomycete Cheimonophyllum candidissimum (Berk & Curt.) Sing., Z. Naturforsch. 5Oc, 473-475. STARK, ANKE,T., MOECK, STEGLICH, A., U., W., KIRFEL,A., WILL,G. (1988), Lentinellic acid, a biologically active protoilludane derivative from Lentinellus species (Basidiomycetes), 2. Naturforsch. 43c, 177-183. STARK, ANKE,T., MOCEK,U., STEGLICH, A., W. (1991), Omphalone, an antibiotically active benzoquinone derivative from fermentations of Lentinellus omphalodes, Z. Naturforsch. 46c, 989-992. STERNER, BERGMAN, KIHLBERG, WICKO., R., J., BERG, B. (1985), The sesquiterpens of Laccarius vellereus and their role in a proposed chemical defense system, J. Nut. Prod. 48, 279-288. O., CATRE,R., NILSON,L. (1987a). STERNER, Structure-activity relationship for unsaturated dialdehydes. 1. The mutagenic activity of 18 compounds in the Salmonellalmicrosome assay, Mutat. Rex 188, 169-174. STERNER, STEFFAN, STEGLICH, (1987b), O., B., W. Novel azepine derivatives from the pungent mushroom Chalciporus piperatus, Tetrahedron. 43,1075-1082. STERNER, ETZEL,W., MAYER, ANKE,H. O., A., (in press), Omphalotin, a new cyclic peptide with potent nematicidal activity from Omphalotus olearius. 11. Isolation and structure determination, Nut. Prod. Lett. STEWART, R. (1986), A method for generating K. protoplasts from Clitopilus pinsitus, J. Antibiot. 39, 1486-1487. SUGA, SHIO,T., MAEDA,Y. Y., CHIHARA, T., G. (1984), Antitumor activity of lentinan in murine syngeneic and autochthonous hosts and its sup-
pressive effect on 3-methylcholantrene-induced carcinogenesis, Cancer Res. 44,5132-5137. K., T., S., TAKAHASHI, SOMEYA, MURAKI, YOSHIDA, T. (1980), A new keto alkohol, (-)-minthlactone, ( + )-isorninthlactone and minor components in peppermint oil, Agaric. Biol. Chem. 44, 1535-1543. TAKAHASHI, KUSANO, OHTA,T., OHIZUA., G., MI,Y., NOZOE,S. (1989). Fasciculic acids A, B and C as calmodulin antagonists from the mushroom Naematoloma fasciculare, Chem. Pharm. Bull. 37,3247-3250. H., J., TAKEUCHI, IINUMA, IWANAGA, TAKAT., HASHI, S., TAKITA, UMEZAWA, (1969), T., H. Coriolin, a new basidiomycete antibiotic, J. Antibiot. 22, 215-217. THOMPSON, B. (1995), Apoptosis in the pathoC. genesis and treatment of disease, Science 267, 1456-1462. TOMODA, HUANG, H., COA, J., NISHIDA, H., X. H., NAGAO, R., OKUDA, TANAKA, OMUS., H., H., K. RA,S., ARAI, INOUE, (1992), Inhibition of acyl-CoA :Cholesterol acyltransferase activity by cyclodepsipeptide antibiotics, J. Antibiot. 45, 1626-1632. S., Y., TSUGAGOSHI, HASHIMOTO, FUJI, G., KoBAYASHI, H., NOMOTO, ORITA,K. (1984), K., 1 Krestin (PSK), Cancer Treat. Rev. 1 ,131-155. TURNER, B. (1971), Fungal Metabolites. LonW. don: Academic Press. TURNER, B., ALDRIDGE, C. (1983), Fungal W. D. Metabolites 11. London: Academic Press. H., UMEZAWA, AOYAGI, T., SUDA,H., HAMADA, M., TAKEUCHI, (1976), Bestatin, an inhibitor T. of aminopeptidase B, produced by actinomycetes, J. Antibiot. 29, 97-99. VIDARI, DE BERNARDI, VITA-FINZI, G., M., P., FRONZA, (1976), Sesquiterpenes from LacG. tarius blennius, Phytochemistry 15, 1953-1955. VINING, L. C. (1990), Function of secondary metabolites, Annu. Rev. Microbiol. 44, 395-427. VONJAGOW, GRIBBLE, W., TRUMPOWER, G., G. B. L. (1986), Mucidin and strobilurin A are identical and inhibit electron transfer in the cytochrome bcl complex of the mitochondria1 respiratory chain at the same site as myxothiazol, Biochemistry 25, 775-780. WEBER, ANKE, STEFFAN, STEGLICH, W., T., B., W. (1990a), Antibiotics from basidiomycetes. XXXII. Strobilurin E: a new cytostatic and antifungal (E)-pmethoxyacrylate antibiotic from Crepidotus fulvotomentosus Peck, J. Antibiot. 43, 207-212. WEBER, ANKE, BROSS, STEGLICH, W., T., M., W. (1990b), Strobilurin D and Strobilurin F: two new cytostatic and antifungal (E)-pmethoxyacrylate antibiotics from Cyphellopsis anomala, Planta Med. 56, 446-450.
7 References
533
WEBER,W., SEMAR,M., ANKE, T., BROSS, M., STEGLICH, (1992), Tyromycin A: a novel inW. hibitor of leucine and cysteine aminopeptidase from Tyromycetes lacteus, Planta. Med. 58, 5659. WEBER, W., SCHU, P., ANKE, T., VELTEN, R., STEGLICH, (1994), Caloporosid, a new inhiW. bitor of phospholipases C from Caloporus dichrous (Fr.) Ryv., J. Antibiot. 47, 1188-1194. WINKELMANN, (1986), Iron complex products, G. in: Biotechnology 1st Edn., Vol. 4 (REHM,H. J., REED, G., Eds.), pp. 216243. Weinheim: VCH. D., WUNDER, A., ANKE, T., KLOSTERMEYER, STEGLICH, (1996), Lactarane type sesquiterW. penoids as inhibitors of leukotriene biosynthesis and other new metabolites from submerged cultures of Lentinellus cochleatus (pers. ex Fr.) Karst., Z. Naturforsch. Slc, 493-499. YANG,J. S., Su, Y. L., WANG,Y. L., FENG,X. Z., X. Yu, D. Q., LIANG, T. (1991), Two novel protoilludane norsesquiterpenoid esters, armillasin
and armillatin, from Armillaria mellea, Planta Med. 57,478-480. ZAHNER, (1982), Mikrobieller SekundarstoffH. wechsel, in: Handbuch der Biotechnologie (PRAvE, P., FAUST, SITTIG,W., SUKATSCH, U., D. A., Eds.). Wiesbaden: Akademsiche Verlagsgesellschaft. ZAPF, S. (1994), Neue antifungische Antibiotika aus Pilzen. Ph.D. Thesis, University of Kaiserslautern, FRG. ZAPF, S., WERLE,A., ANKE,T., KLOSTERMEYER, D., STEFFAN, STEGLICH, (1995), 9-MethB., W. oxystrobilurine - Bindeglieder zwischen Strobilurinen und Oudemansinen, Angew. Chem. 107, 255-257. ZAPF, S., WUNDER, A., ANKE, T., KLOSTERD., W., R., MEYER, STEGLICH, SHAN, STERNER, O., SCHEUER,W. (1996), (+)-lOa-hydroxy-4muurolen-3-one, a new inhibitor of leukotriene biosynthesis from a Favolaschia species. Comparison with other sesquiterpenes, Z. Naturforsch. 51c, 487-492.
Biotechnology
12 cyc10sp0rins: Recent Developments in Biosynthesis, Pharmacology and Biology, and Clinical Applications
JORG KALLEN VINCENT MIKOL VALERIEF. J. QUESNIAUX MALCOLM WALKINSHAW D.

Basel. Switzerland
ELISABETH SCHNEIDER-SCHERZER KURT SCH~RGENDORFER GERHARD WEBER

Kufstein-Schaftenau, Austria
HANS FLIRI G.
Vitry-sur-Seine, France
536
12 Cyclosporins: Recent Developments in Biosynthesis, Pharmacology and Biology
1 Introduction 538 2 Clinical Applications of Cyclosporins 539 2.1 Introduction 539 2.2 Transplantation 539 2.3 Autoimmune Diseases 540 2.4 Activity Against Tumor Multidrug Resistance 540 2.5 Anti-HIV Activity 540 3 Mode of Immunosuppressive Action 542 3.1 Introduction to T Cell Activation 542 3.1.1 Signal Recognition 542 3.1.1.1 First Signal 542 3.1.1.2 Costimulatory Signals 542 3.1.2 Signal Transduction 544 3.1.3 Gene Activation 545 3.1.4 Cytokine Receptor Expression 545 3.1.5 Release of Cytokines 545 3.2 Current Knowledge of the Mode of Immunosuppressive Action of CsA 546 3.2.1 Cyclosporin Receptors 547 3.2.2 Active Sites on the Drugs 548 3.2.3 Target Proteins of the Drug-Immunophilin Complexes 549 3.2.4 Downstream Effects in the Nucleus 549 3.3 Molecular Evidence for other Biological Activities of Cyclosporins 550 3.3.1 Antiinflammatory Effects 550 3.3.2 Cyclosporins as Anti-HIV Agents 550 3.3.3 Cyclosporins as Drug Resistance Modifiers 551 3.3.4 Antimalarial Activity of Cyclosporins 551 4 Chemistry 551 4.1 Structural Aspects 551 4.2 Chemical Synthesis and Production 552 4.2.1 MeBmt Transformations and Variations 555 4.2.2 Chemical Transformations of the Cyclic Peptide Backbone 555 4.2.3 Selective Ring Opening Reactions 556 4.2.4 Cyclosporins Incorporating other Non-Proteinogenic Amino Acids 556 4.3 Structure-Activity Relationships 556 5 Structural Investigations of Cyclosporins, Cyclophilins, and their Complexes 559 5.1 Introduction 559 5.2 Structural Investigations of Uncomplexed Cyclosporins 562 5.2.1 The 3D-Structure of Uncomplexed CsA in Apolar Environment 562 5.2.2 The 3D-Structure of the Uncomplexed Peptolide SDZ 214-103 in Apolar Environment 562 5.2.3 The 3D-Structure of Uncomplexed CsA in Polar Environment 563 5.3 The 3D-Structure of Cyclophilin A Complexed with a Tetrapeptide 563 5.4 The Peptidyl-Prolyl Isomerase Active Site of Cyclophilin A 564 5.5 The 3D-Structure of Cyclophilin A Complexed with CsA 566 5.6 The X-Ray Structures of Cyclophilin A Complexed with Cyclosporin Derivatives 568 5.6.1 [MeBm2t]-Cyclosporin 568 5.6.2 MeIle4-Cs (SDZ NIM 811) 569 5.6.3 The 3D-Structure of SDZ 214-103 Bound to Cyclophilin A 569 5.7 The X-Ray Structure of Cyclophilin B Complexed with a Cyclosporin Derivative 570 5.8 The X-Ray Structure of Cyclophilin C Complexed with CsA 571 5.9 Conclusions 571
Contents
537
6 Biosynthesis of Cyclosporins 572 6.1 Introduction 572 6.2 Cyclosporin Synthetase 572 6.2.1 Enzymatic Activities of Cyclosporin Synthetase 572 6.2.2 Characterization of the Enzyme 573 6.2.3 Isolation and Characterization of the Cyclosporin Synthetase Gene 574 6.2.4 Manipulation of the Cloned Cyclosporin Synthetase Gene by Integrative Transformation 575 6.2.5 Mechanistic Aspects of Biosynthesis 577 6.2.6 Biosynthesis of Cyclosporin Variants 580 6.2.7 Related Enzymes: Peptolide SDZ 214-103 Synthetase 580 6.3 Alanine Racemase 580 6.3.1 Alanine Racemizing Activity 580 6.3.2 Characterization of the Enzyme 581 6.4 Bmt-Synthesizing Enzymes 581 6.4.1 Polyketide Origin of Bmt 581 6.4.2 Identification of the Basic Assembly Product and Characterization of Bmt-Polyketide Synthase 581 6.4.3 The Transformation Process 583 7 References 583
538
1 Introduction
Cyclosporins are a family of hydrophobic cyclic undecapeptides with a remarkable spectrum of diverse biological activities. The first member of this class to be discovered was named cyclosporin A (CsA; structure shown in Fig. 1).To this date, some 30 members of this family have been isolated from
natural sources. In addition, a close analog named SDZ 214-103, incorporating a lactone function in place of a peptide bond and exhibiting a similar biological profile was also discovered from natural sources (Fig. 2). The discovery and structure elucidation of cyclosporins have been amply reviewed (WENGER 1986; WENGER al., 1986; VON WARTBURG et and TRABER,1988 FLIRI and WENGER, 1990).
c
n c I H
0
F g 1 Structure of cyclosporin A i. .
(Sandimmun"), including the numbering system.
F g 2 Structure o SDZ 214-103. i. . f
2 Clinical Applications of Cyclosporins
539
In this chapter, emphasis is given to some more recent aspects of chemistry, biosynthesis, and biological activity.

2.1 Introduction
Cyclosporin A was initially isolated as an antifungal antibiotic. It was later shown to possess immunosuppressive properties of high therapeutic value. Since 1983, cyclosporin A, under the trade name Sandimmunm, has been in clinical use worldwide to prevent rejection of organ transplants. It has subsequently been approved for the therapy of certain autoimmune diseases. Since the time of market introduction of SandimmunB, many additional biological activities of cyclosporins have been discovered, some of which may lead to novel clinical applications of cyclosporin A or of non-immunosuppressive analogs. At the time of market introduction of Sandimmunm, the mechanism by which this drug mediates immunosuppression was not understood at the molecular level nor was a receptor known. Since then, not only was a whole family of receptors discovered (i.e., the cyclophilins), but a possible role of these proteins for protein folding and cellular protein traffic has emerged. Much of what is known today about cyclosporins, cyclophilins and their biochemistry was greatly aided by the discovery of FK506, an immunosuppressive macrolide. This compound elicited much interest because, like cyclosporin A, it was a T cell selective immunosuppressant, but much more potent. This activity was soon shown to be based on a mechanism identical to that of cyclosporin. Search for FK506 receptors led to the discovery of the FK506 binding proteins (FKBPs), a novel protein family with no homologies to cyclophilins, yet with many properties in common. Like the cyclophilins, FKBPs appear to have a functional role in protein folding. Unveiling the mechanism of immunosuppressive activity of Sandimmunm
and FK506 was greatly facilitated by the availability of a third compound, rapamycin, which binds to the same receptors as FK506, yet exhibits a different spectrum of biological activities. A detailed account of these aspects is given in Sects. 3.2. and 3.3. In this section present clinical applications of Sandimmunm are discussed. They include the following indications: allograft rejection, Behsets uveitis, rheumatoid arthritis, aplastic anemia (NDAs pending), nephrotic syndrome, atopic dermatitis (NDAs pending), psoriasis vulgaris.
L.2 Transplantation
Sandimmunm is a reversible inhibitor of the transcription of interleukin 2 (IL-2) and several other lymphokines, most notably in helper T lymphocytes (see Sect. 3.2). As a consequence, it suppresses the activation and/or maturation of various cell types, in particular those involved in cell-mediated immunity. Because of these properties, Sandimmunm has become the first-line immunosuppressant for prophylaxis and therapy of transplant rejection. In fact, the modern era of transplantation surgery was only possible after the availability of cyclosporin. The first patient to receive a kidney graft under CsA treatment was reported in 1978 (CALNE al., 1978). Soon et thereafter, transplantations of liver, heart, and combined lung-heart commenced (ERNST,1991; BARRY,1992; KAHAN,1992; TSANG al., 1992). In 1991, only in Germany et 450 liver transplantations were performed (HOPFet al., 1992). Organ availability has become a major limiting factor and numerous patients die while awaiting a donor organ. Therefore, organ preservation techniques have become an important aspect in the area of transplantation surgery. Histocompatibility matching, besides immunosuppression, is the key factor contributing to long-term graft survival. Currently, expected 10-year first graft survival rates for kidneys from HLA-identical siblings, l-haplotype-matched relative, and cadaver donors are 74, 51, and 40%, respec-
540
tively (BARRY, 1992). The survival probability after lung transplantation is approximately 65% after 1 year and 50% after 3 years (ERNST,1991). Major problems encountered in transplantation surgery are technical difficulties during operation, serious infections, and acute rejection episodes during the first postoperative period, and chronic (long-term) rejection. Side effects can be classified into those associated to immunosuppression (lymphoproliferative disorders, infectious diseases caused by bacterial and fungal pathogens as well as viruses), and other adverse effects which are specific for the immunosuppressive drugs used. For Sandimmun" these include primarily impairment of renal function, hypertension, hirsutism, and gingival hyperplasia (MASON, 1989). Neurological and gastrointestinal effects are also common in Sandimmun" recipients but are usually mild to moderate and resolve on dosage reduction.
tage in a number of indications is the steroidsparing effect of Sandimmun". To avoid relapse after control of active disease, patients should continue receiving SandimmunB maintenance therapy at the lowest effective dose.
2.4 Activity Against Tumor Multidrug Resistance

Cellular resistance to cytotoxic drugs is often the cause of inefficient treatment of cancer with potent antitumor drugs. While many mechanisms of resistance occur, the mechanism of "multidrug resistance" (MDR) has received particular attention (ENDICOTT and LING, 1989). Most often, this type of resistance extends to several anticancer drugs of unrelated structural classes and mechanisms of action. A common feature of MDR is overexpression of a particular class of transmembrane glycoproteins called P-glycoproteins (Pgp) which serve as transport proteins rapidly effluxing antitumor drugs out of the tumor cells as soon as they have entered through the membrane. As a consequence, Pgp transporters decrease intracellular drug concentrations below their active threshold. Numerous in vitro studies have described agents which can restore the sensitivity of MDR tumor cells, including cyclosporin A at clinically achievable concentrations (TWENTYMAN, 1992). Moreover, this effect can be dissociated from immunosuppression, as nonimmunosuppressive analogs have been shown to retain resistance modifier activity and some are even more potent than cyclosporin A. One such analog from Sandoz Pharma AG called SDZ PSC 833 is approximately tenfold more potent than Sandimmun" as a resistance modifier and is currently undergoing clinical trials (BOESCH al., 1991). The strucet ture of SDZ PSC 833 is shown in Fig. 3.
2.3 Autoimmune Diseases

Since SandimmunB not only suppresses cell-mediated immunity but also humoral immune responses and inhibits chronic inflammatory reactions it appeared very promising in the treatment of autoimmune diseases. Prospective controlled trials performed in patients with autoimmune diseases have recentet ly been reviewed (FREY,1990; FAULDS al., 1993). Efficacy could be proven for the following diseases: Endogenous uveitis, rheumatoid arthritis, Sjogren's syndrome, myasthenia gravis, psoriasis, atopic dermatitis, Crohn's disease. The drug is considered as a first-line therapy in patients with moderate or severe aplastic anemia who are not eligible for bone marrow transplantation. It may also be of benefit in patients with primary biliary cirrhosis and intractable pyoderma gangrenosum. Sandimmun" does not appear to be effective in patients with allergic contact dermatitis, multiple sclerosis, or amyotropic lateral sclerosis. Successful application in insulin-dependent diabetes will depend on the development of diagnostic tools indicating early disease onset before beta cell destruction has progressed too far and clinically overt diabetes is present. The most significant advan-
2.5 Anti-HIV Activity

A possible beneficial effect of Sandimmun" in HIV disease has been proposed as early as 1986 (ANDRIEU al., 1986). The rationale is et that activation of CD4+ cells which is required for HIV replication (ZACKet al., 1990
541
Fig. 3. Structure of SDZ PSC 833.
I
clinical research laboratories at the Sandoz Research Institute of Sandoz Pharma AG in Vienna, Austria. Compounds were evaluated for antiviral, cytotoxic, and for immunosuppressive activity i v i m . It was found that n some non-immunosuppressive analogs of SandimmunB were equal or even superior in their antiviral activity without being cytotoxic. One such analog, SDZ NIM 811 (Fig. 4), is comparable to Sandimmunm re-
STEVENSON al., 1990) is inhibited by CsA. et In addition, CsA would inhibit the initiation of an autoimmune process involving killing of HIV infected lymphocytes by cytotoxic cells and may also counteract HIV-induced apoptotic cell death of CD4+ cells (HABESHAW et al., 1990). A thorough investigation of a series of immunosuppressive and non-immunosuppressive cyclosporins was performed in the pre-
Fig. 4. Structure of SDZ NIM 811.
542
garding oral bioavailability and pharmacokinetics in animals and appears to be of lower et nephrotoxicity (ROSENWIRTH al., 1994).
3 Mode of Immunosuppressive Action

In addition to its clinical use as an immunosuppressant CsA has also been widely employed as an experimental tool for basic research. It has helped to understand the biochemical events needed to translate a signal from the T cell surface to the nucleus and the pathophysiological processes involving lymphocyte activation in a variety of diseases. Early immunological studies revealed that CsA exerts specific effects on T cell lymphoet kine transcription (KRONKE al., 1984). Because T cells are prominent in the cellular imf mune response, studies on the mechanism o immunosuppression by CsA have mainly focused on its role in regulating gene expression in T lymphocytes. To place CsA activity in perspective a summary on T cell activation is first given below.
and cytotoxic suppressor functions. The multiple steps involved in this process, namely signal recognition, signal transduction to the nucleus, resulting in gene activation, expression of growth factor receptors, growth factor synthesis and cell proliferation, are briefly reviewed (RYFFEL,1989).
3.1.1 Signal Recognition 3.1.1.1 First Signal

After uptake and limited proteolysis, the antigen processed by the antigen presenting cell is recognized in the context of the major histocompatibility complex (MHC) by the antigen receptor on T lymphocytes (Fig. 5). The binding of antigen to the T cell receptor is an absolute requirement for T cell activation under physiological conditions. The T cell receptor is a multicomponent structure consisting of the clonotypic (Y and p (or y and 8) chains and the invariant CD3 subunits y, S,E, and 7. The complete assembly of all components is required for cell surface expression, and thus for antigen receptor function. The 5 and 7 subunits of CD3 most likely transduce to the cytoplasm the activating signals originating from antigen recognition by the T cell receptor. Antibodies against the CD3 complex can induce T cell functional responses that are identical to antigen-induced responses, regardless of antigen specificity. In addition, the two transmembrane proteins CD4 and CD8 expressed on helper and cytotoxic T cells participate in the interaction between the T cell and the antigen presenting cell by binding to MHC class I1 and I molecules, respectively. Originally, they were called coreceptors because their association with an intracellular enzyme facilitates signalet ing during T cell activation (JANEWAY al., 1989).
3.1 Introduction to T Cell Activation

A schematic representation of the cellular immune response with emphasis on the central role of the activated T lymphocytes is shown in Fig. 5. For T cell activation, the antigen receptor on the T cell surface interacts with the processed antigen exposed in the proper histocompatibility context on the surface of the antigen presenting cell ( A P C cf. Fig. 5). In the presence of additional accessory interactions between the T cell and the antigen presenting cell, antigen recognition leads to biochemical events which finally result in proliferation, differentiation, and maturation of the T cell to T effector cells with specific immunological function, e.g., helper
3.1.1.2 Costimulatory Signals

In addition to the molecular interactions between the T cell receptor, CD3, CD4, or CD8 and the antigen presenting cell, costimu-
543
W W Modulation of calcineurin phosphatase activity (other kinases/phosp.hatases ?)
p70S6kinase
Fig. 5. CsA and FK506 both interfere, by binding to their respective immunophilins, with the function of intracellular molecules that transmit calcium-associated signals between the T-cell receptor (TCR) and the activation of lymphokine genes (IL-2) in the nucleus. Transcriptional regulation of IL-2 gene expression is modulated by the combination of transcription factors (e.g., NF-AT, NFKB, OTF-1) interacting with their corresponding recognition sites at the IL-2 promoter. These DNA/protein complexes, together with RNA polymerase I1 (RNA pol 11), result in the antigen-inducible transcription of IL-2. Potential intervention sites for the pentameric complex (calcineurin A (p61), B (p19), calmodulin (p17), immunophilin, drug), involving, e.g., modification and translocation of antigen-inducible transcription factors (NF-AT NFKB (p50, p65)), are indicated (11). CsA and FK506 interfere with the Go to G1 transition of the cell cycle, whereas raparnycin interferes with the G1 to S transition (for details, see text) (adapted from BAUMANN and BOREL,1992).
RNA pol II
544
latory signals contribute to the T cell receptor-driven proliferative response. These costimulatory signals are thought to be delivered by antigen presenting cells but not by tissue cells and might play an important role in the selfhon-self discrimination by influencing the consequences of the T cell receptor engagement (LIU and LINSLEY, 1992). Conversely, i v i m stimulation of T cells in the absence n of costimulation might lead to functional inactivation of the T cells, i.e., clonal anergy, or to activation-induced cell death (apoptosis). The molecular basis for T cell costimulation is not yet fully understood. The accessory molecules are membrane proteins including the lymphoid adhesion molecules. They are invariant and bind to ligands expressed on the surface of other cells such as antigen presenting cells or target cells, thereby increasing the strength of adhesion between these cells and the T cells. In addition, accessory molecules may transduce biological signals to the T cell cytosol. CD2, e.g., which interacts with the leukocyte function-associated antigen3 (LFA-3, CD58) augments specific signaling through the T cell receptor. In contrast, CD28 binding to the B7-1 and B7-2 molecules on antigen presenting cells initiate a signaling pathway which is distinct from the pathway emanating from the T cell receptor.
3.1.2 Signal Transduction

The antigen binding to the T cell receptor/ CD3 complex in combination with costimulatory signals triggers a highly complicated signal transduction pathway which finally results in the pleiotropic T cell activation program. Although the molecular details of this program are still largely unknown, an attempt to summarize the current state of our knowledge with a focus on the mechanism of action of known immunosuppressants is schematically represented in Fig. 5 (IZQUIERDO CANand TRELL, 1992; ABRAHAM al., 1992). et One of the earliest events after antigen binding is the activation of protein tyrosine kinases which initiates a cascade of downstream biochemical events (SEFTON and CAMPBELL, 1991). Since none of the subunits
of the T cell receptorKD3 complex reveals intrinsic catalytic domains, the protein tyrosine phosphorylation of CD3 is likely to be mediated by intracellular protein tyrosine kinases. Two members of the src family of protein tyrosine kinases have been identified in this context: ~59""which is physically associated with the T cell receptorKD3 complex, and ~ 5 6 ' 'which is associated with CD4 or ~ CD8 and becomes activated upon coligation of CD4 or CD8 with the T cell receptor. How this signal is being transmitted from the T cell receptor-proximal protein tyrosine kinase activities to the cytoplasmic protein serinel threonine kinase and phosphatase cascades is less clear. In addition, the phosphorylation of tyrosine residues in the cytoplasmic domains of receptors such as CD3 6 is required for the recruitment of several cellular enzymes like phospholipase C (PLC) to the cytoplasmic domain of the receptors. The yl and 2 isoforms of the phosphatidyl inositol-specific PLC depend on tyrosine phosphorylation for their activation. Once activated, PLC increases the hydrolysis of inositol phospholipids and produces two second messenger molecules. One, inositol triphosphate (IP3) binds to intracellular vesicles that store calcium ions (Ca'+) causing them to release calcium into the cytoplasm. Ca2+ ions then bind to a small protein, calmodulin, which acts as a regulatory subunit for other enzymes essential for T cell activation such as the serine/threonine phosphatase calcineurin. Another second messenger molecule is diacylglycerol (DAG), a lipid molecule that remains in the membrane where it activates protein kinase C. Once activated, protein kinase C is translocated from the cytosol to the plasma membrane where it phosphorylates membranebound proteins. This event subsequently directs the modification of a set of other signal transmitting proteins. Finally, phosphorylation and dephosphorylation of certain transcription factors such as NFKB (nuclear factor for K-light chain expression in B cells) and NF-AT (nuclear factor of activated T cells) induce or repress the transcription of their target genes such as the IL-2 gene. The effect of the modification is either direct by increasing or decreasing the affinity of the transcription factors for their specific binding site on
545
the DNA or indirect via additional protein- DNA synthesis and cell division. Late genes protein interactions (HUNTER and KARIN, require both DNA synthesis and cell division 1992). for their expression. They are synthesized during the G2 phase of the cell cycle. The manyfold mechanisms used by T cells 3.1.3 Gene Activation to regulate their immediate, early, and late activation genes include alteration of the The metabolic events resulting from expo- transcriptional rate, of initiation and terminasure of T cells to antigen culminate in blast tion of transcription, and of mRNA stability. transformation and progression through the It is the group of early genes that is most sencell cycle. Activated T cells express novel sitive to immunosuppressants like CsA and functions such as lymphokine secretion and FK506. The expression of this group of genes lytic capability which are related to their ef- is mainly regulated at the level of transcripfector role in the immune response. They intion initiation. Transcription is initiated once crease in volume and undergo rapid increases the transcription factors such as NFKB, jun, in phosphatidyl inositol metabolism, in cyto- fos, or oct-1, -2 that exist as precursors in the plasmic pH, intracellular free calcium concen- cytoplasm have been translocated into the nutration, and in the serinehhreonine and tyro- cleus and are activated by phosphorylation or sine phosphorylation pattern of various cellu- dephosphorylation. This enables them to bind lar proteins. Progression through the cell cy- to their specific regulatory elements and parcle is associated with a general increase in ticipate in a functional transcription complex protein, lipid and RNA synthesis and with an with RNA polymerase I1 (HUNTER and KAincrease of the mRNA level for a variety of RIN, 1992). activation-related genes. Many of these genes encode lymphokines and surface receptors necessary for the expansion and the immune 3.1.4 Cytokine Receptor functions of the activated T cells. About 70- Expression 100 genes are activated in T cells during the differentiation program which is taking place T cells express a series of cytokine recepat times ranging from 15 min to 14 d following tors on their membrane which allow them to stimulation (CRABTREE, 1989). respond to various cytokines including IL-1, By analogy to viral systems the T cell acti- IL-2, IL-4, IL-7, IL-9, IL-10, and IL-12. Some vated genes can be divided into three groups: of these receptors seem to be upregulated the immediate, the early, and the late genes. during T cell activation. The receptor for IL-2, Immediate genes (e.g., c-fos,c-myc) are tran- a major growth factor for T cells, is composed scribed after activation with no need for proof three subunits (a,p, and y) in its hightein synthesis. Their products appear very affinity form (TAKESHITA al., 1992). The p et soon after stimulation, usually within 10- and y chains associate to form receptors of in30 min. Transcription of the immediate genes termediate activity which are expressed on takes place during the transition from the resting T cells. Upon T cell activation by anquiescent state (Go) into the G1 phase of the tigen or by IL-2 itself, the expression of the a cell cycle. Immediate genes are usually iden- chain is upregulated and contritified by their ability to be transcribed in the butes to form high-affinity receptors. The IL-4 presence of a protein synthesis inhibitor such receptor 130 kDa protein is also upregulated as cycloheximide or anisomycin. Early genes, during T cell activation by mitogenic ligands which include the lymphokines (e.g., IL-2, IL-3, or by IL-4 (ARMITAGE al., 1990). et IL-4, interferon- y, GM-CSF, and IL-2 receptor a) are transcribed after the immediate genes and require postactivation protein syn- 3.1.5 Release of Cytokines thesis. The early genes are involved in the mid to late G1 phase of the cell cycle and inSeveral cytokines are produced by T cells clude genes coding for products required for upon activation, namely IFN-y, IL-2, IL-3,
546
IL-4, IL-5, IL-6, IL-9, IL-10, TNFa, and GMCSF. On the other hand, antigen presenting cells release IL-1, IL-6, IL-12, IL-13, TNFa and p and other cytokines. These cytokines have multiple regulatory effects such as the autocrine stimulation of helper T cells and the differentiation of cytotoxic and suppressor T lymphocytes, natural killer cells, and B lymphocytes. A prominent cytokine for further T cell activation is IL-2 since T cell activation can be induced by IL-2 itself, even in the absence of antigen, in vitro.
3.2 Current Knowledge of the Mode of Immunosuppressive Action of CsA

In the following section, our current knowledge as to how CsA interferes with the immune response at the molecular level is reviewed with some reference to two other immunosuppressive drugs of microbial origin, FK506 and rapamycin. Like CsA, FK506 and rapamycin were initially discovered and char-
acterized as antifungal antibiotics. The structures of FK506 and rapamycin, together with that of cyclosporin A, are shown in Fig. 6. FK506 is a neutral macrolactone produced by Streptomyces tsukubaensis, a soil microorganism collected in the Tsukuba area of northern Japan (KINO et al., 1987). FK506 prolongs the survival of skin and organ allografts in experimental animal models and is active at approximately one tenth of the CsA dose required for the same effects. The initial clinical trials with FK506 showed a remarkable effect in liver transplantation and in rescuing drug-resistant rejection in organ transplantation. FK506 has recently been approved in the US for the indication of transplantation (trade name PrograF). Like CsA, FK506 interferes with the process of T cell activation by specifically inhibiting the transcription of lymphokines (TOCCIet al., 1989). Rapamycin, a macrolide isolated from cultures of the soil microorganism Streptomyces hygroscopicus originates from Easter Island. Although the immunosuppressive properties of rapamycin were recognized early (MARTEL et al., 1977) they have been intensively investigated only recently, primarily after the discovery of FK506
Cyclosporin
Immunophilin:
FK506
FKBP
Rapamycin
FKBP
Cyclophilin
Potential effector
Calcineurin (+ Calmodulin, Ca2+)
Calcineurin (+ Calmodulin, Ca2+)
Inhibition
of cell cycle:
Go
-..
GI
Fig. 6. Dual domain concept for the immunosuppressants of microbial origin CsA, FK506,and rapamycin and (adapted from BAUMANN BOREL,1992).
547
and high conservation of the different cyclophilins across species suggests an important role for the normal cell function. All the members of the cyclophilin family show enzyme activity as peptidyl-prolyl cis-trans isomerases (rotamases), a property which enables them to accelerate the cis-trans isomerization of peptide bonds involving a prolyl residue. Interconversion of cis and trans conformers at peptide bonds to proline has been recognized as a rate determining step of protein folding in vitro. Therefore, it is not unlikely that the rotamase activity of cyclophil3.2.1 Cyclosporin Receptors n ins might also facilitate protein folding i vivo. This activity of cyclophilins is potently A specific, saturable and reversible binding inhibited by CsA (FISCHER al., 1989; KERN et of CsA on mononuclear blood leukocytes was et al., 1993) and some of its analogs, dependshown early (RYFFELet al., 1982). Although ing on the respective binding affinity of the membrane binding specific for CsA was re- compounds. The 40 kDa cyclophilin-like procently reported (CACALANO al., 1992) no tein has lower affinity to CsA and its rotamet membrane-bound receptors have been iso- ase activity is also less sensitive to inhibition lated to this point. On the other hand, accu- by this drug. FK506 binds to a separate group mulation of CsA within the cell suggested the of ubiquitous immunophilins termed FKexistence of an intracellular receptor. Indeed, binding proteins, FKBPs (SIEKIERKA al., et CsA, FK506, and rapamycin have been shown 1989b). Both a cytosolic form, FKBP12, and a to bind to cytosolic proteins termed immuno- membrane-associated form, FKBP13, sharing philins (SCHREIBER, 1991). CsA predomi- the same binding and enzymatic sites were renantly binds to cyclophilins (CUP), a family ported (HAYANO al., 1991; JINet al., 1991). et of highly conserved proteins comprising both A higher molecular mass protein, FKBP59, ubiquitous and tissue-specific proteins was shown to be related to and associated (SCHREIBER and CRABTREE, 1992). Cyclo- with heat shock proteins and with the cortiphilin A, the first protein identified, is a ubi- costeroid receptor (TAIet al., 1992). Strikingquitous cytosolic 18 kDa protein (HAND- ly, the FKBPs also exhibit peptidyl-prolyl cisSCHUMACHER et al., 1984). Cyclophilins B, c, trans isomerase activity which is blocked et and D, all with a molecular mass of about when FK506 is bound (SIEKIERKA al., 22kDa were reported later (HASELet al., 1989a). Neither CsA nor FK506 appear to 1991; CARONI al., 1991; PRICEet al., 1991; cross-react for the binding to their respective et BERGSMAet al., 1991). Cyclophilin B con- immunophilins. Since both drugs block the intains an endoplasmic reticulum retention sig- duction of lymphokine gene transcription at nal and is located in calcium-containing intra- the early stage of antigen-induced helper T cellular vesicles. Cyclophilin C seems to have cell activation, it has been postulated that T a more restricted tissue distribution with low cell activation requires the separate activity of expression in lymphoid tissue. Murine cyclo- both immunophilins. In addition, the rotamphilin C was reported to be highly expressed ase activity and the resulting protein folding in kidney and was thus suggested to be in- was thought to be required for nuclear transvolved in CsA nephrotoxicity (FRIEDMAN port of some factors involved in signal transand WEISSMAN, 1991). However, this was not duction. However, there seems to be no corobserved with human cyclophilin C relation between inhibition of rotamase activ(SCHNEIDER al., 1994). In addition, a ity and inhibition of T cell activation, as deet 40 kDa protein binding to CsA was also monstrated with the antagonists of FK506 shown to share homology with cyclophilin A and CsA, rapamycin and MeVa14-Cs or SDZ et (KIEFER al., 1992). The relative abundance NIM 811 (ZENKE al., 1993). et and because of the striking structural similarity of the two compounds. Rapamycin is effective in many experimental transplantation models. A single dose of rapamycin leads to almost indefinite survival of cardiac allografts in the rat. It also showed activity in several murine autoimmune disease models (MORRIS, 1993). Rapamycin inhibits T cell activation at concentrations comparable to those of FK506 but with a completely different mechanism (DUMONT al., 1990a, b). et
548
It has been shown by competition experiments that FK506 and rapamycin bind to a common intracellular receptor since the two drugs act as reciprocal antagonists (DUMONT et al., 1990a). Indeed, rapamycin binds to all the FKBPs described above. In contrast, FKBP25 which shares homology with FKBP12 in its C-terminal region but differs in the N-terminal part selectively binds rapamycin (GALATet al., 1992). Rapamycin inhibits the rotamase activity of the FKBPs but does not inhibit lymphokine transcription (DuMONT et al., 1990b). This clearly demonstrates that the inhibition of the rotamase activity of FKBP is insufficient per se to mediate the biological effect of FK506. The relevance of the different isoforms of cyclophilins or FKBPs for inhibiting T cell activation was addressed in two different ways. (1) Overexpression of CYP-A or CYP-B but not of CYP-C increased T cell sensitivity to CsA. Similarly, overexpression of FKBP12 but not of FKBP13 or FKBP25 increased T cell sensitivity to FK506 (BRAM al., 1993). et Subcellular localization of CYP-B and CYP-C was also essential for their activity (BRAM et al., 1993). (2) By studying the binding of human cyclophilins A, B, and C to a series of cyclosporin derivatives it could be determined that the three cyclophilins recognize very similar sites on the CsA molecule. However, CYP-C can accommodate larger residues at position 2 of the cyclosporin molecule than CYP-A (SCHNEIDERet al., 1994). Therefore, a series of cyclosporin derivatives modified at position 2 could be selected that bind CYP-C up to tenfold better than CYP-A in comparison with CsA. Within this series of derivatives, the immunosuppressive activity measured by inhibition of IL-2 gene transcription correlated better with the binding to CYP-A than to CYP-C (QUESNIAUX al., et unpublished observations). Both approaches suggest that CYP-C might be less essential than CYP-A for T cell sensitivity to CsA.
3.2.2 Active Sites on the Drugs

The current model of the mode of action of immunosuppressant macrolides suggests that FK506 and rapamycin act via two regulatory
domains: an immunophilin(FKBP)-binding domain which is shared between FK506 and rapamycin and an effector domain which is specific for each drug (BIERERet al., 1991) and accounts for their different activities. The same dual domain concept has also been demonstrated for CsA (Fig. 6). The binding site for cyclophilin was first mapped by immunochemical methods (QUESNIAUX al., 1987) et and later confirmed by NMR analysis (WEBER et al., 1991) and X-ray crystallography (PFLUEGL al., 1993; MIKOL al., 1993; see et et also Sect. 5.5). Cyclophilin A was shown to bind to residues 1, 2, 9, 10, and 11 of CsA, residues which were recognized early on to be essential for CsA immunosuppressive activity. The binding of CsA analogs to cyclophilin A also showed some correlation with their immunosuppressive activity (HANDSCHUMACHER et al., 1984; QUESNIAUX et al., 1988; DURE-I-I-E al., 1988). To define the effector et site of the cyclosporin A molecule several cell-permeable, cyclophilin-binding, but nonimmunosuppressive cyclosporin derivatives were selected. These analogs that are modified in the effector site of CsA, like MeVa14-Cs or SDZ NIM 811, antagonize the immunosuppressive effect of CsA on T cells (ZENKE al., 1993). They effectively inhibit et the cis-trans isomerase activity of cyclophilin providing again compelling evidence that inhibition of rotamase activity is insufficient for explaining CsA immunosuppression (SIGAL et al., 1991; BAUMANN al., 1992). However, et these results do not suggest that rotamase activity in general is irrelevant for the cell. Although the biological role of the two highly conserved families of rotamases, cyclophilins and FKBPS, is poorly understood at this time, there are a few examples for their role in protein transport. For example, the product of the Drosophila Nina A gene, a membranebound isomerase homologous to CYP-A, selectively transports rhodopsins in photoreceptor cells (STAMMES al., 1992). Similarly, et preliminary evidence suggests that cellular cyclophilin appears to be required for nuclear transport of DNA transcripts of HIV-1 (RoSENWIRTH et d., 1994; See Sect. 3.3.2).
549
3.2.3 Target Proteins of the Drug-Immunophilin Complexes

The structure-activity relationship of a large number of CsA or FK506 derivatives showed that the binding of CsA or FK506 to their respective immunophilins is necessary but per se not sufficient to inhibit T cell activation and thus raised the question of the molecular target of the drug-immunophilin complexes, CsAlCYP and FK506/FKBP. Indeed, CsA and FK506, when attached to their respective immunophilins could be part of a macromolecular complex and act on different primary targets or on different sites of a common primary target. Such a target would most probably be a component of the signal transduction pathway the activation of which ultimately leads to lymphokine transcription. Recent biochemical experiments support this model. The complexes of FK506/FKBP12 or CsA/CYP-A but not of rapamycin/FKBP12 were found to interact with calcineurin, a calciumkalmodulin-dependent protein phosphatase (LIUet al., 1991). Calcineurin is composed of two subunits, the catalytic, calmodulin binding 61 kDa A subunit and the regulatory, calcium binding 19 kDa B subunit. The drug-immunophilin complexes bind to and modulate the serinehhreonine phosphatase activity of calcineurin in vitro. However, calcineurin does not bind uncomplexed CsA, FK506, cyclophilin, or FKBP. In T cell lysates containing natural immunophilins both CsA and FK506 inhibit calcineurin activity at concentrations which effectively block IL-2 production in activated T cells (FRUMAN al., et 1992a). Rapamycin has no effect on calcineurin activity. Furthermore, excess concentrations of rapamycin compete the effects of FK506, apparently by displacing FK506 from FKBP. Cyclophilins A, B, C and CYP-40, when complexed with CsA, bind calcineurin. However, the cyclophilin B complex is 2-5fold more potent than that of cyclophilin A to inhibit calcineurin phosphatase activity (SWANSON al., 1992). Of the FKBP comet plexes only FK5061FKBP12 seems to inhibit calcineurin. Interestingly, CsA/CYP-A and FK506/FKBP12 compete for binding to calcineurin despite absence of obvious structural similarities.
The pharmacological relevance of the inhibition of calcineurin phosphatase activity by CsA or FK506 for immunosuppression has been substantiated by the good correlation between calcineurin activity and IL-2 production in T cells which is dose-dependently inhibited by CsA or FK506 (FRUMAN al., et 1992b). In addition, overexpression of the calcineurin catalytic subunit in the Jurkat T cell line rendered the transfected cells more resistant to the immunosuppressive effects of CsA and FK506 (OKEEFEet al., 1992; CLIPSTONE and CRABTREE, 1992). These results clearly showed that calcineurin is a target for drug-immunophilin complexes in vivo and suggested a physiological role for calcineurin in T cell activation. Since calcineurin is present in small amounts in T cells, it might be a rate determining enzyme in the T cell activation pathway. Such a situation would provide a rationale for the profound effects of CsA and FK506 on T cell activation. A 77 kDa cyclophilin C-associated protein which binds CYP-C with high affinity in the absence but not in the presence of CsA was described (FRIEDMAN al., 1993). The role et of this protein is still unknown. A cellular function for FKBP12 has been reported recently, following the observation that FKBP12 co-purifies with the ryanodine receptor (BRILLANTES al., 1994). Four ryanoet dine receptors associate to form intracellular Ca2+ release channels of the sarcoplasmic and endoplasmic reticula. FKBP12 was shown to stabilize channel gating by improving conductance increasing mean open time. This effect was reversed by both FK506 and rapamycin suggesting that blocking the cis-trans isomerase catalytic site is sufficient for inhibition by the drugs.
3.2.4 Downstream Effects in the Nucleus

Both CsA and FK506 act specifically on activation pathways mediated by the T cell receptor that induce an increase in intracellular Ca2+ concentration. The actual molecular target of the Ca2+-dependent serinekhreonine phosphatase calcineurin is not known.
550
Highly attractive candidates include cytosolic components of the transcription complex that need to be assembled in the nucleus and/or activated for expression of early T cell activation genes. Calcineurin may modify the phosphorylation of downstream components such as the antigen-inducible transcription factors NFKB (BAUMANN al., 1991) and NF-AT et (EMMEL al., 1989) which are essential for et IL-2 transcription. Both NFKB and NF-AT are present as precursors in the cytoplasm. They require protein modification such as phosphorylation or dephosphorylation to be translocated into the nucleus where they bind to the DNA and participate in the formation o a functional transcription complex leading f to the transcription of the IL-2 gene. It has been shown recently that CsA and FK506 inhibit dephosphorylation of NF-AT (MCCAFFREY et al., 1993). CsA and FK506 bound to their respective immunophilin most probably interfere at this level by inhibiting the phosphatase activity of calcineurin. This might directly or indirectly alter the protein modification of the transcription factors and consequently block lymphokine transcription (SCHREIBER and CRABTREE,1992). Rapamycin is not active on this pathway. As implicated by the mutually antagonistic activity of rapamycin and FK506 the mechanism of rapamycin action in lymphoid cells is likely to depend on immunophilin/drug complex formation. Rapamycin does not affect the transcription of genes involved in early activation of T cells, such as IL-2, but appears instead to block later events leading to T cell activation, such as the signal transduction pathway driven by the interaction of IL-2 with the high-affinity IL-2 receptor. More generally, rapamycin was shown to inhibit growth factor receptor-mediated activation signals in a number of different cells as well as in vivo after myelodepression (QUESNIAUX et al., 1994). The rapamycin/FKBP complex and the putative associated effector molecules interfere with the activation of the p70 S6 protein kinase in response to growth factor stimulation (CHUNGet al., 1992). This is a key regulatory step in the cell cycle progression from G1 to S phase. The direct target of rapamycin appears to be a crucial element linking growth factor receptors to subsequent
intracellular processes that regulate proliferation (PRICEet al., 1992) and has been identified recently by SCHREIBER collaborators and (BROWN al., 1994). Previously, it had been et suspected that it might be a proximate upstream activator of the p70 S6 protein kinase such as an activating p70 S6 kinase-kinase or a regulator of such an enzyme.
3.3 Molecular Evidence for other Biological Activities of Cyclosporins 3.3.1 Antiinflammatory Effects
Crossreactivity of antiserum against human IL-8 with cyclophilins led to the finding that IL-8 binds CsA but not some of its non-immunosuppressive analogs. Although IL-8 bears some sequence similarities to cyclophilin, it has no rotamase activity (BANGet al., 1993) and the relevance of these findings still needs further clarification. Cyclophilin A shows proinflammatory activity and leukocyte chemotactic activity which can be inhibited by CsA but not by a non-immunosuppressive analog (Xu et al., 1992; SHERRYet al., 1992). CYP-A is secreted by macrophages in response to endotoxin and it was proposed that CYP-A may function as a cytokine (SHERRY al., 1992). et FKBP was also shown to display some leukocyte chemotactic activity which is inhibited by FK506 (LEIVAand LY-ITLE, 1992). Cyclophilin 40 was shown to share homology with P59, a member of the steroid receptor complex (KIEFFERet al., 1993).
3.3.2 Cyclosporins as Anti-HIV Agents

The clinical potential of the anti-HIV activity of cyclosporins, in particular of SDZ NIM 811, was briefly discussed in Sect. 2.4 without addressing possible mechanisms of this effect which is clearly different from that of all other anti-HIV agents described to date. In cell-
4 Chemistry
551
free assay SDZ NIM 811 does not inhibit reverse transcriptase, protease, or integrase. There is some evidence that the cyclosporinsensitive step in HIV replication is an event after virus penetration, possibly nuclear translocation of viral DNA (ROSENWIRTH al., et 1994). It has been reported recently that some cyclophilins bind to the HIV-1 protein p55gg as well as the capsid protein p24 (LUBAN et al., 1993). In a cell-free system, this interaction is disrupted by both cyclosporin A and SDZ NIM 811. Furthermore, inhibition of the gag-cyclophilin complex formation by cyclosporins correlates with the cyclophilin binding capacity of the compounds but not with their immunosuppressive potential. The same correlation was found for the anti-HIV activity of these derivatives (ROSENWIRTH al., et 1994). Whether inhibition of gag-cyclophilin complex formation by cyclosporins explains the antiviral effect remains to be elucidated.
3.3.3 Cyclosporins as Drug Resistance Modifiers

The clinical potential of the non-immunosuppressive cyclosporin SDZ PSC 833 to sensitize multidrug resistant tumor cells to chemotherapeutic agents was described in Sect. 2.3. This resistance phenomenon (MDR) is thought to be associated with the p-glycoprotein transporter. Direct binding of CsA to pglycoprotein could be demonstrated by using a photoaffinity-labeled cyclosporin derivative (FOXWELL al., 1989). Interestingly, while et the immunosuppressive activity of cyclosporins is mediated by complex formation with cyclophilin (see Sect. 3.2.3) SDZ PSC 833 has no measurable affinity to cyclophilin. Similarly, other cyclosporins not binding to cyclophilins (e.g., cyclosporin H) also bind to pglycoprotein. Resistance modifier activity of cyclosporins appears thus to be a cyclophilinindependent effect.
vivo was reported as early as 1981 by THOMMEN-SCOTT. Interestingly, treatment was most effective when started at a time when parasitemia was already established suggesting a direct toxic effect on the parasite. Recently, with the availability of cyclosporin analogs binding to cyclophilins but devoid of immunosuppressive activity (such as SDZ NIM 811) as well as of cyclosporins without any measurable affinity to cyclophilins, it has become possible to discern between cyclosporin effects mediated by inhibition of calcineurin by the cyclosporin-cyclophilin complex (e.g., immunosuppression), mediated by cyclophilin binding only (such as, e.g., the antiHIV activity of SDZ NIM 811), from effects that are independent of cyclophilin (such as, e.g., the MDR activity of SDZ PSC 833 which does not bind to cyclophilin). With these reagents, a re-investigation of antimalarial activities of cyclosporins was performed (BELL et al., 1994). While CsA-sensitive peptidyl-prolyl cis-trans isomerase activity could be detected in extracts of P. falciparum the highest activity against P. falciparum was exhibited by SDZ PSC 833 suggesting that inhibition of rotamase activity may not be the lethal target of cyclosporins in P, falciparum.
4 Chemistry
4.1 Structural Aspects
Cyclosporins are composed of 11 aliphatic lipophilic amino acids of which four are leucines and three do not occur in mammalian proteins. In CsA these are: (4R)-4-[(E)-2-butenyl]-4-methyl-~-threonine (Bmt) in position 1, (L)-cY-amino butyric acid in position 2, and (D)-alanine in position 8. Of the 11 peptide bonds 7 are N-methylated. This feature has several important implications: Firstly, the Nmethylated peptide bonds and the cyclic structure of the molecule render cyclosporins stable towards mammalian digestive and systemic proteases (SandimmunB is metabolized extensively in animals and man, however, ex-
3.3.4 Antimalarial Activity of Cyclosporins

Activity of cyclosporin A against Plasmodium spp. in vitro and in rodent models in
552
clusively by cytochrome P450-mediated oxidative transformations). Therefore, cyclosporins are not only well absorbed when given by the oral route, high and long lasting plasma levels are commonly obtained. These properties are essential prerequisites for a successful drug. A second consequence of the N-methylation pattern is a certain conformational rigidity in non-polar environment characterized by intramolecular hydrogen bonds and the lipophilic side chains being oriented towards the (hydrophobic) environment. The high number of leucines in the cyclosporins is striking, especially in view of the well recognized role of leucines in proteinprotein interactions. Their role for biological activity will be discussed below. The structures of cyclosporins in apolar as well as polar solution have been determined by NMR spectrospic methods as well as by Xray crystallography. This work is extensively discussed in Sect. 5. In non-polar solvents, the conformation of cyclosporins is very similar to that found in the crystalline state. Cyclosporin A in polar environment exhibits a different conformation from that found in apolar systems (KO and DALVIT,1992). Cyclosporin A is not soluble enough in aqueous solution to allow a full 3D-structure determination by NMR but the H-NMR spectra in water indicate the presence of a family of conformations. A single predominant conformation of cyclosporin is observed only in complex with certain metal ions (KOECK et al., 1992) or in complex with cyclophilin. There exist derivatives, however, which are more water-soluble than CSA and some modifications (e.g., a methyl group in position 3) of cyclosporin seem to stabilize one conformation in water. Recently, the 3D-structure of such a derivative, namely of [D-MeSer3-DSer-(O-Gly)X]-cyclosporin, DMSO and in in water has been determined by NMR (WENGER et al., 1994). Fig. 7 shows the conformation of this derivative in DMSO which is identical to the one in water. Strikingly, this structure is very similar to the one found for CsA bound to cyclophilin (Fig. 8) thus giving evidence for the first time that CsA adopts the cyclophilin-bound conformation (among many other conformations) in aqueous solution. An induced-fit hypothesis for the
Fig. 7. NMR-structure of [D-MeSer3-~-Ser-(OGly)R]-cyclosporin,not bound to a receptor protein, in DMSO (identical to the one in water).
mode of binding of cyclosporins to cyclophilin would thus not seem necessary. To date, some 30 cyclosporins have been isolated from natural sources. Their structures are shown in Tab. 1.
4.2 Chemical Synthesis and
Production
The method of choice for the production of cyclosporins on a large scale is fermentation. Invariably, complex mixtures containing the desired cyclosporin along with a plethora of minor metabolites are obtained. To achieve sufficient purity of the final product repeated chromatography operations are necessary. For the production of cyclosporins on a large scale (tons), substantial investments in largescale chromatography facilities are therefore
4 Chemistry
553
Tab. 1 Cyclosporins Isolated from Fermentation Brothsa .
Name
Amino Acid in Position

1
2 Abu Ala Thr Val
3 Sar
4 Me Leu
5
Val
6
Me Leu
7 Ala
8 (D)-Aia
9 Me Leu
10
Me Leu
11
CsA CsB csc CsD CsE CsF CsG CsH CSI CsK CSL CsM CsN cso CSP CsQ CsR css CsT csu csv csw csx CSY csz Cs26 Cs27 Cs28 Cs29 Cs30 Cs31 Cs32 FR 901495
a
Me Bmt
Me Val
Val Deoxy-MeBmt Abu2 Nva (D)-MeVal Val Deoxy-MeBmt Val2 Bmt Nva Nva Me Nva Leu Thr Bmt Val Leu Thr Val Leu Leu Abu Thr Nva Nva Me Aocb Bmt Me Leu Nva Val Leu Val Leu Leu Leu Leu
Nva Leu
Me Ile Me Leu Val Ile Thr GIY (D)-Ser Leu Leu
Only residues different from those in Cyclosporin A are given. MeAoc = N-Methyl-(L)-amino octanoic acid.
554
necessary. For structure-activity studies, a large number of analogs has been prepared by total synthesis according to the method developed by Wenger (WENGER,1983) or by semisynthetic transformations of natural cyclosporins. In addition, a solid phase-based methodology has been developed in the preclinical research laboratories of Sandoz Pharma AG allowing the preparation of almost any analog on a small scale in a relatively short time (BOBE,unpublished work). Semisynthesis of cyclosporins involves several main approaches: Modification of the MeBmt aliphatic side chain, chemical transformations of the cyclic peptide backbone, and modifications based on selective ring opening reactions.
4.2.1 MeBmt Transformations and Variations

Fig. 8. X-ray structure of CsA when bound to CYPA, crystallized from an aqueous solution. There is
only one intramolecular hydrogen bond, namely between the hydroxyl group of MeBmt' and Leu4(CO). All amide bonds are in the trans conformation.
The olefinic double bond of MeBmt offers a number of possibilities of selective chemical transformations. In the preclinical research laboratories of Sandoz Pharma AG many analogs have been obtained by ozonolysis
1.
ii
Fig. 9. Some chemical transformations of the aliphatic side chain of MeBmt. R1 =protective group (most often acetyl); reagents: i =ozone, ii = R2-CH=P(~henyl)~; k = N-bromo-succinimide. For references, see text.
558
CH,
CY
mono- and dithio amides. The oxygen atoms replaced by sulfur atoms during these reactions are indicated with an asterisk. Cleavage of the thioamides can be achieved by alkylation at the sulfur atom, followed by acid hydrolysis.
Fig. 13. Treatment of cyclosporin with phosphorous pentasulfide or Lawessons reagent forms mixtures of
Me
As has been outlined in Sect. 3.2, the immunosuppressive activity of cyclosporins is based on an unusually complex mechanism which in a first step requires binding of the drug to a receptor of the family of cyclophilins and in a second step binding of this complex to protein phosphatase 2B (calcineurin), thereby inhibiting its catalytic activity. This activity is crucially dependent on cyclosporin; cyclophilin without any bound drug does not interact with calcineurin. There is also evidence that binding of the cyclosporin-cyclophilin complex to calcineurin involves both residues of cyclosporin as well as part of the cyclophilin molecule. This means that the molecule causing immunosuppression is the entire cyclosporin-cyclophilin complex. Consequently, cyclosporin structure-activity relationships must be analyzed both in terms of
cyclophilin binding and in terms of mediating the calcineurin interaction of the complex. Much insight into the nature of this interaction was gained from analyzing the crystal structures of many cyclosporin-cyclophilin complexes (Sects. 55-58). As indicated above, both CsA and SDZ 214-103 contain a strikingly high number of leucines. In the cyclophilin complexes of these molecules the side chains of the leucines in position 4,6, and 10 are exposed next to each other on the surface; they form a leucine cluster on the protein surface. It has long been recognized that leucine side chains in a specific three-dimensional disposition on the surface of a protein play a crucial role in mediating protein-protein interactions, a prominent example being the dimerization of transcription factors through leucine zipper domains (LAND-
5 Structural Investigations of Cyclosporins, Cyclophilins, and their Complexes
559
HN Q I
HN
CH3
Fig. 14. Cyclosporins incorporating rigid dipeptide fragments at the the p t u r n region in the absence of cyclophilin. For references, see text.
Fig. 15. Design and synthesis of a "calcineurin-bridging" ligand (ALBERG SCHREIBER, and 1993). The bicyclic amino acid is incorporated into the cyclosporin molecule in place of alanine' and (D)-alaninex.
SCHULTZ et al., 1989). The presence of a leucine cluster on the surface of the cyclosporincyclophilin complex prompted an analysis of the role of these leucines by substituting each of them for valine or alanine. A shortened version of the outcome of this analysis is shown in Tab. 2. From these results it is evident that in positions 4 and 6 the cyclosporin leucine side
chains have crucial functions for the calcineurin interaction; they can be substituted for valine or alanine without much affecting the affinity of the compounds for cyclophilin. However, the affinity of their respective complexes to calcineurin has been lost as evidenced by the lack of immunosuppressive properties of these derivatives. In line with this interpretation none of these compounds
561
This protein has 165 amino acids and is found in the cytoplasm. Cyclophilin B has 208 residues and is found in the endoplasmic reticulum. Cyclophilin C has 194 residues and is thought to have some tissue specificity for kidney, at least in the mouse (FRIEDMAN and WEISSMAN, 1991). There is no sequence homology between cyclophilins and FKBPs and no obvious three-dimensional structural similarity. Despite the chemical dissimilarity of the ligands and the structural differences between the proteins there is an intriguing overlap of biochemical and biological activity between the two immunophilin families (see Sect. 3.2). The common target of their complexes with the cognate drugs, calcineurin, is a heterodimer composed of subunit A (61 kDa) and subunit B (19 kDa) which has been shown to have affinity only for the immunophilin-drug complexes but not for the drug alone or the immunophilin alone (Lru et al., 1991). Furthermore, the binding of the FKBP12-FK506 complex to calcineurin competes with the binding of the CsA-CYP-A complex. Natural ligands for cyclophilins or FKBPs which could regulate phosphatase activity have not yet been discovered. Another puzzling coincidence between the two immunophilin families is their shared peptidyl-prolyl isomerase (PPIase) activity which led to the re-discovery of cyclophilin (TAKAHASHI al., 1989; et FISCHER al., 1989) as an enzyme catalyzing et protein folding in vitro. For proteins to adopt the correctly folded conformation the X-Pro amide bonds must be in the correct cis or trans conformation. Using model peptide substrates it has been found that immunophilins lower the energy of activation for isomerization of this amide bond by over 6 kcal .mol from about 20 kcal.mol-' down to 14 kcalemol-' (PARK et al., 1992). Both FKBP and cyclophilin have also been shown to accelerate the refolding of a number of proteins in vitro, presumably by catalyzing the rate determining step of proline isomerization (SCHOENBRUNNER al., 1991; et FRANSSON al., 1992). Specific cellular taret gets of the PPIases are not yet known but it has been suggested that they play a role in the folding of newly synthesized proteins (GETHING and SAMBROOK, 1992). Blocking the cis-
trans isomerase activity is, however, insufficient to induce immunosuppression (SIGAL et al., 1991). There is a growing body of available 3Dstructural information on immunophilins and their ligands. This work is directed at trying to understand both the biological and enzymatic activity. Protein X-ray crystal structures of cyclophilin A complexed with a tetrapeptide substrate (KALLEN al., 1991), a dipepet tide substrate (KE et al., 1993a), without substrate (KE et al., 1991), with CsA (PFLUEGL et al., 1993; MIKOLet al., 1993), and with a cyclosporin derivative modified in position 1 (KE et al., 1994; MIKOLet al., 1994b) have also been published as well as the X-ray structure of a Fab-CsA complex (ALTSCHUH et al., 1992). The 3D-structure of the CsA-CYPA complex has also been determined by NMR techniques (THERIAULT al., 1993), et as well as the 3D-structure of a water soluble cyclosporin derivative (WENGER et al., 1994).
5.2 Structural Investigations of Uncomplexed Cyclosporins 5.2.1 The 3D-Structure of Uncomplexed CsA in Apolar Environment
Prior to the studies of the conformation of CsA bound to CYP-A the 3D-structure of free CsA had been determined. Because of the very low solubility of CsA in aqueous solution, crystallization and NMR studies were done using apolar solvents. The NMR structure in chloroform at 20C (Fig. 17) is virtually identical to the X-ray crystal structure (LOOSLIet al., 1985). The backbone forms a twisted psheet that involves residues 1, 2, 5, 6, 7, and 11 and a type 11' turn at Sar3 and MeLeu4. There is a cis-peptide bond between MeLeu' and MeLeu". All the four amide protons are involved in hydrogen bonds (three transannular ones): Abu2(NH)- Va15(CO), Va15(NH)-Abu2(CO), and
562
Fig. 18. X-ray structure of SDZ 214-103 (not bound to cyclophilin), crystallized from diethyl ether.
Fig. 17. Conformation of cyclosporin A in chloroform.
Ala7(NH)-MeVal"(CO), and one additiona1 hydrogen bond D-AlaX(NH) -MeLeu6(CO).
Thr2(CO)-Leu5(NH), Ala7(NH)-Leu"(CO), and Ala7(CO) -Leu "(NH). The hydroxyl group of MeBmt' is not involved in an intramolecular hydrogen bond, but makes a packing interaction in the crystal with Sar3(CO) of a neighboring molecule.
5.2.2 The 3D-Structure of the Uncomplexed Peptolide SDZ 214-103 in Apolar Environment
The peptolide SDZ 214-103 differs chemically from CsA in the following way: Thr' instead of Abu2, Leu' instead of Val', D-Hiv' instead of D-Ala*, and Leu" instead of MeLeu". The X-ray structure (Fig. 18) of the uncomplexed peptolide (crystallized from an organic solvent) is very different from the one of uncomplexed CsA (Fig. 30) or CsA bound to CYP-A (Fig. 8): There is a cis amide bond between residues 3 and 4 and a Pbend involving residues 7, 8, 9, 10. The following intramolecular hydrogen bonds are formed:
5.2.3 The 3D-Structure of Uncomplexed CsA in Polar Environment

CsA is not soluble enough in aqueous solution to allow a full 3D-structure determination by NMR but the 'H-NMR spectra o f CsA in water indicate the presence of a family of conformations (CsA adopts a single, predominant conformation only if complexed with certain metal cations (KOECK et al., 1992) or when bound to cyclophilin). There exist cyclosporin derivatives, however, that are more water soluble than CsA. Furthermore, introduction of a substituent in position 3 seems to stabilize one conformation in wa-
563
ter. Recently, the 3D-structure of such a derivative, namely of ~-MeSer~-~-Ser-(O-Gly)*cyclosporin in DMSO and in water, has been determined by NMR (WENGERet al., 1994). The derivatization in position 8 makes the derivative more water-soluble and the substitution of D-MeSer for Sar in position 3 stabilizes a single conformation. Fig. 7 shows the conformation of this derivative in DMSO which is identical to the one in water. Strikingly, this structure is very similar to the one found for cyclosporins bound to cyclophilin (Fig. 8) thus giving evidence for the first time that cyclosporins preadopt the cyclophilinbound conformation (among many other con9 formations) in aqueous solution. An in- Fig. 1 . The architecture of CYP-A consists of an duced-fit hypothesis for the mode of binding eight-stranded antiparallel /?-barrel the ends of of cyclosporins to cyclophilin is thus not re- which are closed off by two a-helices. The tetrapeptide (a substrate for the PPIase activity of CYP-A) quired. binds to the outside of the P-barrel. Selected Ca-
5.3 The 3D-Structure of Cyclophilin A Complexed with a Tetrapeptide

At Sandoz Pharma AG the first three-dimensional X-ray structure of CYP-A with a model substrate (N-acetyl-Ala-Ala-Pro-Alaamidomethyl coumarin) bound to its active site (two complexes per asymmetric unit) was elucidated (KALLENet al., 1991; KALLENand WALKINSHAW, 1992). The X-ray structure of unliganded CYP-A (1 molecule per asymmetric unit) was determined by KE (KE et al., 1991; KE, 1992). The same group also determined the structure of a CYP-A complex with the dipeptide substrate Ala-Pro (KE et al., 1993a). The structure of CYP-A was determined in a collaborative effort using NMR and X-ray methods: The secondary structure of CYP-A was determined by NMR methods while simultaneously the tertiary structure was determined by X-ray methods (KALLEN et al., 1991; KE et al., 1991). Via chemical shift changes of certain residues in CYP-A upon complexation of the protein with CsA the NMR technique was also able to identify residues probably interacting with cyclosporin (KALLENet al., 1991).
positions of CYP-A are indicated by their sequence numbers.
Fig. 20. The complex CYP-A-tetrapeptide rotated by 90with respect to Fig. 19 around a vertical axis in the picture plane.
Cyclophilin is an approximate1 spherical molecule with a radius of ca. 17 (Figs. 19, 20). The main structural feature is the eightstranded antiparallel P-barrel consisting of two roughly perpendicular four-stranded @ sheets with a +3, -1, -2, +1, -2, -3 topology. Both ends of the barrel are closed off with an a-helix. The cyclophilin Pbarrel is similar in some respects to the superfamily of
564
proteins involved in ligand transport, including retinol binding protein (RBP), bilin binding protein, plactoglobulin, fatty acid binding protein, and P2 myelin (COWAN al., 1990). et Most of these molecules encapsulate their ligand in the /%barrel core. In contrast, the barrel core in CYP-A is tightly packed with hydrophobic residues and the ligand binding site is on the outside of the barrel. The topology of CYP-A also differs from the simple [ + l]n-up-and-down fold found in the RBP class of proteins or the [ -3, 1, 11-Greek key topology that is most frequently found in /%barrel proteins.
+ +
Active Site of Cyclophilin A
5 4 The Peptidyl-Prolyl Isomerase .

Two X-ray structures of CYP-A-substrate complexes have been published a complex of CYP-A with the tetrapeptide N-acetyl-AlaAla-Pro-Ala-amidomethylcoumarin (NacAAPA-amc) (KALLENet al., 1991) and a complex of CYP-A with the dipeptide AlaPro (KE et al., 1993a). In both cases the AlaPro amide bond adopts a cis conformation. In the following, details of the CYP-A tetrapeptide structure will be given (Fig. 21). CYP-A residues which have one atom within 4 8, of any atom of the active site-bound NacAAPA-amc are: Arg55, Ile5', Phew, Gln63, Ala"', AsnIo2,Gln"', Phe'l3, Leu'22, His'26, and Arg'48. The enzyme active site is a channel sitting on top of two antiparallel P-strands (with contacts from residues Phem, Gln"', and Phe1I3). Two loops protrude out from the surface of the barrel and provide a distinctive grooved protein surface. One loop from residues 101 to 110 contains the contact residues Ala"' and AsnIo2. A narrow pass separates this loop from the second which comprises residues 69 to 74. Another important topological feature of the binding site is the wall composed of residues 118-126 in a close to helical conformation. Leu122 and His'26 are in contact with the peptide substrate. The cis-proline of Nac-AAPA-amc sits in a rather deep pocket made principally by Leu'22, His'26, Phe1I3, Phew, and Met6'. There are three hydrogen bonds formed between the peptide and CYP-A (Fig. 21).
Fig. 21. Details of the interactions of the tetrapeptide with the enzyme active site of CYP-A. Intermolecular hydrogen bonds are formed between the main chain N and 0 atoms of Ala' (from the peptide) and the main chain N and 0 atoms of Asn'' (from CYP-A) as well as between the carbonyl oxygen of Pro3 (from the peptide) and NH1 and NH2 of Arg55 (from CYP-A).
The guanidinium group of Arg55 forms hydrogen bonds to carbonyl oxygen of Pro3 in the peptide. It also makes a hydrogen bond to Gln63 which in turn is hydrogen bonded to the side chain of Gln"'. The other two direct substrate-protein hydrogen bonds are in the form of a short stretch of antiparallel P-sheet between the main chain N and 0 atoms of Ala2 and the main chain N and 0 atoms of Asnlo2. The formation of a short stretch of antiparallel sheet is a common feature in many enzyme inhibitor complexes including aspartate proteases and serine proteases (BLUNDELL al., 1987). The side chain of et His'26 is close to the alanyl-prolyl cis amide bond, however, the X-ray refinement suggests a conformation in which the histidine side chain preferentially hydrogen bonds to solvent water and protein main chain rather than to the substrate. The mechanism of isomerase action is not yet clear, however, the Ala-Pro cis amide bond is significantly twisted out of plane. Until now, four indepen-
565
dent determinations of the w angle have been performed that was found to vary between 20 and 45. This is consistent with a mechanism of catalysis by distortion (HARRISON and STEIN,1990; PARKet al., 1992; LIU et al., 1990; ROSENet al., 1990) in which the immunophilin would bind the X-Pro amide bond of the substrate with a twisted, high-energy conformation. The transition state could also be stabilized by hydrogen bonding to the proline amide nitrogen (KOFRON al., 1991) possiet bly via a water molecule or His126 Arg55. or Based on his X-ray structure of the CYPA-AP complex (not showing a significantly distorted amide bond) KE proposed a mechanism in which the transition state is stabilized via a hydrogen bond of a water molecule to the carbonyl oxygen of the amide bond (KE et al., 1993a). A comparison of the active site of CYP-A when complexed with a substrate (,-f. Fig. 21) and in the uncomplexed form (Fig. 22) shows that there is practically no structural change for CYP-A. The biggest movement occurs for the side chain of Arg55 and shows a movement of its guanidyl group by about 2 A. The only well-ordered water molecule (with a crystallographic B-factor c 40 A) that has to
*RG5
HIS^
fig. 22. Details of the enzyme active site of CYP-A in the unliganded form. Only selected water molecules with B-factors <40 A are indicated (spheres) as well as their hydrogen bonds to atoms from CYP-A.
Fig. 23. A close-up view of the CsA-CYPA complex which shows the distinction between residues of CsA making mostly contacts with CYP-A (the CYP-A-binding region) and residues available for interactions with calcineurin (the effector region).
566
be displaced upon binding of the Ala-Pro moiety of the tetrapeptide is the one hydrogen-bonded to the main chain 0 and N of Asn'02. These hydrogen bonds are replaced, in the complex, by hydrogen bonds between Ala2 of the tetrapeptide and Asn'02 (cf. Fig. 21). Interestingly, the water molecules hydrogen bonded to Hiss4 and His'26 are conserved in the two structures.
5.5 The 3D-Structure of Cyclophilin A Complexed with CsA

The NMR structure of CsA when bound to CYP-A had been determined (WEBERet al., 1991; NERI et al., 1991) before the 3D-structure of CYP-A was known and was found to be substantially different from the conformation of uncomplexed CsA as determined by NMR in chloroform or by X-ray crystallography. The main features of the CYP-A-bound conformation as determined by NMR are that all amide bonds are in the rruns conformation and at first no intramolecular hydrogen bonds were found (subsequently, the X-ray structure of the CsA-CYP-A complex has shown that there is an intramolecular hydrogen bond between the hydroxyl group of MeBmt' and the carbonyl oxygen of MeLeu4). This contrasts with the structure of uncomplexed CsA in the crystal (cf. Fig. 30) or in chloroform where the amide bond between MeLeu' and MeLeu" is cis and the four free NH groups are all involved in intramolecular hydrogen bonds (LOOSLI al., 1985). et Using this NMR structure of CsA and 32 intermolecular NOES (nuclear Overhauser effects) between CsA and CYP-A as distance constraints CsA was subsequently docked into CYP-A (SPITZFADEN al., 1992). The et refined docked structure of the complex showed that 13 residues of cyclophilin are in contact with CsA. This docked structure agreed well with the structure-activity hypothesis based on the binding of a series of CsA derivatives and highlighted the importance of residues 10, 11, 1, 2, 3 (Fig. 23) for CYP-A binding (QUESNIAUX al., 1987). The direcet tion of the chain in the docked CsA is in the
opposite direction to that observed in the tetrapeptide-cyclophilin complex. Argss is also involved as a hydrogen bond donor in both structures, to the carbonyl oxygens of MeLeu" in CsA and Pro3 in the peptide. An important result of the collaborative X-ray, NMR, and modelling work on the CsA-CYP-A complex was to show conclusively that the cyclosporin binding site was identical to the peptidyl-prolyl isomerase active site (cf. Figs. 19 and 24). The NMR structure of the full CsA-CYPA complex (THERIAULT al., 1993) and the et X-ray structure of a decameric CsA-CYP-A complex have now been solved (PFLUEGL et al., 1993). Both structures are very similar and broadly confirm the results of the previous docking studies with the following corrections: There is actually an intramolecular hydrogen bond between the hydroxyl group of MeBmt and the carbonyl oxygen of MeLeu4 and an intermolecular hydrogen bond between Abu2(NH) of CsA and Asnlo2(CO)of CYP-A, both not found in the docking model. The biological relevance of the decameric form (with a pentamer of CsA-CYP-A complexes per asymmetric unit; Fig. 25) is unclear. Direct interactions between the effector domain of cyclosporin and calcineurin would practically be impossible since about 80% of the CsA surface are buried in the decameric complex. Subsequently, it was possible to grow crystals and solve the X-ray structure of a monomeric CsA-CYP-A complex at 2.1 8, resolution (MIKOLet al., 1993). In contrast to the decameric structure there are no intermolecular contacts between CsA and neighboring CYP-A molecules in the crystal. The monomeric CsA-CYP-A complex shows the following features (cf. Fig. 26): The binding pocket for CsA is a mainly hydrophobic crevice formed by the following 13 residues of CYPA which are within 4.08, of CsA: Arg", Phem, Met6', Gln63, G ~ Y ' ~ , Ala"', Asn'OZ, Ala103, Glu"', Phe113, TrpI2', Leu'22, and His'26. The binding site rests on three of the antiparallel strands of the eight-stranded p barrel involving Phe6', Met61, Gln63, Phe1I3, Gln"l, and Arg". Other clusters of residues (Trp12', Leu'22, His'26), (Ala'", Asn'", Ala'03), and ( G ~ Y ' ~ ) located on three separe
567
Fig. 24. The monomeric CsA-CYP-A complex viewed in the same orientation as the CYP-A-tetrapeptide complex in Fig. 19 showing that the cyclosporin binding site is identical to the peptidylprolyl active site (but CsA and the tetrapeptide bind with opposite polarities from N- to C-terminus).
arate loop regions which protrude some 10 A-15 A from the surface of the barrel forming a deeply grooved surface (cf. Fig. 24). CsA docks into CYP-A like a coin going part way into a slot machine; only one rim of the circular cyclosporin (formed by residues 9,10, 11, 1, 2, 3) is in contact with CYP-A. The complementarity of fit is particularly good for MeVal" which lies in the center of this binding rim. The valine side chain fits snugly into the deep proline-binding pocket formed by Phe6", Met61, Phe1I3, and Leu'22 (Fig. 26). There are five direct hydrogen bonds between CsA and CYP-A: MeBmt'(C0)G ~ ~ I ~ ~ (notH ) ( N unambiguously identified in the NMR structure of the complex), Abu2(NH)-Asn'02(CO), MeLeu'(C0)Trp12'(NE), M ~ L ~ U " ( C O ) - A ~ ~ ~ ~ ( N H ~ ) , and M ~ L ~ U " ( C O ) - A ~ ~ ~(notN H ~ ) in ~ ( reported the NMR structure of the complex). There are five clearly identified water molecules mediating interactions between CsA and MeBmt'(CO)-Wat'8-His54(NE2), CYP-A: MeBmt'(C0) - Wat" - Wat" - Asn7'(CO), A~u'(CO)-W~~~~-T~~~~(CO), and MeLeu6(C0)-Watx7-Watlo7-Args5(NH2).
Fig. 25. The pentamer half of the decameric CsA-CYP-A complex.
568
Fig. 26. Details of the interactions between CsA and CYP-A in the monomeric complex.
Fig. 27. A superposition (using C, of CYP-A) of CsA (black) and [MeBm2t]-cyclosporin (grey).
5.6 The X-Ray Structures of Cyclophilin A Complexed with Cyclosporin Derivatives

Using the crystal form with one CsA-CYPA complex per asymmetric unit and crossseeding techniques it was possible to obtain crystals and to solve the structures of 15 complexes of cyclophilin A and different cyclosporin derivatives. For all derivatives studied (variations in positions 1, 2, 3,4, 8) the backbone conformation of the Cs-ring is practically unchanged (as well as the interactions with CYP-A). There are also practically no changes in the CYP-A-structure. In the following, the results from two complexes will be summarized.
1991). The X-ray structure of CYP-A/ ([MeBm2t]-Cs) (MIKOLet al., 1994b) shows that a change with respect to the structure of CsA/CYP-A has occurred for the side chains of residues 1 and 1 0 The C, of MeLeu and of MeBm2t move apart by about 0.9 8, to accommodate the presence of the additional methyl group on MeBm2t (cf. Fig. 27). This small structural change must be beneficial for the interaction with calcineurin A/B. On the other hand, the affinity for CYP-A has decreased (as compared to CsA), although all the structurally mediated interactions to CYP-A are conserved. This finding suggests that for [MeBm2t]-Cs the equilibrium (attained in aqueous solution for the uncomplexed ligand) between nonbinding and binding conformation is shifted to the nonbinding side, indicating a higher kinetic barrier for complex formation.
5.6.1 [MeBm,t] -Cyclosporin

[MeBm2t]-Cs differs from CsA in having two (instead of one) methyl groups on the C, of residue 1 (AEBIet al., 1990). Its affinity for CYP-A is about 1% of that of CsA whereas its immunosuppressive activity i v i m is only n reduced to 30% of that of CsA (SEAL et al.,
5.6.2 MeIle4-Cs (SDZ NIM 811)

MeIle4-Cs has the same affinity for CYP-A as CsA but no immunosuppressive activity. The X-ray structure (J. KALLENet al., to be submitted) shows that the only change with respect to the structure of CsA/CYP-A has
569
Fig. 28. A superposition (using C, of CYP-A) of CsA (black) and MeIle4-cyclosporin (SDZ NIM 811) (grey). The substitutionof Ile for Leu in position 4, without any other structural change, leads to a < 1000-fold reduced immunosuppressive activity.
Fig. 29. The NMR-structure (a family of 20 structures with low NOE violations of maximally 0.29 A) of SDZ 214-103 l ~ ~ to (2YP-A. n d
occurred for the side chain of residue 4 in that Leu is replaced by Ile (Fig. 28). The mere presence of a branched (vs. unbranched) C, for residue 4 is thus sufficient to drastically impair binding to calcineurin A/B (which must, therefore, have a tight-binding pocket for this region (PAPAGEORGIOU al., et 1994); cf. the discussion in Sect. 4.3.
structures of free CsA and free SDZ 214-103 are quite different (cf. Figs. 18 and 30).
5.6.3 The 3D-Structure of SDZ 214-103 Bound to Cyclophilin A

The 3D-structure of CYP-A bound SDZ 214-103 has been determined by NMR methods (WIDMER al., unpublished results). 95 et structurally relevant intramolecular NOES were used in structure calculations with the program DIANA. The non-hydrogen atoms of a family of 20 structures (with low residual NOE-violations of maximally 0.29 A) are depicted in Fig. 29. A comparison with the Xray structure of CsA when bound to CYP-A (Fig. 30) shows that the backbone conformations are very similar (slight differences might be due to measurement errors of the two techniques). On the other hand, the X-ray
5.7 The X-Ray Structure of Cyclophilin B Complexed with a Cyclosporin Derivative

The affinity of cyclophilin B for CsA is approximately 10-fold higher as that of cyclophilin A (SCHNEIDER al., 1994). Similarly, et the complex CypB-CsA is known to be 5- to 10-fold more effective than CsA-CYP-A in inhibiting the SerlThr phosphatase activity of calcineurin (SWANSON al., 1992). At Sanet doz Pharma AG the X-ray structure of a complex of CypB with a cyclosporin derivative modified in position 8 ([0-(choliny1ester)-DSerx]-cyclosporin) was determined in order to elucidate possible structural origins responsible for this difference (MIKOLet al., 1994a).
570
F g 31. The architecture of CypB (in complex with i. cyclosporin), viewed in the same orientation as CYP-A in Fig. 8. The major structural differences with respect to CYP-A occur at the N- and C-termini and in the loops 19-24 and 152-164 (residue numbering of CypB). F g 3 . The X-ray structure of CsA (not bound to i. 0
cyclophilin) crystallized from acetone. C , atoms are indicated by the residue numbers (1-11) and hydrogen bonds are indicated by dashed lines. All the four amide protons are involved in hydrogen bonds (three transannular ones: Abu*(NH)-ValS(CO), Va15(NH)-Abu2(CO), and Ala'(NH)-MeVal'*(CO), and one additional hydrogen bond DAlaH(NH) MeLeu6(CO)). -
The overall structures of CypB and CYP-A are relatively similar (cf. Figs. 31, 32, 19, and 20). However, significant differences occur in two loops (residues 19-24 and 152-164 of CypB) and at the N- and C-termini. The active site regions of CypB and CYP-A are very similar. Indeed, there are practically no differences for any residues of the cyclosporinbinding pocket within 5.0 A of cyclosporin (rmsd for all these non-hydrogen atoms is less than 0.15 A). The cholinylester derivatization in position 8 could not be seen in the electron density map, probably because of too high conformational mobility. The binding and conformation of the cyclosporin residues are
F g 3 . The complex CypB-CsA rotated by 90" i. 2 with respect to Fig. 30 around a vertical axis in the Picture Plane, to be compared with the view of depicted in Fig. 20.
571
practically identical for the two complexes. Candidates for an explanation of the increased potency of CypB-Cs complexes are the following residues of CypB: Arg, Lys113, Ala2x, and the loop containing Arg5X.
5.8 The X-Ray Structure of Cyclophilin C Complexed with CsA

The structure of CypC (KE et al., 1993b) shows that it is similar to CYP-A, with the exception of the loops Asp7-Lys9, Met70-Ile76, and Gln79-Thr189. The cyclosporin binding pockets are practically identical as well as the conformation of cyclosporin itself for the complexes CypC-CsA and CsA-CYP-A.
5.9 Conclusions
Cyclosporins can adopt a variety of conformations depending on the molecular environment. If the environment is hydrophobic, the number of intramolecular hydrogen bonds is maximized by adopting a conformation as shown in Fig. 30 (all four amide protons are involved in intramolecular hydrogen bonds). In a polar environment, on the other hand, the seven N-methyl groups are shielded from the solvent by adopting a conformation as shown in Fig. 8 (there is just one intramolecular hydrogen bond between the hydroxyl group of MeBmt and the carbonyl oxygen of MeLeu4). The transition between these two conformations resembles the inversion of a glove. The NMR structure of a water-soluble cyclosporin derivative in a polar solvent in the absence of cyclophilin shows that cyclosporins can preadopt the cyclophilin-bound conformation in the absence of the protein and there is no need for an induced-fit hypothesis to explain the binding of cyclosporins to cyclophilins. The CYP-A-bound backbone conformation seems to be an energetically favorable one. It is robust against perturbations of cyclosporin side chains: In all cyclosporin derivatives analyzed structurally until now the backbone conformation is practically identical in all cases; (see, e.g., Fig. 29
for an extreme derivatization of CsA). The cyclosporin structure is also preserved in complexes with different receptors (as seen in the complex structures with CYP-A, CypB, CypC). The division of the cyclosporin molecule into a binding region (making binding interactions with cyclophilin) and an effector region (important for the immunosuppressive effect by mediating interaction of the cyclophilin complex with calcineurin) is visualized in Fig. 23. In the effector region there are some sensitive positions, i.e., residues in very close contact with calcineurin such as, e.g., residue 4 The mere change from MeLeu4 to MeIle4 without any other structural change is sufficient to completely abolish immunosuppressive activity (cf. Fig. 28). CsA alone cannot exert an inhibitory effect on calcineurin, it needs to be complexed with cyclophilin. Hence, there must be crucial direct interactions between cyclophilin and calcineurin residues also. Evidence for this hypothesis is based on the different inhibitory potencies of cyclophilins A, B, and C when complexed with CsA (FLIRIet al., 1993). Extensive site-directed mutagenesis experiments on CYP-A have been done (ETZKORN al., et 1994) in order to find such crucial residues (e.g., the CYP-A mutant Argl48Glu complexed with CsA shows a 20-fold improved inhibition of calcineurin). The X-ray structure of CypB (the complex CypB-CsA is at least 10-fold more effective than CsA-CYP-A in inhibiting calcineurin) suggests also that the loop containing ArgISX, which corresponds to the loop containing Arg4Xin CYP-A, might be important for modulating the interactions with calcineurin (apart from the possible candidates ArggO,Lys113,and Ala2x of CypB). A final answer to these questions will be given, of course, by the experimentally determined 3D-structure of the CsA-CYP-A-calcineurin complex. The additional availability of a 3Dstructure of the FKBP12-FK506-calcineurin complex would also answer the question of how the obviously structurally dissimilar CsA-CYP-A and FKBP12-FK506 complexes can both compete for interaction with calcineurin.
512
6 Biosynthesis of Cyclosporins
6.1 Introduction
Some structural features of cyclosporin A are suggestive of a non-ribosomal biosynthetic origin: the unusual amino acids (4R)-4[(E)-2-butenyl]-4-methyl-~-threonine(Bmt, position l), a-amino butyric acid (position 2), and D-alanine (position 8) as well as the cyclic structure and the N-methylation of seven peptide bonds. Indeed, cyclosporins are biosynthesized by an extraordinary large multienzyme called cyclosporin synthetase. However, efficient cyclosporin biosynthesis not only needs a high activity of this central enzyme but also of pathways providing the unusual components as well as a sufficiently high internal pool of the methyl group donor S-adenosyl-methionine (Tab. 3). Whereas aamino butyric acid is most likely derived from the common amino acid pool (SENNet al., 1991) Bmt and D-alanine have to be supplied by separate pathways. Hence, general interest in cyclosporin biosynthesis has been focused
on three enzymes or enzymatic pathways, respectively: (1) cyclosporin synthetase, (2) alanine racemase, (3) Bmt-synthesizing enzymes.
6.2 Cyclosporin Synthetase 6.2.1 Enzymatic Activities of Cyclosporin Synthetase

Enzymatic activity in protein fractions of Tolypocladium niveum that could be ascribed to cyclosporin synthetase was first detected et by ZOCHER al. (1986). Initial in vitro studies led to the enzymatic synthesis of cyclo(D-alanyl-N-methylleucyl) diketopiperazine (D-DKP) rather than to the synthesis of cyclosporin A. The reason for this is not known, but considering the finding that mutants of T. niveum expressing non-functional cyclosporin synthetase and, therefore, lacking the competence to produce cyclosporin A produce DDKP instead (DITTMANN al., 1990) it can et be concluded that D-DKP is a bypass-product of cyclosporin synthetase under non-optimal conditions. First in vifro enzymatic synthesis
Tab. 3. Requirements for Cyclosporin Biosynthesis
Source Common amino acid pool (L)-Alanine Acetate, SAM, NADPH 0 2 . Nz c,-Pool Glucose Pantothenate
Enzyme(s)
Building Units and Cofactors for Cs-Biosynthesis Alanine, a-amino butyric acid, glycine, leucine, valine (D)-Alanine Bmt SAM ATP (MgCIJ 4'-Phosphopantetheine covalently attached to cyclosporin synthetase
Alanine racemase Bmt-polyketide synthase Transformation enzymes SAM Synthetase
Cyclosporin Synthetase
1
Cyclosporins
573
of the complete cyclosporin molecule from et al., 1990; MACCABE al., 1991; GUTIERet the constituent amino acids, ATP, MgC12, and REZ et al., 1991), enniatins (HAESE et al., S-adenosyl-methionine as a methyl group do- 1993), gramicidin S (TURGAY al., 1992; et nor was successful in 1987 (BILLICHand HORI et al., 1989; KRAETZSCHMAR al., et ZOCHER, 1987). 1989), tyrocidine A (WECKERMANN al., et Whereas enzymes synthesizing other D-ami- 1988), HC-toxin (SCOTT-CRAIG al., 1992), et no acid-containing peptides, e.g., gramicidin S and surfactin (COSMINA al., 1993) it beet and tyrocidine (KLEINKAUF VON D ~ H R - came clear that such enzymes are composed and EN, 1990), harbor an integral epimerase func- of domains. Each of these domains is respontion which epimerizes the respective L-amino sible for the recognition and activation of one acid into the D-form following its activation, amino acid and for the peptidation reaction cyclosporin synthetase rather incorporates D- with the amino acid activated by the neighalanine in position 8 only if already supplied boring domain. In the case of methylated enin its D-form. Hence, epimerization of t-alan- niatins the corresponding domain harbors an ine has to be carried out by a distinct enzyme additional module responsible for the methyl(see Tab. 3). In contrast, the methyltransfer- ation step (HAESE et al., 1993). Taking the ase activity for the N-methylation of the pep- molecular masses of all these peptide synthettide bonds is an integral part of the purified ases into account a molecular mass for cycloenzyme (LAWEN and ZOCHER,1990). Pro- sporin synthetase of at least 1.6 MDa can be teolysis experiments indicated the existence extrapolated. of several methyltransferase domains in one enzyme molecule, probably one for each of the seven methylation steps, which has been 6.2.3 Isolation and confirmed by cloning and sequencing the Characterization of the gene (see below). Alltogether, cyclosporin synthetase catalyzes at least 40 partial reac- Cyclosporin Synthetase Gene tion steps: 11 aminoadenylation reactions, 11 The gene coding for cyclosporin synthetase transthiolation reactions, 7 N-methylation reactions, 10 elongation reactions, and the fi- has recently been cloned and sequenced (LEITNER al., 1994; WEBERet al., 1994). In et nal cyclization reaction. order to obtain partial amino acid sequences to derive specific oligonucleotide probes the cyclosporin synthetase was purified from my6.2.2 Characterization of the celia of T. niveum and partially digested with Enzyme endoproteinases. 18 fragments were isolated, purified, and used for sequence determinaPurification and analysis by SDS-PAGE tion. One of these fragments was identified by clearly demonstrated that the complete reac- photoaffinity labeling with S-adenosyl-metion sequence of cyclosporin biosynthesis is thionine and a second fragment by its capacicoded by a single multienzyme polypeptide ty to activate L-alanine (LAWENand Zoet (LAWEN ZOCHER,1990) but lacking ade- CHER, 1990; LEITNER al., 1994; WEBER et and quate molecular weight markers it was a diffi- al., 1994). The cyclosporin synthetase gene (simA) cult task to determine the molecular mass of this giant protein. Analytical ultracentrifuga- was cloned using an oligonucleotide probe tion indicated that the enzyme most likely has derived from one of these amino acid sea discus-like structure with a diameter of quences. A genomic library of T. niveum about 33 nm, a thickness of 4.6 nm, and a mo- ATCC 34921 in XEMBL3 was screened, alet lowing determination of a nucleotide selecular mass of about 1.4MDa (SCHMIDT al., 1992). From cloning and sequencing stu- quence of 46889bp with an ORF of dies of other peptide synthetases responsible 45823 bp. The gene product (CYSYN) correfor biosynthesis of, e.g., 6-(a-L-aminoadipy1)- sponds to 15281 amino acids with a predicted L-cysteinyl-D-valine (SMITH al., 1990; DIEZ molecular mass of 1.7MDa. A data bank et
574
search showed characteristic similarities to known peptide synthetases. 11 regions of CYSYN show high similarity. Two different types are found within the 11 domains of CYSYN. The first type (type I) is about 1000 amino acids in size and very similar to the one already detected in other multifunctional peptide synthetases. This type I is shown in Fig. 34 (non-filled rectangle). The second type (type 11) is larger than the first type and includes an approximately 447 amino acid polypeptide (Fig. 33 and 34). This polypeptide sequence has probably N-methyltransferase activity as demonstrated by several lines of evidence: (1) An experimentally derived amino acid sequence of a 45 kDa fragment of cyclosporin synthetase having methyltransferase activity was found in the deduced amino acid sequence at a position corresponding to one of these additional sequences (Fig. 33). (2) This additional sequence is very similar to a corresponding sequence found in the enniatin synthetase (HAESE et al., 1993). (3) All seven sequences show similarity to a postulated consensus sequence for S-adenosyl-methionine binding and all seven CYSYN sequences are highly conserved in this region (LEITNER al., 1994; WEBERet al., 1994). et There are seven type I1 and four type I domains. All ll domains contain in the same relative position the putative amino acid binding and phosphopantetheine acid attachment sites found in other peptide synthetases. The 130000 Da cyclosporin synthetase fragment could be characterized by its capacity to activate L-alanine. This permits an assignment of its N-terminal sequence to CYSYN (Fig. 33). The fragment corresponds well to the 11th domain (LEITNER al., 1994; WEBERet al., et 1994). L-Alanine is known as the last amino acid added to the growing peptide chain (DITTMANN al., 1994). et The order of domains - with or without a putative methyltransferase activity - corresponds to the biosynthetic order of methylated and non-methylated amino acids (DITTMANN et al., 1994). The authors, therefore, concluded that there is a correspondence of the order of the constitutive amino acids of cyclosporin A and the order of the 11 domains as shown in Fig. 33 (LEITNERet al., 1994; WEBERet al., 1994). A similar colinear-
ity has been shown for other peptide synthetases (LAWEN al., 1992). et The C-terminal end of domain 11 is not the C-terminus of CYSYN. There are approximately 500 amino acid residues of non-domain character. As the biosynthesis of cyclosporins also includes ring closure of the peptide one could speculate that this non-domain (i.e., showing no significant homology to domains 1-11) peptide sequence harbors this function (WEBERet al., 1994).
6.2.4 Manipulation of the Cloned Cyclosporin Synthetase Gene by Integrative Transformation

The description of the simA gene and the correlation of the order of protein domains and constituent amino acids of cyclosporin will enable the construction of new fungal strains by exchange of domain-specific parts of the simA gene by gene replacement (Fig. 35). Necessary for such experiments is a transformation system which allows the re-introduction of in v i m manipulated DNA by homologous recombination. Plasmids composed of A. nidulans promoters fused to a bacterial hygromycin phosphotransferase gene have been used successfully in a number of fungal species (FINCHAM, 1989). Such plasmids can also be used for the transformation of T. niveum but there is a high proportion of multiple tandem integrations of the plasmid DNA. Transformation systems for T. niveum were described by LEITNERand WEBER (1994). The authors isolated the cyclophilin gene as a source of a homologous promoter element and called the isolated gene cpfA. The entire gene consists of 890 bp, including the three introns of 220 bp, 57 bp, and 60 bp respectively. The gene codes for a protein (CFT) with a molecular mass of 19569 Da. The cDNA corresponding to cptA is similar to the cyclophilin cDNA of Neurosporu crussu (80%) which was used as a probe to isolate the T. niveum gene. At the amino acid level the similarity is also 80%. The promoter of the T. niveum gene was used for plasmid constructions in which this promoter is fused to a bacterial hy-
Sfi SPeI f h I
575
i
SCaI
Xbal
iscaI
SCaI
NotI Xba
MT
I
in
lIlmllI
30000 40000
Gala
loo00
unnn,
Fig. 33. Structure of the cyclosporin synthetase gene (simA) and the derived polypeptide (CYSYN). A partial restriction map of the 47 kb gene region is shown. The structure of the derived translation product is illustrated by horizontal boxes. There are two different types of domains which are described in more detail in Fig. 34. The grey parts identify the N-methyltransferase subdomains. The box labeled with a C (horizontal lines) indicates the about 500 amino acid long C-terminal part of the translation product. The two boxes with vertical lines indicate the postujated positions of the N-methyltransferase fragment (MT) and the L-alanine activating fragment (L-ala) based on the N-terminal sequences and the observed molecular masses (WEBERet al., 1994). TYPE I Fig. 3 . Relationship of the two do4
a b c d e f g h i
main types. The two types of domains are aligned in order to illustrate their relationships. a to I illustrate similar amino acid sequences. The grey part of the lower box stands for the N-methyltransferase fragment. The small box within this region symbolizes putative S-adenosyl methionine binding sites. Further putative binding sites for AMP and DhosDhoDantethein cofactor are aiso inhicited (AMP, PP) (WEBERet al., 1994).
1
AMP
TYPE U
gromycin phosphotransferase gene. A fragment containing 2.1 kb upstream of the putative start codon of the cprA gene was amplified by PCR and ligated with a 1.76 kb ClaIXbaI fragment from pCSN44. This 1.76 kb pCSN44 fragment covers a bacterial hygromycin phosphotransferase gene and the fragment with the transcriptional terminator of the A. nidulans trpC gene (STABENet al., 1989). The resulting plasmid was called pSIM10. Protoplasts of T. niveum could be
transformed with purified plasmid pSIMlO DNA in the presence of PEG and hygromycin-resistant colonies were obtained (LEITNER and WEBER,1994; WEBERet al., 1994). As it was intended to use the transformation system to manipulate the gene for the cyclosporin synthetase of T. niveum, derivatives of pSIMlO were constructed containing internal fragments of this gene. The first example, pSIM11, contains a 3.6 kb XhoI restriction fragment of the simA gene. A crossover be-
n.
l
81
1. B
I sl
PP
576

Cyclosporin A
MeBmt Abu
Sar
MeLeu Val
MeBmt
Abu
Sar
Val
Val
Cyclosporin Q
Fig. 35. Theoretical reprogramming of cyclosporin synthetase. The partial structure of the cyclosporin synthetase gene is shown schematically. The gene regions coding for domains are shown as boxes and labeled by the names of the corresponding amino acids. A fragment of the gene coding for a domain which activates valine (hatched box) is modified in vitro by adding the flanking regions of another part of the gene coding for a domain which activates leucine. Due to this sequences the DNA coding for the valine-specific domain can replace the DNA coding for the leucine-specific domain following recombination. The recombined gene codes for a cyclosporin synthetase that preferentially produces cyclosporin Q instead of cyclosporin A. This is a hypothetical scheme.
tween the cloned and the genomic version of DNA should lead to insertion of the plasmid DNA and to a partial duplication of the target DNA. As the cloned DNA does not contain the 5 or 3 end of the gene the insertion inactivates the gene. T. niveum protoplasts were transformed with pSIMll following linearization of the plasmid DNA. Out of 81 pSIM11 transformants 50 (62%) were found that do not produce cyclosporin. Several of the transformants which lost the ability to produce cyclosporin were verified to contain the expected fragment sizes by Southern hybridizations (LEITNER and WEBER, 1994; WEBERet al., 1994). The second example, pSIM13, includes a 2.1 kb EcoRI fragment of the cyclosporin synthetase gene cloned into pUC18. A 3.7 kb
XhoI fragment including the promoter of the cyclophilin gene, the hygromycin phosphotransferase gene, and a A. nidulans trpC transcription terminator fragment as inserted into the central XhoI site of the cloned simA fragment. Following transformation, this construct can be inserted into the genomic DNA by a double crossover (Fig. 36). pSIM13 was used for transformation of T. niveum ATCC 34921 protoplasts and the transformants were analyzed for cyclosporin production. DNA from pSIM13 transformants was analyzed by Southern hybridization. For the three different restriction enzymes used to digest the DNA the sites expected for the DNA of a mutant generated by gene replacement were identified (Fig. 37) (WEBERet al., 1994). The high frequency of transformants which do not produce cyclosporin indicates that only one copy of the simA gene is present in the genome of T. niveum. A cyclosporin non-producing mutant of T. inflatum Cyb156 accumulating the cyclosporin precursor amino acid et Bmt has been described by SANGLIER al. (1990). This mutant showed reduced sporulation and reverted to cyclosporin formation at high frequency. In contrast, the pSIM11transformants show normal morphology and growth characteristics. Transformants were also analyzed for accumulation of Bmt but only small amounts could be detected even if high-yielding T. niveum strains were used for transformation (LEITNERand WEBER, 1994; WEBERet al., 1994). The transformation system described proved to be a powerful tool for gene disruption of the cyclosporin synthetase (simA) gene of T. niveum. It is intended to use these methods for gene replacement experiments in which parts of the simA gene coding for amino acid specific domains are exchanged. The mutated genes will direct the synthesis of new cyclosporins or of cyclosporins which are up to now only by-products of the biosynthesis.
6.2.5 Mechanistic Aspects of Biosynthesis

Detailed studies of the mechanism of cyclosporin biosynthesis provided strong evi-
577
Fig. 3. Gene disruption with 6 pSIMl1. The upper part of the figure shows the XbaI fragment of the Tolypocladium niveum simA gene. The XhoI fragment cloned in pSIMll is indicated by vertical lines. The middle part of the figure shows pSIM11, the hph coding region is indicated by a box with diagonal lines. The lower part shows the predicted structure of the DNA following a single recombination event in the region cloned in pSIM11. The restriction fragments hybridizing with the labeled XhoI fragment are indicated (WEBERet al., 1994).
,Sd
ji
"")( I
XhO
NdI XhoI
aI
10.6kb
4.8b
-2.1kb 0.95kb
Fig. 37. Gene disruption with the Nhel EcoRI fragment of pSIM13. The upper part of the figure shows a region of the Tolypocladium niveum simA gene. The EcoRI fragment cloned and disrupted in pSIM13 is indicated by a box with vertical lines. The middle part of the figure shows the NheI EcoRI fragment of pSIM13 used for transformation; the hph coding region is indicated by a box with diagonal lines. The lower part shows the predicted structure of the DNA following a double recombination event in the region cloned in pSIM13. The restriction fragments hybridizing with the labeled EcoRI fragment if the DNA is digested with EcoRI or EcoRV or SpeI are indicated. For SalI, e.g., this is a 3.4 kb fragment for the nondisrupted gene and a 7.1 kb fragment following gene disruption (WEBERet al., 1994).
3.4kb
V
EcoRV EcoRV
27r .1b
7.1kb
4:R
dence that binding of D-alanine is the initial reaction of biosynthesis, followed by the stepwise synthesis of a single linear undecapeptide precursor which is finally cyclized to cyclosporin A (DITTMANN al., 1994); (cf. Fig. et 38). However, the details of the whole biosyn-
thetic cycle are still unclear. As early as 1973, a thiotemplate mechanism had been suggested f o r the nonribosomal biosynthesis of peptides (LALANDand ZIMMER, 1973). Peptide synthetases activate their respective substrate amino acids in two steps, involving am-
578
inoacyl adenylates and thioesters as reactive intermediates, as shown in the following equation:
rin synthetase is rather low. This is reflected by the finding of more than 30 naturally occurring variants of cyclosporin A synthesized et by T.niveurn (TRABER al., 1987); (see Tab. l), the relative amount of the cyclosporin varE,+aa+MgATP*-+E(aaAMP)+E-S -aa iants being considerably dependent on the Ea enzyme, adenylation site fermentation conditions (KOBEL and TRAE - S - enzyme, amino acid binding site BER, 1982). Strikingly, only some positions, aa amino acid especially position 2, show variability in their amino acid incorporation whereas other posiThe assembly of the peptide chain is achieved tions, especially positions 3 and 8, are relaby repeated transpeptidation and transthiolatively invariant. Detailed in vitro studies tion reactions facilitated by 4 -phosphopanteshowed a somewhat divergent behavior of the theine as a carrier which interacts with the SH isolated enzyme: Position 1, e.g., was found to groups of the peripheral amino acid activashow substantial flexibility and position 8 extion centers (KLEINKAUF VON DOHREN, and hibits low substrate specificity as well (LA1990). The 4 -phosphopantetheine molecule and TRABER, has a length of 2 nm and has been postulated WEN et al., 1989,1992; LAWEN 1993). Even incorporation of p-alanine in poto reach up to six of the active centers of a synthetase (KLEINKAUF VON DOHREN, sition 8 (and 7, respectively) was shown to be and 1987). Although 4-phosphopantetheine is possible, causing formation of a 34-membered also an essential component of cyclosporin ring in contrast to the 33-membered ring of synthetase (LAWENand ZOCHER,1990), in cyclosporins (LAWENet al., 1994). However, view of the size of this enzyme the model of a only few of these in vitro accessible comcentral swinging phosphopantetheine arm ap- pounds can be synthesized in vivo by precurpears not applicable. Recent results with sor directed biosynthesis (TRABERet al., et gramicidin S synthetase as well as with surfac- 1988, 1994; HENSENS al., 1992) perhaps tin synthetase strongly indicate that the cru- due to unsuitable physiological conditions cial amino acid for binding of the substrate and/or metabolic utilization of the added preamino acid is serine (SCHLUMBOHM al., cursors. et 1991; VOLLENBROICH al., 1993; DSOUZA et et al., 1993), leading the authors to suggest a modified version of the thiotemplate mechanism in which each substrate amino acid is bound by the SH group of a phosphopantetheine arm which in turn is attached to the crucial serine residue of the respective activation center. This concept is further supported Fig. 3 . Model for cyclosporin biosynthesis by a b 8 by the finding of the consensus amino acid se- modified thiotemplate mechanism. Cyclosporin quence for attachment of phosphopante- synthetase activates its respective substrate amino theine in each of the 11 domains of cyclospo- acids in two steps involving adenylation (indicated rin synthetase (see Sect. 6.2.3). A model for as a l , a2, etc.) followed by thio-esterification, most cyclosporin biosynthesis, including the con- probably by a phosphopantetheine arm attached to et a serine symbolized by -0- (SCHLUMBOHM cept of a modified thiotemplate mechanism, is al., 1991). 7 out of 11 amino acids are methylated at shown in Fig. 38. this stage by integral methyl transfer units. Repeated transpeptidation and transthiolation reactions, (D)-alanine being the N-terminal amino acid (DIVMAN al., 1994), lead to the assembly of the et linear undecapeptide precursor which is finally cyclized to cyclosporin. The numbers 1-11 do not correlate with the numbering system of the chemical nomenclature but rather with the sequence of reactions of the biosynthetic pathway.
6.2.6 Biosynthesis of Cyclosporin Variants

As common in nonribosomal peptide synthesis, the amino acid specificity of cyclospo-
579
,3
Domains 1-1 1
"
I
a 3 SAM (leu)
- REPEATED ACTIVATION AND

ELONGATION REACTlONS
N = N - ~ u ~ ~ s
al, ...., allaminoacidsl-11
CYCLIZATION
(involvement of possible)
c = c - terminus @@ methyltransfaase - unit

C-t~sequence
R1,...., R11 side chains of al, ...., a1 1

S A M = S - adenosyl methionine
.5
ICYCMSPORINI
PP = phosphopantotbein arm
580
12 Cyclosporins:Recent Developments in Biosynthesis, Pharmacology and Biology
6.2.7 Related Enzymes: Peptolide SDZ 214-103 Synthetase
of C. oligospermum (Corda) Bonorden (HOFFMANN al., 1994), the producer of the et SDZ 214-103 which contains an ester linked D-2-hydroxyisovaleric acid instead of D-alaIn addition to the multitude of cyclosporin nine (cf. Fig. 2). In view of this result, it seems variants which are synthesized by cyclosporin very likely that the exclusive and indispensasynthetase in vitro and/or in vivo, the exis- ble role of the detectable alanine racemase is tence of many related compounds with more supplying D-alanine for cyclosporin biosynpronounced structural deviations, synthesized thesis. In line with this assumption, inhibition by homologous multienzymes, can be ex- of alanine racemase activity in vivo destroys pected. One example is SDZ 214-103 (Fig. 2), the competence of T. niveum for biosynthesis the cyclosporin-related peptolide produced of cyclosporin A, shown by feeding experiby the fungus Cylindrotrichum oligospermum ments with 3-fluoro-(~)-alanine, a well (Corda) Bonorden (HOFFMANN al., 1994). known inhibitor of prokaryotic alanine raet et The multienzyme responsible for biosynthesis cemases (HENSENS al., 1992). Since activaof this compound was isolated from C. oligo- tion of D-alanine is the first step in cyclospoet spermum, found to have a similar molecular rin biosynthesis (DITTMANN al., 1994) and weight as cyclosporin synthetase and to show the affinity of cyclosporin synthetase for Det cross reactivity with cyclosporin synthetase- alanine is remarkably high (HOFFMANN al., specific antibodies (LAWEN et al., 1991, 1994) the role of this enzyme seems not to be 1992). Despite the obvious similarity of the restricted only to supply one of the structural enzymes as well as of their respective prod- components of cyclosporin A but it appears ucts, cyclosporin synthetase and SDZ 214-103 to be rather a rate-limiting pacemaker of the synthetase can not substitute enzymatically whole biosynthetic cycle. This view is supfor each other. Furthermore, detailed in vitro ported by the absence of free D-alanine in exincorporation studies point to a more retracts of cyclosporin producing strains of T. stricted substrate specificity of SDZ 214-103 niveum. synthetase (LAWEN and TRABER,1993).
6.3 Alanine Racemase 6.3.1 Alanine Racemizing Activity

As mentioned above, cyclosporin synthetase is the first example of a peptide synthetase which lacks an integral activity epimerizing L-alanine to D-alanine. Hence, the existence of a distinct alanine racemase had to be postulated. This was somewhat unexpected since the existence of alanine racemases had previously been restricted to prokaryotic organisms where they are primarily involved in cell wall biosynthesis (ADAMS,1972; WALSH et al., 1985; SODA et al., 1986). Enzymatic analysis of crude extracts of T. niveum indeed led to the detection of an alanine racemizing activity and allowed purification and characterization of the respective protein (HOFFMANN et al., 1994). Interestingly, no corresponding activity could be detected in strains
6.3.2 Characterization of the Enzyme

T.niveum enzyme consists of several subunits
Like many of the prokaryotic homologs the
of a molecular mass of about 40000 Da (estimated by SDS-gel electrophoresis) and depends on pyridoxal phosphate as the exclusive cofactor (HOFFMANN al., 1994). Conet sequently, the enzyme is susceptible to the classical racemase suicide inhibitors in vivo (HENSENS al., 1992) as well as in vitro, as et shown for D- and t-(l-aminoethy1)-phosphonate (HOFFMANN al., 1994). In vitro et studies with a purified enzyme fraction further revealed that the specificity of the racemase for L- and D-alanine is high but not exclusive. Compared to the reaction velocity of the epimerization of L-alanine, L-serine, 2-~-aminobutyric acid, and L-leucine, e.g., are racemized with relative reaction rates of 23%, 15%, and 13%, respectively (HOFFMANN al., et
581
1994). Whether the enzyme is related to the prokaryotic homologs in an evolutionary sense will be elucidated by cloning and sequencing the corresponding gene which is in et al., progress (K. SCH~RGENDORFER unpublished results).
6.4.2 Identification of the Basic Assembly Product and Characterization of Bmt-Polyketide Synthase
NMR analysis of cyclosporin A produced by feeding experiments with [l-'3C,'X02]acetate demonstrated retention of the acetate6.4 Bmt-Synthesizing Enzymes derived "0-isotope during biosynthesis (measured as an upfield shift of the *3C-signal 6.4.1 Polyketide Origin of Bmt et in position 3 of Bmt; OFFENZELLER al., 1993). Assuming that all condensation, reducBmt with its long aliphatic chain is the most tion and dehydration steps as well as the meunusual amino acid of cyclosporins. The thylation reaction occur in the course of the structure suggested an acetate-derived bio- basic assembly, this result was a first conclusynthetic pathway, confirmed by KOBEL et al. sive proof of 3(R)-hydroxy4(R)-methyl(1983) who were able to show by feeding ex- 6(E)-octenoic acid to be the biosynthetic key periments that four "C-labeled acetate units intermediate (Fig. 39). Subsequently, detailed are coupled in a head-to-tail fashion and enzymatic studies led to the identification of processed to finally yield Bmt. Methionine this compound as a product of cell free enwas identified as the source of the methyl zyme fractions from T. niveurn (OFFENZELgroup in position 4, whereas the origin of the LER et al., 1993). The respective Bmt-polykeamino group remained unknown. These re- tide synthase has been characterized and its sults were later confirmed by feeding experi- isolation, which is in progress, will allow its ments using selectively I3C-labeled glucose thorough characterization and provide the (SENNet al., 1991). Hence, Bmt can be clas- basis for cloning the corresponding gene. One interesting feature of the enzyme is that it resified as a polyketide. These findings led to a model for the bio- leases its product as a coenzyme A thioester et synthetic pathway of Bmt which, in accor- (OFFENZELLER a]., 1993), a process also dance with the current concepts of polyketide known from fatty acid synthases of several biosynthesis, most likely takes place in at fungi (LYNEN,1980). Details of the basic asleast two distinct phases (for recent reviews, sembly process, especially the stage of the 1993; JORDAN and methylation reaction (see possible routes in see KATZ and DONADIO, SPENCER, 1993; O'HAGAN,1991; ROBINSON, Fig. 39), remain to be elucidated and are un1991; HOPWOOD SHERMAN, and 1990; SIMP- der investigation. SON, 1989): (1) a basic assembly process involving the coupling of four acetate units, the respective reduction and dehydration steps as 6.4.3 The Transformation Process well as the methylation reaction; and (2) a According to the current understanding of transformation process introducing the amino et group (OFFENZELLER al., 1993). Based on Bmt-biosynthesis transformation most likely this concept it was obvious that identification takes place by hydroxylation of the basic asof the basic assembly product would provide sembly product at the C2-atom with subsethe first clues for elucidation of the biosyn- quent oxidation and transamination to Bmt (cf. Fig. 39). It seems unlikely that these thetic route. transformation reactions are performed by Bmt polyketide synthase itself suggesting the involvement of additional enzymes in Bmt biosynthesis, namely of a hydroxylase, an oxidase, and a transaminase. This model is supported by i vivo incorporation of the amn
TRANSFORMATION
$=OH.wl
Fig. 39. Biosyntheticpathway to Bmt. Abbreviations:S-adenosylmethionine, SAM; acyl carrier protein, ACP.
7 References
583
BANG, H., BRUNE,K., NAGER, C., FEIGE, U. (1993), Interleukin-8 is a cyclosporin A binding protein, Experientia 49, 533-538. BARRY, M. (1992), Immunosuppressive drugs in J. renal transplantation. A review of the regimens, Drugs 44,554-566. BAUMANN, BOREL,J. F. (1992), MCcanismes G., Note added in proof molkculaires de Iaction des agents immunosuppresseurs, mtdecine/sciences 8, 366-371. Enzyme-bound 3-0x0-4-hexenoic acid BAUMANN, GEISSE,S., SULLIVAN, (1991), G., M. could be identified as the direct methylation Cyclosporin A and FK-506 both affect DNA M., precursor (OFFENZELLER, SAUTER,G., binding of regulatory nuclear proteins to the huTOTSCHING, Su, Z., MOSER,H., TRABER, man interleukin-2 promoter, New Biologist 3, K., R., SCHNEIDER-SCHERZER, (1996). BioE. 270-278. synthesis of the unusual amino acid (4R)-4- BAUMANN, ZENKE, WENGER, M., HIESG., G., R. [(E)-2-butenyl]-4-methyl-~-threonine cyof v. TAND, P. c.,QUESNIAUX, F. J., ANDERSEN, M. E., SCHREIER, H. (1992), Molecular mechaclosporin A: enzymatic analysis of the reacnisms of immunosuppression, J. Autoimmun. 5 tion sequence including identification of the (SUPPI.A), 67-72. methylation precursor in a polyketide pathBELL,A., WERNLI, FRANKLIN, M. (1994). B., R. way (Biochernisfry 35,84014412). Roles of peptidyl-prolyl cis-trans isomerase and calcineurin in the mechanisms of antimalarial action of cyclosporin A, FK506, and rapamycin, Biochem. Pharmacol. 48,495-503. BERGSMA, J., EDER,C., GROSS, KERSTEN, D. M., H., SYLVESTER, APPELBAUM, CUSIMAD., E., NO, D., LIVI,G. P., MCLAUGHLIN, M., KAM. ABRAHAM, T., KARNITZ, M., SECRIST, P., R. L. J. SYAN, K. ET AL. (1991), The cyclophilin multiLEIBSON,P. J. (1992), Signal transduction gene family of peptidyl-prolyl isomerases. Charthrough the T-cell antigen receptor, Trends Bioacterization of three separate human isoforms, J. chem. Sci. 17,434-438. Biol. Chem. 266,23 204-23 214. E. ADAMS, (1972), Amino acid racemases and epi- BIERER, E., JIN,Y. J., FRUMAN, A., CALVO, B. D. B. merases, in: The Enzymes (BOYER, D., Ed.), S. V., BURAKOFF, J. (1991), FK 506 and rapamy3rd Edn., pp. 479-507. New York: Academic cin: molecular probes of T-lymphocyte activaPress. tion, Transplant. Proc. 23,2850-2855. AEBI,J. D., DEYO,D. T., SUN, Q., GUILLAUME,BILLICH, ZOCHER, (1987), Enzymatic synC. A., R. D., DUNLAP, RICH,D. H. (1990), Synthesis, B., thesis of cyclosporin A, J. Biol. Chem. 262, conformation, and immunosuppressive activities 17258-17259. of three analogues of cyclosporin A modified in T. J., S. I., BLUNDELL, L., COOPER, FOUNDLING, the 1-position, J. Med. Chem. 33, 999-1009. JONES, M., ATRASH, SZELKE, (1987), D. B., M. S. ALBERG, G., SCHREIBER, L. (1993), StrucD. On the rational design of renin inhibitors: X-ray ture-based design of a cyclophilin-calcineurin studies of aspartic proteinases complexed with bridging ligand, Science 262, 248-250. transition state analogues, Biochemistry 26, J. ALTSCHUH, VIX,O., REES,B., THIERRY, C. D., 5585-5595. (1992), A conformation of cyclosporin A in D., C., B., aqueous environment revealed by the X-ray BOESCH, GAVERIAUX, JACHEZ, POURTIER MANZANEDO, BOLLINGER, LOOR, A., P., structure of a cyclosporin-Fab complex, Science F. (1991), In vivo circumvention of P-glycopro256,92-94. tein-mediated multidrug resistance of tumor ANDRIEU, M., EVEN,P., VENET, A. (1986), J. cells with SDZ PSC 833, Cancer Res. 51, 4226AIDS and related syndromes as a viral-induced 4233. autoimmune disease of the immune system: an D. P. R. anti-MHC I1 disorder. Therapeutic implications, BRAM, J., HUNG, T., MARTIN, K., SCHREIG. BER,S. L., CRABTREE, R. (1993), IdentificaAIDS Res. 2,163-174. tion of the immunophilins capable of mediating M. R. ARMITAGE, J., BECKMANN, P., IDZERDA, R. inhibition of signal transduction by cyclosporin L., ALPERT, FANSLOW, C. (1990), ReguA., W. A and FK506: roles of calcineurin binding and lation of interleukin 4 receptors on human T cellular location, Mol. Cell Biol. 13, 47604769. cells, Int. Immunol. 2, 1039-1045.
ino group of N-labeled glutamate into Bmt which indicates participation of transaminases and in the biosynthetic route (R. TRABER K. MEMMERT, personal communication).
7 References
584
12 Cyclosporins: Recent Developments in Biosynthesis, Pharmacology and Biology DIEZ, B., GUTIERREZ, BARREDO, L., VON S., J. SOLINGEN, VAN DER VOORT, L. H. M., P., MARTIN, F. (1990), The cluster of Penicillin J. biosynthetic genes, J. Biol. Chem. 265, 1635816365. DITTMANN, LAWEN, ZOCHER,R., KLEINJ., A., KAUF, (1990), Isolation and partial characterH. ization of cyclosporin synthetase from a cyclosporin non-producing mutant of Beauveria nivea, Biol. Chem. Hoppe-Seyler 371,829-834. DITTMANN, WENGER, M., KLEINKAUF, J., R. H., LAWEN, (1994), Mechanism of cyclosporin A A. biosynthesis. Evidence for synthesis via a single linear undecapeptide precursor, J. Biol. Chem. 269,2841-2846. DUMONT, J., MELINO, R., STARUCH, J., F. M. M. S. P. N. KOPRAK, L., FISCHER, A., SIGAL, H. (1990a), The immunosuppressive macrolides FK-506 and rapamycin act as reciprocal antagonists in murine T cells, J. Immunol. 144, 14181424. DUMONT, J., STARUCH, J., KOPRAK, L., F. M. S. M. N. MELINO, R., SIGAL, H. (1990b), Distinct mechanisms of suppression of murine T cell activation by the related macrolides FK-506 and rapamycin, J. Immunol. 144,251-258. DURETTE, L., BOGER, DUMONT, J., FIREP. J., F. STONE, FRANKSHUN, A,, LIN,C. S., MER., R. LINO, M. R., PESSOLANO, A. (1988), A study A. of the correlation between cyclophilin binding and in vitro immunosuppressive activity of cyclosporin A and analogues, Transplant. Proc. 20, 51-57. EBERLE, K., NUNINGER, (1992), Synthesis of M. F. the main metabolite (OL-17) of cyclosporin A, J. Org. Chem. 57,2689-2691. EBERLE, K., NUNINGER, (1993), Preparation M. F. of (~-cysteine)8-cyclosporinvia intramolecular sulfur transfer reaction, J. Org. Chem. 58, 673677. EMMEL, A., VERWEIJ, L., DURAND, B., E. C. D. HIGGINS, M., LACY,E., CRABTREE, R. K. G. (1989), Cyclosporin A specifically inhibits function of nuclear proteins involved in T cell activation, Science 246, 1617-1620. ENDICOTI-, A., LING,V. (1989), The biochemisJ. try of P-glycoprotein mediated multidrug resistance, Annu. Rev. Biochem. 58, 137-171. ERNST,P. (1991), Lung transplantation: The current state of knowledge, Union Med. Can. 120, 64-66. ETZKORN, CHANG, STOLZ, WALSH, T. F., Z., L., C. (1994), Cyclophilin residues that affect noncompetitive inhibition of the protein serine phosphatase activity of calcineurin by the cyclophilin-cyclosporin A complex, Biochemistry 33, 2380-2388.
BRILLANTES, A.-M. B., ONDRIAS, SCOTT,A., K., KOBRINSKY, ONDRIASOVA, MOSCHELE., E., LA, M. C., JAYARAMAN, LANDERS, T., M., EHRLICH, E., MARKS, R. (1994), StabilizaB. A. tion of calcium release channel (ryanodine receptor) function by FK506-binding protein, Cell 77,513-523. BROWN, J., ALBERS, W., SHIN, B., ICHIE. M. T. KAWA, K., CURTIS, T., LANE, K. W. S., SCHREIBER, S. L. (1994), A mammalian protein targeted by G1-arresting rapamycin-receptor complex, Nature 369,756-758. CACALANO, A., CHEN,B. X., CLEVELAND, N. W. L., ERLANGER, F. (1992), Evidence for a B. functional receptor for cyclosporin A on the surface of lymphocytes, Proc. Natl. Acad. Sci. USA 89,43534357. CALNE, Y., WHITE, J., THIRU, EVANS, R. D. S., D. B., MCMASTER, DUNN, C., CRADDOCK, P., D. G. N., PENTLOW, D., ROLLES, (1978), CyB. K. closporin A in patients receiving renal allografts from cadaver donors, Lancet 2 (8104-5), 12231227. CARONI,P., ROTHENFLUH, MCGLYNN, A., E., SCHNEIDER, (1991), S-cyclophilin.New memC. ber of the cyclophilin family associated with the secretory pathway, J. Biol. Chem. 266, 1073910742. CHUNG, Kuo, C. J., CRABTREE, R., BLENIS, J., G. J. (1992), Rapamycin-FKBP specifically blocks growth-dependent activation of and signaling by the 70 kD S6 protein kinases, Cell 69, 12271336. CLIPSTONE, A., CRABTREE, R. (1992), IdenN. G. tification of calcineurin as a key signalling enzyme in T-lymphocyte activation, Nature 357, 695-697. COSMINA, RODRIGUEZ, DE FERRA, P., F., F., M., G., GRANDI, PEREGO, VENEMA, VAN G., SINDEREN, (1993), Sequence and analysis of D. the genetic locus responsible for surfactin synthesis in Bacillus subtilis, Mol. Microbiol. 8,821831. COWAN, W., NEWCOMER, E., JONES,T. A. S. M. (1990), Crystallographic refinement of human serum retinol binding protein at 2 8, resolution, Proteins 8, 44-61. CRABTREE, R. (1989), Contingent genetic regulG. atory events in T lymphocyte activation, Science 243,355-361. D'SOUZA,C., NAKANO, M., CORBELL, ZuM. N., BER,P. (1993), Amino-acylation site mutations in amino acid-activating domains o surfactin f synthetase: Effects on surfactin production and competence development in Bacillus subtilis, J. Bacteriol. 175, 3502-3510.
7 References
585
FAULDS, Go, A. S., BENFIELD, (1993), CyD., P. death in murine T cell hybridomas, Eur. J. Imclosporin. A review of its pharmacodynamic and munol. 22, 2513-2517. pharmacokinetic properties and therapeutic use GALAT,A., LANE,W. S., STANDAERT, F., R. in immunoregulatory disorders, Drugs 45, 953S. SCHREIBER, L. (1992), A rapamycin-selective 1040. 25-kDa immunophilin, Biochemistry 31, 2427FINCHAM (1989), Transformation in fungi, Microb. 2434. Rev. 53, 148-170. GETHING, J., SAMBROOK, (1992), Protein M. J. FISCHER, WITTMANN-LIEBOLD,LANG,K., G., B., folding in the cell, Nature 355, 33-45. KIEFHABER, SCHMID, X. (1989), Cyclo- GUTIERREZ, DIEZ, B., MONTENEGRO, T., F. S., E., philin and peptidyl-prolyl cis-trans isomerase are MARTIN, F. (1991), Characterization of the J. probably identical proteins, Nature 337, 476Cephalosporium acremonium pcbAB gene en478. coding cu-aminoadipyl-cysteinyl-valine synthetFLIRI, G., WENGER, M. (1990), Cyclosporins: H. R. ase, a large multidomain peptide synthetase: synthetic studies, structure-activity relationLinkage to the pcbC gene as a cluster of early ships, biosynthesis, and mode of action, in: Biocephalosporin biosynthetic genes and evidence chemistry of Peptide Antibiotics (KLEINKAUF, of multiple functional domains, J. Bacteriol. 173, H., H., VON DOHREN, Eds.), 1st Edn., pp. 2462354-2365. 287. Berlin, New York: Walter de Gruyter. HABESHAW, A., DALGLEISH, G., BOUNTIFF, J. A. FLIRI, G., BAUMANN, ENZ,A., KALLEN, H. G., J., A. D., L. L., NEWELL, L., WILKS, WALKER, C., LUYTEN,M., MIKOL,V., MOVVA,R., QUESMANCA, (1990), AIDS pathogenesis: HIV enF. NIAUX, F. J., SCHREIER, H., WALKINV. M. velope and its interaction with cell proteins, ImSHAW, M. D., WENGER, M., ZENKE, ZUR. G., munol. Today 11,418425. RINI, M. G. (1993), Cyclosporins. Structure-acHAESE, A,, SCHUBERT, M., HERRMANN, A., tivity relationships, Ann. N. Y. Acad. Sci. 696, ZOCHER, (1993), Molecular characterization R. 47-53. of the enniatin synthetase gene encoding a mulFOXWELL, M. J., LING,V., RYFFEL, (1989), B. B. tifunctional enzyme catalyzing N-methyl-depsiIdentification of the multidrug resistance related peptide formation in Fusarium scirpi, Mol. Microbiol. 7, 905-914. p-glycoprotein as a cyclosporin binding protein, Mol. Pharmacol. 36,543-546. HANDSCHUMACHER, E., HARDING, W., R. M. FRANSSON, FRESKGARD, O., HERBERTSSON, RICE, J., DRUGGE,R. J., SPEICHER, W. C., P. D. (1984), Cyclophilin: A specific cytosolic binding A., P., H., JOHANSSON, JONASSON, MARTENSprotein for cyclosporin A, Science 226, 544-547. SON, L. G., SVENSSON, JONSSON, H., M., B. U. CARLSSON, (1992), Cis-trans isornerization is HARRISON, K., STEIN, L. (199O), Mechanistic R. R. rate-determining in the reactivation of denastudies of peptidyl prolyl cis-trans isomerase: tured human carbonic anhydrase I1 as evidenced evidence for catalysis by distortion, Biochemistry by proline isomerase, FEBS Lett. 2%, 90-94. 29, 1684-1689. FREY,F. J. (199O), Cyclosporin in autoimmune dis- HASEL,K. W., GLASS, R., GODBOUT, SUTJ. M., eases, Schweiz. Med. Wochenschr. 120, 772-786. CLIFFE, G. (1991), An endoplasmic reticulumJ. specific cyclophilin, Mol. Cell Biol. 11, 3484FRIEDMAN, WEISSMAN, (1991), Two cytoplasJ., I. mic candidates for immunophilin action are re3491. vealed by affinity for a new cyclophilin: one in HAYANO, TAKAHASHI, KATO,S., MAKI, T., N., N., the presence and one in the absence of CsA, Cell SUZUKI, (1991). Two distinct forms of peptiM. 66,799-806. dylprolyl-cis-trans-isomeraseare expressed sepaFRIEDMAN, TRAHEY, WEISSMAN, (1993), J., M., I. rately in periplasmic and cytoplasmic compartCloning and characterization of cyclophilin C-asments of Escherichia coli cells, Biochemistry 30, 3041-3048. sociated protein: a candidate natural cellular ligand for cyclophilin C, Proc. Natl. Acad. Sci. HENSENS, D., WHITE,R. F., GOEGELMAN, 0. R. USA 90,6815-6819. T., INAMINE, S., PATCHETT, A. (1992), E. A. B. The preparation of [2-deutero-3-fluoro-~FRUMAN, A., KLEE,C. B., BIERER, E., BuD. ala*]cyclosporin A by directed biosynthesis, J. s. RAKOFF, J. (1992a), Calcineurin phosphatase activity in T lymphocytes is inhibited by FK 506 Antibiot. 45,133-135. and cyclosporin A, Proc. Natl. Acad. Sci. USA HOFFMANN, SCHNEIDER-SCHERZER K., E., KLEIN89,3686-3690. KAUF, H., ZOCHER, (1994), Purification and R. characterization of an eukaryotic alanine racemFRUMAN, A., MATHER, E., BURAKOFF, J., D. P. S. ase, acting as a key enzyme in cyclosporin bioBIERER, E. (1992b), Correlation of calcineuB. synthesis. J. Biol. Chem. 269, 12710-12714. rin phosphatase activity and programmed cell
586
12 Cyclosporins: Recent Developments in Biosynthesis, Pharmacology and Biology T-cell cyclophilin A at 2.5 8, resolution, Proc. Natl. Acad. Sci. USA 88,9483-9487. KE, H. M., MAYROSE, CAO,W. (1993a), CrysD., tal structure of cyclophilin A complexed with substrate Ala-Pro suggests a solvent-assisted mechanism of cis-trans isomerization, Proc. Natl. Acad. Sci. USA 90,3324-3328. KE, H. M., ZHAO, Y., Luo, F., WEISSMAN, I., FRIEDMAN, (1993b), Crystal structure of muJ. rine cyclophilin C complexed with immunosuppressive drug cyclosporin A, Proc. Natl. Acad. Sci. USA 90, 11850-11854. KE, H. M., MAYROSE, BELSHWA, J., ALD., P. S. BERG,D. G., SCHREIBER, L., CHANG,Z., ETZKORN, Ho, S., WALSH,C. T. (1994), F., Crystal structures of cyclophilin A complexed with cyclosporin A and N-methyl-[(E)-2-butenyl]-4,4-dimethylthreoninecyclosporin A, Structure 2, 33-44. KERN, D., DRAKENBERG, WIKSTROM, T., M., FORSEN, BANG, FISCHER, (1993), The S., H., G. cishans interconversion of the calcium regulating hormone calcitonin is catalyzed by cyclophilin, FEBS Lett. 323, 198-202. T. KIEFFER,L. J., THALHAMMER, HANDSCHUR. MACHER, E. (1992), Isolation and characterization of a 40-kDa cyclophilin-related protein, J. Biol. Chem. 267,5503-5507. KIEFFER, J., SENG,T. W., LI, W., OSTERMAN, L. D. G., HANDSCHUMACHER, BAYNEY, R. E., R. M. (1993), Cyclophilin-40, a protein with homology to the P59 component of the steroid receptor complex. Cloning of the cDNA and further characterization, J. Biol. Chem. 268, 1230312310. KINO,T., HATANAKA, HASHIMOTO, NIH., M., SHIYAMA, M., GOTO,T., OKUHARA, KOHM., SAKA, AOKI,H., IMANAKA, (1987), FKM., H. 506, A novel immunosuppressant isolated from a Streptomyces. I. Fermentation, isolation, and physico-chemical and biological characteristics, J. Antibiot. 40, 1249-1255. H. KLEINKAUF, VON DOHREN, (1987), BiosynH., thesis of peptide antibiotics, Annu. Rev. Microbiol. 41, 259-289. KLEINKAUF, VON DOHREN, (1990), NonriH., H. bosomal biosynthesis of peptide antibiotics, Eur. J. Biochem. 192, 1-15. C. KO, S. Y., DALVIT, (1992), Conformation of cyclosporin A in polar solvents, Int. J. Pept. Protein Res. 40, 380-382. KOBEL,H., TRABER, (1982), Directed biosynR. thesis of cyclosporins, Eur. J. Appl. Microbiol. Biotechnol. 14, 237-240. VOGES,R.(1983), ConKOBEL, LOOSLI, R., H., H. tribution to knowledge of the biosynthesis of cyclosporin A, Experientia 39,873-876.
HOPF,U., NEUHAUS, KONIG, BAUDITZ, P., V., J., KUTHER, S., SCHMIDT, A., STEFFEN, R., C. BLUMHARDT, BECHSTEIN, O., NEUG., W. R. HAUS, ET AL. (1992), Orthotopic liver transplantation in hepatic cirrhosis: on the problem of infection of the transplant with persistent hepatitis viruses, Z. Gastroenterol. 30,576-582. HOPWOOD, A., SHERMAN, H. (1990), MolecD. D. ular genetics of polyketides and its comparison to fatty acid biosynthesis, Annu. Rev. Genet. 24, 37-66. Y., T., HORI,K., YAMAMOTO, MINETOKI, KUROTS., K., su, T., KANDA, MIURA, OKAMURA, M., FURUYAMA, SAITO, Y. (1989), Molecular J., cloning and nucleotide sequence of the Gramicidin S synthetase 1 gene, J. Biochem. 1%, 639645. HUNTER, KARIN, (1992), The regulation of T., M. transcription by phosphorylation, Cell 70, 777789. IZQUIERDO, CANTRELL, A. (1992), T-cell M., D. activation, Trends Cell Biol. 2 (9), 268-271. JANEWAY, A., ROJO,J., SAIZAWA, DIANC. K., P., ZANI,U., PORTOLES, TITE, J., HAQUE,S., JONES,B. (1989), The Co-receptor function of murine CD4, Immunol. Rev. 109,77-92. JIN,Y., ALBERS, W., LANE, S., BIERER, M. W. B. S. S. E., SCHREIBER, L., BURAKOFF, J. (1991), Molecular cloning of a membrane-associated human FK506- and rapamycin-binding protein, FKBP-13, Proc. Natl. Acad. Sci. USA 88, 66776681. P. J. JORDAN, M., SPENCER, B. (1993), The biosynthesis of tetraketides: Enzymology, mechanism, and molecular programming, Biochem. SOC. Trans. 21,222-228. KAHAN, D. (1992), Immunosuppressive therapy, B. Curr. Opin. Immunol. 4, 553-560. KALLEN, WALKINSHAW, D. (1992), The XJ., M. ray structure of a tetrapeptide bound to the active site of human cyclophilin A, FEBS Lett. 300, 286-290. KALLEN,J., SPITZFADEN, ZURINI,M. G., C., WIDER, WIDMER, WUTHRICH, WALG., H., K., KINSHAW, M. D. (1991), Structure Of human Cyclophilin and its binding site for cyclosporin A determined by X-ray crystallography and NMR spectroscopy, Nature 353, 276-279. KATZ, L., DONADIO, (1993), Polyketide syntheS. sis - prospects for hybrid antibiotics, Annu. Rev. Microbiol. 47, 875-912. KE, H. M. (1992), Similarities and differences between human cyclophilin A and other beta-barre1 structures. Structural refinement at 1.63 A resolution, J. Mol. Biol. 228, 539-550. KE, H. M., ZYDOWSKY, D., LIU,J. WALSH, T. L. C. (1991), Crystal structure of recombinant human
7 References
587
KOECK, KESSLER, SEEBACH, THALER, LEE,J. P., DUNLAP, RICH, H. (1990), SyntheM., H., D., B. D. A. (1992), Novel backbone conformation of cysis and immunosuppressive activities of conforclosporin A: The complex with lithium chloride, mationally restricted cyclosporin lactam anaJ. Am. Chem. SOC. 114, 2676-2686. logues, Int. J. Pept. Protein Res. 35, 481494. KOFRON, J. L., KUZMIC, KISHORE, COLON LEITNER, WEBER,G. (1994), Disruption of the P., V., E., BONILLA, RICH, D. H. (1991), DeterminaE., cyclosporin synthetase gene of Tolypocladium tion of kinetic constants for peptidyl prolyl cisniveum, Curr. Genet. 26, 461467. trans isomerases by an improved spectropho- LEITNER, SCHNEIDER-SCHERZER, E., E., SCHOERtometric assay [published erratum appears in GENDORFER, K., WEBER, G. (1994), CYClOSpOBiochemistry 1991 Nov 5; 30 (44): 108181, Biorin synthetase, European Patent Application chemistry 30, 6127-6134. 578616. KRAETZSCHMAR, KRAUSE, MARAHIEL, J., M., M. LEIVA,M. C., LYTTLE, C. R. (1992), Leukocyte A. (1989), Gramicidin S biosynthesis operon chemotactic activity of FKBP and inhibition by containing the structural genes grsA and grsB FK506, Biochem. Biophys. Res. Commun. 186, has an open reading frame encoding a protein 1178-1 183. homologous to fatty acid thioesterase, J. Bacteri- LIU,Y., LINSLEY, S. (1992), Costimulation of TP. 01. 171, 5422-5429, cell growth, Curr. Opin. Immunol. 4, 265-270. KRONKE, M., LEONARD, J., DEPPER,J. M., LIU,J., ALBERS, W., CHEN, SCHREIBER, W. M. C., S. ARYA,F., WONG-STAAL, GALLO,R. C., F., L., WALSH,C. T. (1990), Cloning, expression, WALDMANN, A., GREENE, C. (1984), CyT. W. and purification of human cyclophilin in Escherclosporin A inhibits T-cell growth factor gene ichia coli and assessment of the catalytic role of expression at the level of mRNA transcription, cysteines by site-directed mutagenesis, Proc. Proc. Natl. Acad. Sci. USA 81, 5214-5218. Natl. Acad. Sci. USA 87, 2304-2308. LALAND, G., ZIMMER, S. T.-L. (1973), The protein LIU,J., FARMER, D. JR., LANE,W. S., FRIEDJ. thiotemplate mechanism of synthesis for the MAN, J., WEISSMAN, SCHREIBER, L. (1991). I., S. peptide antibiotics produced by Bacillus brevis, Calcineurin is a common target of cyclophilinEssays Biochem. 9, 31-42. cyclosporin A and FKBP-FK506 complexes, Cell 66,807-815. LANDSCHULTZ,W. H., JOHNSON, P. F., MCKNIGHT, L. (1989), The DNA binding do- LOOSLI, R., KESSLER, OSCHKINAT, WES. H. H., H., main of the rat liver nuclear protein C/EBP is BER,H. P., PETCHER, (1985), The conformaT. bipartite, Science 243, 1681-1688. tion of cyclosporin A in the crystal and in solution, Helv. Chim. Acta 68, 682-704. LAWEN, TRABER, (1993), Substrate specificA., R. ities of cyclosporin synthetase and peptolide LUBAN, BOSSOLT, L., FRANKE, K., KALJ., K. E. SDZ 214-103 synthetase. Comparison of the subG. PANA, V., GOFF, S. P. (1993), Human immustrate specificities of the related multifunctional nodeficiency virus type 1 Gag protein binds to polypeptides, J. Biol. Chem. 268,20452-20465. cyclophilins A and B, Cell 73, 1067-1078. LAWEN, ZOCHER, (1990), Cyclosporin syn- LYNEN, (1980), On the structure of fatty acid A., R. F. thetase. The most complex peptide synthesizing synthase of yeast, Eur. J. Biochem. 112, 431multienzyme polypeptide so far described, J. 442. Antibiot. 42, 1283-1289. MACCABE, P., VAN LIEMPT, PALISSA, A. H., H., LAWEN, TRABER, GEYL,D., ZOCHER, A., R., R., RIACH,M. B. R., PFEIFER, VON DOHREN, E., H. KLEINKAUF, (1989), Cell-free biosynthesis of H., KINGHORN, R. (1991), S(L-a-aminoadiJ. new cyclosporins, J. Antibiot. 42, 1283-1289. pyl)-L-cysteinyl-D-valine synthetase from Aspergillus nidulans, J. Biol. Chem. 266, 12646LAWEN, TRABER, GEYL, (1991), In vitro A., R., D. biosynthesis of [Thr2,Leus,D-Hiv8,Leu'o]-cy- 12954. closporin A, a cyclosporin-related peptolide, MARTEL,R. R., KLICIUS, GALET,S. (1977), InJ., with immunosuppressive activity by a multienhibition of the immune response by rapamycin, a new antifungal antibiotic, Can. J. Physiol. Pharzyme polypeptide, J. Biol. Chem. 266, 1556715 570. macol. 55,48-51. LAWEN, D I ~ M A NJ., SCHMIDT, RIESNER, MASON,J. (1989), Pharmacology of cyclosporine A., N, B., (Sandimmune) VII. Pathophysiology and toxiH. D., KLEINKAUF, (1992), Enzymatic biosyncology of cyclosporine in humans and animals, thesis of cyclosporin A and analogues, Biochimie 74,511-516. Pharmacol. Rev. 42,423-434. P. B. LAWEN, TRABER, REUILLE, PONELLE, MCCAFFREY, G., PERRINO, A,, SODERLING, A., R., R., T. R., RAO,A. (1993), NF-ATp, a T lymphocyte M. (1994), In vitro biosynthesis of ring-extended DNA-binding protein that is a target for calcicyclosporins, Biochem. J. 300,395-399.
588
neurin and immunosuppressive drugs, J. Biol. PATCHETT,A. A., TAUB, D., HENSENS,0. D., GOEGELMAN, T., YANG,L. H., DUMONT, R. F. Chem. 268,3747-3752. L., J., PETERSON, SIGAL,N. H. (1992). AnaMIKOL,V., KALLEN,J., PFLUEGL,G., WALKINlogues of cyclosporin A modified at the D-Ala8 SHAW, M. D. (1993), X-ray structure of a monposition, J. Antibiot. 45, 94-102. omeric cyclophilin A-cyclosporin A crystal comG., J., T., plex at 2.1 8, resolution, J. Mol. Biol. 234, 1119- PFLUEGL, KALLEN, SCHIRMER, JANSONM. M. IUS,J. N., ZURINI, G., WALKINSHAW, D. 1130. (1993), X-ray structure of a decameric cyclophilMIKOL, V., KALLEN,J., WALKINSHAW, D. M. in-cyclosporin crystal complex, Nature 361, 91(1994~), X-ray structure of a cyclophilin B/cy94. closporin complex: comparison with cyclophilin L. A and delineation of its calcineurin binding do- PRICE,E. R., ZYDOWSKY, D., JIN,M. J., BAKER, main. Proc. Natl. Acad. Sci. USA 91, 5183C. H., MCKEON,F. D., WALSH,C. T. (1991), 5186. Human cyclophilin B: A second cyclophilin gene encodes a peptidyl-prolyl isomerase with a signal MIKOL, V., KALLEN,J., WALKINSHAW, D. M. sequence, Proc. Natl. Acad. Sci. USA 88, 1903(1994b), The X-ray structure of (MeBm2t)'-cy1907. closporin complexed with cyclophilin A provides an explanation for its anomalously high immu- PRICE,D. J., GROVE, R., CALVO, AVRUCH, J. V., nosuppressive activity, Protein Eng. 7 (5), 597J., BIERER, E. (1992), Rapamycin-induced inB. 603. hibition of the 70-kilodalton S6 proteinkinase, Science 257,973-917. R. MORRIS, E. (1993), New small molecule immunosuppressants for transplantation: review of es- QUESNIAUX, F. J., SCHREIER, H., WENGER, V. M. sential concepts, J. Heart Lung Transplant. 12, R. M., HIESTAND, C., HARDING, W., VAN P. M. S275-S286. REGENMORTEL, H. V. (1987), Cyclophilin M. binds to the region of cyclosporine involved in NERI, P., GEMMECKER, ZYDOWSKY, D., G., L. its immunosuppressive activity, Eur. J. Immunol. WALSH, T., FESIK,S. W. (1991), NMR studies C. 17, 1359-1365. of [U-(13)C]cyclosporin A bound to human cyclophilin B, FEBS Lett. 290, 195-199. QUESNIAUX, F. J., SCHREIER, H., WENGER, V. M. R. M., HIESTAND, C., HARDING, W., VAN P. M. O'HAGAN,D. (1991), The Polyketide Metabolites. REGENMORTEL, H. V. (1988), Molecular M. (MELLOR,J., Ed.). Chichester: Ellis Horwood characteristics of cyclophilin-cyclosporine interSeries in Organic Chemistry. action, Transplantation 46, 23-28. J., R. O'KEEFE,S. J., TAMURA, KINCAID, L., TocV. S., C., CI, M. J., O'NEILL, E. A. (1992), FK-506- and QUESNIAUX, F. J., WEHRLI, WIOLAND, SCHULER, SCHREIER, H. (1994), Effects W., M. CsA-sensitive activation of the interleukin-2 of rapamycin on hematopoiesis, Transplant. promoter by calcineurin, Nature 357, 692-694. Proc. 26, 3135-3140. OFFENZELLER, Su, Z., SANTER, MOSER, M., G., D., H., TRABER,R., MEMMERT, SCHNEIDER- RICH,D. H., SUN,C. Q., GUILLAUME, DUNK., LAP, B., EVANS, A., WEBER,A. E. (1989), D. SCHERZER, (1993), Biosynthesis of the unE. usual amino acid (4R)-4-[(E)-2-buteny1]-4-me- Synthesis, biological activity, and conformationa1 analysis of (25, 3R, 4s)-MeBmt '-cyclosporin, thyl-L-threonine of cyclosporin A. Identification a novel 1-position epimer of cyclosporin A, J. of 3(R)-hydroxy-4(R)-methyl-6(E)-octenoic acid Med. Chem. 32, 1982-1987. as a key intermediate by enzymatic in vitro synthesis and by in vivo labeling techniques, J. Biol. ROBINSON, (1991), Polyketide synthase comA. Chem. 268,26 127-26 134. plexes: Their structure and function in antibiotic PAPAGEORGIOU, FLORINETH, FRENCH, C., A., R. biosynthesis, Philos. Trans. R. Soc. London B 332,107-114. (1994), Calcineurin has a very tight-binding pocket for the side chain of residue 4 of cyclo- ROSEN,M. K., STANDAERT, F., GALAT,A., NAR. sporin, Bioorg. Med. Chem. Lett. 4, 267-272. KATSUKA, M., SCHREIBER, s. L. ( l w ) , InhibiPARK, B., MEIER,G. P. (1989), A semisynthetic S. tion of FKBP rotamase activity by immunosupapproach to olefinic analogues of amino acid pressant FK506: twisted amide surrogate, Science 248, 863-866. one (MeBmt) in cyclosporin A, Tetrahedron Lett. 30,42154218. ROSENWIRTH, BILLICH, DATEMA, DoB., A., R., PARK,S. T., ALDAPE,R. A., FUTER, DECENO., NATSCH, P., HAMMERSCHMID, HARRISON, F., zO, M. T., LIVINGSTON, J. (1992), PPIase caD. R. K., HIESTAND, C., JAKSCHE, MAYER, P. H., talysis by human FK506-binding protein proP., PEICHL,P., QUESNIAUX, F. J., SCHATZ, V. ceeds through a conformational twist mechaF., WENGER, M., WOLFF,B., ZENKE, ZuR. G., nism, J. Biol. Chem. 267, 3316-3324. RINI, M. G. (1994), Inhibition of HIV-1 replica-
7 References
589
tion by SDZ NIM 811, a non-immunosuppressive cyclosporin A analogue. Antimicrob. Agents Chemother. 38, 1763-1772. RYFFEL,B. (1989), Pharmacology of cyclosporine. VI. Cellular activation: Regulation of intracellular events by cyclosporine, Pharmacol. Rev. 41, RYFFEL,B., GOETZ, U., HEUBERGER, (1982), B. Cyclosporin receptors on human lymphocytes, J. SANGLIER, J., TRABER, BUCK,R. H., HOFJ. R., MA, H., KOBEL, H. (1990), Isolation of (4R)-4-[(E)-2-butenyl]-4-methyl-~-threonine, the characteristic structure element of cyclosporins, from a blocked mutant of Tolypocladium inflatum, J. Antibiot. 43,707-714. SCHLUMBOHM, STEIN,T., ULLRICH, VATW., C., ER, J., MARAHIEL, A., KRUFT,V., WITTM. B. MANN-LIEBOLD, (1991), An active serine is involved in covalent substrate amino acid binding at each reaction center of gramicidin S synthetase, J. Biol. Chem. 266,23135-23 141. SCHMIDT, RIESNER,D., LAWEN, B., A., KLEINKAuF, H. (1992), Cyclosporin synthetase is a 1 4MDa multienzyme polypeptide. Re-evalua. tion of the molecular mass of various peptide synthetases, FEBS Lett. 307,355-360. SCHNEIDER, CHARARA, H., N., SCHMITZ,R., S., M. WEHRLI, MIKOL,V., ZURINI, G., QUESNIAUX, F. J., MOVVA,N. R. (1994), Human V. cyclophilin C: Primary structure, tissue distribution, and determination of binding specificity for cyclosporins. Biochemistry 33,8218-8224. SCHOENBRUNNER,R., MAYER, TROPSCHUG, E. S., G., N., F. M., FISCHER, TAKAHASHI, SCHMID, X. (1991), Catalysis of protein folding by cyclophilins from different species, J. Biol. Chem. 266,
Immunol. 129,1978-1982.
407-423.
SEEBACH, BECK,A. K., BOSSLER, G., GERD., H. BER, c., KO, s. Y., MURTIASHAW, W., NAEF, c. R., SHODA,S., THALER, WENGER,R. M. A., (1993), Modification of cyclosporine-A (CS) Generation of an enolate at the sarcosine residue and reactions with electrophiles, Helv. Chim. Acta 76,1564-1590. %!=TON, M., CAMPBELL, A. (1991),The role B. M. of tyrosine protein phosphorylation in lymphocyte activation, Annu. Rev. Cell Biol. 7 , 257SENN,H., WEBER, C., KOBEL,H., TRABER,R. (1991), Selective 13C-labellingof cyclosporin A, Eur. J. Biochem. 199,653-658. SHERRY, YARLETT, STRUPP,A., CERAMI, B., N., A. (1992), Identification of cyclophilin as a proinflammatory secretory product of lipopolysaccharide-activated macrophages, Proc. Natl. Acad. Sci. USA 89, 3511-3515. SIEKIERKA, J., HUNG,S. H., POE, M., LIN,C. S., J. SIGAL,N. H. (1989a),A cytosolic binding protein for the immunosuppressant FK506 has peptidyl-prolyl isomerase activity but is distinct from cyclophilin, Nature 341,755-757. SIEKIERKA,J., STARUCH, J., HUNG, H. Y., J. M. S. SIGAL,N. H. (1989b), FK-506, a potent novel immunosuppressive agent, binds to a cytosolic protein which is distinct from the cyclosporin Abinding protein, cyclophilin, J. Immunol. 143,
274.
SIGAL, H., DUMONT, J., DUREITE, P. L., SIEN. F. J. L., KIERKA, J., PETERSON, RICH,D. H., DUNM. M. LAP, B. E., STARUCH, J., MELINO, R., KOPRAK, L. ET AL. (1991),Is cyclophilin inS. volved in the immunosuppressive and nephrotoxic mechanism of action of cyclosporin A? J. Exp. Med. 173,619-628. 3630-3635. T. SCHREIBER, L. (1991),Chemistry and biology of SIMPSON, J. (1989), Biological chemistry. Part S. (11) biosynthesis, Annu. Rep. Prog. Chem. Sect. the immunophilins and their immunosuppressive B 85,321-351. ligands, Science 251,283-287. SMITH, J., EARL,A. J., TURNER, (1990),The D. G. SCHREIBER,. L., CRABTREE, R. (1992),The S G. multifunctional peptide synthetase performing mechanism of action of cyclosporin A and the first step of penicillin biosynthesis in PenicilFK506, Imrnunol. Today 13,136-142. lium chrysogenum is a 421073 Dalton protein SCOTT-CRAIG, S., PANACCIONE, G., PoJ. D. similar to Bacillus brevis peptide antibiotic synCARD,J.-A., WALTON,J. D. (1992), A large, thetases, EMBO J. 9,2743-2750. multifunctional cyclic peptide synthetase catalyzH., K. ing HC-toxin production in the plant pathogenic SODA,K., TANAKA, TANIZAWA,(1986),Pyridoxyl phosphate enzymes catalyzing racemizafungus Cochliobolus carbonum encoded by a tion, in: Vitamin B6, Pyridoxyl Phosphate: 15.7-kb open reading frame, J. Biol. Chem. 267, Chemical, Biochemical and Medical Aspects. 26044-26049. D., Coenzymes and Cofactors (DOLPHIN, POULSEEBACH, KO, S. Y., KESSLER, KOECK, D., H., M., O., REGGELIN, SCHMIEDER, WALKINSHAW, SON, R., AVRAMOVIC, Eds.), pp. 223-251. M., P., New York: Wiley & Sons. H. M. D., BOELSTERLI, J., BEVEC,D. (1991), C., Thiocyclosporins: Preparation, solution and SPITZFADEN, WEBER, H. P., BRAUN, W., KALLEN, WIDER, WIDMER, WALKINJ., G., H., crystal structure, and immunosuppressive activiSHAW, M. D., WUTHRICH, (1992),CyclospoK. ty, Helv. Chim. Acta 74,1953-1990.
1580-1583.
590
rin A-cyclophilin complex formation. A model based on X-ray and NMR data, FEBS Lett. 300, 291-300. STABEN, JENSEN, B., SINGER, POLLOCK, C., P. M., J., SCHECHTMAN, KINSEY,J., SELKER,E. M., (1989), Use of a bacterial hygromycin resistance gene as a dominant selectable marker in Neurospora crassa transformation, Fung. Genet. Newslett. 36, 79-81. STAMNES, A,, RUTHERFORD, L., ZUKER, M. S. C. S. (1992), Cyclophilins: A new family of proteins involved in intracellular folding, Trends Cell Biol. 2, 272-276. STEVENSON, STANWICK, L., DEMPSEY, M., T. M. P., LAMONICA, A. (1990), HIV-1 replication C. is controlled at the level of T cell activation and proviral integration, EMBO J. 9, 1551-1560. SWANSON, K., BORN, T., ZYDOWSKY, D., S. L. H. C. CHO, H., CHANG, Y., WALSH, T. RUSNAK, F. (1992), Cyclosporin-mediated inhibition of bovine calcineurin by cyclophilins A and B, Proc. Natl. Acad. Sci. USA 89,3741-3745. TAI,P., ALBERS, W., CHANG, FABER, E., M. H., L. SCHREIBER,. L. (1992), Association of a 59S kilodalton immunophilin with the glucocorticoid receptor complex, Science 256, 1315-1318. TAKAHASHI, HAYANO, SUZUKI, (1989), N., T., M. Peptidyl-prolyl cis-trans isomerase is the cyclosporin A-binding protein cyclophilin, Nature 337,473-475. TAKESHITA, ASAO,H., OHTANI, ISHII,N., T., K., KUMAKI, TANAKA, MUNAKATA, NAS., N., H., KAMURA, SUGAMURA, (1992), Cloning M., K. of the gamma chain of the human IL-2 receptor, Science 257,379-382. THERIAULT, LOGAN,T. M., MEADOWS, Y., R., E. T. Yu, L., OLEJNICZAK, T., HOLZMAN, F., SIMMER, L., FESIK,S. W. (1993), Solution R. structure of the cyclosporin Ncyclophilin complex by NMR, Nature 361, 88-91. THOMMEN-SCOTT, (1981), Antimalarial activity K. of cyclosporin A, Agents Actions 11,770-773. TOCCI, J., MATKOVICH, A., COLLIER, A., M. D. K. KWOK,P., DUMONT, J., LIN, S., DEGUDICIF. J. BUS, S., SIEKIERKA, J., CHIN,J., HUTCHINSON, N. I. (1989), The immunosuppressant FK.506 selectively inhibits expression of early T cell activation genes, J. Immunol. 143, 718-726. TRABER, HOFMANN, LOOSLI,H. R., PONR., H., A. ELLE, M., VON WARTBURG, (1987), Neue Cyclosporine aus Tolypocladium inflatum. Die Cyclosporine K-Z, Helv. Chim. Acta 70, 13-36. TRABER, HOFMANN, KOBEL,H. (1988), CyR., H., closporins - new analogues by precursor directed biosynthesis, J. Antibiot. 42, 591-597. TRABER, KOBEL,H., LOOSLI, R., SENN,H., R., H. ROSENWIRTH, B., LAWEN, A. (1994),
[MeIle4]cyclosporin,a novel natural cyclosporin with anti-HIV activity: Structural elucidation, biosynthesis, and biological properties, Antiviral Chem. Chemother. 5,331-339. TSANG, V., HODSON, M. E., YACOUB,M. H. (1992), Lung transplantation for cystic fibrosis, Br. Med. Bull. 48,949-971. TURGAY,K., KRAUSE, M., MARAHIEL,M. A. (1992), Four homologous domains in the primary structure of GrsB are related to domains in a superfamily of adenylate-forming enzymes, Mol. Microbiol. 6, 529-546. TWENTYMAN, (1992), Cyclosporins as drug P. R. resistance modifiers, Biochem. Pharmacol. 43, 109-117. VOLLENBROICH, KLUGE, B., DSOUZA, C., D., ZUBER, VATER,J. (1993), Analysis of a muP., tant amino acid-activating domain of surfactin synthetase bearing a serine-to-alanine substitution at the site of carboxyl thioester formation, FEBS Lett. 325, 220-224. VON WARTBURG, TRABER, (1988), CycloA., R. sporins. Fungal metabolites with immunosuppressive activities, Progr. Med. Chem. 25 (ELLIS, G. P., WEST,G. B., Eds.), pp. 1-33. Amsterdam: Elsevier Science Publishers, B.V., Biomedical Division. WALSH,C. T., BADET,R., DAUB,E., ESAKI,N., GALAKATOS, (1985), Bacterial alanine raN. cemases: Targets for antibacterial agents, Spec. Publ. - R. SOC.Chem. 55 (SCI-RSC Med. Chem. Symp. 3rd), 193-209. WEBER, C., WIDER, VON FREYBERG, TRAG., B., BER,R., BRAUN, WIDMER, WUTHRICH, W., H., K. (1991), The NMR structure of cyclosporin A bound to cyclophilin in aqueous solution, Biochemistry 30,65634574, WEBER, SCHOERGENDORFER, SCHNEIDERG., K., E. SCHERZER, LEITNER, (1994), The peptide E., synthetase catalyzing cyclosporin production in Tolypocladium niveum is encoded by a giant 45.8-kilobase open reading frame, Curr. Genet. 26,120-125. WECKERMANN, FUERBASS, MARAHIEL, R., R., M. A. (1988), Complete nucleotide sequence of the tycA gene encoding the tyrocidine synthetase 1 from Bacillus brevis, Nucleic Acids Res. 16, 11841. WENGER, M. (1983), Synthesis of cyclosporine, R. Helv. Chim. Acta 67, 502-525. WENGER, M. (1989), Pharmacology of cyclospoR. rin (Sandimmune). 11. Chemistry, Pharmacol. Rev. 41, 243-247. WENGER,R. M. (1986), Cyclosporine and analogues: Structural requirements for immunosuppressive activity, Transplant. Proc. 18, 213-218.
7 References
591
J. S. S. WENGER,R. M., PAYNE, G., SCHREIER, H. ZACK, A., ARRIGO, J., WEITSMAN, R., Go, T. M. A. S., HAISLIP, CHEN,I. S. Y. (1990), HIVA., (1986), Cyclosporine: Chemistry, structure-acentry into quiescent primary lymphocytes: motivity relationships and mode of action, Progr. lecular analysis reveals a labile, latent viral strucClin. Biochem. Med. 3, 157-191. ture, Cell 61, 213-222. WENGER,R. M., FRANCE,J., BOVERMANN, G., WALLISER, L., WIDMER, A., WIDMER,H. G., R. (1994), The 3D structure of a cyclosporin ana- ZENKE,G., BAUMANN, WENGER, M., HIESV. TAND,P. C., QUESNIAUX, F. J., ANDERSEN, logue in water is nearly identical to the cycloE., SCHREIER, H. (1993), Molecular mechaM. philin-bound conformation, FEBS Lett. 340, nisms of immunosuppression by cyclosporins, 255-259. Ann. N. Y. Acad. Sci. 685, 330-335. WITZEL, B. (1989), New cyclosporin analogues with modified C-9 amino acids, US Patent no. ZOCHER, NIHIRA,T., PAUL, E., MADRY,N., R., 4798823. PEETERS, H., KLEINKAUF, (1986), BiosynH. Xu, Q., LEIVA,M. C., FISCHKOFF, A., HANDS. thesis of cyclosporin A: Partial purification and SCHUMACHER, E., LYTTLE, C. R. (1992), R. properties of a multifunctional enzyme from ToLeukocyte chemotactic activity of cyclophilin, J. Iypocladium inflatum, Biochemistry 25, 550-553. Biol. Chem. 267, 11968-1 1971.
Biotechnology
l3 Secondary Products from Plant

Cell Cultures
JOCHEN BERLIN
Braunschweig, Federal Republic of Germany
1 Introduction 595 2 Some General Conclusions and Suggestions Based on the Present Biotechnological Impact of Plant Cell Cultures as Producers 596 3 Secondary Product Formation in Suspension Cultures 598 3.1 Products Accumulating at High or Good Levels in Suspension Cultures 598 3.1.1 Cinnamic Acid Derivatives 598 3.1.2 Naphthoquinones and Anthraquinones 601 3.1.3 Protoberberines and Benzophenanthridine Alkaloids 603 3.1.4 Monoterpene Indole Alkaloids 606 3.1.5 Anthocyanins and Betalains 609 3.1.6 Steroidal Compounds 611 3.1.7 Immunologically Active Polysaccharides 612 3.2 Products of Commercial Interest Accumulating in Traces or not at all in Suspension Cultures 612 3.2.1 Morphinan Alkaloids 613 3.2.2 Tropane Alkaloids 613 3.2.3 Quinoline Alkaloids 614 3.2.4 Antitumor Compounds 614 3.2.5 Cardiac Glycosides 615 3.2.6 Vanillin and Vanilla Aroma 615 4 Secondary Product Formation in Hairy Root Cultures 619 5 Plant Tissue Cultures as a Source of New Chemicals? 621 6 Biotransformations with Cultured Plant Cells 621 6.1 Arbutin 622 6.2 Biotransformation of Cardiac Glycosides 622 7 Metabolic Engineering of Secondary Pathways in Cultured Cells 623 7.1 Serotonin Biosynthesis in Peganum harmala 624 7.2 Affecting Nicotine Alkaloid Biosynthesis in Tobacco 624
594
13 Secondary Products from Plant Cell Cultures
7.3 Enhancing Scopolamine Production in Atropa belladonna 625 7.4 Does Genetic Engineering Improve the Potential of Plant Tissue Cultures as Producers of Interesting Metabolites? 626 8 Conclusions and Outlook 627 9 References 628
1 Introduction
595
1 Introduction
Higher plants produce a great variety of secondary metabolites, some of which are an indispensable source of commercially important compounds. These include pharmaceuticals (such as steroids, alkaloids, and glucosides), natural flavors, fragrances, dyes, and gums (such as natural rubber). As only a small portion of plants have as yet been analyzed for secondary products, it is not surprising that isolation and characterization of new compounds continue unabatedly. The array of assays for detecting biological activities of natural compounds has been enlarged and simplified, and has been paralleled by an increase in assay sensitivity. Thus, the screening of plant extracts for new biologically active compounds still seems to be a promising approach today (BALANDRIN al., 1985). et However, it should also be mentioned that the screening of some 40000 plants by the NCI (National Cancer Institute, USA) yielded only three really novel anticancer compounds (vincristine, vinblastine, and tax01). It is doubtful whether this result helps to attract more industrial investments to this area of plant research. Five Rauwolfia alkaloids with tranquilizing and antihypertensive activities and the three anticancer substances were the only plant-derived substances with therapeutic efficacy and utitility, proven beyond doubt (TYLER, 1988), which have been introduced to the American market during the last 40 years. All other plant-derived compounds used as drugs today have been known for a much longer time and their production is well established. Thus, a general concern of pharmacognosists is how to make plant analyses more efficient with respect to drug development (TYLER,1988). If the plant product under consideration cannot be synthesized chemically it has to be isolated from wild or field-grown plants. Consequently, plants synthesizing products in high demand are usually grown in large-scale plantations. Since the establishment of cell cultures of any plant species is nowadays routine in most cases, plant cell biomass can also be produced in huge fermentors. Since in addition, reasonable amounts of secondary me-
tabolites have been produced from some plant cell cultures it seems obvious to suggest that commercially interesting compounds might be produced in a factory type production in large bioreactors, such as those used in the production of microbial drugs. The advocates of the new technology have greatly overestimated the problems with agriculturally grown plant material while the biological, technical, legal, and financial problems of tissue culture technology have been generally underestimated. Interestingly, advocates of the tissue culture technology have come mainly from universities and public research centers, while in industry (perhaps with the exception of some Japanese industries) the interest was in general very meager worldwide. Thus, one can name immediately two reasons why the tissue culture technology was in a difficult position from the beginning: Firstly, the lack of new plant compounds with very special, extraordinary pharmacological characteristics for which it would be worth-while developing novel tissue culture processes; secondly, the difficulty of replacing an established and legally approved production process by a new technique. It is indeed difficult to replace a conventional, approved production process based on the extraction of fieldgrown plants by a plant tissue culture production process. This could only happen if the calculation of costs showed an extreme advantage for tissue cultures, and if the procedure of obtaining a new legal approval required for a product from a new source is not too time-consuming and expensive. If a new product is under consideration, the superiority of the new drug must be clear and evident to justify the long and expensive procedures needed nowadays for introducing a new pharmaceutical to the market. In view of these two obstacles we have to leave it to industry and biotechnological companies to decide whether they consider production by tissue cultures as an attractive alternative. For each product under consideration the decision may be quite different from company to company and from country to country. Updates of almost all the products found in plant tissue cultures were published by ELLIS (1988), CONSTABEL VASIL (1988), and and
596
BANTHORPE (1994). However, for many of the compounds mentioned in these reviews it is not readily clear whether they are of any biotechnological importance. Therefore, I will concentrate on some groups of compounds about which sufficient information is available in order to evaluate the biotechnological impact of plant cell cultures as producers. The main objective of my previous review (BERLIN, 1986) was to describe the experimental methods by which high yielding cultures were obtained and to demonstrate that the same techniques that could be applied to some pathways failed for others. Thus the aim was to provide the reader with a better understanding of what is possible today and what might become possible tomorrow. In this updated review I will expand on this aspect, explaining why certain compounds are easily produced in cultured cells while other metabolites have proved recalcitrant to all known techniques for improved production. As I will describe mainly the same groups of compounds as in 1986 in the first edition of Biotechnology, the extent of progress with respect to production levels during the last 10 years will become apparent. Some compounds have been replaced by others which are presently of more biotechnological interest; the new developments are included in this review. The possibility of altering production characteristics of pathways by genetic engineering opened a new area. This will be analyzed critically to whether it will improve the situation of tissue cultures as producers of commercially interesting compounds.
2 Some General Conclusions and Suggestions Based on the Present Biotechnological Impact of Plant Cell Cultures as Producers
PETERSEN and ALFERMANN (1993) have already described special features of plant cell
cultures and the various culture types and techniques in great detail in Vol. 1 of this multi-volume comprehensive treatise. Due to their recent review and those of other (PARR, 1989; CHARLWOODand RHODES, 1990; PAYNEet al., 1991; BUITELAAR al., 1992; et BANTHORPE, 1994; MISAWA, 1994) it is not necessary to repeat here the general aspects and specific characteristics of plant cell cultures and plant secondary metabolism out1986). lined in my previous article (BERLIN, However, before reviewing individual culture systems and documenting the state of the art, I would like to make some general statements which may be regarded as suggestions for future research programs and which gain supf port from the analyses o the individual systems. As indicated in Sect. 1, plant tissue cultures are presently not well accepted as a source of commercially interesting products. Indeed, the impact of tissue culture technology has not increased, but rather decreased despite all efforts and scientific progress. Some think that this technique is a futile approach; others believe that the scientific breakthrough has not yet been achieved for a true evaluation of the potential of plant cell cultures as producers. This prevents a larger engagement of companies. Thus, a critical review should analyze in which areas research efforts must be intensified and identify which approaches can be reduced today. The analysis of the field after 20 years of rather intensive research allows indeed some clear conclusions to be made about promising directions and about what should no longer be tried. Today it is possible to analyze the scientific usefulness of systems for improving the general standing of plant tissue cultures as tools of biotechnology. The term scientific usefulness means that biotechnologically relevant studies must not be restricted to cultures of plant species synthesizing commercially important compounds. Meaningful studies on model systems include studies on the regulation and expression of metabolic pathways and the possibilities of their manipulation. These are presently at least as important for the future of this field as working on so-called commercially attractive pathways. The seemingly most attractive pathways yielding com-
2 Some General Conclusions and Suggestions Based on the Present Biotechnological Impact
597
mercially useful products can presently not be expressed very well in morphologically undifferentiated cells, and thus such cultures are not very suitable for biochemical or molecular studies. In the past the efforts of most tissue culture groups were aimed at establishing highly productive plant cell cultures by conventional techniques. The product level was the most important goal. These studies included the initiation of many individual cultures from different explants of one or more plants, of one or more species or varieties. The cultures were grown on media with different phytohormone compositions. Screening or selection, as well as trials for enhancing productivity by media variation or by the use of inducer compounds have been part of a good program for optimizing a culture system. The application of the same techniques led in some cases to product levels in the range of g L-' within a few days, while in other cases amounts from zero to a few pg L were produced within a month. Some products were generally found at high levels while others were always found at low levels or even lacking in a culture. These findings were confirmed by laboratories from all over the world. Indeed, most tissue cultures of one plant species, after full adaptation to the culture conditions, have very similar production characteristics, independent of the laboratory where they were established. The extent of expression of a pathway is usually not restricted to one plant species. If, for example, a pathway is highly repressed in the cultures of one plant species, it is most likely that cultures of other plant species containing the same or related pathways will also be poor producers. If there are clear indications that a certain pathway is poorly expressed in cultured cells, it is unlikely that one can change this through the analyses of many more individually established cultures from various explants, by the application of screening and selection, by media variation, or by the use of elicitors (BERLIN, 1988). Thus, it does not make sense to waste time with such poorly producing cultures, unless new approaches can be applied. There are some reports in the literature claiming production characteristics which seem to contradict all other experi-
-'
ences. If this is a biotechnologically relevant finding, e.g., a true variant line, similar lines can easily be isolated by other researchers by employing a corresponding screening, selection, or an extended culture initiation program (BERLIN,1988). If seemingly unique lines with surprisingly good or different production characteristics cannot be detected by other groups despite all efforts it is certainly a transient trait, not useful for any biotechnological purposes. Important and stable production improvements are usually confirmed by independent laboratories. As pointed out, some pathways are spontaneously well expressed in cultured cells allowing product formation rates in the range of g L-' in a rather short culture period. A prerequisite for such biotechnologically relevant levels is that product formation occurs in rapidly growing cells, parallels growth, or is easily induced in a production medium. If a well expressed product is of commercial interest (shikonin, berberines), industry can substantially improve the production rates in shake flasks or small fermentors by optimizing all process parameters. If industry has taken over a process, it does not make much sense for research groups to continue with that culture system in shake flask systems or small bioreactors if the only aim is to demonstrate that another medium or elicitor, or an altered fermentation protocol improves productivity. Unfortunately, most products found at biotechnologically relevant levels in cultured cells are not of great commercial importance. Thus, it is even more questionable to spend too much time on further productivity improvements for such compounds. Nowadays, the main question should be whether a previously established product expression level, or its manipulation by changing the culture conditions, allows meaningful biochemical and molecular studies. Without knowledge of the enzymes involved in the pathways, without recognizing the rate-limiting steps within complex and branched pathways, and without identifying the factors controlling the often organ-specific expression of pathways, it is unlikely that plant tissue culture based production processes (especially in view of the general obstacles discussed in Sect. 1) will more often be considered as an
598
attractive alternative. From this point of view, all research which aims to elucidate specific or general features of pathway regulation with the ultimate goal of manipulating pathway expression is important in keeping the tissue culture technology alive. Thus, it is hoped that more and more biotechnologically orientated groups use their established tissue culture systems for such purposes during the next decade.
has been gathered for an evaluation. Subsequently, I will give an overview of the seemingly commercially most important pathways which have remained recalcitrant to improvements despite all efforts. For information about pathways not mentioned here the reader is referred to 34 special reviews in the book of CONSTABEL VASIL(1988) which coand vers nearly all groups of secondary compounds found in plant tissue cultures.
3 Secondary Product Formation in Suspension Cultures

Well-growing cell suspension cultures have been regarded for a long time as the only relevant system for biotechnological production processes. Most recently, rapidly growing hairy root cultures with their root-specific production characteristics have also been grown in larger bioreactors (CURTIS,1993). There are also a few groups who believe that immobilized plant cells are better suited for plant cell production processes than freely suspended cells (TANAKA, 1994). However, only cell suspension cultures have been scaled up to volumes of industrial relevance. For example, suspension cultures of Echinuceu purpureu have successfully been grown in bioreactors of up to 75 m3 (WESTPHAL, 1990). Thus, without a doubt they remain the most attractive system from a technological point of view. The disadvantage of rapidly growing, morphologically undifferentiated callus cultures and cell suspensions is that in such a cell state only a small part of the biosynthetic potential is expressed. Nevertheless, some products are found in suspension cultures at levels which exceed those of the differentiated plant. The few pathways which are extremely well expressed in suspension cultures will be described first, followed by a few examples of pathways which yield reasonable product levels. For the latter group I have chosen compounds of at least some biotechnological interest and about which sufficient information
3.1 Products Accumulating at High or Good Levels in Suspension Cultures 3.1.1 Cinnamic Acid Derivatives
Esters and amides of cinnamic acids, mainly of caffeic acid, form a major, widespread group of phenylpropanoid metabolites in plants. This group of compounds accumulates spontaneously, often at high levels, in cell suspension cultures. Two groups independently reported the accumulation of high levels of rosmarinic acid in Coleus blurnei (Fig. 1) (RAZZAQUEand ELLIS, 1977; ZENK al., et 1977a). The cultures were initiated on the widely used B5 medium with 2,4-D and kinetin as phytohormones. These cultures were maintained in suspension for several years without loosing their capacity for synthesizing and accumulating rosmarinic acid. RAZZAQUE and ELLIS (1977) reported the accumulation of 81 1% rosmarinic acid (corresponding to 1.2-1.5gL-') on the B5 growth medium. The synthesis of rosmarinic acid can be stimulated by increasing the sucrose levels to 5 4 % . Under these conditions, the content of rosmarinic acid increased up to 15% (ZENK et ai., 1977a). The specific values exceed those of the various organs of the plants by a f factor o 5. From their experiments in a volume of 25 mL ZENK'S group calculated a yield of 3.6 g L rosmarinic acid within 13 days (0.3 g L-l d-'). However, their first attempts at scaling up rosmarinic acid production in a 3 0 L airlift bioreactor showed that shake flask experiments are not necessarily transferable to larger-scale fermentations
-'
599
tures. This production level justifies a serious consideration of establishing a culture process as an alternative to harvesting from lower producing plants. However, the pharmacological efficacy of rosmarinic acid was evidently not high enough to encourage the Nattermann Company to further continue work on this compound. Indeed, the company has givRosmarinic acid en up all previous activities in plant tissue culColeus blumei tures, and presently there seems to be no Anchusa officinalis commercial interest in rosmarinic acid. Returning to the discussion in Sect. 2, which of the points raised are applicable to rosmarinic acid? (1) Rosmarinic acid belongs to the group of compounds which is spontaneously accumulated at reasonably high to high levels in typical plant cell culture media. Verbascoside R = Rhamnose It is not only produced at high levels by C. Syringa vulgaris blumei but also by several other plant species which are able to biosynthesize rosmarinic 0 et acid (WHITAKER al., 1984). Cultures of Anchusa officinalis spontaneously accumulated 6% rosmarinic acid (DE-EKNAMKUL and ELLIS, 1984). (2) The production of rosmarinic acid can be greatly improved by very Caffeoyl putrescine and simple media variation (DE-EKNAMKUL Nicotiana tabacum ELLIS, 1985a, b). (3) As with most spontaFig. 1. Caffeic acid derivatives accumulating at very neous high producer lines production is rathhigh levels in cultured cells (see text). er stable or can easily be re-established by controlled culture conditions. The fact that high levels of rosmarinic acid since productivity dropped to 15% of the can be easily produced by suitable tissue culshake flask experiments. As rosmarinic acid tures was well established in 1985. The hope has recently been shown to have good anti- for a commercial process was abandoned two phlogistic activity, the Nattermann Company, or three years later. Thus, even higher proCologne, investigated this culture further. duction levels than those reported by ULThey established new lines of C. blumei which BRICH et al. (1985) will not change the situaspontaneously synthesized rosmarinic acid tion. A recent report that an optimized mewith yields of 200 mg L-' on the growth medium resulted in product levels of 6.4 g L-' in et dium. This level could be enhanced 20-fold by a culture of Salvia officinalis (HIPPOLYTE transferring the cells to a simple production al., 1992) is thus only a strong confirmation of medium with an optimized concentration of previous findings. In terms of biotechnologsucrose as the main stimulator for maximum ical relevance, it is clear that the importance production (ULBRICH al., 1985). As ex- of rosmarinic acid for the field is not so much et pected from the experience with microbial a question of product levels. More important fermentations, these yields were further in- is how this system is used to determine at the creased in 30 L airlift or stirred reactors up to molecular level why this pathway is so well 21% or 5.6 g L-'. The output of rosmarinic expressed and why, for example, sucrose (UL- sometimes acts as a strong inducer. Media acid was thus raised to 0.93 g L d variation experiments aimed elucidating the BRICH et al., 1985). This was at that time the highest amount of a defined secondary me- physiological or biochemical basis of obtabolite ever produced with plant cell cul- served production increases are of course also
COOH
-' -'
600
still important. Thus, it has recently been been confirmed. Cultures of Hydrophifa erecet noted that the effect of sucrose depends on tu (HENRY al., 1987) or Leucosceptrum juet the carbohydrate level in the medium at the ponicum (INAGAKI al., 1991) yielded prodtime when phosphate limitation occurs uct levels in the range of 2 g L-' without op(GERTLOWSKI and PETERSEN,1993). This timization. The analyses of cultures of several observation might explain why not all rosmar- other plant species at the callus level indicate inic acid cell cultures react to the same extent that many more highly effective systems for to increased sucrose supply. At least two the production of verbascoside can be estabet groups have used cultures of A. officinalis and lished (DELL et al., 1989; INAGAKI al., C. bfumei for identifying all enzymes involved 1991). However, there is currently no comin rosmarinic acid biosynthesis (MIZUKAMI mercial interest in the production of these and ELLIS,1991; PETERSEN al., 1993) and compounds by tissue cultures. Though the et some of these enzymes have been purified so pathway is well expressed in cultured cells it that cloning of the corresponding genes could has not yet been used for biochemical and now proceed. Culture systems expressing a regulatory studies. biosynthetic pathway quite well under all culThe presence of various hydroxycinnamoyl ture conditions may not be very suitable for putrescines (Fig. 1) in cultured cells of Nicothe identification of regulatory factors con- tiunu tabacum was first reported by MIZUSAtrolling expression. Recently, it has been KI et al. (1971). That these compounds have shown that rosmarinic acid and the enzymes received some more attention during the last involved in its biosynthesis can be induced in years is due to accidental finding. When seLithospermum erythrorhizon and Orthosi- lecting the p-fluorophenylalanine resistant phon aristutus by elicitors such as yeast and cell line TX4 from a widely distributed XD methyl jasmonate from nearly zero up to ca. (TX1) line (N. tubucum cv. Xanthi) PALMER 1.5% of dry mass (MIZUKAMI al., 1993; Su- and WIDHOLM et (1975) noted that the levels of MARYONO et al., 1991). Identification of the phenolics were increased manifold in TX4 factors which allow spontaneous overproduc- cells. The phenolics were identified as hytion of rosmarinic acid in cultures of some droxycinnamoyl putrescines with caffeoyl puplant species, and clarification of why other trescine as main component (Fig. 1) (BERLIN plant cells require inducer compounds will et al., 1982). TX1 cells accumulated between hopefully provide some useful hints for ma- 0.6-1 % hydroxycinnamoyl putrescines on the nipulating pathway controls. growth medium, while TX4 cells contained up Another caffeoyl derivative is verbascoside to 10% of these compounds on a dry mass ba(1983) established sis. This system has been used in two direc(acteoside) (Fig. 1). ELLIS various lines of Syringa vulgaris on B5 me- tions: optimization of product formation and dium, all of which are produced spontaneous- comparison of the biochemical differences ly high levels of up to 1.4 g L - verbascoside. which lead to the different productivities. Rapidly growing suspension cultures of S. Growth limiting conditions (e.g., phosphate vulgaris were found to contain a higher spe- limitation) stimulated product formation cific content of verbascoside (15%) than cal- (SCHIELet al., 1984a). It is a frequently oblus cultures ( 5 4 % ) (ELLIS1983, 1985). Due served phenomenon that growth and secondto reports that verbascoside is a biologically ary metabolism are countercurrent processes active compound with antibacterial, antiviral, in cultured cells. Thus, even the formation of antihypertensive, and immunosuppressive products accumulating at high levels on the properties, the interest in verbascoside-pro- growth medium can often be strongly enducing tissue cultures has increased, especial- hanced by growth limiting conditions. The nely because plants generally contain only low gative effect of accumulated phosphate on the amounts of this compound (see literature synthesis of hydroxycinnamoyl putrescines cited by INAGAKI al., 1991). The initial ob- suggested that the employment of fedbatch et servation of ELLIS that verbascoside belongs fermentation would give highest yields. The to the group of compounds whose production productivity of the high producing variant is favored under cell culture conditions has TX4 was indeed increased to 1.5 g L-' by a
'
601
70 L fed-batch fermentation with phosphate et as a limiting nutrient (SCHIEL al., 1984b). Shake flask and batch fermentation of TX4 et cells usually yielded 0.8-1.2 g L (BERLIN al., 1982). The yield of TX1 cells was increased by the fedbatch techniques from 160200mgL- to 300-400mgL-. This shows the importance of using the best possible line for product optimizations. TX4 cells were the first biochemically selected variant line which showed overproduction of secondary metabolites and were thus interesting for biotechnological studies. It is also noteworthy that this highly productive variant line has maintained its production potential for more than et 15 years (MEURER-GRIMES al., 1989). 1986) no furSince the last review (BERLIN, ther studies regarding production improvements of hydroxycinnamoyl putrescines have been performed. Though hydroxycinnamoyl putrescines are of no commercial interest the results obtained with TX4 cells during product optimization stimulated the analysis of other culture systems. A biochemical comparison of the low and high producing tobacco lines showed that enhanced hydroxycinnamoyl putrescine formation was due to distinctly enhanced activities of biosynthetic enzymes providing the cinnamoyl and amine moieties in TX4 cells (BERLIN al., 1982) while the activity of the et conjugating enzyme was similar in TX1 and et TX4 cells (MEURER-GRIMES al., 1989). Though the biochemical comparison of parent and variant lines provided some clues as to the biochemical requirements of overproducing lines, we did not analyze the system at the molecular level. One reason for this was the finding that not only the hydroxycinnamoyl putrescine pathway but also other enzyme activities such as tyrosine decarboxylase (WALKER al., 1986) were altered in the p et fluorophenylalanine resistant cell line. Another reason was the fact that not only one but several enzyme activities should be coenhanced in order to improve hydroxycinnamoyl putrescine production by genetic techniques. It has recently been shown that phenylalanine ammonia lyase is the rate-determining step in phenyl propanoid biosynthesis (BATE et al., 1994). Overexpression of this enzyme in transgenic tobacco callus, however,
did not enhance the content of the major compound, the flavonoid rutin, but instead increased the levels of many other phenolics which were only present in trace amounts in wild-type callus (BATE et al., 1994).
3.1.2 Naphthoquinones and Anthraquinones

Quinones comprise a large group of secondary metabolites that are widely distributed in the plant kingdom. Since they are colored and, therefore, visible compounds, they have been a favored target of tissue culture research. Naphthoquinones and anthraquinones sometimes accumulate in cultured cells at levels far exceeding the amounts found in the intact plant. Some of these structures represent the active components of drugs. They are also important as natural dyes. The red shikonin pigments of the cork layer of the roots of Lithospermum erythrorhizon are derivatives of 1,4-naphthoquinones (Fig. 2). These compounds have been used medicinally in Japan for the treatment of burns and skin disease and are now mainly used as a dye for lipsticks and for staining
OH 0
Shikonins R = H, or aliphatic acids Lithospermum erythrorhizon

0
mo-R
CHzOH 0 OH
Anthraquinones (e.g. Lucidin primveroside, R = Glucose-Xylose) Morinda citrifolia Galium mollugo

Fig. 2. Naphthoquinones and anthraquinones accumulating at very high levels in cultured cells (see text).
602
silk. The plants have to be grown for 3-4 individual shikonins was different (FUJITA, years before a yield of 1-2% shikonin is 1988). The altered ratio of the various shikonachieved in the roots (FUJITA,1988). The to- in derivatives in the cell cultures is no probtal amount of Lithospermum roots used each lem if the mixture of compounds is to be used year in Japan is 10000 kg. From this an an- for cosmetic purposes or as a dye. However, if nual demand of 150 kg shikonins can be cal- one wants to replace a known approved plant culated. As the plant cannot be grown in drug by a tissue culture extract extensive evalcommercial quantities in Japan it has to be uation of the pharmacological properties and imported from Korea and China. It is often equivalence studies are required. Such studies argued by Japanese scientists in industry (Ko- have recently been initiated for L. erythrorhizon cultures by the Mitsui Company (OZAKI MAMINE et al., 1991) that due to the geoet graphic situation the indigenous supply of et al., 1990; SUZUKI al., 1991). There are scattered reports in the literature plant material is not sufficient, and that plant tissue culture technology is, therefore, per- of increased shikonin production in L. eryhaps a more attractive alternative in Japan throrhizon cultures by in situ extraction, elicithan elsewhere. This would explain the com- tation, or media variation. Though the obmercial production of shikonins from L. ery- served effects are well demonstrated their fithrorhizon cell suspension cultures by the nal levels remain far below those reported by Mitsui Company. Callus cultures of L. ery- FUJITAet al. Such studies would deserve throrhizon were found to accumulate shikon- more attention if the production levels at the in derivatives (TABATA al., 1974). By re- Mitsui Company could additionally be imet peated analytical screening over a period of proved by the newly recommended techtwo years two highly productive strains con- niques. This remark seems to be especially taining 20-fold increased levels of 1mg g valid in view of the report of the Mitsui Comfresh mass (ca. 10 mg g -' dry mass) were iso- pany that two-phase cultures did not improve lated (MIZUKAMI al., 1978). FUJITAet al. productivity of their high yielding line (DENO et (1981a, b) investigated the effects of all media et al., 1987). In general, naphthoquinones and constituents on growth and production. They benzoquinones seem to belong to the groups developed a production medium yielding of compounds which might readily be formed 1.4 g L-' shikonin derivatives within 23 d or in cultured cells of various plant species (Fuet 12% on a dry mass basis. Combining the two KUI et al., 1983; INOUE al., 1984). It is evimedia in a two-stage process (1st stage 200 L dent that the highly expressed naphthoquingrowth medium, 2nd stage 750 L production one biosynthetic pathway would be a good medium) the yield was increased to 3.7 g g -' system for biochemical and molecular studies. dry mass inoculum within 23 d (FUJITA al., The first results on the regulation of shikonin et 1982). By screening protoplast derived clones biosynthesis have been presented by HEIDE they isolated lines with an improved growth et al. (1989), showing that the ratio of p-hyand higher productivity. The best line had a droxybenzoic acid geranyltransferase and p specific content of 23.2% and yielded 6.45 g hydroxybenzoic acid glucosyltransferase acshikonin per g inoculum (FUJITAet al., 1985). tivities is one of the regulatory controls of shiProduction characteristics during the fermen- konin biosynthesis. Anthraquinones in higher plants are tation process were optimized by high density cultivation, fedbatch technique, controlled formed either via the acetate polymalonate f oxygen supply, and by the development o a pathway or via the o-succinylbenzoic acid rotating cylindrical bioreactor (FUJITA and pathway. Their production in cell cultures has and HARA,1985; FUJITA,1988; TAKAHASHI been reviewed in great detail by KOBLITZ FUJITA,1991). Since 1983 the Mitsui Compa- (1988). Highly productive suspension cultures ny produces shikonins by this technology for of plant species (e.g., of Cassia spp. or Rhamcosmetics and dyes (TAKAHASHI FUJITA, nus spp.) producing anthraquinones via the and 1991). Comparison of the composition of shi- acetate polymalonate pathway have never konins extracted from cell cultures with that been reported (VANDEN BERGet al., 1988). of various roots showed that the ratio of the However, it has been documented in numer-
-'
603
ous publications that the o-succinylbenzoic nylbenzoic acid derived anthraquinones. 17 of acid derived anthraquinones are well ex- the cultures yielded anthraquinone levels pressed in rapidly growing cell suspension higher than those found in the corresponding cultures and accumulate sometimes at explants. It was shown that nutritional and horf traordinary levels. A good example are cell monal requirements o the various anthraquicultures of Morinda citrifolia. ZENKet al. none producing cultures, even those of the (1975) tested a large variety of nutritional fac- one family, may be quite different. Cultures tors for their effect on growth and anthraqui- of Rubia cordifolia maintained their high proand none production. They established a produc- ductivity when scaled up to 75 L (SUZUKI tion medium yielding 2.5 g L-' anthraqui- MATSUMOTO, 1988), and cultures of Rubia nones corresponding to more than 10% of tinctorum have been studied at San-Ei Chemdry mass which exceeds the concentration of ical Industries (ODAKEet al., 1991) for the the root by a factor of 10. Important for high development of a commercial production production levels were the replacement of the process of anthraquinone pigments (e.g., of phytohormone 2,4-D by NAA and an in- alizarin, purpurin). It is clear that these culcrease of the sucrose level to 7%. The anthra- ture systems are also suitable for biochemical quinones of cell cultures of M. citrifolia con- and regulatory studies. The first enzymes insist of a mixture of at least 12 aglyca and glu- volved in the biosynthesis of o-succinylbenzoand cosides (LEISTNER, 1975; INOUE al., 1981). ic acid (SIMANTIRAS LEISTNER,1989) et The main components are lucidin derivatives and its further metabolism (SIEWEKEand 1992) have been studied in cell (Fig. 2). Some of them have not yet been LEISTNER, found in the intact plant (INOUEet al., 1981). cultures of Galium spp. The cultures of G. The spontaneous high production of these an- mollugo have also been used to overproduce thraquinones were maintained in various shikimic acid. The addition of glyphosate inbioreactors (WAGNER and VOGELMANN,hibits the formation of o-succinylbenzoic acid 1977). The yields of anthraquinones in an air- and anthraquinones in these cultures thus lift reactor were 30% higher than in experi- causing the accumulation of the biosynthetic ments with shake flasks. An interesting exam- precursor, shikimic acid (10% of dry mass, and AMRHEIN, ple of manipulating the expression of path- 1.2 g L-') (STEINRUCKEN ways in cultured cell was reported for cell sus- 1980). pension cultures of Morinda lucida. In photoautotrophic cell cultures (chlorophyllous, no sugar in the medium) lipoquinones were 3.1.3 Protoberberines and the main components while anthraquinones were not found. When these cultures were Benzophenanthridine Alkaloids transferred into the dark and sugar was added Isoquinoline alkaloids represent one of the to the medium, anthraquinone biosynthesis was induced and lipoquinone formation was largest groups of alkaloids in the plant kingdom. Common to all these alkaloids is that repressed (IGBAVBOA al., 1985). et The results achieved with M. citrifolia cul- they are derived from tyrosine via (S)-norcotures can also be obtained with cultures of claurine as a central intermediate (RUEFFER Galium mollugo (BAUCH and LEISTNER, and ZENK,1987) and not, as initially thought, 1978). A BS-NAA medium with 7% sucrose via (S)-norlaudanosoline synthase (RUEFFER gave the highest yields with ca. 2 g L-' within et al., 1981). Since the pathway is highly 14 days. Lucidin primveroside (Fig. 2) was the branched, a great variety of very different main component. While the pathway of an- structures results from this central intermethraquinones was readily expressed in cul- diate. Some of the branches are readily extured cells, the biosynthesis of iridoids repressed in cultured cells, while others such as mained repressed under all culture condi- the morphinan alkaloids remain mostly retions. SCHULTEet al. (1984) optimized cell pressed. There are many reports showing that prosuspension culture media of 19 different Rubiaceae species for optimal yields of o-succi- toberberine alkaloids (Fig. 3) spontaneously
604
els of up to 1.7 g L-' were readily achieved in normal growth medium. The Coptis line described by YAMADAand SATO (1982) contained mainly berberine and only trace amounts of other alkaloids. By screening small cell aggregates for high berberine (yelProtoberberine Alkaloids low) producing clones cell lines were estabBerberine R'+ R2= CH2 lished which produced more than 1g berberR'= CH,, R2= H Columbamine ine L or 10% on a dry weight basis (SATO R' = H, R2= CH, Jatrorhizine and YAMADA, 1984). e.g., Coptis japonica Another important species spontaneously Berberis species accumulating high levels of protoberberines is Thalictrum rugosum Thulictrum. While cell cultures of Thulictrum minus secrete most berberine into the meet dium (NAGAKAWA al., 1984) those of T. flavum, T. dipterocarpum or T. rugosum accumulate the alkaloids within the cells (SUZUKI et al., 1988; PIEHL al., 1988). Interestingly, et when cells of T. rugosum were transferred to 0 fresh medium lacking phosphate, protoberR3 berines were released into the medium (BERBenzophenanthridine alkaloids LIN et al., 1988a). There are other reports inR + R2= R3+ R4= CH2 ' Dihydrodicating that cultures of plant species with the sanguinarine capability to biosynthesize protoberberines R' + R2= CH2, R3+ R4= CH, Dihydrochelerythrine will spontaneously yield cultures accumulate.g., Papaver somniferum ing these colored alkaloids in high amounts Eschscholtzia californica (0.2-1 g L-'). The levels of protoberberines are so high that the Mitsui Company has deFig. 3. Structures of protoberberine and benzocided to develop a production process for phenanthridine alkaloids. berberine with Coptis juponicu cell cultures (MATSUBARA FUJITA,1991). As in the and case of shikonins the medium was first optimaccumulate at high levels in cultured cells. ized. For example, a 10-fold increase of Cu2+ Various cultures of Berberis spp. accumulated increased the berberine content by 40% (Mobetween 0.2 and 1.7 g protoberberine alka- RIMOTO et al., 1988). The addition of very loids, mainly jatrorrhizine, on a growth me- low levels of gibberelic acid increased the dium with 3.5% sucrose (HINZ and ZENK, content by 30% (HARA et al., 1988). Since 1981). BREULING al. (1985) optimized the cell aggregate screening was regarded as very et production to 3 g L-' in a 20 L airlift bioreac- time-consuming and not very efficient, prototor. Independently, two Japanese groups re- plasts with high berberine content were isoported the accumulation of high levels of pro- lated with a cell sorter (HARAet al., 1989). toberberines in cultures of Coptis juponicu All selected lines contained higher levels than (FUKUIet al., 1982; YAMADAand SATO, the original culture. The next step was to de1982). FUKUI al. established a line accumu- velop culture conditions for a high density et lating berberine, jatrorrhizine, palmatine, and culture (MATSUBARA al., 1989) so that 70 g et coptisine at a ratio of 50: 22 :22 :6. Interesting- dry mass per L (corresponding to ca. 700g ly, the content of alkaloids increased gradual- fresh weight) were obtained. This required inly during subculturing from 8-15% which was creasing the nutrition and oxygen supply by paralled by increased growth (FUKUI al., fedbatch-perfusion cultivation (MATSUBARA et 1982). This shows that protoberberine forma- and FUJITA, 1991). Today, the Mitsui Compation is indeed a favored pathway in these rap- ny produces berberines by a high density conidly growing cell cultures. Thus, alkaloid lev- tinuous culture method over a period of sev-
-'
605
era1 months with a yield of 0.65 g L- d-. This is 8 times more than the value obtained in batch cultures (MATSUBARA FUJITA, and 1991). This demonstration of the potential of plant cell culture process optimization by the bioengineers suggests that further attempts to optimize protoberberine production should start from initial product levels as high as those published by the Mitsui Company. It was speculated that cultures which release their compounds into the medium might be favored by industry because this would diminish the problem of sacrificing the slowly growing plant cells for product extraction. In the case of protoberberines, an alternative production process seemed to be possible with the highly productive T. minus line. T. minus cells were probably tested by the Mitsui Company in an attempt to develop a new bioassay-based screening method for high berberine-producing cell colonies (SUZUKI et al., 1987). The real advantage of T. minus cells can only be exploited if cells are immobilized for berberine production. KOBAYASHI al. et (1988) obtained a production rate of 50 mg L d - in batch and semicontinuous bioreactors over a period of at least 60 days of cultivation. This was undoubtedly an exciting result. However, in view of the improvements obtained with the C. juponicu culture, it is clear that T. minus cultures are not yet competitive for use in a commercial process. Berberine chloride is used in Japan as a medicine for intestinal disorders and treatments of abnormal zymosis and has previously been obtained by extraction from the roots of C. juponica and the cortex of Phellodendendron umurens (MATSUBARA and FUJITA, 1991). Berberine produced with tissue culture has not yet been approved as a drug in Japan. However, pharmacological studies towards this goal are underway (SUZUKI et al., 1993a, b). Another group of isoquinoline alkaloids which accumulate spontaneously in cultured cells are benzophenanthridine (Fig. 3) and the related protopine alkaloids. Cell cultures of most Papaveraceae contain alkaloids of this group. FURUYA al. (1972) were the et first to describe the occurrence of sanguinarines, protopines, and the aporphine magnoflorine in callus cultures of Pupuver somnife-
rum. When they analyzed the alkaloid pattern of callus cultures of 11 other species of Papaveraceae, almost identical alkaloid spectra were found (IKUTAet al., 1974). Since then, the occurrence of these alkaloids- has been confirmed in numerous reports. Cell culture conditions seem to favor the synthesis of benzophenanthridine alkaloids which are found in the corresponding intact plants often at rather low levels (WILLIAMSand ELLIS, 1993). However, there is a significant difference between the production of benzophenanthridine and protoberberine alkaloids in cultured cells. While in C. juponicu, e.g., protoberberine production increased with inet creasing growth rate (FUKUI al., 1982), benzophenanthridine synthesis decreased with better growth (BERLIN al., 1985). A et suspension culture of P. somniferum with a growth cycle of 28 d contained nearly 6% sanguinarines (360 mg L-) after the 5th subcultivation. During further subcultivation the yield decreased to 200 mg L after 20 subcultures and to 20mgL- after 35 subcultures. Biomass production of this culture increased from 6 g in 28 d to 11 g in 10 d (BERLIN et al., 1985). The biotechnological interest in benzophenanthridine alkaloids increased only when it was noted that these alkaloids can be induced greatly by stress factors and elicitors. Thus, the yields of benzophenanthridine alkaloids of suspension cultures of Eschscholtziu culifornicu were increased 10-fold to 150 mg L by increasing the sucrose concentration in the medium to 8% (BERLIN al., 1983). The observation of et EILERT al. (1985) that sanguinarine levels et of P. somniferum were enhanced from 0.01% to 2.9% by treatment with fungal elicitors received even more interest. The advantage of the elicitor treatment in comparison with the altered production media is that the cells respond more rapidly, and the induced alkaloids in Pupuver cell cultures are released into the medium (EILERT al., 1985; CLINEand et COSCIA,1988). In the case of E. culifornicu, the alkaloids, induced by various fungal and yeast preparations evidently remained within the cells (SCHUMACHER a]., 1987). These et authors reported product levels in the same range (160 mg L-) as were found in the sucrose induced cells (BERLIN et al., 1983).
606
There was, however, one significant difference: the elicited cells contained the quarternary alkaloids while the sucrose-induced cells contained the corresponding dihydro forms. group did It is interesting to note that ZENKS not consider elicitor technology to be helpful for increasing productivity for commercial purposes (SCHUMACHER al., 1987). This et was based on the finding that lines containing high levels of the alkaloids in the unelicited state did not produce higher levels after elicitation than low yielding lines. This technology has nevertheless been used in efforts of developing a commercial production process for sanguinarine. This compound has an antibiotic activity against oral microorganisms causet ing periodontal disease (SOUTHARD al., 1984). The chances of plant cell cultures to be used as a source were improved by the fact that cultured cells contained much higher levels of sanguinarine than tissues of intact plants and that an established production process did not exist. Consequently, research has concentrated on the optimization and scale-up of sanguinarine production. It was shown that the mildly elicited P. somniferum could be re-elicited after a period of regrowth suggesting that a semicontinuous production process with reet elicitation could be established (TYLER al., 1988). Due to the fact that a great portion of the elicited alkaloids were released, the suitability of surface-immobilized cells with adsorption of the released alkaloid to a resin was tested (KURZ et al., 1990). The total yields, however, were reduced. The highest yield published up to now is 300 mg L- in a 300L airlift reactor (PARKet al., 1992). The cultures were grown from ca. 20 to 180 g fresh mass per L and were elicited after 7 d or when ca. 2 g L- glucose were left in the medium, and were harvested two days later. The authors believed that further improvements might be possible by optimizing the amount of dissolved oxygen in the medium. If, in addition, conditions for elicitation of a high density culture could be established, a commercial production would be possible. There have also been efforts to select cell lines with higher production potential for sanguinarine (SONGSTAD al., 1990). The staet bility of these cell lines during production and
other characteristics important for the suitability of a selected line for scale tlp purposes remain to be shown. Although a commercial enterprise (Vipont) has been involved in the development of this process, it is unclear whether an industrial production will result. Further attempts to optimize sanguinarine production with cultures of P. somniferum, E. californica, or Sanguinaria canadensis should be left to industry if product level improvements are the only goal. They know how much the productivity has to be enhanced for a tissue culture process to be superior to conventional extraction of field grown plant material. It is evident that the culture systems producing protoberberines and benzophenanthridines are useful systems for studying the enzymology of these pathways. All 13 enzymes required for the synthesis of berberine have been identified. A summary of the present knowledge, derived mainly from studies group at the University of Munich of ZENKS group at the University and from YAMADAS of Kyoto, has been presented by HASHIMOTO and YAMADA(1994). The enzymology and molecular biology of benzophenanthridine alkaloid biosynthesis has been reviewed by KUTCHAN and ZENK (1993). Thus, the biotechnological value of the cultures producing these alkaloids does not only lie in their impressive productivities but also in their potential to provide a deep insight into the regulation of the expression of these complex and branched pathways.
3.1.4 Monoterpene Indole Alkaloids

About 1200 alkaloids derived from tryptophan have been isolated from higher plants, which corresponds to about one quarter of all alkaloids (GRBGER,1980). Several of these alkaloids from Rauwolfia, Catharanthus, and Cinchona are used medicinally. Therefore, it was of great interest to see whether these alkaloids also accumulate in cultured cells. The dimeric monoterpene indole alkaloids, vinblastine and vincristine, used as antileukemia agents, are only present in trace amounts in
607
the whole Catharanthus roseus plant. Indeed, these two compounds were initially the main reason why cultures of C. roseus received so much attention. These two alkaloids as well as the medicinally used alkaloids of Cinchona spp. will be discussed later in Sect. 3.2 in connection with poorly expressed compounds. More than 50 monoterpene indole alkaloids have been isolated from various Apocynaceae, e.g., Catharanthus and Rauwolfia spp. The chemical identification of their complex alkaloid mixtures was achieved mainly by three groups (STOCKIGTand SOLL, 1980; KOHLet al., 1982; KUTNEY al., 1983). A list et of almost all monoterpene indole alkaloids found in plant tissue cultures of various plant species is given by ELLIS(1988). Most of the alkaloids accumulate at low levels in the cultures. Sufficiently confirmed quantitative data allowing an evaluation of the biotechnological relevance are most often not available. Most studies regarding product levels have concentrated on the optimization of ajmalicine, serpentine (oxidized form of ajmalicine) and catharanthine (Fig. 4). There has been one report that a cell suspension culture of Rauwolfia serpentina accumulated 1.6 g raucaffricine per liter of medium (SCHOBELet al., 1989). However, this level dramatically exceeds the levels of all other monoterpene indole alkaloids ever measured in cultured cells and has not yet been confirmed in further publications by this or any other group working with R. serpentina cell cultures.
Ajmalicine
Catharanthine
Fig. 4. Monoterpene indole alkaloids of Catharanthus roseus.
Therefore, this claim cannot be regarded as being biotechnologically relevant. Cell cultures of R. serpentina were scaled up to 75 m3 by the Diversa Company and no alkaloids were found in the rapidly growing cell cultures (WESTPHAL, 1990). Indeed, it is sometimes very difficult to evaluate the impact of product level claims, and it is futile to discuss such claims here with respect to their biotechnological impact. Information regarding this issue can be found in several special reviews covering the monoterpene alkaloids in cultured cells (BALSEVICH, 1988; DELUCAand et KURZ, 1988; VAN DEN HEIJDEN al., 1989; MORENOet al., 1995) or in the article of ELLIS (1988). The biotechnological progress on the improvement of product levels of ajmalicinelserpentine and catharanthine, about which reports from many different laboratories are available, is discussed below. Callus cultures of C. roseus were found to contain low levels of monoterpene indole alkaloids (CAREW, 1975). However, suspension cultures accumulated no or only trace amounts of these compounds when grown on a growth medium with 2,4-D as phytohormone. ZENKS group was the first to develop a production medium (ZENKet al., 1977b). When the cells were transferred from the growth to a production medium, alkaloid formation was resumed after 3-5 d. This medium has successfully been used by several other laboratories. Of 458 independently established cell lines, 312 (approx. 75%) produced alkaloids when transferred to ZENKS production medium (KURZet al., 1980). This production medium, however, had no special composition; the sole transfer of the cells into a 2,4D-free medium allows the accumulation of reasonable levels of indole alkaloids (ZENK et al., 1977b; KNOBLOCH BERLIN, and 1980; PAREILLEUX VINAS,1984). The increase and of sucrose to 5 4 % had an additional beneficial effect. The phytohormone composition (indole acetic acid and benzylaminopurine) of ZENKS medium may have an additional stimulatory effect. The stimulatory effect of the production medium probably depends on the physiological state of the cells at the time of transfer from the growth to the production medium. The level of phosphate accumulated in the cells seems to play an important role in
608
the inducibility of alkaloid formation in a phosphate-free production medium (KNoBLOCH and BERLIN,1983; SCHIEL et al., 1987). Suspension cultures of C. roseus grew rapidly, were easy to scale up in fermentors and yields of 30-70 mg L - serpentine were et usually obtained within 20-30 days (ZENK al., 1977b WAGNER VOGELMANN, and 1977; SMART al., 1982). There have been several et efforts to isolate higher producing cell lines using screening methods. The group at the NRC in Saskatoon established more than 2000 individual lines from three C. roseus cultivars and screened them for alkaloid patterns and levels (KURZet al., 1985). They found a substantial variation among the lines, and one variety seemed to be more suitable for catharanthine formation. However, this tremendous effort did not evidently result in the isolation of an exeptionally high producing line. ZENK'Sgroup has screened established lines for high yielding clones using specific radioimmunoassays and the fluorescence microscope (ZENKet al., 1977b; DEUS and ZENK, 1982). Indeed, they found clones having accumulated high levels of serpentine or ajmalicine of which theoretical levels of up to 400 mg L-' were calculated. The isolated clones were, however, unstable, and alkaloid levels rapidly dropped to the values of the unscreened culture (DEUS-NEUMANNand ZENK,1984). The behavior of the clones indicates that the they were not true variants as claimed but instead were cell colonies in a transient physiological state. Reasons for the failure to select true variants in this system by the applied screening method have been discussed in detail (BERLINand SASSE, 1985; BERLIN,1986). For many years cultures of C. roseus were grown on media with 2,4-D. Alkaloid formation thus required transfer to a production medium. MORRIS(1986) found that cultures of C. roseus can not only be established and maintained on a medium with NANkinetin lacking 2,4-D but that they also produce alkaloids on this medium. These findings have also improved the chances of selecting clones with higher production. As predicted at a meeting in 1988 (BERLIN, 1990), high yielding clones were isolated by consecutive repeated
'
analytical screening of cultures growing on NAA/kinetin medium. Again, researchers of the Mitsui Company (FUJITA al., 1990) obet tained this success. Repeated analytical screening improved the specific catharanthine content from 0.1% to 0.7%, and these lines have maintained their superior productivity for at least 10 subcultures. Additionally, a biochemical selection for 5-methyl-tryptophan-tolerant cell lines was applied. (FUJITA et al., 1990). Although the specific catharanthine content was further increased, it is not clear why these cell lines form higher levels of alkaloids. The productivity of the isolated lines was further optimized by optimizing the medium composition, inoculum density, and fermentation parameters. Finally, the 5-MTtolerant line produced 230 mg L - catharanthine in a 1.7 L bioreactor within one week (specific content 1.1% of dry mass). There have been tremendous efforts to stimulate the production of ajmalicine and serpentine in cultures of C. roseus. Many approaches have been used such as the addition of various biotic and abiotic elicitors, the use of high and low density culture, feeding and elicitation, alteration of media composition, immobilization, etc. (for reviews, see VAN DEN HEIJDENet al., 1989; MORENOet al., 1995). Despite the many reports claiming to have demonstrated increased alkaloid production, one has to realize that the product levels have not been improved since the first findings 15-20 years ago. Thus, the product levels of bioreactor studies remain at 1050 mg L-' ajmalicine or serpentine (corresponding to 0.1 to 0.5% of dry mass) (SCHIEL and BERLIN, 1987; SCRAGG et al., 1989; SCRAGGet al., 1990; SCHLATMANN al., et 1994). One reason for this undoubtedly poor progress could be the fact that most often only a single factor was analyzed at any one time. Furthermore, analysis of the various factors did not take advantage of the previously improved culture levels. A comparison of the many studies done in universities and research centers in an attempt to optimize monoterpene alkaloid production in C. 10seus with the industrial approach, e.g., of the Mitsui Company reveals that the former group of scientists is interested in detecting factors influencing alkaloid levels, but not re-
'
609
ally in improving product levels. Despite nearly 20 years of bioreactor studies with C. roseus it has been reported that the inconsistencies and the often contradictory results are likely due to insufficient characterization of the inoculum material (VAN GULIKet al., 1994). The authors suggested to use identical inoculum material when analyzing the effects of factors on growth and alkaloid production. For industrial scientists, productivity improvement is a step-by-step enhancement of product levels per volume and time and this requires that each step is reproducible. Thus, it seems that the maximum potential of C. roseus cultures for serpentine and ajmalicine production may only be found when a commercial enterprise takes over. A first step would be a screening of cultures growing on NAA kinetin medium for true variants overproducing ajmalicine or serpentine. However, a comparison of the production level improvements for berberine, shikonins, and catharanthine by the Mitsui Company shows that the maximum yield possible is mainly determined by the biology of the cell. The rather low levels of the monoterpene indole alkaloids found to date suggest that at best levels in the range of 200-500 mg L-' within 1-2 weeks might be produced by an optimized culture. According to an economic assessment of the production of ajmalicine by C. rosew cultures, the present levels have to be increased at least by a factor of 40 to make the process competitive (DRAPEAU al., 1987). et In contrast to the poor progress at the product level cell cultures of C. roseus have been used quite well for studies on the regulation of alkaloid formation. The correlation of enzyme activity pattern and product accumulation curves and the conclusions drawn by various groups are not, however, always consistent. Therefore, only facts which are generally accepted are mentioned here. It has repeatedly been shown that tryptophan decarboxylase is induced when alkaloids are produced. However, it has also been shown that induction of this enzyme does not necessarily lead to enhanced alkaloid formation (for a review see VAN DEN HEIJDENet al., 1989). Nevertheless, since tryptophan decarboxylase is readily inducible two groups have cloned the cDNA of this enzyme from C. roseus
(DELUCAet al., 1989; GODDIJN,1992). The negative effect of phytohormones on alkaloid formation can perhaps now be explained by the finding that auxins down-regulate transcription levels of the tryptophan decarboxylase gene (GODDIJN al., 1992). Geraniol-lOet hydroxylase is regarded as a regulatory enzyme for the monoterpene moiety (secologanin) of the indole alkaloids. Indeed, this enzyme is also induced during initiation of alkaloid biosynthesis (SCHIELet al., 1987). It has already been purified and its cloning is in progress (MEIJERet al., 1993a). The enzyme and the gene coding for strictosidine synthase which connects tryptamine and secologanin have been isolated (KUTCHAN,1993). in some lines the enzyme was present in nonproducing cells, in others it was induced when alkaloid formation was elicited. Molecular studies suggest a coordinated regulation of the genes coding for tryptophan decarboxylase and strictosidine synthase (PASQUALI et al., 1992). Detailed overviews of the enzymology and regulation of monoterpene indole alkaloids have been given recently (DELUCA, 1993; MEIJER et al., 1993b). The results of molecular studies should show whether genetic improvements of the biosynthesis of certain monoterpene alkaloids are an attainable goal.
3.1.5 Anthocyanins and Betalains

Anthocyanins have been found in many cultured cells of many plants (Daucus, Haplopappus, Catharanthus, Petunia, Mathiola, Euphorbia, Perilla, Vitis, Aralia, and many others). For some cultures the individual anthocyanin components have been identified, for others only the total amount of the pigments has been given (ELLIS,1988; SEITZand HINDERER,1988; YAMAMOTO, 1991). In a few culture systems anthocyanins are formed in dark grown cells, but in most cases light is required for optimal production (SAKAMOTO et al., 1994). Due to their low toxicity, anthocyanins are widely used in food additives and also as a dye for silk (YAMAMOTO, 1991). While anthocyanin extracts are rather inexpensive the prices for individual anthocyanins seem to be very high which would justify the
610
use of tissue culture based processes. Again, Japanese scientists and industries are leaders in the development of biotechnologically relevant culture systems. Since anthocyanin producing cells are easily detected in callus cultures they have become an early target for visual screening (KINNERSLEY and DOUGALL, 1980). The most impressive success was reported by YAMAMOTO et al. (1982) when they established high yielding cell lines of Euphorbiu milli by consecutive repeated screening. Only after 24 clonal selections were stable lines found; this required permanent screening over a period of many months. NOZUE et al. (1987) were also successful in selecting high anthocyanin yielding cell lines from sweet potato by consecutive repeated screening. This shows that even for compounds formed in the growth medium, stable clones are only obtained by repeated screenings. This result casts some doubt to claims of the instability of high producing variant lines if permanent screening was not applied over a very long period (BERLIN, 1990). As for many other secondary. metabolites, the culture medium can also be optimized for anthocyanin production. The optimization of E. miflii cultures by the Nippon Paint Corporation improved the productivity 4.5-fold to 32 mg L-' d-' (YAMAMOTO et al., 1989). The main component of the E. rniffii cell culture is cyanidin-3-arabinoside (Fig. 5 ) (YAMAMOTO, 1991). A great disadvantage of the Euphorbiu system in commercial production is the requirement of light for pigment production. The same holds true for a highly productive culture of Perrillu frutescens (ZHONG and YOSHIDA, 1995) which produces up to 5.8 g L-' anthocyanins in Iod under optimized culture conditions and light. Therefore, the best line for the commercial production of anthocyanins may be cultures of Aruliu cordutu which produce high levels in the dark. As with the cultures of Euphorbiu and Zpomeu, continuous cell aggregate cloning led to a highly productive and fast growing line (SAKAMOTO al., 1994). 90% of all et cells of the cloned line produced anthocyanins, a result which is only possible if product formation and growth are closely interrelated. Optimization of the basic inorganic salt concentration, the carbohydrate concentration,
R = arabinoside Cyanidin-3-arabinoside(Euphorbia millii)
HO
HOOC
A!k N
H
I
COOH
Betanidin, R = H Betanin, R = Glucose (various Centrospermae)
Fig. 5. Structures of anthocyanin and betacyanin.
the ratio NO,/NH,+, and the total nitrogen levels improved the anthocyanin content to 10.3% of dry mass. This culture was scaled up to 300 L in a 500 L jar fermentor (KOBAYASHI et al., 1993). After 1 6 d of cultivation 69 kg fresh mass with 545 g anthocyanin (corresponding to 17.2% of dry mass) were obtained. The further improvement of productivity during scale-up was evidently due to the controlled supply of C 0 2 to the cultures. Efforts to identify the anthocyanins present in this high yielding culture system are underway. As is typical in Japan the project is being done with close cooperation between researchers from university and industry (Kyowa Hakko Kogyo Co. and Tonen Co.). Tissue culture systems may only become competitive for the production of individual anthocyanins and not for mixtures of anthocyanins. Therefore, it might be useful to look not only at total yields during selection and media optimization but also at whether such methods affect the percentage of any one anthocyanin in a complex mixture ( D o and CORMIER, 1991). As shown in the previous sections optimized culture systems have often successfully
611
been applied for biochemical and regulatory studies by those interested in their biotechnological application. Although this has not yet been the case, our knowledge about their biosynthesis and the regulation of this pathway far exceeds that of all other secondary pathways (for reviews see DIXON and LAMB, 1993; et 1990; FORKMANN, DOONER al., 1991; KOES et al., 1994). Thus several genetic manipulations of these pathways have been made with the aim of altering pigmentation of flowers (MEYERet al., 1987; VAN DER KROL, 1990). Other pigments often found in cell cultures of Centrospermae are the betalains which include red betacyanins and yellow betaxanthins (BOHM and RINK, 1988). Cultures of Chenopodium rubrum (BERLINet al., 1986), Phytolaca americana (SAKUTA al., 1987), et Amaranthus tricolor (BIANCO-COLOMAS and HUGUES,1990), Beta vulgaris (GIROD and et ZRYD,1991), and Portulaca (KISHIMA al., 1991) produce rather high levels of betacyanins under light conditions. Betacyanin formation is more or less growth-rela#ed, which explains the spontaneous accumubtion of these pigments in cell cultures. By visual screening higher productive lines were isolated from the parent callus line and media compositions were optimized. Often, quantification was based only on the absorption at 535nm. In the case of Chenopodium rubrum we identified amaranthin, celosianin, and betanin as et the main betacyanins (Fig. 5 ) (BERLIN al., 1986). Under optimized cultured conditions 35-45 mg L betacyanins were produced in a modified Murashige & Skoog growth medium supplemented with tyrosine (BERLIN et al., 1986). From this line a superior subline was selected by screening. This line produced not only 3 4 times more betacyanins but also had an altered pigmentation due to an altered ratio of betacyanins (with celosianin as the main compound) (BERLIN,unpublished results). This line has not yet been optimized for highest product levels but was successfully used for the detection of enzymes involved in celosianin biosynthesis (BOKERN al., 1991). et The question of whether further optimizations are justified arises if the only aim is to make betacyanin tissue culture production more competitive. Although cell cultures of
Beta vulgaris contain higher levels of betacyanins than all plant tissues tested, it is unlikely that cell cultures will be used for the production of betacyanins since the marked price for red beet extracts is extremely low (LEATHERS al., 1992). et
3.1.6 Steroidal Compounds

There are numerous reports on the occurrence of sterols, steroids, sapogenins, saponins, and steroidal alkaloids (ELLIS,1988). In my previous review (BERLIN,1986) the efforts to optimize diosgenin production in cultures of Dioscorea deltoidea were described in some detail. As no significant progress towards a commercial production has been reported since then, I will concentrate here on the ginseng saponins, which were only briefly mentioned in the last review. The roots of Panax ginseng C.A. Meyer have widely been used as a tonic and precious medicine in oriental countries since ancient times. It is regarded as an adjuvant to prevent health disorders and is considered to be a miraculous drug for preserving health and promoting longevity (MISAWA, 1994). The active component with proven pharmacological effects are saponins, e.g., ginsenoside-Rb and -Rg (Fig. 6). Due to the worldwide increased demand for ginseng extracts and the fact that a cultivation period of 5-6 years is necessary before harvest, production of ginseng extracts or saponins using tissue cultures was regarded as an alternative source of supply. FURUYAS group analyzed a number of callus cultures of Panax ginseng with different phythohormone requirements (FURUYA, 1988), optimized
&
-I
R3
RZ
Fig. 6. Saponin aglycone of Panax ginseng. R, and R3 are often diglycosides in Rb-ginsenosides. In Rgginsenoides R2 is glucosylated.
612
their growth conditions, and compared their of the immunologically active components of saponin patterns with those of P. ginseng Echinacea drugs, two polysaccharides were roots. The patterns were quite comparable isolated and characterized (PROKSCHand 1987). The large amounts of these and levels of total saponins reached up to WAGNER, 0.7% of dry mass or 50mgL-' within 4 polysaccharides needed for further in vivo exweeks in a 30 L jar fermenter. It was shown periments were difficult to obtain from the that some morphological differentiation in- plant material. Therefore, cell cultures of E. creased saponin formation, and thus a rapidly purpurea were established. The cultures also growing embryo-like cell line was selected for produced two major active polysaccharides 1991). A bio- and released them into the culture medium; scale-up to 20 m3 (USHIYAMA, mass production of 700 mg L-' d -* was ob- however, they were not identical with the poet lysaccharides of the intact plant (WAGNER tained. Since 1988 the use of the tissue culture ma- al., 1989). These cultures were optimized on a terial has been approved for the Nitto Denko laboratory scale and then scaled up to 60m3 Co. by the Japan Ministry of Welfare and in a cascade of bioreactors (WESTPHAL, Health. The extracts from the tissue culture 1990). Especially noteworthy was that the material have been added to wines, tonic specific yield was improved 2-3-fold and that drinks, soups, herbal liqueurs, and other food the cultivation periods in various fermentors preparations since 1989 (USHIYAMA, 1991). were greatly reduced during the optimization For the approval, a comparison of the consti- by the process engineers (WAGNERet al., 1990). The cooperation betuents of plant extracts and plant tissue cul- 1989; WESTPHAL, ture derived material was necessary. In addi- tween researchers of the University of Mution, some biological assays (Ames test, acute nich, the pharmaceutical company Lomavirulence test) and dietary tests with livestock pharm, Emmenthal, and the bioengineering feeding containing high amounts of the tissue company DIVERSA, Hamburg, led to the culture material had to be performed. It is a production of commercially relevant levels of little astonishing that the chemical compari- a tissue culture specific drug which entered son of plant and marketed tissue culture ex- clinical trials. However, DIVERSA has retracts does not contain details about the sa- cently been taken over by Phyton, mainly ponins and the ratio of Rb:Rg saponins known for its engagement in developing a (USHIYAMA, 1991). In view of the large-scale production process for taxol@(palitaxel) and production of ginseng extract using an em- thus it is not clear whether production of the bryo-like suspension culture originally estab- immunologically active polysaccharides will group (USHIYAMA, continue. lished by FURUYA'S 1991), it is also somewhat surprising that one of the latest reports from FURUYA'S group described the growth of an embryo culture in various 3 L bioreactors (ASAKAet al., 1993). 3.2 Products of Commercial This culture produced 0.1 mg L-' d sapon- Interest Accumulating in Traces or ins.
-'
not at all in Suspension Cultures
3.1.7 Immunologically Active Poly saccharides

There are a few reports that plant cell cultures sometimes release considerable amounts of polysaccharides into the culture medium. These polysaccharides however, have rarely been analyzed for biological activity. During investigations on the identification
Large efforts have been made in the past to express compounds which are widely used as pharmaceuticals in cell cultures. These include morphinan alkaloids, tropane alkaloids, quinoline alkaloids, and cardiac glycosides. In addition, substantial efforts have been made to develop cell culture systems producing antitumor compounds. In this section the progress in increasing the product levels of these compounds will be analyzed. In contrast to
613
the previous review (BERLIN, 1986) a special section on monoterpenoids will not be given here. Production of these compounds has remained low with callus and suspension cultures (for a review see MULDER-KRIEGER et 1994). So far, tissue al., 1988; BANTHORPE, culture systems have not contributed very much to the analysis of lower terpenoid biosynthesis and its regulation, despite the fact that even a low producing tissue culture can be a good source for related enzymes (BANTHORPE, 1994). Organ cultures are usually much better producers of lower terpenoids (CHARLWOOD al., 1990; BANTHORPE, et 1994). Nevertheless, the development of a biotechnologically viable process for these compounds is unlikely.
3.2.1 Morphinan Alkaloids

A number of reports have claimed that morphinan alkaloids are present in callus and cell suspension cultures of Papaver somniferum and P. bracteatum (for a review see KAMO and MAHLBERG,1988; KAMIMURA, 1991). However, there are also several reports where no morphinan alkaloids were detected (BERLIN, 1986; KAMO and MAHLBERG, 1988). The elicitation of poppy cell cultures was not initially planned for the production of sanguinarines but instead for the induction of morphinan alkaloids. EILERT al. (1985) et clearly stated that morphinan alkaloids could not be induced by any of the elicitors tested. Claims in the literature of substantial morphinan alkaloid levels of up to 0.15% dry mass are of no biotechnological relevance since such data were not confirmed in further investigations. Evidently, some morphological differentiation is required for expression of the morphinan alkaloid pathway. Thus, freshly initiated cultures having retained the capacity to redifferentiate when stressed by elicitors or altered phytohormone composition, are able to synthesize and accumulate low levels of these alkaloids (KAMO and MAHLBERG, 1988). Such cultures usually exhibit poor growth and are not very stable, and thus no efforts have been made to produce these alkaloids in larger volumes. The most recent report on morphinan alkaloids in cell cultures
claims a production of 2.5 mg morphine and 3.0 mg codeine per g dry mass (SIAH and DoRAN, 1991). However, considering that these very high levels correspond only to 1014 mg L-' after a period'of 56 d using phytohormone-free medium which must be replaced several times, it is clear that such findings are of no biotechnological relevance. They only show that under extreme stress conditions the morphinan branch is weakly expressed. Progress in the production of morphinan alkaloids by undifferentiated, rapidly growing cell suspension cultures will, therefore, be made only if factors are identified which prevent expression of the pathway in better growing cultured cells. A breakthrough could indeed result from the recent detection of the enzyme channelling (R)-reticuline via salutaridine into the morphinan alkaloid and ZENK, 1993). This branch (GERARDY membrane-bound microsomal enzyme is evidently only expressed in tissues and cells able to synthesize morphinan alkaloids. This might, however, be changed in the near future by genetic engineering.
3.2.2 Tropane Alkaloids

Hyoscyamine and scopolamine are the most important medicinal tropane alkaloids. Hence, several groups have worked on the production of these compounds in cultured cells. However, the alkaloid levels were found to be extremely low in callus and suspension cultures of Hyoscyamus, Atropa, Datura, Duboisia, and Scopolia. By using a squash technique in screening for alkaloid producing and (1982) seclones, YAMADA HASHIMOTO lected a cell line of Hyoscyamus niger containing just 0.01-0.02% hyoscyamine per g dry weight in suspension cultures. Scopolamine levels were 10-fold lower. Also media variation experiments were not successful in increasing tropane alkaloid levels. Evidently, some root formation is required to express the tropane alkaloid pathway to a reasonable extent (see Sect. 4). There has only been one report (BALLICA and RYU, 1994) describing a selected cell suspension culture derived from stem cells of Datura stramonium which seems to produce
614
up to 80 mg L total tropane alkaloids within 25 days. According to the authors, an integrated approach (selection, media optimization, elicitation, precursor feeding, optimization of process parameters) is necessary for enhancing productivity in cell cultures. However, it was not clearly shown that a combination of all factors positively affecting productivity was indeed necessary to obtain the above yield. In any case it does not make much sense to continue studies on the optimization of tropane alkaloid formation in suspension cultures for production purposes. For studying the biosynthesis and regulation of this pathway, highly productive hairy root cultures of all these species are available (see Sect. 4). Using hairy root cultures, the initial steps of tropanelnicotine alkaloid formation have been identified, enzymes have been purified, and gene cloning is in progress (HASHIMOTO and YAMADA, 1994).
-'
3.2.4 Antitumor Compounds
The search for antitumor compounds in tissue cultures has been a favored goal of many researchers. The great interest in cell cultures of Cathurunthus roseus resulted from the fact that this plant species synthesizes in very low amounts (0.0005% of dry mass) the expensive dimeric monoterpene indole alkaloids vinblastine (Fig. 7) and vincristine which are the most active agents used in the treatment of certain forms of cancer (MISAWA and ENDO, 1988). HIRATA al. (1989) determined a vinet blastine content of 1.4 mg g of dry mass in leaves of C. roseus, a level which would reduce the need for a tissue culture derived process. It is now generally accepted that cell suspension cultures of C. roseus are not able to synthesize the dimeric alkaloids; vindoline, one of the monomeric precursors, is not formed in cultured cells but in leaves (DECAROLIS and DELUCA, 1993). Research has gone into two directions - development of 3.2.3 Quinoline Alkaloids shoot culture systems and semi-syntheses from the two monomers. However, as with Cinchona trees have been grown in planta- many shoot culture systems, productivity is tions since more than 130 years for the pro- presently 100-fold lower than in the leaves of duction of bark containing the antimalarial, intact plants (HIRATAet al., 1989). More effiantifever compound quinine. The related cient is the approach of coupling vindoline compund quinidine is used as a drug against and catharanthine (derived from plant tissue cardiac arrhythmia. Though the compounds culture) by chemical or enzymatic means to have repeatedly been found in cell suspension form 3 ',4 '-anhydrovinblastine and vinblaset cultures of Cinchona succirubra, C. pubes- tine (GOODBODY al., 1988). The coupling cens, and C. ledgeriana at levels of up to 0,9% of catharanthine and vinblastine with peroxi(KOBLITZ al., 1983), suspension cultures of dases or'.ferric ions and subsequent reduction et Cinchona could not be further developed for yielded 50-70% anhydrovinblastine and 12biotechnological purposes. Attempts to in- 28% vinblastine (DICOSMO,1990). The reand crease the usually rather low levels by screen- sults obtained by GOODBODY coworkers ing and media variation were not very suc- at Allelix looked promising from a commercessful (HARKESet al., 1985). In addition, cial point of view. Nevertheless, the project growth of cultures of this species is generally was discontinued. Evidently, there was no slow (WJNSMA and VERPOORTE, 1988). need for developing a tissue-culture-based Highest alkaloid production was obtained commercial process. High levels of catharanwith cultures showing some degree of differ- thine and vindoline for chemical coupling are and readily isolated as by-products from C. roseus entiation, e.g., roots, shoots (WIJNSMA VERPOORTE,1988), and compact globular plants. Since vindoline is not found in cell susstructures (HOEKSTRA al., 1990). Due to pension cultures, tissue cultures are not et the difficulties of establishing reasonably pro- needed for efficient synthesis of vinblastine. ductive culture systems, the progress in the Presently, there is a tremendous interest in elucidation of the biosynthetic pathway lead- developing a tissue culture process for the ing to quinoline alkaloids has remained slow production of taxole (taxol is a registered (WIJNSMA VERPOORTE, and 1988). trademark of Bristol Myers-Squibb for pacli-
-'
615
Vin blastine (Catharanthus roseus)

0
\=/
Taxol (Taxus spp.)
5-rnethoxypodopyllotoxin-4~-D-glucoside (Linum flavum)
OCH3
Triptolide (Tripterygiumwilfordii)
Fig. 7. Antitumor compounds of some higher plants.
taxel) (Fig. 7). Detected during the screening program of the National Cancer Institute (NCI), it was considered the most interesting compound among 11OOOO tested structures (NICOLAOUet al., 1994). Taxol has been
proven to be a very effective drug against several forms of cancer. It was approved by the US Food and Drug Administration (FDA) for the treatment of ovarian cancer in 1992. The chemistry and the mbde of action of tax01 has been reviewed recently (NICOLAOU et al., 1994; KINGSTON,1994). A significant problem is a shortage in taxol supply (CRAGG et al., 1993). To isolate 1 kg taxol loo00 kg of dried bark from 3000 Taxus brevifolia trees have to be extracted. As nearly 2 g taxol are needed for the treatment of one patient it is evident that research has been initiated to,improve taxol production in trees, to look for alternative sources of taxol, and to develop a total or semi-synthetic chemical synthesis (NICOLAOU et al., 1994; KINGSTON, 1994). Presently, taxol is produced commercially by semi-synthesis from 10-deacteyl baccatin 111 from needles of young Taxus trees grown in plantations. One alternative is to produce tax01 by using tissue cultures and, indeed, as pointed out by EDGINGTON (1991) if ever there was a plain target for plant tissue culture, this is it. In 1991, two American companies, Phyton Inc. and ESCAgenetics, announced that the development of their tissue culture-based processes was to be completed within 2-5 years. It was impossible to estimate how realistic these claims were, because both companies did not release scientifically sound information about their processes. Thus, conclusions could only be drawn from the results published by researchers from universities and from the improvements of systems with similar productivity obtained when industry was involved. In the meantime, ESCAgenetics dropped out of the race, while Phyton licensed their tissue culture process to Bristol-Myers Squibb. It has to be seen whether the huge fermentation facilities of Phyton in Germany will be used for commercial taxol production. In contrast to the American companies, the Mitsui Company has published some of their research results on taxol production in Taxus cultures (YuKIMUNE et al., 1996). According to the published data, the Japanese company again is ahead of all others. Early publications on taxol production in tissue cultures of various Taxus species looked not very promising and did not sup-
616
port the taxol levels (153 mg L-' after 6 and weeks) claimed in the patent of BRINGI KADKADE (1993). WICKREMESINHE ARand TECA (1993) induced callus cultures of various Taxus spp. and screened them for taxol production. The levels ranged from O.OOO10.0131% of dry mass, and older browning callus produced more taxol than young pale callus. When they established cell suspension cultures from taxol producing callus, the specific taxol content could not be determined because the taxol peak was overlaid by other compounds (WICKREMESINHE ARTECA, and 1994a). From 28 g dry mass 120 pg pure taxol (4.3 x 10-'%) were extracted. Thus quantitative data based only on HPLC determinations should be regarded with caution. The group of DICOSMO Toronto University published at several articles regarding the optimization of taxol production in Taxus callus and suspension cultures. They reported initial yields of 0.02% for a culture of Taxus cuspidata calli (FETT-NETO et al., 1992). Subsequently, it was shown that the medium composition affects taxol production and growth. As only percent of controls were given, it is not clear whether the experiments resulted in really higher productivity (FETT-NETO al., 1993). et When the cells were cultivated in shake flasks productivity went down by a factor of 10 in rapidly growing cultures (FETT-NETOet al., 1993), and in a more productive suspension culture levels of 0.15 mg L-' were detected (FETT-NETO al., 1994a). Taxol production et could only be improved a little by feeding of potential precursors and thus, levels of 0.01 % dry mass were found (FETT-NETO et al., 1994b). Substantially higher levels were reported by two groups from Cornell University, Ithaca (the city where Phyton is located); during culture media optimization experiments they reached levels of up to 15 mg L-' taxol using lines of T. baccata, T. canadensis, and T. cuspidata (HIRASUNA al., 1996; et KETCHUM and GIBSON, press). However, in substantial variations of yields in individual experiments were observed making interpretations quite difficult. To overcome the oscillations of taxol production might be a severe problem in developing a large-scale process. In a bioreactor experiment with a working volume of 600 mL levels of up to 22 mg L -
'
or 1.1 mg L-'d-' taxol were found of which 90% were extracellular (PESTCHANKER al., et 1996). A breakthrough maybe was the finding that methyl jasmonate is an inducer of taxol biosynthesis in cultures of various Taxus species (MIRJALILI LINDEN, and 1996). Addition of 10pM methyl jasmonate increased taxol productivity 19-fold, from 0.2 mg L - to 3.4 and coworkers (1996) mg L-'. YUKIMUNE added higher concentrations of methyl jasmonate (100 pM), used a cell line of T. media with a higher capacity for taxol biosynthesis, and observed taxol accumulation after induction for a longer period. Although taxol levels increased only 5-fold compared with controls, a productivity of 110 mg L-' within 14 d was obtained. The specific content of taxol obtained was 0,6% and thus exceeded by far the specific contents found in any tissue of the intact plant. Recently, transformed phytohormone-independent callus cultures were found to contain taxol levels up to 16 pg g-' dry mass (HAN et al., 1994). Embryo culture has also been tested for taxol production (FLORES et al., 1993). However, taxol and taxane yields were not very different from those reported initially for callus and suspension cultures. A recent publication suggests roots of hydroponically grown Taxus plants as a source for taxol and related taxanes (WICKREMESINHE and ARTECA,1994b), which would indicate that transformed root cultures should also be a suitable source for these compounds. Indeed, a recent patent application of Celex-Laboratories (1994) describes the production of taxol via hairy root cultures. Studying the published literature on taxol production in cell culture systems (for a review, see JAZIRI et al., in press), one might conclude that most cell lines of all Taxus species contain low levels of taxol, that some more productive lines can be obtained by screening, that methyl jasmonate is presently the only unambigously proven enhancer of taxol synthesis, and that production levels of higher producing lines often oscillate to an unacceptable extent. Nevertheless, if the published production rates can reproducibly be obtained in large volumes of several thousand liters and if they are optimized further by biochemical engineers, culture systems may compete with
'
617
semi-synthesis. More knowledge about the biochemistry, regulation and limiting steps of the taxol pathway may eventually help to engineer lines with improved productivity (SRINIVASAN al., 1996). They provide et some evidence that taxol biosynthesis is located in plastids. They also conclude that the conversion of phenylalanine to phenylisoserine is a rate-limiting step of taxol biosynthesis. It might be important to stimulate the flux of the primary precursors into the taxane skeleton. Previously, it was assumed that mevalonate is the precursor of the isoprenoids. et However, the findings of EISENREICH al. (1996) suggest that the isoprenoid moieties of the taxane molecule are derived from a yet unknown precursor which is likely to be derived from a novel pathway of isoprenoid biosynthesis (using triose phosphate-type compounds and activated acetaldehyde), recently detected by ROHMERet al. (1993). CROTEAUS group demonstrated that taxa4(5),11(12)-diene is the first intermediate in taxane biosynthesis, and recently they have purified the corresponding enzyme taxadiene synthase cyclizing the universal diterpene precursor geranylgeranyl pyrophosphate to taxa-4(5),11(12)-diene in a single step (HEZARI et al., 1995). Due to the importance of taxol it is predicted that further progress in the elucidation of taxol biosynthesis and its regulation will soon be made and that genetically engineered lines might be created in the near future. Another important, commercially used antitumor compound is etoposide, a semisynthetic podophyllotoxin. Production of podophyllotoxin by tissue cultures of Podophyllurn peltatum was first attempted by KADKADE (1982). However, the levels remained disappointingly low. The interest in podophyllotoxins was revived when root cultures of Linum flavum were found to contain high levels (1% of dry mass) of 5-methoxypodophyllotoxin (Fig. 7) and its glucoside (BERLIN al., et 1988b). Two Dutch groups confirmed this result and showed that suspension cultures of L. flavum also contained some podophyllotoxins (VAN UDEN et al., 1990; WICHERSet al., 1991). However, reasonable productivity required some morphological differentiation, e.g., root formation (VANUDENet al., 1991).
Other podophyllotoxin producing cell cultures are those of L. album (SMOLLNY al., et et 1993) and P. hexandrum (WOERDENBAG al., 1990). The levels of podophyllotoxins vary from 0.05 to 0.3% of dry mass depending on the culture conditions and the tendency to differentiate. Thus root cultures (producing at least 1% podophyllotoxins in dry weight) seem to be most suitable for future research efforts, e.g., for biochemical studies (OOSTHAM et al., 1993). A recent review by MISAWAand ENDO (1988) describes the productivity of cell culture systems for other potential antitumor compounds produced by plants and plant cell cultures. These include camptothecine, homoharringtonine, and maytansine. It must be concluded that the levels found in the culture systems are far too low to be attractive from a biotechnological point of view. The same holds true for tripdiolide and triptolide (Fig. 7) produced at 0.01% of dry mass by Tripterygium species (TAKAYAMA, 1994).
3.2.5 Cardiac Glycosides

The formation of cardiac glycosides in undifferentiated cell cultures of various Digitalis species was found to be very low or even lacking in rapidly growing suspension cultures. In slowly growing green callus cultures or in embryogenic cultures some cardiac glycoside accumulation was observed (for review see LUCKNER and DIETTRICH,1988). However, low- or non-producing plant cell suspension cultures of Digitalis fanata are effective in the biotransformation of cardenolides and for this reason they are still rather attractive from a biotechnological point of view (see Sect. 6.2).
3.2.6 Vanillin and Vanilla Aroma

Vanilla is probably the most widely used flavor in food industries with a worldwide annual consumption of 1200-1 500 t in 1993 (HAVKIN-FRENKEL, The prices for one 1994). kg natural vanillin from cured beans of Vunillaplanifolia amount to $3000-4000 while the price for synthetic vanillin is less than $20 per
618
kg. The flavor of natural vanilla extracts from beans of different origin varies. Though vanillin is the most important constituent of the aroma other compounds seem to affect substantially the taste of this flavor. Due to the high price of natural vanilla flavor plant tissue cultures, mainly of V. plunifoliu, have been regarded as an alternative source for the production of vanilla aroma and vanillin. The first publication on phenylpropanoid metabolism in V.plunifolia showed that vanillin was not produced under normal culture condi1990)Formations (FUNKand BRODELIUS, tion of vanillic acid, however, was observed when 3,4-(methy1enedioxy)-cinnamic acid, an inhibitor of p-coumarate CoA ligase, was added to the suspension cultures. This led to a shift from lignin to benzoate biosynthesis. After feeding potential cinnamic acid precursors there were indications that the cell culture must contain enzymes involved in benzoate biosynthesis. Finally, it was found that kinetin is an efficient elicitor of vanillic acid formation (FUNK and BRODELIUS,1992). Vanillic acid levels of up to 0.1% of dry mass were detected. Thus, according to the literature, one would assume that cultures of V. plunifoliu are of no commercial interest. It was, therefore, somewhat surprising that ESCAgenetics (KNUTH and SAHAI, 1991) filed a patent in 1988 (issued in 1991)for the production of vanilla aroma (PhytoVanillaTM)and biosynthesized vanillin (PhytovanillinTM)by suspension cultures of V. frugruns. The aroma compounds are secreted into the medium and bind to absorbents. The patent states that 16-18 mg vanillin per L of medium were produced within ca. 45 days. A comparison of the vanilla flavor profiles of beans with that of tissue cultures as presented in the patents reveals that vanillin is the main component in both extracts, but that otherwise the composition of the two extracts is quite different. Thus, one can conclude that the suspension cultures produce a new vanilla aroma for which a market may or may not exist. ESCAgenetics announced that they would be able to produce tissue culture flavors such as vanilla at an incredibly low price of $50-100 kg- (MOSHY et al., 1 8 ) STAHLHUT 99. (1993) reported that the vanilla flavor from suspension cultures of V. fragruns exited the lab and en-
tered commercial development. Unfortunately, the company was not willing to support their announcements by providing any scientific data or comments of their marketed product for this review. Thus, some doubts remained as to whether PhytoVanillaTMis considerably more economic to produce than natural vanilla extract (GOLDSTEIN, ESCAgenetics). It has to be seen whether another company will continue the process of vanilla aroma production after the shutting down of ESCAgenetics. Based on the literature and the patent it is now clear that vanillin production can be induced and optimized in cultured cells of Vunilla spp. by media composition, selection, elicitation, and feeding of suitable precursors. The highest specific yields (0.16% vanillin on a dry mass basis) were reported for an embryo culture of V. plunifoliu grown in small bioreactors (KNORR et al., 1993; HAVKIN1994). A comparison of an HPLC FRENKEL, extract of the embryo culture and a Bourbons vanilla bean extract exhibited much more similarities of the components than shown in the patents of ESCAgenetics. HAVKIN-FRENKEL (1994) also compared the production costs. If the cultures produce 2% vanillin in large vessels, the estimated costs for the production of 1 kg vanillin would be between $500-1000 (investment for bioreactors not included). Although the embryo culture system contains the highest published content of vanillin reported up to now it is commercially not yet competitive. In this context it should be mentioned that alternative biotechnological approaches for the production of vanillin 19) have been investigated (CHEETHAM, 9 3 . For example, root tissue of V. pfunifofiuconverts ferulic acid to vanillin at production rates of 400 mg kg d root tissue and concentrations of 7 g kg- roots were obtained (WESTCOTT al., 1994). This suggests that et hairy root cultures should be tested for the production of vanillin by biotransformation. The production of vanillin and related compounds may also be improved in the near future on the basis of present studies on the enzymology of the formation of benzoic acids from cinnamic acids (LOSCHERand HEIDE, 1994).
- -
4 Secondary Product Formation in Hairy Root Cultures
619
are accessible to biochemical and molecular studies. Indeed, they are also amenable to genetic manipulations. The first report on the initiation of hairy root cultures for production purposes was in 1985 when FLORES and FILNER showed that Hyoscyamus spp. hairy root cultures proThe attempts at enhancing product forma- duced 0.5% tropane alkaloids on a dry mass tion in rapidly growing suspension cultures basis and those of tobacco contained more showed that some pathways are spontaneous- than 3% nicotine. This was followed by a ly well expressed under such conditions. Fur- huge number of reports on the initiation of thermore, their product formation can easily metabolite producing hairy root cultures. be optimized by rather simple methods. Poor- Thus, this review cannot cover all the producly expressed pathways, however, do not re- tive hairy root culture systems which have spond well to manipulations aimed at obtain- been described in the recent years (for a reing productivity improvements. There are un- view or literature see SIGNSand FLORES, fortunately many more compounds, not men- 1990; RHODES al., 1990 PAYNE al., 1991; et et tioned in the previous sections, which are SAITOet al., 1992; TOIVONEN, 1993; BANonly formed at low levels in undifferentiated THORPE, 1994). Naturally, plant species which cell cultures. After induction of some mor- do not produce the desired commercially imphological differentiation, however, enhanced portant compounds in suspension cultures biosynthesis of many of these compounds was were the main target. The greatest biotechnoobserved. Organ cultures of plant cells, e.g., logical efforts have been put into hairy root root and shoot cultures, are known for quite cultures of tropane alkaloid producing Solasome time. However, from a biotechnological naceae. According to the literature several point of view these cultures were of low inter- hundred individually transformed hairy root est because they are difficult to establish, they cultures have been established and analyzed often grow slowly and are sometimes very un- with respect to their ability to form tropane stable. An overview of the nevertheless sub- alkaloid (KNOPPet al., 1988; PARR et al., stantial studies on the accumulation of sec- 1990) and hairy root cultures of all known ondary compounds by organized shoot and plants (e.g., Arropa, Datura, Hyoscyamus, root plant cultures has been given by CHARL- Brugmansia, Duboisia, Scopolia) producing WOOD et al. (1990). The image of organ culthese alkaloids have been established. In gentures has greatly changed since the finding eral, it was found that hairy root cultures conthat for most dicotyledonous plants rapidly tained roughly the same amount of alkaloids growing hairy root cultures can be initiated as that found in the roots of the correspondby transformation with Agrobacterium rhizo- ing intact plant species. Root cultures may genes. Besides the progress in the biochemis- show some variation in growth rate, alkaloid try and molecular biology of plant secondary content, and productivity (MANO et al., pathways the broad application of hairy root 1989). Thus, screening for a high producing cultures has been the most important stimulus line seems to be worthwhile. However, the for the field. The special characteristics of productivity differences between 60 different transformed hairy root cultures in compari- lines of one species were only 2-3-fold on avson with normal root cultures have been re- erage. The medium composition is also imviewed by RHODESet al. (1990). They are portant for growth and alkaloid production et phytohormone-independent, exhibit rapid (MANOet al., 1989; BERLIN al., 1990). In growth and a high degree of genetic and bio- particular, the amount of nitrogen and the rachemical stability. Their biosynthetic capacity tio NOq/NH$ seem to affect branching of is equivalent to the corresponding plant root, roots and product formation. With the addiand they can be grown in bioreactors. The tion of phytohormones to the growth medium most important biotechnological value of of root cultures, more callus-like root cultures these cultures is perhaps that many pathways are formed which produce much lower levels
4 Secondary Product Formation in Hairy Root Cultures
620
of alkaloids. In contrast to suspension cultures where improved production is often observed under growth limiting conditions hairy root cultures produce best when root formation and growth is optimal. In agreement with this elicitors (stress factors) had no great effect on product levels although it induced phytoalexin production (FURZE al., 1991). et In general, it was found that the specific production of secondary metabolites cannot be manipulated as readily as in productive suspension cultures (BERLINet al., 1990 TOIVONEN, 1993). In theory, all dicot plants should be susceptible to transformation by A. rhitogenes and should yield good growing hairy root cultures. However, some plant species have been found to be rather recalcitrant. For example, transformation of Pupuver somniferum with A. rhizogenes resulted in phytohormone-independent cell suspension cultures rather than root cultures (WILLIAMSand ELLIS, 1993). Since normal root cultures of Linum fravum produce high levels of 5-methoxypodophylotoxin (BERLINet al., 1988b) it was logical to initiate hairy root cultures of this species. A successful transformation has been reported only once (OOSTHAM al., 1993). et However, this culture exhibited a lower growth rate than normal root cultures. Our own efforts to establish hairy root cultures of L. flavum have not yet been successful (BERLIN and KUZOVKINA, unpublished results). In addition to the usefulness of hairy root cultures in biochemical and molecular studies, they might also be useful for the production of desired metabolites in large-scale volumes due to their good productivity. A recent review on the cultivation of root cultures in bioreactors (CURTIS, 1993) shows that the process engineers are just at the beginning of developing suitable reactors and operation schemes. Several hairy root cultures have been grown in bioreactors with capacities of 1-20L. For example, hairy root cultures of Coleus forskohlii grown in 20 L glass jar fermentors produced forskolin (a novel heart active and blood pressure-lowering compound) with a yield of 14mgL-' after 3 weeks (KROMBHOLZ al., 1992). In suspension culet tures the forskolin content was extremely low. Many different reactor configurations have
been used for growing hairy root cultures. Good results were obtained with mechanically agitated fermentors (KONDOet al., 1989; BUITELAAR al., 1991). High growth rates et were also achieved in reactors in which the roots were fixed to supports of stainless steel or polyurethane foam and sprayed with nutrient (WILSONet al., 1990; Dr IORIO et al., 1992). A technical problem is the scale-up of hairy root cultures from small to very large bioreactors. Hairy roots form usually large aggregates which cannot be pumped through pipes from one reactor to another. We suggested (BERLIN al., 1990) to grow the phyet tohormone-independent hairy root cultures initially in the presence of phytohormones. The resulting callus-root culture with substantially shorter roots would be used for scale up. In the last bioreactor full root formation could then be reinduced by omission of the et phytohormone. TAKAYAMAal. (1994) demonstrated the feasibility of this approach for a hairy root culture of Hyoscyumus niger which was scaled up in the presence of 0.3mgL-' NAA in an agitated bioreactor. They obtained a root suspension which could be transferred through pipes (13 mm internal diameter) from one reactor to the next. Tropane alkaloid production (3% of dry mass) reached the original levels after transfer to the phytohormone-free medium. Sometimes the products of hairy root cultures are spontaneously released into the medium, from which they can easily be recovered. A scopolamine secreting line of Duboisiu leichhurdtii was cultivated in a bioreactor with continuous exchange of the medium. The product was recovered by using a XAD-2 column. Under optimized conditions (stainless steel mesh as a support, turbine-blade reactor, two-stage culture for growth and product release) a total of 1.3 g L-' scopolamine was recovered from the XAD-column during 11 weeks of continuous operation (MURANAKA al., 1993). High levels of et thiophenes are only produced in root cultures of Tugetes ssp. BUITELAAR al. (1991) grew et ' hairy roots of 7.putulu with 1.6% thiophenes on a dry mass basis in various bioreactors and continuously harvested 70% of the lipophilic compounds in a two-liquid-phase system. Interestingly, product formation and secretion
6 Biotransformations with Cultured Plant Cells
621
of secondary products into the medium was often enhanced when two-phase culture or absorption to resins was used (TOIVONEN, 1993).
5 Plant Tissue Cultures as a Source of New Chemicals?

Plant tissue cultures are not usually used as producers for commercially interesting compounds since alternative production methods (field grown plants, chemical synthesis) have been established by industry. The observation that under cell culture conditions new metabolites sometimes accumulate which had not previously been detected in the intact plant suggested that plant cell cultures may be used as a source of new chemicals. In addition, components which are present in the intact plant at low levels were found to be present at high levels in cultured cells in some cases, for example, sanguinarine in poppy plants (WILLIAMS ELLIS, 1993). Dedifferentiaand tion of the cultured plant cell leads to a regulatory state which is not present in the developing plant, and this may lead to the expreset sion of novel product patterns (ARENS al., 1982). The idea of screening plant cell cultures for novel pharmacologically active compounds was mainly followed by the Nattermann Company in Cologne and researchers of the Pharmaceutical Institute of the University of Munich. Indeed, some new biologically active compounds were detected by such screenings (ARENSet al., 1986). However, from a biotechnological point of view this approach would only be successful if the new compound had unique (superior) pharmacological characteristics and if product levels were high enough to encourage further development. Taking into account the number of extracts which had to be screened by the NCI program before a unique compound such as taxol was detected, and that product patterns and formation are much lower in cultured cells than in intact plants (BERLIN, 1986), the
chances of a successful outcome of this approach were questionable from the beginning. In the meantime, the Nattermann Company has given up this approach. Although the number of new compounds found exclusively or originally in tissue cultures has steadily increased (RUYTERand STOCKIGT, 1989), the importance of cell cultures as a source of novel compounds has not increased.
6 Biotransformations with Cultured Plant Cells

The previous sections dealt mainly with the de novo synthesis of secondary metabolites by cell suspension and hairy root cultures. The production of valuable plant products from cheap precursors by biotransformation is another possibility of using plant tissue cultures. Indeed, the 12-p-hydroxylation of p-methyldigitoxin to P-methyldigoxin by cell suspension cultures of Digitalis lanata was expected to become the first commercial tissue culture process (REINHARD ALFERMANN, and 1980). This example showed that even cultures unable to synthesize a particular compound de novo can carry out specific reactions of the corresponding pathway. Many other examples demonstrate the potential of plant cell cultures to biotransform exogenous substrates (for a review see SUGAand HIRATA,1990). However, the main question remains as to whether plant cell cultures can compete with the well-known capabilities of microorganisms in the field of biotransformations. Thus, one should look for plant-specific biotransformation reactions which are not performed by microorganisms. Glycosylation of foreign products, for example, is a reaction which is not performed by bacteria. Plant cells should also be superior to microorganisms if stereospecific enzymatic reactions of plant specific pathways are required. The two examples given below demonstrate how plant cell cultures can be used for biotechnologically relevant biotransformations.
622
6.1 Arbutin
tent in therapy, and therefore conversion of digitoxin to digoxin by 12-P-hydroxylation Arbutin has been used as a urethral disin- was regarded as a biotechnologically useful fectant for many years and has been shown to biotransformation reaction. Cell cultures of be a potent suppressor of melanin synthesis in D. lanata, although unable to produce cardiac human skin. It is thus used as a skin depig- glycosides de novo, were found to contain mentation agent. Although arbutin is found high 12-phydroxylation activity (REINHARD 1980). However, p-methylat rather high levels (54%) in several mem- and ALFERMANN, bers of the Ericaceae, it is presently produced digitoxin was initially found to be the only by chemical synthesis. The Japanese Shiseido substrate which was exclusively converted Company is trying to develop an alternative into a digoxin derivative. With all other digiproduction process by the glucosylation of hy- toxins, side reactions were observed. The use droquinone added to cell suspension cultures of pmethyldigitoxin as a substrate seemed to of Catharanthus roseus, although arbutin is be feasible since it was already being pronot a natural constituent of this plant (Yo- duced by Boehringer Mannheim. This compaKOYAMA and YANAGI, 1991). First the cul- ny initially was interested in applying cell culture stage during which highest glucosylation tures of D. lanata in the biotransformation of occurs was determined. Then the medium pmethyldigitoxin. However, the idea of using composition was optimized, and the biotrans- this technology for production was dropped formation efficiency of various strains and in the early 1980s. Nevertheless, research on and cultures were compared. By fed-batch cultiva- this system has continued. REINHARD cotion and continuous feeding of the high densi- workers at the University of Tubingen optimty cultures with substrate (14 d old, 5 L im- ized the biotransformation up to a 210 L peller driven bioreactor) arbutin formation working volume in an airlift reactor (REINwas improved to nearly 1OgL-' within 4 HARD et al., 1989). During a semicontinuous days. A specific content of 45% arbutin per g process (6 repeated batch cultivations) 513 g dry mass was tolerated by the cells which is pmethyldigoxin was produced from 641 g Pthe highest specific content of a secondary methyldigitoxin within 3 months. Since the metabolite reported for cultured plant cells. It product is released into the medium further was stated that the production costs for arbu- improvements in shortening the production tin by chemical synthesis or biotransforma- period are expected. Optimization of the biotion in tissue cultures are very similar (Yo- transformation process did not only include KOYAMA and YANAGI, 1991). They are rath- studies on optimal substrate supply and meer optimistic that the biotechnological pro- dium composition but also selection of an imduction of useful glucosides will become in- proved strain for highest hydroxylation cacreasingly inexpensive relative to chemical pacity by repeated cell aggregate cloning et synthesis. Recently, it was shown that cell sus- (REINHARD al., 1989). pension cultures of Rauwolfia serpentina are When digitoxin was added to cell cultures also able to form efficiently arbutin from hy- of D. lanata two reactions occurred 12-phydroquinone (18 g L-' in 7 d) (LU'TTERBACH droxylation and 16'-O-glucosylation yielding and STOCKIGT, 1992). deacetyllanatoside C as the main product. KREIS and REINHARD (1990) developed a two-stage semicontinuous process by transferring a selected cell line into a production me6.2 Biotransformation of Cardiac dium (8% glucose) for the biotransformation Glycosides reaction. On average 400 mg L-' deacetyllanatoside C were produced from 600mg digiDigitalis lanata plants contain two main toxin in a 20 L airlift fermentor within 7 d. To cardiac glycosides which upon hydrolysis achieve a high biotransformation of digitoxin yield digoxin and digitoxin, drugs which are into one major cardenolide, it was important widely used in the treatment of heart dis- to change from a growth to a production meeases. Digitoxin is used to a much lesser ex- dium. In contrast to pmethyldigoxin, deace-
7 Metabolic Engineering of Secondary Pathways in Cultured Cells
623
tyllanatoside C is mainly found within the cells. The main goal, however, was to develop a biotransformation production process from digitoxin to digoxin by preventing the 16'-0glucosylation. Since the optimum temperature of 12-p-hydroxylase is around 20C while that of glucosyltransferase is 37"C, lower temperatures (e.g., 19OC) would therefore favor hydroxylation (KREISand REINHARD, 1992). Deacetyllanatoside is mainly stored in the vacuoles of the cells while digoxin is released into the medium. The lower the cell density was, the lower was the storage capacity for deacetyllanatoside and the higher was the percentage of digoxin formation. Based on this knowledge a semicontinuous production process in a 300 L airlift reactor was developed with 8% glucose as a production medium and incubation times of 40-60 h. From 0.8 mmol digitoxin, ca. 0.6 mmol digoxin were obtained, 80% of which was found in the medium (KREISand REINHARD, 1992). The latter result demonstrates in particular the importance of understanding the physiology and biochemistry of a pathway if natural products are used in biotransformation processes (KREISet al., 1993).

Many reviews have pointed out that more knowledge of the regulation of secondary pathways at the enzyme and gene level is needed before the true potential of cultured plant cells for metabolite production can be recognized and exploited. The above analysis of the field clearly supports this view. Despite all the progress, the fact remains that the biology of the cell is the limiting factor since only those pathways which are spontaneously well expressed under culture conditions or easily inducible by simple modifications are accessible for the development of biotechnological processes. This can only be changed if we succeed in altering the expression of pathways by
genetic engineering techniques. Indeed, several groups worldwide have used their culture systems (as indicated in the previous sections) not only for optimizing product levels but also for biochemical and molecular studies. Thus, our knowledge of the enzymology of several pathways has greatly increased during the last few years. The genes coding for several of these enzymes have been cloned. For the phenylpropanoid/flavonoid pathway, which has been analyzed the most extensively, the first regulatory elements have been detected (FORKMANN, 1993; KOES et al., 1994). Thus, we seem to be at the beginning of an era when aimed manipulations of secondary pathways might become possible. This might also tremendously increase the production potential of plant cell cultures and thus their biotechnological impact. However, progress in this direction is likely to be slow since even in microbial systems the success of metabolic engineering is not yet overwhelming (CAMERON and TONG,1993; STEPHANOPOLOUS and SINSKY, 1993). In order to devise rational approaches for metabolic engineering a knowledge of the limiting steps of the pathways is required (STEPHANOPOULOS and SINSKY, 1993). Locating such critical steps is often rather difficult. In the case of non-expressed pathways in plant cells, a one-step manipulation will usually not be sufficient. Nevertheless, the first genetic manipulations of secondary pathways in plant cells have been performed. However, only a few of them have been directed towards the production of secondary metabolites. In most cases, the establishment of transgenic plants rather than cultures with altered characteristics has been attempted. As the expression of secondary pathways is quite different in intact plants and in cultured cells, the metabolic effects of a genetic transformation in these systems will also be different. This review focusses on the manipulation of product formation in cultured cells which might later be grown in bioreactors. Thus, only the results of transgenic cell cultures which were created for the overproduction of desired compounds by various biosynthetic pathways will be described in some detail. Nevertheless, it should be mentioned that genetic engineering of transgenic plants with desired characteristics is in
624
full progress. For example, Calgene and Suntory have announced the development of blue and TANAKA, 1994). A genroses (HOLTON eral overview of metabolic engineering of commercially useful biosynthetic pathways in transgenic plants and plant cells were given (1993) and by KISHOREand SOMERVILLE NESSLER (1994).
of Catharanthus roseus (GODDIJN, 1992) was introduced into P. harmala cells under the control of the 35s promoter of the Cauliflower Mosaic Virus via Agrobacterium tumefaciens (BERLIN al., 1993). Several cell suset pension cultures were obtained with constitutively expressed TDC activities of around 30 pkat/mg protein and serotonin levels of 12% of dry mass; the serotonin levels of the controls were below 0.1%. It was also shown that in these transgenic cultures tryptophan 7.1 Serotonin Biosynthesis in supply is the next rate-limiting factor (BERPeganum harmala LIN et al., 1993) which has to be overcome to obtain even higher levels of serotonin. AlCell cultures of Peganum harmala loose though serotonin biosynthesis is not repretheir capability of synthesizing serotonin sentative of a typical secondary pathway be(Fig. 8) during prolonged subcultivation. It cause of its simplicity, this example shows was found that this was due to the lack of that removal of a rate-limiting step (clearly tryptophan decarboxylase (TDC) activity in identified by the feeding experiments with fully undifferentiated cell suspension cultures. tryptamine) can help to increase the producThe second and final enzymatic step of sero- tion rate. tonin biosynthesis, the 5-hydroxylation of The same tdc gene was also overexpressed tryptamine, however, remains highly ex- in C. roseus, the plant from which it was depressed in all P. harmala cell cultures (SASSE rived. However, as expected from literature et al., 1987; COURTOIS al., 1988). Thus, it data, the constitutive expression of TDC only et was a rational approach to introduce a consti- led to a large accumulation of tryptamine and tutively expressed tdc gene into P. harmala did not affect alkaloid levels (GODDIJN, cell suspension cultures to restore or maintain 1992). Feeding of tryptamine does not enserotonin formation. A corresponding cDNA hance alkaloid formation in Catharanthus cell suspension cultures. The overproduction of tryptamine by the constitutive expression of the tdc gene had evidently no negative effect on tryptophan supply for protein synthesis. Anthranilate synthase levels (the proposed regulatory enzyme of tryptophan biosyntheH sis) were unchanged in transgenic tobacco Serotonin et plants (POULSEN al., 1994). An interesting, yet inexplicable observation was made by SONGSTAD al. (1991). They found that rdcet transgenic tobacco plants not only overproN duce tryptamine but also produce high levels of tyramine, although the engineered enzyme Anabasine does not accept tyrosine as substrate.
d?
Scopolamine
FH~OH
Fig. 8. Secondary metabolites whose levels were enhanced in cultured cells by metabolic engineering.
7.2 Affecting Nicotine Alkaloid Biosynthesis in Tobacco

The regulation of nicotine biosynthesis is relatively well understood (for a review see HASHIMOTO YAMADA, and 1994). The activ-
625
ity levels of putrescine N-methyltransferase zyme had been targeted to leucoplasts, the (PMT) determine whether reasonable levels site of lysine biosynthesis in root cells. High of nicotine can be formed in tobacco suspen- levels of cadaverine (0.5% of dry mass; consion cultures. As this enzyme is also a key en- trol cultures, ca. 0.01%) and anabasine (0.5% zyme in tropane alkaloid biosynthesis, great of dry mass; controls, 04.02%) were found in efforts have been made to clone this gene, the most productive root culture (HEReven by researchers from industry. Philip MINGHAUS et al., 1996). Feeding of lysine enMorris (NAKATANI MALIK,1992) filed a hanced both cadaverine and anabasine levels and patent on the production of transgenic tobac- to more than 1% of dry mass while metabolco plants overexpressing PMT. However, re- ite levels of control cultures were hardly afports on such transgenic plants overproducing fected (HERMINGHAUS al., 1996). Nicotine et PMT have not yet been published. was still the main alkaloid in these cultures. The first attempts to affect nicotine biosyn- However, anabasine became the second most thesis genetically have used the gene of a abundant alkaloid. Improving the internal yeast ornithine decarboxylase (odc) (HAMILL supply of lysine by altering the feedback conet al., 1990). This enzyme activity, however, is trol of lysine biosynthesis (SHAULand GALnot generally regarded as a rate-limiting step ILI, 1991) might lead to even higher levels of of the pathway since feeding putrescine does anabasine. Lysine decarboxylase expressed in not affect nicotine levels. Transgenic root cul- the leaves resulted in the production of cadtures of Nicotiana rustica constitutively ex- averine only (HERMINGHAUS al., 1996). et pressing the yeast gene under the control of This shows that genetic manipulation at one the CaMV35S promoter contained distinctly biosynthetic node will lead to significant imenhanced ODC activity. However, the high provements of product levels only when sevenzyme activity did not result in correspond- eral prerequisites are fulfilled. Firstly, the taring high levels of nicotine as product levels get enzyme must be the rate-limiting step; were only increased 2-fold. In view of the fact secondly, the engineered enzyme must be sufthat putrescine is not a limiting intermediate ficiently supplied with substrate; the reaction in nicotine biosynthesis, the outcome was not product must be transported to the site from so surprising. It is not yet clear whether the where it is channelled into the related path2-fold increase is really due to the increased way. The latter means that the other enzymes enzyme activity. Feeding of the cultures with of the pathway should be expressed quite ornithine could provide more insight into the well. Therefore, genetic manipulations in uninfluence of the engineered ODC on the nico- differentiated cells, in organ cultures, or in whole plants might yield completely different tine levels. Anabasine (Fig. 8) is an analog of nicotine results depending on the overall expression of in which the N-methyl pyrrolinium ring is re- a biosynthetic pathway in various tissues. placed by a A l-piperideinium ring. This ring is derived from lysinekadaverine, and it has clearly been shown that cadaverine is the 7.3 Enhancing Scopolamine rate-limiting intermediate in anabasine biosynthesis (WALTON and BELSHAW,1988). Production in Atropa belladonna Therefore, a bacterial lysine decarboxylase (Idc) gene was introduced into tobacco root Hyoscyamine 6p-hydroxylase catalyzes the cultures (BERLIN al., 1994, HERMINGHAUS oxidative reactions in the biosynthetic pathet et al., 1996). When the gene was placed under way leading from hyoscyamine to scopolamthe control of the CaMV35S promoter and ine (Fig. 8). A cDNA coding for this enzyme fused to the coding sequence of the transit with epoxidating and hydroxylating activities peptide of the small subunit of ribulose di- was cloned using RNA from a root culture of phosphate carboxylase, many root cultures Hyoscyamus niger (reviewed by HASHIMOTO 1994). The cDNA was cloned with LDC activity were obtained. This en- and YAMADA, zyme activity was not detected in any of the into a binary vector for plant transformation control cultures. It was shown that the en- under the control of the CaMV35S promoter
626
and integrated into Agrobacterium rhizogenes 15834 (HASHIMOTO al., 1993). Hairy root et cultures were obtained by infecting Atropa belladonna leaf disks. Hyoscyamine 6P-hydroxylase activities were 3- to 5-fold increased in the transgenic root cultures compared to the controls. Scopolamine levels of the controls were ca. 0.05% of dry mass, hyoscyamine being the main alkaloid. In the best transgenic lines scopolamine levels were up to 5-fold higher while the hyoscyamine content was greatly reduced (HASHIMOTO al., et 1993). However, there is evidence that the expression of the engineered enzyme is not yet optimal in root tissues. When the vector was introduced into an Agrobacterium tumefaciens strain transgenic A. belladonna plants which accumulated high levels of scopolamine in leaves and stems were obtained; the amount of scopolamine formation in the roots of the same plants was not very high (YUN et al., 1992).
7.4 Does Genetic Engineering Improve the Potential of Plant Tissue Cultures as Producers of Interesting Metabolites?
' The examples given above show that the levels of end products of biosynthetic pathways can be modified by genetic engineeriqg methods. Best results will be obtained when the overproduced enzyme activity is clearly the rate-limiting step of the pathway. However, usually a pathway has a series of ratelimiting steps. Thus, if one step has been removed by genetic engineering, the next (e.g., substrate supply) may soon become active and limit the extent of overproduction. The systems we have been working on (serotonin and anabasine biosynthesis) could be used to check whether by a second transformation the second rate-limiting step might also be overcome. The genes controlling lysine and tryptophan biosynthesis have been cloned. On the other hand, it is clear that the identification of rate-limiting steps of pathways is usually not as easy as in our systems where simple feeding experiments provided the in-
formation. Thus, one can understand and share the sceptical view of HASHIMOTO and YAMADA(1994) that single site manipulations of secondary pathways will usually not be sufficient for high increases of pathway end product levels. On the other hand, one should realize that we are just at the beginning of a new era. Therefore, conclusions should not be drawn before the first double transformants have been obtained. In addition, the product levels which are not overwhelming so far may increase further when expression and targeting of the engineered protein is optimized. The greatest potential of metabolic engineering for improving the standing of plant cell cultures will, however, only become apparent if we succeed in identifying regulatory genes which control the expression of whole pathways in cultured cells. It is possible that pathways which are not usually in operation might become active in undifferentiated cells. We are still far away from the identification of elements which regulate the tissue- and organ-specific initiation of a whole secondary pathway. However, a promoter analysis and the search for transduction signals and transcription factors involved in the regulation of secondary pathways have already been and started (DIXON LAMB,1990 KOESet al., 1994). Thus, it may be possible to express more pathways in cultured cells in the not too distant future. In some situations it may be desirable to turn off or fortify one branch of a pathway so that more or less carbon enters that branch. Overexpression of a tryptophan decarboxylase gene in Brassica napus led to the accumulation of high levels of tryptamine, while the levels of unwanted indole glucosinolates in seeds of the transgenic plants were only 3% of that found in seeds of untransformed plants (CHAVADEJ al., 1994). Reduction of et metabolite levels has also been achieved by expressing antisense RNA which is complementary to the mRNA encoding a pathway enzyme. Antisense expression of a chalcone synthase gene resulted in distinctly decreased flower pigmentation (VAN DER KROL et al., 1988). Overexpression of an engineered enzyme may lead to an accumulation of intermediates
8 Conclusions and Outlook
627
of a pathway to high levels. Examples are the overexpression of the amino acid decarboxylases leading to high accumulation levels of the corresponding amines (BERLIN et al., 1994). In addition new products may be formed in cultured cells. For example, formation of resveratrol in tobacco occurred only if stilbene synthase genes from Aruchis hypoet gueu or Vitis viniferu were expressed (HAIN al., 1990 1993). Expression of a coriander desaturase results in the formation of petroselinic acid, a novel product of tobacco (CAHOON et al., 1992). However, engineering of novel enzyme reactions in a plant cell will only be useful if a corresponding substrate is naturally formed in the transgenic cells. Expression of a strictosidine synthase gene, coding for a key enzyme of monoterpene alkaloid biosynthesis, in tobacco, led to an active enzyme without function because its substrates tryptamine and secologanin are not formed in tobacco (MCKNIGHT al., 1991). et The enzymes of secondary metabolism in plant cell cultures have been regarded as a pot of gold (ZENK, 1991). There is no doubt that certain plant cell culture systems have been most important for recent progress in the enzymology of secondary pathways. It remains to be seen whether these enzymes are also a pot of gold for biotechnology purposes. However, it is now possible to create plant cell cultures which overproduce certain enzymes of secondary metabolism by genetic engineering. When the enzyme itself is the target compound for overproduction, plant cell cultures are not considered to be competitive (KUTCHAN al., 1994). These et authors recommended the heterologous expression of plant genes in insect cultures as a more convenient approach, because of their higher productivity and secretion of the enzymes into the medium from which they are easily purified. Using a cell culture of Spodopteru frugiperdu (Sf9) up to 4 mg L- strictosidine synthase of berberine bridge enzyme was produced (KUTCHAN al., 1994). The et real potential of transgenic plant cell cultures as enzyme producers, however, has not yet been exploited. Using targeting signals and suitable promoters it may be possible to produce comparable levels of a desired protein in plant cells.
8 Conclusions and Outlook

Substantial progress has been made in the biochemistry and molecular biology of secondary pathways of higher plants since the first edition of Biotechnology. Many novel enzyme reactions have been detected, and a number of genes encoding components of secondary metabolism have been cloned. Successful alterations of secondary pathways by genetic transformation have been achieved. With the availability of stable and rapidly growing hairy root cultures, the number of products found at high levels in cultured cells has increased greatly. The technological feasibility of large-scale fermentations of suspension cultures has been clearly demonstrated. The technical problems of growing hairy root cultures in huge bioreactors are surmountable. However, despite all scientific progress, the enthusiasm of industry for plant cell cultures as producers has not yet correspondingly increased. Indeed, there is no reason to believe that industries will change their reserved attitude in the foreseeable future. One can only hope that this conclusion is wrong, and that the following statement made at the International Tissue Culture Congress is correct: Scientists from universities and research centers have not yet realized what is really going on industry with respect to the tissue culture technology (SMITH,1995). Although this optimistic view is shared by some other American colleagues (TATICEK al., et 1994) the future does not look so promising. As a consequence of the overall low interest of the industrial sector in plant secondary metabolites and plant tissue culture technology, the governmental support for this field of biotechnology was reduced. Several leading groups of biotechnologically orientated governmental research centers, e.g., in England and Canada, were recently forced to leave the field. Thus, this field will remain an exciting area of basic research but not of biotechnology. It remains to be seen whether the future progress in understanding and manipulating regulatory controls of secondary pathways will eventually enhance the biotechnological
628
BANTHORPE, V. (1994), Secondary metabolism D. in plant tissue culture: scope and limitations, Nut. Prod. Rep. 11, 303-328. BATE,N. J., ORR,J., NI, W., MEROMI, NADA., R. LER-HASSAR, DOERNER, W., DIXON, T., P. Y. A., LAMB, J, ELKIND, (1994), Quantitative C. relationship between phenylalanine ammonialyase levels and phenylpropanoid accumulation in transgenic tobacco identifies rate-determining step in natural product synthesis, Proc. Natl. Acknowledgement Acad. Sci. USA 91,7608-7612. BAUCH,H. J., LEISTNER, (1978), Aromatic meE. 1 would like to thank Dr. HARA,Mitsui Co., tabolites in cell suspension cultures of Galium Dr. SMITH,Urbana, and Dr. TAKAYAMA, mollugo, Planta Med. 33, 105-127. Shizuoka, for providing me with new and un- BERLIN, (1986), Secondary products from plant J. published information. Special thanks are due cell cultures, in: Biotechnology, Vol. 4, 1st Edn. (REHM, H.-J., REED, G., Eds.), pp. 630-658. to Dr. HAVKIN-FRENKEL, David Michael & Weinheim: VCH. Co,, for her helpful advice regarding flavor BERLIN,J. (1988), Formation of secondary metaproduction in tissue cultures. bolites in cultured plant cells and its impact on pharmacy, in: Biotechnology in Agriculture and Y. Forestry, Vol. 4 (BAJAJ, P. S., Ed.), pp. 37-59. Berlin: Springer-Verlag. BERLIN, (1990), Screening and selection for varJ. iant cell lines with increased levels of secondary metabolites, in: Secondary Products from Plant B. Tissue Culture (CHARLWOOD, V., RHODES, ARENS, H., BORBE, H. O., ULBRICH, B., M. J. C., Eds.), pp. 119-137. Oxford: ClarenJ. STOCKIGT, (1982), Detection of pericine, a don. new CNS-active indole alkaloid from Picralima J., nitida cell suspension culture by opiate receptor BERLIN, SASE, F. (1985), Selection and screening techniques for plant cell cultures, Adv. Biobinding studies, Planta Med 40,218-223. ARENS, ULBRICH, FISCHER, PARNHAM, chem. Eng. 31, 99-132. H., B., H., J., K. M. J., ROMER,A. (1986), Novel antiinflammato- BERLIN, KNOBLOCH, H., HOKE, G., WITTE, L. (1982), Biochemical characterization of two ry flavonoids from Podophyllum verspille cell tobacco cell lines with high and low yields of cinculture, Planta Med. 52, 468-473. namoyl putrescines, J. Nut. Prod. 45, 83-87. ASAKA,I., 11, I., HIROTANI, ASADA,Y., FuM., J., J., RUYA, T. (1993), Production of ginsenoside sa- BERLIN, FORCHE,E., WRAY,V., HAMMER, HOSEL,W. (1983), Formation of benzophenanponins by culturing ginseng (Panax ginseng) emthridine alkaloids by suspension cultures of bryogenic tissues in bioreactors, Biotechnol. Eschscholtzia californica Z. Naturforsch. 38c, Lett. 15, 1259-1264. 346-352. BALANDRIN, F., KLOCKE, A., WURTELE, M. J. E. S., BOLLINGER, H. (1985), Natural plant BERLIN,J., BEIER,H., FECKER,L., FORCHE,E., W. Noe, W., SASSE, F., SCHIEL,O., WRAY, V. chemicals: Sources of industrial and medicinal (1985), Conventional and new approaches to inmaterials, Science 228, 1154-1160. crease alkaloid production of plant cell cultures, BALLICA, RYU,D. D. Y. (1994), Tropane alkaR., in: Primary and Secondary Metabolism of Plant loid production from Datura stramonium: An inCell Cultures (NEUMANN, H., BARZ, W., K. tegrated approach to bioprocess optimization of REINHARD, E., Eds.), pp. 272-280. Berlin: plant cultivation, in: Advances in Plant BiotechSpringer-Verlag. nology. Studies in Plant Sciences, Vol. 4 (RYu, S., D. D. Y., FURASAKI,Eds.), pp. 221-254. Am- BERLIN,J., SIEG, S., STRACK,D., BOKERN,M., sterdam: Elsevier. HARMS,H. (1986). Production of betalains by BALSEVICH, (1988), Monoterpene indole alkaJ. suspension cultures of Chenopodium rubrum, loids from Apocynaceae other than CatharanPlant Cell Tissue Organ Cult. 5, 163-174. thus roseus, in: Cell Culture and Somatic Cell Ge- BERLIN, J., MOLLENSCHOTT, C., WRAY, V. netics, Vol. 5 Phytochemicals in Plant Cell Cul(1988a), Triggered efflux of protoberberine afF., tures (CONSTABEL, VASIL, I. K., Eds.), kaloids from cell suspension cultures of Thalicpp. 371-384. San Diego: Academic Press. trum rugosum, Biotechnol. Lett. 10, 193-198.
importance of plant cell cultures as producers. The decisive question, which can only be answered by each individual company, is whether the commercial importance of any plant-tissue-culture-derived product is high enough to justify the undoubtedly high investments needed for large-scale production.
References
9 References
629
BERLIN, J., BEDORF, N., MOLLENSCHOTT, C., WRAY,V., SASSE,F., HOFLE, G. (1988b), On the podophyllotoxins of root cultures of Linum flavum, Planta Med. 54, 204-206. BERLIN, MOLLENSCHOTT, GREIDZIAK, J., C., N., ERDOGAN, KUZOVKINA, (1990), Affecting S., I. secondary product formation in suspension and hairy root cultures - a comparison, in: Progress in Plant Cellular and Molecular Biology (NIJKAMP, J. J., VAN DER PLAS,L. H. W., VAN H. AARTRIJK, Eds.), pp. 763-768. Dordrecht: J., Kluwer. J., C., BERLIN, RUGENHAGEN, DIETZE,P., FECKER, L. F., GODDIJN, J. M., HOGE,J. H. C. 0. (1993), Increased production of serotonin by suspension and root cultures of Peganum harmala transformed with a tryptophan decarboxylase cDNA from Catharanthus roseus, Transgen. Res. 2, 336-344. BERLIN,J., FECKER,S., HERMINGHAUS, RUC., GENHAGEN (1994), Genetic modification of plant secondary metabolism: alteration of product levels by overexpression of amino acid decarboxylases, in: Advances in Plant Biotechnology. Studies in Plant Science, Vol. 4 (RYu, D. D. Y., FURASAKI, Eds.), pp. 57-81. Amsterdam: S., Elsevier. BIANCO-COLOMAS, HUGUES, (1990), EstabJ., M. lishment and characterization of a betacyanin producing cell line of Amaranthus tricolor: Inductive effects of light and cytokinins, J. Plant Physiol. 136,734-739. BOHM,H., RINK, (1988), Betalains, in: Cell CulE. ture and Somatic Cell Genetics of Plants, Vol. 5: Phytochemicals in Plant Cell Cultures (CONSTABEL, F., VASIL,I. K., Eds.), pp. 449-463. San Diego: Academic Press. BOKERN, WRAY, STRACK, (1991), AccuM., V., D. mulation of phenolic conjugates and betacyanins, and changes in the activities of enzymes involved in feruloylglucose metabolism in cell suspension cultures of Chenopodium rubrum, Planta 184, 261-270. BREULING, ALFERMANN, W., REINHARD, M., A. E. (1985), Cultivation of cell cultures of Berberis wilsonae in 20-L airlift bioreactors, Plant Cell Rep. 4,220-223. BRINGI, KADKADE, (1993), Enhanced proV., P. duction of taxol and taxanes by cell cultures of T a u s species, WO 93/17121. BUITELAAR, M., LANGENHOFF, A. M., R. A. J. HEIDSTRA, TRAMPER, (1991), Growth and R., thiophene production by hairy root cultures of Tagetes patula in various two-liquid-phase bioreactors, Enzyme Microb. Technol. 13, 487-494. BUITELAAR, M., TRAMPER,(1992), Strategies R. J. to improve the production of secondary meta-
bolites with plant cell cultures: a literature review, J. Biotechnol. 23, 111-141. CAHOON, B., SHANKLIN, OHLROGGE, B. E. J., J. (1992), Expression of a coriander saturase results in petroselinic acid production in transgenic tobacco, Proc. Natl. Acad. Sci. USA 89, 1118411 188. CAMERON, D., TONG, T. (1993), Cellular and D. I. metabolic engineering - an overview, Appl. Biochem. Biotechnol. 38, 105-140. CAREW, P. (1975), Tissue culture studies of CaD. tharanthus roseus, in: The Catharanthus Alkaloids (TAYLOR, I., FARNSWORTH, R., W. N. Eds.), pp. 193-208. New York: Marcel Dekker. Celex-Laboratories (1994), Production of Taxus sp. hairy root culture - taxol, taxane and diterpene production by Agrobacterium rhizogenes transformed yew culture, WO 94/20606. CHARLWOOD, V., RHODES, J. C. (1990), SecB. M. ondary products from plant tissue culture, Proc. Phytochem. Soc., Vol. 30, Oxford Clarendon. CHARLWOOD, V., CHARLWOOD, A,, MOLIB. K. J. NA-TORRES, (1990), Accumulation of secondary compounds by organized plant cultures, in: Secondary Products from Plant Tissue Culture (CHARLWOOD, V., RHODES,M. J. C., Eds.), B. pp. 167-200. Oxford: Clarendon. CHAVADEJ, BRISSON, MCNEIL,J. N., DES., N., LUCA, V. (1994), Re-direction of tryptophan leads to production of low indole glucosinolate canola, Proc. Natl. Aad. Sci. USA 91, 21662170. CHEETHAM, S. J. (1993), The use of biotransforP. mations for the production of flavours and frangrances, TIBTECH 11,478-488. CLINE, . D., COSCIA,C. J. (1988), Stimulation of S sanguinarine production by combined fungal elicitation and hormonal deprivation in cell suspension cultures of Papaver bracteatum, Plant Physiol. 86, 161-165. CONSTABEL, VASIL,I. K. (Eds.) (1988), Cell F., Culture and Somatic Cell Genetics of Plants, Vol. 5: Phytochernicals in Plant Cell Cultures. San Diego: Academic Press. COURTOIS, D., YVERNEL,D., FLORIN,B., PETIARD, V. (1988), Conversion of tryptamine to serotonin by cell supension cultures of Peganum harmala, Phytochemistry 27, 3137-3141. CRAGG, M., SCHEPARTZ, A., SUFFNESS, G. S. M., GREVER, R. (1993), The taxol supply crisis. M. New NCI policies for handling the large-scale production of novel anticancer and anti-HIV agents, J. Nut. Prod. 56, 1657-1668. CURTIS, R. (1993), Cultivation of roots in bioW. reactors, Curr. Opin. Biotechnol. 4, 205-210. DE CAROLIS, DE LUCA,V. (1993), Purification, E., characterization, and kinetic analysis of a 2-0x0-
630
13 Secondary Products from Plant Cell Cultures and evaluation, Appl. Microbiol. Biotechnol. 37, 457-462. DIXON, A., LAMB,C. J. (1990), Regulation of R. secondary metabolism at the biochemical and genetic levels, in: Secondary Products from Plant B. Tissue Culture (CHARLWOOD, V., RHODES, M. J. C., Eds.), pp.103-118. Oxford Clarendon. Do, C. B., CORMIER, (1991), Effects of high amF. monium concentrations on growth and anthocyanin formation in grape (Vitis vinifera L.) cell suspension cultured in a production medium, Plant Cell Tissue Organ Cult. 27, 169-174. DOONER, K., ROBBINS, P., JORGENSEN, H. T. R. A. (1991), Genetic and developmental control of anthocyanin biosynthesis, Annu. Rev. Genet. 25, 173-1 99. DRAPEAU, D., BLANCH, W., WILKE, C. R. H. (1987), Economic assessment of plant cell culture for the production of ajmalicine, Biotechnol. Bioeng. 30,946-953. S. EDGINGTON, M. (1991), Taxol - out of the woods, Bio/Technology 9,933-938. EISENREICH, MENHARD, HYLANDS, J., W., B., P. ZENK,M. H., BACHER, (1996), Studies on A. the biosynthesis of taxol: the taxane carbon skeleton is not of mevalonoid origin, Proc. Natl. Acad. Sci. USA 93,6431-6436. F. EILERT,U., KURZ, W. G. W., CONSTABEL, (1985), Stimulation of sanguinarine accumulation in Papaver somniferum cells by fungal homogenates, J. Plant Physiol. 119, 65-76. ELLIS,B. E. (1983), Production of hydroxyphenylethanol glycosides in suspension cultures of Syringa vulgaris, Phytochemistry 22, 1941-1943. ELLIS,B. E. (1985), Metabolism of caffeoyl derivatives in plant cell cultures, in: Primary and Secondary Metabolism of Plant Cell Cultures (NEUMANN, H., BARZ,W., REINHARD, Eds.), K. E., pp. 164-173, Berlin: Springer-Verlag. ELLIS,B. E. (1988), Natural products from plant tissue culture, Nut. Prod. Rep. 5 , 581-612. F., W. FEIT-NETO, A. G., DICOSMO, REYNOLDS, F., SAKATA, (1992), Cell culture of Taxus as a K. source of the antineoplastic drug taxol and related taxanes, Bio/Technology 10, 1572-1575. FETT-NETO,A. G., MELANSON, J., SAKATA, S. K., DICOSMO, (1993), Improved growth and taxol F. yield in developing calli of Taxus cuspidata by medium composition modification, Bio/TechnolOgy 11,731-734. FEIT-NETO, A. G., ZHANG, Y., DICOSMO, W. F. (1994a), Kinetics of taxol production, growth and nutrient uptake in cell suspensions of Taxus cuspidata, Biotechnol. Bioeng. 44,205-210. FEIT-NETO,A. G., MELANSON, J., NICHOLSON, S. S. A., PENNINGTON, DICOSMO, (1994b), J. J., F.
glutarate-dependent dioxygenase involved in vindoline biosynthesis from Catharanthus roseus, J. Biol. Chem. 268,5504-5511. DE-EKNAMKUL, ELLIS,B. E. (1984), RosmarW., inic acid production and growth characteristics of Anchusa officinalis cell suspension cultures, Planta Med. 50, 346-350. DE-EKNAMKUL, ELLIS,B. E. (1985a), Effects W., of auxins and cytokinins on growth and rosmarinic acid formation in cell suspension cultures of Anchusa offcinalis, Plant Cell Rep. 4, 50-53. DE-EKNAMKUL, ELLIS,B. E. (1985b), Effects W., of macronutrients on growth and rosmarinic acid formation in cell suspension cultures of Anchusa officinalis, Plant Cell Rep. 4, 46-49. DELL,B., ELSEGOOD, L., GHISALBERTI, L. C. E. (1989), Production of verbascoside in callus tissue of Eremophila spp., Phytochemistry 28, 1871-1872. DELUCA,V. (1993), Enzymology of indole alkaloids, in: Methods of Plant Biochemistry, Vol. 9 (DEY,P. M., HARBORNE, B., Eds.), pp. 345J. 368. London: Academic Press. DELUCA,V., KURZ,W. G. W. (1988), Monoterpene indole alkaloids (Catharanthus alkaloids), in: Cell Culture and Somatic Cell Genetics of Plants, Vol. 5: Phytochemicals in Plant Cell CulF., tures (CONSTABEL, VASIL, I. K., Eds.), pp. 385-402. San Diego: Academic Press. DELUCA, MARINEAU, BRISSON, (1989), V., C., N. Molecular cloning and analysis of a cDNA encoding plant tryptophan decarboxylase: comparison with animal dopa decarboxylase, Proc. Natl. Acad. Sci. USA 86, 9969-9973. DENO,H., SUGA,C., MORIMOTO, FUJITA, T., Y. (1987), Production of shikonin derivatives by cell suspension cultures of Lithospermurn erythrorhizon. VI. Production of shikonin derivatives by a two layer culture containing an organic solvent, Plant Cell Rep. 6, 197-199. DEUS, B., ZENK,M. H. (1982), Exploitation of plant cells for the production of natural compounds, Biotechnol. Bioeng. 24, 1965-1974. DEUS-NEUMANN, ZENK, H. (1984), InstabilB., M. ity of alkaloid production in Catharanthus roseus cell suspension cultures, Planta Med. 50, 427431. DICOSMO, (1990), Strategies to improve yields F. of secondary metabolites to industrially interesting levels, in: Progress in Plant Cellular and MoH. lecular Biology (NIJKAMP, J. J., VAN DER PLAS, L. H. W., VAN AARTRIJK, Eds.), J., pp. 717-724. Dordrecht: Kluwer. DI IORIO, A., CHEETHAM, D., WEATHERS, A. R. P. J. (1992), Growth of transformed roots in a nutrient mist bioreactor: Reactor performance
9 References
631
Improved taxol yield by aromatic and amino Plant Cell Culture 1982 (FUJIWARA, Ed.), A., acid feeding to cell cultures of Taxus cuspidata, pp. 313-314. Tokyo: Maruzen Press. Biotechnol. Bioeng. 44,967-971. FUKUI, TSUKADA, MIZUKAMI, TABAH., M., H., TA, M.(1983), Formation of stereoisomeric mixFLORES, E., FILNER, (1985), Metabolic relaH. P. tures of naphthoquinone derivatives in Echium tionships of putrescine, GABA and alkaloids in lycopsis callus cultures, Phytochemistry 22,453cell and root cultures of Solanaceae, in: Primary 456. and Secondary Metabolism of Plant Cell Cultures P. (NEUMANN, H., BARZ,W., REINHARD, K. E., FUNK, C., BRODELIUS, (1990), Influence of growth regulators and elicitor on phenylpropanEds.), pp. 174-185. Berlin: Springer-Verlag. oid metabolism in suspension cultures of Vanilla FLORES, WAGNER, J., FLORES, E. (1993), T., L. H. planifolia, Phytochemistry 29,845-848. Embryo culture and taxane production in Taxus FUNK, BRODELIUS, (1992), Phenylpropanoid C., P. spp., In Vitro Cell. Dev. Biol. 29P,160-165. metabolism in suspension cultures of Vanilla FORKMANN, (1993), Control of pigmentation in G. planifolia. IV. Induction of vanillic acid formanatural and transgenic plants, Curr. Opin. Biotion, Plant Physiol. 99, 256-262. technol. 4, 159-165. T. FUJITA,Y. (1988), Shikonin: Production by (Li- FURUYA, (1988), Saponins (Ginseng saponins), in: Cell Culture and Somatic Cell Genetics of thospermum erythrorhizon) cell cultures, in: BioPlants, Vol. 5: Phytochemicals in Plant Cell Cultechnology in Agriculture and Forestry, Vol. 4 F., tures (CONSTABEL, VASIL, I. K., Eds.), (BAJAJ, Y. P. S., Ed.), pp. 225-236. Berlin: pp. 213-236. San Diego: Academic Press. Springer-Verlag. A., K. T., FUJITA, HARA, (1985), The effective produc- FURUYA, IKUTA, SYONO, (1972). AlkaY., Y. loids from callus tissue of Papaver somniferum, tion of shikonin by cultures with an increased 1 Phytochemistry 1 ,3041-3044. cell population, Agric. Biol. Chem. 49, 2071FURZE, M., RHODES, J. C., PARR,A. J., RoJ. M. 2075. I. D. BINS, J., WHITEHEAD,M., THRELFALL, R. FUJITA,Y., HARA, Y., OGINO, T., SUGA, C. R. (1991), Abiotic factors elicit sesquiterpenoid (1981a), Production of shikonin derivatives by phytoalexin production but not alkaloid produccell suspension cultures of Lithospermum erytion in transformed root cultures of Datura strathrorhizon. I. Effects of nitrogen sources on the monium, Plant Cell Rep. 10, 111-114. production of shikonin derivatives, Plant Cell GERARDY, ZENK,M. H. (1993), Formation of R., Rep. 1 59-60. , salutaridine from (R)-reticuline by a membrane FUJITA, HARA,Y., SUGA,C., MORIMOTO, Y., T. bound cytochrome-P-450 enzyme from Papaver (1981b), Production of shikonin derivatives by somniferurn, Phytochemistry 32,79-86. cell suspension cultures of Lithospermum ery- GERTLOWSKI,,PETERSEN, (1993), Influence C . M. throrhizon. 11. New medium for the production of the carbon source on growth and rosmarinic of shikonin derivatives, Plant Cell Rep. 1, 61acid production in suspension cultures of Coleus 63. blumei, Plant Cell Tissue Organ Cult. 34, 183Y. FUJITA, TABATA, NISHI,A., YAMADA, Y., M., 190. (1982), New medium and production of second- GIROD, A., ZRYD P. (1991), Secondary metabP. J. ary compounds with the two-stage culture metholism in cultured red beet (Beta vulgaris L.) ods, in: Plant Tissue Culture 1982 (FUJIWARA, cells. Differential regulation of betaxanthins and A., Ed.), pp. 399-400. Tokyo: Maruzen Press. betacyanin biosynthesis, Plant Cell Tissue Organ FUJITA, TAKAHASHI, YAMADA, (1985), Y., S., Y. Cult. 25, 1-12. Selection of cell lines with high productivity of GODDIJN, J. M. (1992), Regulation of terpenoid 0. shikonin derivatives by protoplast culture of Liindole alkaloid biosynthesis in Catharanthus rothospermum erythrorhizon cells, Agric. Biol. seus: The tryptophan decarboxylase gene. PhD Chem. 49,1755-1759. Thesis, Rijksuniversiteit Leiden. FUJITA, HARA,Y., MORIMOTO, MISAWA, GODDIJN, J. M., DE KAM, R. J., ZANETTI, Y., T., 0. A., M. (1990), Semisynthetic production of vinblasSCHILPEROORT, A., HOGE,J. H. C. (1992), R. tine involving cell cultures of Catharanthus roAuxin rapidly down-regulates transcription of seus, in: Progress in Plant Cellular and Molecular the tryptophan decarboxylase gene from CathaH. Biology (NIJKAMP, J. J., VAN DER PLAS, L. ranthus roseus Plant Mol. Biol. 18, 1113-1120. J., H. W., VAN AARTRIJK, Eds.), pp.763-768. A. T., J., GOODBODY, E., ENDO, VUKOVIC, KUTDordrecht: Kluwer. M. NEY, J. P., CHOI, L. s. L., MISAWA, (1988), Enzymic coupling of catharanthine and vindolFUKUI, NAGAKAWA, TSUDA, TABATA, H., K., S., ine to form 3 ,4 -anhydrovinblastineby horseraM. (1982), Production of isoquinoline alkaloids dish peroxidase, Planta Med. 54, 136-140. by cell suspension cultures of Coptis japonica, in:
632
ultures I3 Secondary Products from Plant Cell CI HEIDE,L., NISHIOKA, N., FUKUI, TABATA, H., M. (1989), Enzymatic regulation of shikonin biosynthesis in Lithospermum erythrorhizon cell cultures, Phytochemistry 28, 1873-1877. HENRY,M., ROUSSEL, L., ANDARY, (1987), J. C. Verbascoside production in callus and suspension cultures of Hygrophila erecta, Phytochemistry 26, 1961-1963. HERMINGHAUS, THOLL,D., RUGENHAGEN, S., C., FECKER, L. F., LEUSCHNER, C., BERLIN, J. (1996), Improved metabolic action of a bacterial lysine decarboxylase gene in tobacco hairy root cultures by its fusion to a rbcS transit peptide coding sequence, Transgen. Res. 5, 193-201. HEZARI,M., LEWIS,N. G., CROTEAU, (1995), R. Purification and characterization of taxa-4 (5),11(12)-diene synthase from Pacific yew (Taxus brevifolia) that catalyzes the first committed step of taxol biosynthesis, Arch. Biochem. Biophys. 322,437444. HINZ,H., ZENK,M. H. (1981), Production of protoberberine alkaloids by cell suspension cultures of Berberis species, Naturwissenschafien 68,620621. HIPPOLYTE,I., MARIN, B., BACCOU,J. C., JONARD, R. (1992), Growth and rosmarinic acid production in cell suspension cultures of Salvia officinalis, Plant Cell Rep. 11, 109-112. HIRASUNA, J., PESTCHANKER, J., SRINIVAST. L. AN, V., SHULER, L. (1996), Taxol production M. in suspension cultures of Taxus baccata, Plant Cell Tiss. Org. Cult. 44, 95-102. HIRATA, KOBAYASHI, M., MIYAMOTO, HAK., K., SHI,T., OKAZAKI, MIURA, (1989), QuanM., Y. titative determination of vinblastine in tissue cultures of Catharanthus roseus by radioimmunoassay, Planta Med. 55, 262-264. HOEKSTRA, S., HARKES, A. A., VERPOORTE, S. P. R., LIBBENGA, R. (1990), Effect of auxin on K. cytodifferentiation and production of quinoline alkaloid in compact globular structures of Cinchona ledgeriana, Plant Cell Rep. 8, 571-574. HOLTON, A., TANAKA, (1994), Blue roses - a T. Y. pigment of our imagination? TIBTECH l2,4041. IGBAVBOA, U., SIEWEKE,H. J., LEISTNER,E., ROWER,I., HUSEMANN, BARZ, W. (1985), W., Alternative formation of anthraquinones and lipoquinones in heterotrophic and photoautotrophic cell suspension cultures of Morinda lucida, Planta 166, 537-544. IKUTA,A., SYONO,K., FURUYA, (1974), AlkaT. loids of callus tissues and redifferentiated plantlets in the Papaveraceae, Phytochemistry 13, 2175-2179. INAGAKI, NISHIMURA, OKADA, MITSUN., H., M.,
GROCER,D. (1980), Alkaloids derived from tryptophan and anthranilic acid, in: Secondary Plant Products. Encyclopedia of Plant Physiology, Vol. 8 (BELL,E. A., CHARLWOOD, V., Eds.), B. pp. 128-159. Berlin: Springer-Verlag. HAIN,R., BIESELER,B., KINDL,H., SCHRODER, G., STOCKER,R. (1990), Expression of a stilbene synthase gene in Nicotiana tabacum results in synthesis of the phytoalexin resveratrol, Plant Mol. Biol. 15, 325-335. HAIN, R., REIF, H. J., KRAUSE,E., LANGEBARTELS, R., KINDL,H., VORNAM, WIESE, W., B., E., P. R. SCHMELZER, SCHREIER, H., STOCKER, H., STENZEL,K. (1993), Disease resistance results from foreign phytoalexin expression in a novel plant, Nature 361, 153-156. HAMILL, D., ROBINS, J., PARR,A. J., EVANS, J. R. D. M., FURZE,J. M., RHODES,M. J. C. (1990), Overexpressing a yeast ornithine decarboxylase gene in transgenic roots of Nicotiana rustica can lead to enhanced nicotine accumulation, Plant Mol. Biol. 15, 27-38. HAN, K. H., FLEMING, WALKER,K., LOPER, P., M., CHILTON, S., MOCEK,U., GORDON,M. W. P., FLOSS,H. G. (1994), Genetic transformation of mature Taxus: an approach to genetically control the in vitro production of the anticancer drug taxol, Plant Sci. 95, 187-196. HARA,Y., YOSHIOKA, MORIMOTO, FUJITA, T., T., Y., YAMADA, (1988), Enhancement of berY. berine production in suspension cultures of Coptis japonica by gibberellic acid treatment, J. Plant Physiol. 133, 12-15. HARA, Y., YAMAGATA, MORIMOTO, HIH., T., RATSUKA, J., YOSHIOKA, T., FUJITA,Y., YAMADA, Y. (1989), Flow cytometric analysis of cellular berberine contents in high- and low-producing cell lines of Coptis japonica, Planta Med. 55,151-154. HARKES, A. A., KRIJBOLDER, LIBBENGA, P. L., K. T., R., WJNSMA,R., NSENGIYAREMGE, VERPOORTE, R. (1985), Influence of various media constituents on the growth of Cinchona ledgerianu tissue cultures and the production of alkaloids and anthraquinones therein, Plant Cell Tissue Organ Cult. 4, 199-214. HASHIMOTO, YAMADA,Y. (1994), Alkaloid T., biogenesis: Molecular aspects, Annu. Rev. Plant Physiol. Plant Mol. Biol. 45, 257-285. HASHIMOTO, YUN, D. J., YAMADA, (1993), T., Y. Production of tropane alkaloids in genetically engineered root cultures, Phytochernistry 32, 713-718. HAVKIN-FRENKEL, (1994), Vanilla flavor proD. duction: Tissue culture and the whole plant. VIII. IA PTC Congress, Firenze (Abstract), S1856.
9 References
633
HASHI, H. (1991), Verbascoside production by cultures of Catharanthus roseus. I. Comparison of enzyme activities and product accumulation, plant cell cultures, Plant Cell Rep. 9, 484-487. Plant Cell Tissue Organ Cult. 2, 333-340. INOUE, NAYESHIRO, INOUYE, ZENK, K., H., H., M. H. (1981), Anthraquinones in cell suspension KNOPP, STRAUSS, WEHRLI, (1988), Root W. E., A., induction on several Solanaceae species by cultures of Morinda citrifolia, Phytochemistry 20, Agrobacterium rhizogenes and determination of 1693-1700. root tropane alkaloid content, Plant Cell Rep. 7 , INOUE, UEDA,S. NAYESHIRO, MORITOME, K., H., 590-593. N., INOUYE, (1984), Biosynthesis of naphthoH. H., quinones and anthraquinones in Streptocarpus KNORR, D., CASTER, C., DORNENBURG, GRAF, S., HAVKIN-FRENKEL, PODSTOLSKI, D., dunnii cell cultures, Phytochemistry 23, 313A., WERRMANN, (1993), Biosynthesis and U. 318. yield improvement of food ingredients from JAZIRI, ZHIRI, Guo, Y. W., DUPONT, P., M., A., J. SHIMOMURA, HAMADA, VANHAELEN, plant cell and tissue cultures, Food Technol. 47 K., H., (12), 57-63. M., HOMES,J. (in press), Taxus sp. cell, tissue M. 0. and organ cultures as alternative sources for tax- KNUTH, E., SAHAI, P. (1991), Flavor composition and method, US Patents No. 5,057,424 and oid production: a literature survey, Plant Cell 5,068,184. Tiss. Org. Cult. Y., H., M. KADKADE, G. (1982), Growth and podopyllo- KOBAYASHI, FUKUI, TABATA, (1988), P. Berberine production by batch and semi-contintoxin production in callus tissues of Podophyluous cultures of immobilized Thalictrum cells in lum peltatum, Plant Sci. Lett. 25, 107-115. an improved bioreactor, Plant Cell Rep. 7 , 249KAMIMURA, (1991), Production of morphinan S. 252. alkaloids, in: Plant Cell Culture in Japan. ProY., M., K., gress in Production of Useful Plant Metabolites KOBAYASHI, AKITA, SAKAMOTO, LIU, H., SHIGEOKA, KOYANA, KAWAMURA, T., T., by Japanese Enterprises Using Plant Cell Culture T. M., FURUYA, (1993), Large-scale production Technology (KOMAMINE, MISAWA, DIA., M., of anthocyanin by Aralia cordata cell suspension COSMO, Eds.), pp. 27-38. Tokyo: CMC Co. F., cultures, Appl. Microbiol. Biotechnol. 40, 215KAMO,K. K., MAHLBERG, G. (1988), MorphiP. 218. nan alkaloids: Biosynthesis in plant (Papaver H. spp.) tissue cultures, in: Biotechnology in Agri- KOBLITZ, (1988), Anthraquinones, in: Cell Culture and Somatic Cell Genetics of Plants, Vol. 5: culture and Forestry, Vol. 4 (BAJAJ,Y. P. S., Production of Phytochemicals in Plant Cell CulEd.), pp. 251-263. Berlin: Springer-Verlag. tures (CONSTABEL, VASIL, I. K., Eds.), F., KETCHUM, E. B., GIBSON, M. (in press) PaR. D. pp. 113-142. San Diego: Academic Press. clitaxel production in suspension cultures of KOBLITZ,H., KOBLITZ,D., SCHMAUDER, P., H. Taxus, Plant Cell Tiss. Org. Cult. GROCER, (1983), Studies on tissue cultures of D. KINGSTON, G. I. (1994), Taxol: the chemistry D. genus Cinchona L.: Alkaloid production in cell and structure relationships of a novel anticancer suspension cultures, Plant Cell Rep. 2, 122-125. agent, TIBTECH 12, 222-227. F., KINNERSLEY, M., DOUGALL, K. (1980), In- KOES, R. E., QUATTROCCHIO, MOL, J. N. M. A. D. (1994), The flavonoid biosynthetic pathway in crease in anthocyanin yield from wild carrot cell plants: function and evolution, BioEssays 16, cultures by a selection based on cell-aggregate 123-132. size, Planta 149, 200-204. Y., K., R., T. KOHL,W., WITTE,B., HOFLE,G. (1982), Alkaloids KISHIMA, NOZALI, AKASHI, ADACHI, from Catharanthus roseus tissue cultures 111, 2. (1991), Light-inducible pigmentation in PortulaNaturforsch. 37b,1346-1351. ca callus; selection of a betalain producing cell KOMAMIME, MISAWA, DICOSMO, (Eds.) A., M., F. line, Plant Cell Rep. 10, 304-307. (1991), Plant Cell Culture in Japan. Progress in KISHORE, M., SOMERVILLE, R. (1993), GeG. C. Production of Useful Plant Metabolites by Japanetic engineering of commercially useful biosynnese Enterprises Using Plant Cell Culture Techthetic pathways in transgenic plants, Curr. Opin. nology. Tokyo: CMC Co. Biotechnol. 4, 152-158. H., KNOBLOCH, H., BERLIN, (1980), Influence of KONDO,O., HONDA, TAYA,M., KOBAYASHI, K. J. T. (1989), Comparison of growth properties of the medium composition on the formation of carrot hairy root in various bioreactors, Appl. secondary compounds in cell suspension cultures Microbiol. Biotechnol. 32,291-294. of Catharanthus roseus, Z. Naturforsch. 35c, KREIS,W., REINHARD, (1990), Two-stage cultiE. 551-556. vation of Digitalis lanata cells: semicontinuous KNOBLOCH, H., BERLIN, (1983), Influence of K. J. production of desacetyllanatoside C in 20-litre phosphate on the formation of the indole alkaairlift bioreactors, J. Biotechnol. 16, 123-126. loids and phenolic compounds in cell suspension
634
L., var. bikores monogerm (red beet). I n Vitro KREIS,W., REINHARD, (1992), 12PHydroxylaE. Dev. Biol. Plant. 28, 39-45. tion of digitoxin by suspension-culturedDigitalis lanata cells: Production of digoxin in 20-litre and LEISTNER, (1975), Isolierung, Identifizierung E. 300-litre air-lift bioreactors, J. Biotechnol. 26, und Biosynthese von Anthrachinonen in Zell257-273. suspensionskulturen von Morinda citrifolia, Planta Med. (Suppl.) 214-224. KREIS,W., HOELZ,H., SUTOR, REINHARD, R., E. R., (1993), Cellular organization of cardenolide bio- LOSCHER, HEIDE,L. (1994), Biosynthesis of p hydroxybenzoate from p-coumarate and p-coutransformation in Digitalis grandiflora, Planta marate-coenzyme A in cell-free extracts of Li191,246-251. thospermum erythrorhizon cell cultures, Plant KROMBHOLZ, MERSINGER, KREIS, W., R., R., Physiol. 106, 271-279. REINHARD, (1992), Production of forskolin E. M., B. by axenic Coleus forskohlii roots cultivated in LUCKNER, DIETTRICH, (1988), Cardenolides, in: Cell Culture and Somatic Cell Genetics shake flasks and 20-L glass jar bioreactors, Planof Plants, Vol. 5: Phytochemicals in Plant Cell ta Med. 58, 328-333. Cultures (CONSTABEL, VASIL,I. K., Eds.), F., IRZ, W. G. W., CHATSON, B., CONSTABEL, K. F., J. KUTNEY, P., CHOI,L. s. L., KOLODZIEJCZYK, pp. 193-212. San Diego: Academic Press. R., J. S. K. B. P., SLEIGH, K., STUART, L., WORTH, R. LUTTERBACH, STOCKIGT, (1992), High-yield formation of arbutin from hydroquinone by cell (1980), Alkaloid production in Catharanthus rosuspension cultures of Rauwolfia serpentina, seus cell cultures: initial studies on cell lines and Helv. Chim. Acta 75, 2009-2011. their alkaloid content, Phytochemistry 19, 2583H., Y. MANO, Y., OHKAWA, YAMADA, (1989), 2587. Production of tropane alkaloids by hairy root KURZ, W. G. W., CHATSON, B., CONSTABEL, R. F. cultures of Duboisia leichhardtii transformed by (1985), Biosynthesis and accumulation of indole Agrobacterium rhizogenes, Plant Sci. 59, 191alkaloids in Catharanthus roseus cultivars, in: 201. Primary and Secondary Metabolism in Plant Cell K., Y. K. Cultures (NEUMANN, H., BARZ, W., REIN- MATSUBARA, FUJITA, (1991), Production of berberine, in: Plant Cell Culture in Japan. ProHARD,E., Eds.), pp. 143-153. Berlin: Springergress in Production of Useful Plant Metabolites Verlag. by Japanese Enterprises Using Plant Cell Culture KURZ, W. G. W., PAIVA,N. L., TYLER,R. T. A., M., Technology (KOMANINE, MISAWA, DI(1990), Biosynthesis of sanguinarine by elicitaCOSMO, Eds.), pp. 3 9 4 . Tokyo: CMC Co. F., tion of surface immobilized cells of Papaver S., Y., K., somniferum, in: Progress in Plant Cellular and MATSUBARA, KITANI, YOSHIOKA, MoY., RIMOTO, T., FUJITA, YAMADA, Y. (1989), H. Molecular Biology (NIJKAMP, J. J., VAN DER High density culture of Coptis japonica cells inPLAS, L. H. W., VAN AARTRIJK, Eds.), J., creases berberine production, J. Chem. Technol. pp. 682-688. Dordrecht: Kluwer. Biotechnol. 46, 61-69. KUTCHAN, M. (1993), Strictosidine: From alkaT. MCKNIGHT, D., BERGEY, R., BURNETTE, T. D. R. loid to enzyme to gene, Phytochemistry 32,493J., NESSLER, C. L. (1991), Expression of an en506. zymatically-active and correctly targeted strictoKUTCHAN, M., ZENK,M. H. (1993), EnzymolT. sidine synthase in transgenic tobacco plants, ogy and molecular biology of benzophenanthriPlanta 185, 148-152. dine biosynthesis, J. Plant Res. (Spec. Issue) 3, MEIJER, H., LOPESCARDOSO, I., VOSKUILA. M. 165-1 73. EN, J. T., DE WAAL, A., VERPOORTE, HOGE, R., KUTCHAN,T. M., BOCK, A., DIETTRICH,H. J. H. C. (1993a), Isolation and characterization (1994), Heterologous expression of the plant of a cDNA clone from Catharanthus roseus enproteins strictosidine synthase and berberine coding NADPH: cytochrome P-450 mono-oxybridge enzyme in insect culture, Phytochemistry genase in plants, Plant J. 4,4740. 35,353-360. MEIJER,A. H., VERPOORTE, HOGE,J. H. C. R., KUTNEY,J. P., AWERYN, CHOI,L. S. L., HONB., (1993b), Regulation of enzymes and genes inDA, T., KOLDZIEJZCYK, LEWIS, G., SATO, P., N. volved in monoterpene indole alkaloid biosynT., SLEIGH, K., STUART, L., WORTH, R., S. K. B. thesis in Catharanthus roseus, J. Plant Res. K. KURZ,W. G. W., CHATSON, B., CONSTABEL, (Spec. Issue) 3, 145-164. F. (1983), Studies in plant tissue culture: The MEURER-GRIMES, BERLIN,J., STRACK,D. B., synthesis and biosynthesis of indole alkaloids, (1989), Hydroxycinnamoyl-CoA:putrescine hyTetrahedron 39,3781-3795. droxycinnamoyltransferase in tobacco cell culLEATHERS, R., DAVIN, ZRYD,J. P. (1992), R. C., tures with high and low levels of caffeoylputresBetalain producing cell cultures of Beta vulgaris cine, Plant Physiol. 89, 488-492.
9 References
635
MEYER, P., HEIDMANN,I., FORKMANN,G., MULDER-KRIEGER, VERPOORTE,R., BAERT., SAEDLER, (1987), A new petunia flower colH. HEIM-SVENDSEN, SCHEFFER, J. C. (1988), A., J. our generated by transformation of a mutant Production of essential oils and flavours in plant with a maize gene, Nature 330, 677-678. cell and tissue cultures. A review, Plant Cell Tissue Organ Cult. 13, 85-154. MIRJALILI, LINDEN, C. (1996) Methyl jasN., J. T., OHKAWA, H., YAMADA, Y. monate induced production of taxol in suspen- MURANAKA, (1993), Continuous production of scopolamine sion cultures of Taxus cuspidata: ethylene interaction and induction models, Biotechnol. Prog. by a culture of Duboisia leichardtii hairy root 12,110-118. clone in a bioreactor system, Appl. Microbiol. Biotechnol. 40,219-223. MISAWA,M. (1994), Plant tissue culture: an alterK., A., H., native for production of useful metabolites, NAGAKAWA, KONAGAI, FUKUI, TABATA, M. (1984), Release and crystallization of F A 0 Agric. Serv. Bull. 108. Rome: FAO. berberine in liquid medium of Thalictrum minus MISAWA,M., ENDO, T. (1988), Antitumor comcell suspension cultures, Plant Cell Rep. 3, 254pounds, in: Cell Culture and Somatic Cell Ge257. netics, Vol. 5: Phytochemicals in Plant Cell CulH., V. tures (CONSTABEL,F., VASIL, I. K., Eds.), NAKATANI, MALIK, S. (1992), Putrescine Nmethyltransferase, recombinant DNA molecules pp. 553-568. San Diego: Academic Press. encoding putrescine N-methyltransferase, and MIZUKAMI, ELLIS, B. E. (1991), Rosmarinic H., transgenic tobacco plants with altered nicotine acid formation and differential expression of tycontent, Eur. Patent Application 9 113283.6, rosine arninotransferase isoforms in Anchusa ofPublication Nr. 0486214 A2. ficinalis cell suspension cultures, Plant Cell Rep. NESSLER,C. R. (1994), Metabolic engineering of 10,321-324. plant secondary products, Transgen. Res. 3, 109MIZUKAMI,H., KONOSHIMA,M., TABATA, M. 115. (1978), Variation in pigment production in LiNICOLAOU, C., DAI,W. M., GUY,R. K. (1994), K. thospermurn erythrorhizon callus cultures, PhyChemie und Biologie von Taxol, Angew. Chem. tochemistry 17, 95-97. 106,3849. MIZUKAMI, TABIRA, ELLIS, E. (1993), NOZUE,M., KAWAI, YOSHITAMA, (1987), SeH., Y., B. J., K. Methyl jasmonate-induced rosmarinic acid biolection of a high anthocyanin-producing cell line synthesis in Lithospermum erythrorhizon cell of sweet potato cell cultures and identification of suspension cultures, Plant Cell Rep. 12, 706the pigment, J. Plant Physiol. US, 81-88. 709. ODAKE,K., ICHI,T., KUSUHARA, (1991), ProK. MIZUSAKI, TANABE, NOGUCHI, TAMAS., Y., M., duction of madder colorants, in: Plant Cell CulKI, E. (1971), p-coumaroyl putrescine, caffeoyl ture in Japan. Progress in the Production of Useputrescine and feruloyl putrescine from callus ful Plant Metabolites by Japanese Enterprises Ustissue culture of Nicotiana tabacum, Phytocheming Plant Tissue Culture Technology (KOMAMistry 10, 1347-1350. INE, A., MISAWA,M., DICOSMO, F., Eds.), MORENO,P. R. H., VAN DER HEIJDEN,R., VERpp. 13&146. Tokyo: CMC Co. POORTE, R. (1995), Cell and tissue cultures of OOSTHAM, M o t , J. N. M., VAN DER PLAS,L. A., Catharanthus roseus: a literature survey. 11. UpH. W. (1993), Establishment of hairy root culdating from 1988-1993, Plant Cell Tiss. Org. tures of Linum flavum producing the lignan 5Cult. 42, 1-25. methoxypodophyllotoxin, Plant Cell Rep. 12, MORIMOTO, HARA, KATO, Y., HIRATSU- 474-477. M., Y., KA, J., YOSHIOKA, FUJITA., T., Y., YAMADA, Y. OZAKI, SUGA,C., YOSHIOKA, MORIMOTO, Y., T., (1988), Berberine production by cultured Coptis T., HARADA, (1990), Evaluation of equivalM. japonica cells in a one stage culture using meence on pharmacological properties between dium with a high copper concentration, Agric. natural crude drugs and their cultured cells Biol. Chem. 52, 1835-1836. based on their components. Accelerative effect MORRIS,P. (1986), Regulation of product synthesis of Lithospermi radix and inhibitory effect of in cell cultures of Catharanthus roseus. 11. ComCoptidis rhizoma on proliferation and granulaparison of production media, Planta Med. 52, tion tissue, Yakugaku Zasshi 110, 268-272. 121-126. PALMER, E., WIDHOLM, M. (1975), CharacterJ. J. ization of carrot and tobacco cell cultures resistMOSHY,R. J., NIEDER, H., SAHAI, P. (1989), M. 0. ant to p-fluorophenylalanine, Plant Physiol. 56, Biotechnology in the flavor and food industry of 233-238. the USA, in: Biotechnology. Challenges for the R. Flavor and Food Industry (LINDSEY, L., WIL- PAREILLEUX, VINAS, (1984), A study on the A., R. LIS,B. J., Eds.), pp. 145-163. London: Elsevier. alkaloid production of resting cell suspensions of
636
Catharanthus roseus in a continuous flow reactor, Appl. Microbiol. Biotechnol. 19, 316-320. PARK, M., YOONS. Y., GILES,K. L., SONGSTAD, J. D., D., D., EPPSTEIN, NOVAKOVSKI, FRIESEN, L., ROEWER, (1992), Production of sanguinarI. ine by suspension cultures of Papaver somniferum in bioreactors, J. Ferment. Bioeng. 74, 292296. PARR,A. J. (1989), The production of secondary metabolites by plant cell cultures, J. Biotechnol. 10, 1-26. PARR,A. J., PAYNE, EAGLES, CHAPMAN, J., J., B. T., ROBINS, J., RHODES, J. C. (1990), VarR. M. iation in tropane alkaloid accumulation within the Solanaceae and strategies for its exploitation, Phytochemistry 29, 2545-2550. PASQUALI, GODDIJN, J. M., DE WALL,A., G., 0. VERPOORTE, SCHILPEROORT, A., HOGE, R., R. J. H. C., MEMELINK, (1992), Coordinated reJ. gulation of two indole alkaloid biosynthetic genes from Catharanthus roseus by auxin and elicitors, Plant Mol. Biol. 18, 1121-1131. PAYNE, F., BRINGI, PRINCE, SHULER, G. V., C., M. L. (1991), Plant Cell and Tissue Culture in Liquid Systems. Munich: Hanser. S. M. PESTCHANKER, J., ROBERTS, C., SHULER, L. L. (1996), Kinetics of taxol production and nutrient use in suspension cultures of Taxus cuspidata in shake flasks and a Wilson-type reactor, Enzyme Microb. Technol. 19,256-260. PETERSEN, ALFERMANN, W. (1993), Plant M., A. cell cultures, in: Biotechnology, Vol. 1,2nd Edn. (REHM, H.-J., REED, G., Eds.), pp. 578-614. Weinheim: VCH. PETERSEN,M., HAUSLER,E., KARWATZKI, B., J. MEINHARD, (1993), Proposed biosynthetic pathway for rosmarinic acid in cell suspension cultures of Coleus blumei, Planta 189, 10-14. PIEHL,G. W., BERLIN,J., MOLLENSCHOTT, C., LEHMANN, (1988), Growth and alkaloid proJ. duction of a cell suspension culture of Thalictrum rugosum in shake flasks and membranestirrer reactors with bubble free aeration, Appl. Microbiol. Biotechnol. 29,456-461. POULSEN, GODDIJN, J. M., HOGE,J. H. C., C., 0. VERPOORTE, (1994), Anthranilate synthase R. and chorismate mutase activities in transgenic tobacco plants overexpressing tryptophan decarboxylase from Catharanthus roseus, Transgen. Res. 3, 43-49. PROKSCH, WAGNER, (1987), Structural analA., H. ysis of a 4-0-methylglucuronoarabinoxylanwith immuno-stimulating activity from Echinacea purpurea, Phytochemistry 26, 1989-1993. RAZZAQUE, ELLIS,B. E. (1977), Rosmarinic A., acid production in Coleus blumei, Planta 137, 287-291.
REINHARD, ALFERMANN, W. (1980), BioE., A. transformation of plant cell cultures, Adv. Biochem. Eng. 16,49-83. REINHARD, KREIS,W., BARTHLEN, HELME., U., BOLD,U. (1989), Semicontinuous cultivation of Digitalis lanata cells: Production of @methyldigoxin in a 300-L airlift bioreactor, Biotechnol. Bioeng. 34,502-508. RHODES,M. J. C., ROBINS, J., HAMILL, D., R. J. PARR, A. J., HILTON, G., WALTON,N. J. M. (1990), Properties of transformed roots, in: Secondary Products of Plant Tissue Culture (CHARLWOOD, V., RHODES,M. J. C., Eds.), B. pp. 201-225. Oxford: Clarendon. ROHMER, KNANI,M., SIMONIN, SUTTER, M., P., B., SAHM,H. (1993), Isoprenoid biosynthesis in bacteria: a novel pathway for the early steps leading to isopentenyl diphosphate, Biochem. J. 295,517-524. RUEFFER, ZENK,M. H. (1987), Distant precurM., sors of benzylisoquinoline alkaloids and their enzymatic formation, 2. Naturforsch. 4212, 319332. RUEFFER,M., EL-SHAGI, H., NAGAKURA, N., ZENK,M. H (1981), S-Norlaudanosoline synthase: the first enzyme in the benzylisoquinoline biosynthetic pathway, FEBS Lett. 129, 5-9. RUYTER, M., STOCKIGT, (1989), Neue NaturC. J. stoffe aus pflanzlichen Zellkulturen, GIT Fachz. Lab. 33,283-293. SAITO,K., YAMAZAKI, MURAKOSHI, (1992), M., I. Transgenic medicinal plants: Agrobacterium-mediated foreign gene transfer and production of secondary metabolites, J. Nut. Prod. 55, 149162. SAKAMOTO, IIDA,K., SAWAMURA, HAJIK., K., RO, K., ASADA,Y., YOSHIKAWA, FURUYA, T., T. (1994), Anthocyanin production in cultured cells of Aralia cordata, Plant Cell Tissue Organ Cult. 36, 21-26. SAKUTA, TAKAGI, KOMAMINE, (1987), M., T., A. Effects of nitrogen source on betacyanin accumulation and growth in suspension cultures of Phytolaca americana, Physiol. Plant. 71, 459463. SASSE,F., WITTE,L., BERLIN, (1987), BiotransJ. formation of tryptamine to serotonin by cell suspension cultures of Peganum harmala, Planta Med. 53, 354-359. SATO,F., YAMADA, (1984), High berberine-proY. ducing cultures of Coptis juponica cells, Phytochemistry 23, 281-285. SCHIEL, BERLIN, (1987), Large-scale fermenO., J. tation and alkaloid production of cell suspension cultures of Catharanthus roseus, Plant Cell Tissue Organ Cult. 8, 153-161.
9 References
637
eine and morphine production in suspended PaSCHIEL, JARCHOW-REDECKER,PIEHL,G. O., K., paver somniferum cultures after removal of exoW., LEHMANN, BERLIN, (1984a), Increased J., J. genous hormones, Plant Cell Rep. 10,349-353. formation of cinnamoyl putrescines by fed-batch fermentation of cell suspension cultures of Nico- SIEWEKE, J., LEISTNER, (1992), o-SuccinylH. E. tiana tabacum, Plant Cell Rep. 3, 18-20. benzoate :CoA ligase from anthraquinone producing cell cultures of Galium mollugo, PhytoSCHIEL, MARTIN, PIEHL, W., NOWAK, O., B., G. J., chemistry 31, 2329-2335. HAMMER, SASSE, F., SCHAER, J., W., LEHMANN, J., BERLIN, (1984b), Some technologiJ. SIGNS, W., FLORES, E. (1990), The biosynM. H. cal aspects on the production of cinnamoyl puthetic potential of plant roots, BioEssays 12, 713. trescines by cell suspension cultures of Nicotiana tabacum, in: 3rd Eur. Congr. Biotechnol., Vol. 1, SIMANTIRAS, LEISTNER, (1989), Formation M., E. pp. 167-172. Weinheim: VCH. of o-succinylbenzoic acid from isochorismic acid O., J. SCHIEL, WITTE, L., BERLIN, (1987), Geranin protein from anthraquinone-producing plant cell suspension cultures, Phytochemistry 28, iol-10-hydroxylase activity and its relation to monoterpene indole alkaloid accumulation in 1381-1382. cell suspension cultures of Catharanthus roseus, SMART, J., MORRIS, FOWLER, W. (1982), N. P., M. 2. Naturforsch. 42c, 1075-1081. Alkaloid production by cells of Catharanthus roseus grown in airlift fermenter systems, in: Plant SCHLATMANN, MORENO, R. H., VINKE, J. E., P. J. Tissue Culture 1982 (FUJIWARAA., Ed.), R., L., TEN HOOPEN,H. J. G., VERPOORTE, pp. 397-398. Tokyo: Maruzen Press. J. HEIJNEN, J. (1994), Effect of oxygen and nutrient limitation on ajmalicine production and SMITH, A. L. (1995), Large-scale production of M. secondary metabolites, in: Current Issues in related enzyme activities in high density cultures of Catharanthus roseus, Biotechnol. Bioeng. 44, Plant Molecular and Cellular Biology (TERZI, M., CELLA,R., FALAVIGNA, Eds.), pp. 669A., 461-468. 674. Dordrecht: Kluwer. SCHUBEL,H., RUYTER, C. M., STOCKIGT, J. (1989), Improved production of raucaffricine by SMOLLNY, T., WICHERS, H., DE RIJK, T., VAN ZWAM, SHASAVARI, ALFERMANN, A., A., A. cultivated Rauwolfia cells, Phytochemistry 28, W. (1993), Formation of lignans in suspension 491494. cultures of Linum album. 3rd Workshop on SCHULTE, EL-SHAGI, ZENK,M. H. (1984), U., H., Primary and Secondary Metabolism of Plants Optimization of 19 Rubiaceae species in cell culand Plant Cell Cultures. Abstract P13, Leiden ture for the production of anthraquinones, Plant University. Cell Rep. 3, 51-54. D. J. SCHUMACHER, M., GUNDLACH, FIEDLER, SONGSTAD, D., GILES,K. L., PARK, M., NovH. H., AKOVSKI, EPPSTEIN, FRIESEN, (1990), D., D., L. F., ZENK,M. H. (1987), Elicitation of benzoUse of nurse cultures to select for Papaver somphenanthridine alkaloid biosynthesis in Eschniferum cell lines capable of enhanced sanguinscholtzia californica cell cultures, Plant Cell Rep. arine production, J. Plant Physiol. 136,236-239. 6,410-413. D. SCRAGG, H., CRESSWELL, C., ASHTON, A. R. S., SONGSTAD, D., KURZ,W. G. W., NESSLER, C. L. (1991), Tyramine accumulation in Nicotiana YORK, BOND, A., FOWLER, W. (1989), A., P. M. tabacum transformed with a chimeric tryptophan Growth and alkaloid production in bioreactors decarboxylase gene, Phytochemistry 30, 3245by a selected Catharanthus roseus cell line, En3246. zyme Microb. Technol. 11, 329-333. G. W. SCRAGG, H., ASHTON, YORK,A., STEPAN- SOUTHARD, L., GROZNIK, J., BOULWARE, A. S., R. T., THORNE, E., WALBORN, E., YANE. D. SARKISSIAN, GREY,D. (1990), Growth of G., KELL, S. L. (1984), Sanguinarine, a new antiCatharanthus roseus suspensions for maximum plaque agent: retentention and plaque specificibiomass and alkaloid accumulation, Enzyme Mity, J. Am. Dent. Assoc. 108, 339-341. crob. Technol. 12,292-298. V., V., SEITZ,H. U., HINDERER, (1988), Anthocyan- SRINIVASAN, CIDDI,V., BRINGI, SHULER, W. M. (1996), Metabolic inhibitors, elicitors, and ins, in: Cell Culture and Somatic Cell Genetics of precursors as tools for probing yield limitation in Plants, Vol. 5: Phytochemicals in Plant Cell Cultaxane production by Taxus chinensis cell culF., tures (CONSTABEL, VASIL, I. K., Eds.), tures, Biotechnol. Prog. 12,457466. pp. 49-76. San Diego: Academic Press. R. SHAUL,O., GALILI,G. (1991), Increased lysine STAHLHUT, (1993), ESCAgenetics Corporation, IAPTC Newslett. 73, 12-15. synthesis in transgenic tobacco plants expressing H. N. a bacterial dihydrodipicolinate synthase in their STEINRUCKEN, C., AMRHEIN, (1980), The herbicide glyphosate is a potent inhibitor of 5chloroplasts, Plant J. 2, 203-209. enolpyruvylshikimic acid 3-phosphate synthase, SIAH,C. L., DORAN, M. (1991), Enhanced codP.
638
13 Secondary Products from Plant Cell Cultures TAKAYAMA, (1994), Tripterygium wilfordii: In S. vitro culture and the production of anticancer compounds tripdiolide and tripdolide, in: Biotechnology in Agriculture and Forestry, Vol. 28 (BAJAJ, Y. P. S., Ed.), pp. 457-468. Berlin: Springer-Verlag. TAKAYAMA, S., TAKIZAWA, KUROYANAGI, N., M. (1994), The method and system for large-scale culture of plant roots using aeration-agitation bioreactor. Strategies for scale-up. VIlI. IA PTC Congr., Firenze (Abstract) S20-22. TANAKA, (1994), Immobilization of plant cells, A. in: Advances in Plant Biotechnology. Studies in Plant Science, Vol. 4 (Ruu, D. D. Y., FURASAKI, S., Eds.), pp. 209-220. Amsterdam: Elsevier. TATICEK, A., LEE, C. W. T., SHULER, L. R. M. (1994), Large-scale insect and plant cell culture, Curr. Opin. Biotechnol. 5, 165-174. TOIVONEN,(1993), Utilization of hairy root culL. tures for production of secondary metabolites, Biotechnol. Prog. 9, 12-20. TYLER,V. E. (1988), Medicinal Plant Research: 1953-1987, Planta Med. 54, 95-100. TYLER,R. T., EILERT,U., RIJNDERS, 0. M., C. I. ROEWER, A., KURZ,W. G. W. (1988), Semicontinuous production of sanguinarine and dihydrosanguinarine by Papaver somniferum L. cell suspension cultures treated with fungal homogenate, Plant Cell Rep. 7,410-413. ULBRICH, WIESNER,W., ARENS, H. (1985), B., Large-scale production of rosmarinic acid from plant cell cultures of Coleus blumei, in: Primary and Secondary Metabolism of Plant Cell Cultures (NEUMANN, H., BARZ, W., REINHARD, K. E., Eds.), pp. 293-303. Berlin: Springer-Verlag. USHIYAMA, (1991), Large-scale culture of ginK. seng, in: Plant Cell Culture in Japan. Progress in the Production of Useful Plant Metabolites by Japanese Enterprises Using Plant Tissue Culture A., M., Technology (KOMAMINE, MISAWA, DICOSMO, F., Eds.), pp. 92-98. Tokyo: CMC CO. VAN DEN BERG,A. J. J., RADEMA, H., LABAM. DIE, R. P. (1988), Effects of light on anthraquinone production in Rhamnus purshiana suspension cultures, Phytochemistry 27, 415417. VAN DEN HEIJDEN,R., VERPOORTE, R., TEN HOOPEN,H. J. G. (1989), Cell and tissue cultures of Catharanthus roseus, Plant Cell Tissue Organ Cult. 18,231-280. VAN DER KROL, A. R., LENTING, E., VEENP. STRA, J., VAN DER MEER,I. M., KOES, R. E., GERATS, G. M., MOL, J. N. M., STUITJE, A. A. R. (1988), An antisense chalcone synthase gene in transgenic plants inhibits flower pigmentation, Nature 333, 866-869. VAN DER KROL, A. R., MUR, L. A., BELD, M., A. MOL,J. N. M., STUITJE, R. (1990), Flavonoid
Biochem. Biophys. Res. Commun. 94, 12071212. STEPHANOPOULOS, SINSKY, J. (1993), MetaG., A. bolic engineering - methodologies and future prospects, TIBTECH 11, 392-396. STOCKIGT, SOLL,H. J. (1980), Indole alkaloids J., from cell suspension cultures of Catharanthus roseus and C. ovalis, Planta Med. 40,22-30. SUGA, HIRATA, (1990), Biotransformation of T., T. exogenous substrates by plant cell cultures, Phytochemistry 29, 2393-2406. SUMARYONO, PROKSCH, HARTMANN, W., P., T., NIMTZ, WRAY,V. (1991), Induction of rosM., marinic acid accumulation in cell suspension cultures of Orthosiphon aristatus after treatment with yeast extract, Phytochemistry 30, 32673271. SUZUKI,H., MATSUMOTO, (1988), AnthraquinT. one: Production by plant cell culture, in: Biotechnology in Agriculture and Forestry, Vol. 4 (BAJAJ, Y. P. S., Ed.), pp. 237-250. Berlin: Springer-Verlag. SUZUKI, YOSHIOKA, HARA,Y., TABATA, T., T., Y. M., FUJITA, (1987), A new bioassay system for screening high berberine-producing cell colonies of Thalictrum minus, Plant Cell Rep. 6, 194196. SUZUKI, NAKAGAWA, FUKUI, TABATA, M., K., H., M. (1988), Alkaloid production in cell suspension cultures of Thalictrum flavum and T.dipterocarpum, Plant Cell Rep. 7, 26-29. SUZUKI,H., SUGA,C., MORIMOTO, HARADA, T., M. (1991), Quantitative analysis of plant hormones, auxins, in biotechnologically cultured products of medicinal plants, Shoyakugaku Zasshi 45, 137-141. SUZUKI, OZAKI, SUGA, MORIMOTO, H., Y., C., T., SATAKE, HARADA, (1993a), Dissolution M., M. tests of Coptis rhizome and cultured cells of Coptis japonica, Shoyakugaku Zasshi 47, 311315. SUZUKI, H., OZAKI, Y., SATAKE,M. (1993b), Changes of components - dissolution from prescription containing Coptis rhizome or cultured cells of Coptis japonica into the detection, Shoyakugaku Zasshi 47, 396-401. TABATA, MIZUKAMI, HIRAOKA, KONM., H., N., OSHIMA, M. (1974), Pigment formation in callus cultures of Lithospermum erythrorhizon, Phytochemistry 13, 927-932. TAKAHASHI, FUJITA, (1991), Production of S., Y. shikonin, in: Plant Cell Culture in Japan. Progress in the Production of Useful Plant Metabolites by Japanese Enterprises Using Plant Cell Culture Technology (KOMANINE, MISAWA, A., M., DICOSMO, Eds.), pp. 72-78. Tokyo: CMC F.,
co.
9 References
639
genes in Petunia: addition of a limited number of gene copies may lead to a suppression of gene expression, Plant Cell 2, 218-221. VAN GULIK, M., NUUTILA, M., VINKE,K. W. A. L., TEN HOOPEN, J. G., HEIJNEN, J. (1994), H. J. Effects of carbon dioxide, air flow rate, and inoculation density on batch growth of Catharanthus roseus cell suspensions in stirred fermenters, Biotechnol. Prog. 10, 335-339. VAN UDEN, W., PRAS, N., VOSSEBELD,E. M., MOL, J. N. M., MALINGRE, M. (1990), ProT. duction of 5-methoxypodophyllotoxin in cell suspension cultures of Linum jlavum, Plant Cell Tissue Organ Cult. 20,81-87. VAN UDEN, W., PRAS, N., HOMAN, B., MALINGRE,T. M. (1991), Improvement of the production of 5-methoxypodophyllotoxin using a new selected root culture of Linum flavum, Plant Cell Tissue Organ Cult. 27, 115-121. WAGNER, VOGELMANN, (1977), Cultivation F., H. of plant cell cultures in bioreactors and formation of secondary metabolites, in: PIant Tissue Culture and its Biotechnological Application (BARZ,W., REINHARD, ZENK,M. H., Eds.), E., pp. 245-252. Heidelberg: Springer-Verlag. WAGNER, H., STUPPNER, H., PUHLMANN, J., BRUMMER, DEPPE,K., ZENK,M. H. (1989), B., Gewinnung von immunologisch aktiven Polysacchariden aus Echinacea-Drogen und -Gewebekulturen, Z. Phytother. 10, 35-38. WALKER, A.,ELLIS,B. E., DUMBROFF, B., M. E. R. DOWNER, G., MARTIN, J. (1986), Changes R. in amines and biosynthetic enzyme activities in p-fluorophenylalanine resistant and wild type tobacco cell cultures, Plant Physiol. 80, 825-828. WALTON,N. J., BELSHAW, J. (1988), The effect N. of cadaverine on the formation of anabasine from lysine in hairy root cultures of Nicotiana hesperis, Plant Cell Rep. 7 115-118. , WESTCOTT,R. J., CHEETHAM, J. S., BARRAP. R. CLOUGH, J. (1994), Use of organized viable plant vanilla aerial roots for the production of natural vanilla, Phytochemistry 35, 135-138. WESTPHAL,K. (1990), Large-scale production of new biologically active compounds in plant cell cultures, in: Progress in Plant Cellular and MoH. lecular Biology (NIJKAMP, J. J., VAN DER PLAS, L. H. W., VAN AARTRIJK, Eds.), J., pp. 601408. Dordrecht: Kluwer. WHITAKER, J., HASHIMOTO, EVANS,D. A. R. T., (1984), Production of secondary metabolite rosmarinic acid by plant cell suspension cultures, Ann. N. Y. Acad. Sci. 435, 364-366. WICHERS, H. J., VERSLUIS-DE HAAN, G. G., MARSMAN, W., HARKES, P. (1991), PodoJ. M. phyllotoxins in plants and cell cultures of Linum jlavum, Phytochemistry 30,3601-3604.
WICKREMESINHE,R. M., ARTECA,R. N. (1993), E. Taxus callus cultures: Initiation, growth, optimization, characterization and taxol production, Plant Cell Tissue Organ Cult. 35, 181-193. WICKREMESINHE, R. M., ARTECA, R. N. E. (1994a), Taxus cell suspension cultures: optimization of growth and production of taxol, J. Plant Physiol. 144,183-188. WICKREMESINHE, R. M., ARTECA, R. N. E. (1994b), Roots of hydroponically plants as a source of taxol and related taxanes, Plant Sci. 101,125-135. WJNSMA, VERPOORTE, (1988), Quinoline alR., R. kaloids of Cinchona, in: In Cell Culture and Somatic Cell Genetics of Plants, Vol. 5: PhytochemF., icals in Plant Cell Cultures (CONSTABEL, VASIL, I. K., Eds.), pp. 337-356. San Diego: Academic Press. WILLIAMS, D., ELLIS,B. E. (1993), Alkaloids R. from Agrobacterium rhizogenes-transformedPapaver somniferum cultures, Phytochemistry 32, 719-723. WILSON,P. D. G., HILTON,M. G., MEEHAN, T. P. H., WASPE,C. R., RHODES,M. J. C. (1990), The cultivation of transformed roots from laboratory to pilot plant, in: Progress in Plant CelIular and H. Molecular Biology (NIJKAMP, J. J., VAN DER PLAS, L. H. W., VAN AARTRIJK,J., Eds.), pp. 700-705. Dordrecht: Kluwer. WOERDENBAG, J., VAN UDEN, W., FRIJLINK, H. H. W., LERK,C. F., PRAS,N., MALINGRE, M. T. (199O), Increased podophyllotoxin production in Podophyllum hexandrum cell suspension cultures after feeding coniferyl alcohol as a p-cyclodextrin complex, Plant Cell Rep. 9,97-100. YAMADA, HASHIMOTO, (1982), Production Y., T. of tropane alkaloids in cultured cells of Hyoscyamus niger, Plant Cell Rep. 1,101-103. YAMADA, SATO, F. (1982), Production of berY., berine in cultured cells of Coptis japonica, Phytochemistry 20, 545-547. YAMAMOTO, (1991), Anthocyanin production in Y. plant cell cultures, in: Plant Cell Culture in Japan. Progress in the Production of Useful Plant Metabolites by Japanese Enterprises Using Plant A., Cell Culture Technology (KOMAMINE, MISF., AWA,M., DICOSMO, Eds.), pp. 114-126. Tokyo: CMC Co. YAMAMOTO, MIZUGUCHI, YAMADA,Y. Y., R., (1982), Selection of a high and stable-pigment producing strain in cultured Euphorbia millii cells, Theor. Appl. Genet. 61, 113-116. YAMAMOTO, KINOSHITA, WATANABE, Y., Y. S., YAMADA, (1989), Anthocyanin production in Y. suspension cultures of high producing cells of Euphorbia millii, Agric. Biol. Chem. 53, 417423.
640
YOKOYAMA, YANAGI, (1991), High-level M., M. production of arbutin by biotransformation, in: Plant Cell Culture in Japan. Progress in the Production of Useful Plant Metabolites by Japanese Enterprises Using Plant Cell Culture Technology (KOMAMINE, MISAWA, DICOSMO, A., M., F., Eds.), .. 78-91. Tokyo: CMC Co. pp. YUKIMUNE, TABATA, HIGASHI, HARA, Y., H., Y., Y. (1996), Methyl jasmonate-induced overproduction of paclitaxel and baccatin 111 in Taxus cell suspension cultures, Nut. Biotechnol. 14, 1129-1132. YUN,D. J., HASHIMOTO, YAMADA, (1992), T., Y. Metabolic engineering of medicinal plants: Transgenic Atropa belladonna with an improved alkaloid composition, Proc. Natl. Acad. Sci. U S A 89,11799-1 1803. ZENK, H. (1991), Chasing enzymes of secondary M. metabolism - Plant cell culture a pot of gold, Phytochemistry 30,3861-3863.
ZENK,M. H., EL-SHAGI, SCHULTE, (1975), H., U. Anthraquinone production by cell suspension cultures of Morinda citrifolia, Planta Med. (Suppl.), 79-101. ZENK, H., EL-SHAGI, ULBRICH, (1977a), M. H., B. Production of rosmarinic acid by cell suspension cultures of Coleus blumei, Naturwissenschaften 64,585-586. ZENK,M. H., EL-SHAGI, ARENS,H., STOCKH., IGT, J., WEILER, W., DEUS,B. (1977b), ForE. mation of indole alkaloids serpentine and ajmalicine in cell suspension cultures of Catharanthus roseus, in: Plant Tissue Culture and its Biotechnological Application (BARZ, W., REINHARD, E., ZENK,M. H., Eds.), pp. 2743. Heidelberg: Springer-Verlag. J. T. ZHONG, J., YOSHIDA, (1995), High density cultivation of Perilla frutescens cell suspensions for anthocyanin production: Effects of Sucrose concentration and inoculum size, Biotechnol. Bioeng. 38,6534158.
Biotechnology
14 Biotechnical Drugs as Antitumor Agents
UDO
GRAFE,KLAUSJURGEN DORNBERGER, HANS-PETER SALUZ

Jena, Germany
1 Introduction 643 2 Prescreening for Antineoplastic Agents 645 3 Classical Anticancer Drugs 647 3.1 Nucleoside Antibiotics and Analogs of Nucleobases 647 3.2 Drugs Binding Non-Covalently to the DNA 649 3.3 Intercalating Antibiotics 650 3.3.1 Ellipticin 650 3.3.2 Actinomycins 650 3.3.3 Quinoxaline Antibiotics 652 3.3.4 Anthracyclines and Related p-Quinones 652 3.4 Inhibitors of Enzymes of DNA Replication and Transcription: DNA Topoisomerases 653 3.5 Agents Forming Covalent Bonds with DNA 656 3.5.1 Mitomycin C 656 3.5.2 Anthramycins 656 3.5.3 Cyclopropane, Aziridines, and Epoxide Compounds 656 3.5.4 Streptonigrin 658 3.5.5 Anthracyclines 658 3.5.6 Bleomycin 661 3.5.7 Macromolecular Antitumor Antibiotics and the Enedyine Family of Cytotoxic Drugs 664 3.5.8 Other Agents Forming Active Oxygen Radicals in the Cells 667 3.6 Inhibitors of Mitosis and the Microtubular System 667 3.7 Reduction of the Side Effects of Highly Toxic Anticancer Agents 669 3.8 Cytotoxic Compounds with Poorly Characterized Mode of Activity 669
642
4 Non-Classical Approaches to Antitumor Drugs 670 4.1 Potentiators of Cytotoxic Antitumor Agents 670 4.2 Inhibitors of Glutathione S Transferase as Enhancers of the Antitumor Activity of Drugs 673 4.3 Antimetastasis Drugs and Inhibitors of Angiogenesis 673 4.4 Antitumor Effects of Immunomodulators 675 4.5 Inhibitors of the Cellular Mitogenic Signal Transduction Pathway 676 4.5.1 Inhibitors of Protein Kinases and Protein Phosphatases 677 4.5.1.1 Inhibitors of Protein Kinase C 679 4.5.1.2 Inhibitors of Other Protein Kinases 679 4.5.1.3 Inhibitors of Tyrosine Protein Kinases 679 4.5.2 Inhibitors Affecting the Metabolism of Phosphoinositols 679 4.5.3 Substances Changing the Morphology of Oncogene-Transformed Cells or Showing Selective Toxicity 682 4.5.4 Inhibitors of rus-Farnesyltransferase 685 4.5.5 Inhibitors of Sexual Hormone Production and Hormone-Receptor Interactions 685 4.5.6 Miscellaneous Drugs with Potential Antitumor Activity 685 5 Closing Remarks 687 6 References 688
I Introduction
643
1 Introduction
Malignant tumors are a major cause of mortality, ranking only second to cardiovascular diseases in industrialized countries. The frequency of cancer increases with age, and the general increase in life span in this century renders the prevention and treatment of cancer a serious problem. The aim of this chapter is to review the structures and modes of action of drugs from microorganisms, such as doxorubicin, bleomycin, mitomycin C, and others which have been used in the past as anticancer chemotherapeutics. The treatment of cancer by microbial metabolites was proposed already in 1955 when the cytotoxic effect of actinomycin D on tumor cells was first described. Subsequently, a plethora of other cytotoxic metabolites of microbial and plant origin was dis-
covered (e.g., vincristine, camptothecin, maytansin, podophyllotoxins, taxol, and taxoids); however, only a few of these have found application (Tab. 1). The high toxicity of most of these' drugs and the nonspecific interaction with normal and tumor cells were the major hurdles for their therapeutic application. Today, a small spectrum of natural products (antibiotics, plant alkaloids, and terpenes), synthetic alkylating agents (nitrosoureas, busulfan), heavy metal complexes (cis-diammine-dichloro-platinum, etc.), and antifolates (5-fluorouracil, methothexate) are used in the treatment of cancer (GALEet al., 1981; WILMAN, 1990). In general, the use of cytotoxic drugs in chemotherapy is accompanied by severe side effects such as gastrointestinal disorders, cardiovascular toxicity, nausea, and vomiting. Many tumors are not susceptible (while slowly dividing) or even become resistant to cytostatic
Tab. 1 Biotechnical and Plant-Derived Drugs as Therapeutic Antitumor Agents .
Name Actinomycin D (Dactinomycin) Daunorubicin Doxorubicin Carminomycin Nogalam ycin Bleomycin (Peplomycin, Phleomycin) Anthramycins Mitomycin C, porfiromycin Endynamicin Taxol, taxoids Vincistine Vinblastin Vindesin Camptothecin Etoposid, teniposid Maytanosides, ansamitocin, geldanamycin
Structural Type
_ _ _ _
Mode of Activity
~~ ~ ~
Chromopeptide Anthracycline
Intercalation into the DNA, inhibition of RNA polymerase Intercalation into the DNA, formation of free radicals, induction of DNA strand breaks, inhibition of topoisomerases Formation of free radicals, DNA strand breaking agent Formation of free radicals and induction of DNA strand breaks Formation of free radicals and induction of DNA strand breaks Formation of free radicals and induction of DNA strand breaks Inhibition of tubulin depolymerization Inhibition of mitosis by interference with tubulin polymerization Topoisomerase I1 inhibitor Topoisomerase I inhibitor Inhibition of mitosis
Glycopeptide Benzodiazepine
Qu in one
Enediyne Diterpene Alkaloids
Terpenoid plant toxins Podophyllum toxins Ansamacrolide
644
drugs under therapeutic conditions (ASZALOS, 1988). In the 1970s it became evident that cytotoxic drugs were not the only answer to the cancer problem. New efforts were needed to uncover the molecular causes of malignant cellular growth and to develop more specifically acting anticancer agents. The similarity of the metabolic pattern of normal and cancer cells, the high diversity and different sensitivity of human tumors to anticancer agents made the search for an anticancer wonder drug a rather hopeless enterprise. This is illustrated by the moderate effects of chemical derivatives on the improvement of the activity of common antif cancer drugs and the reduction o their toxic side effects. More recent efforts towards a more specific delivery of cytoxic drugs, such as the development of immunotoxins or drug encapsulations into liposornes, did not improve this general picture. A major aim of modern anticancer chemotherapy is to reduce the side effects, e.g., nausea and vomiting. Major approaches to both the elucidation of the causes of malignant diseases and the search for new anticancer drugs have been promoted by recent advances of molecular biology and biomedical pharmacy. Although it is not within the scope of this survey to discuss these in detail a short reference is given below. Dividing mammalian cells undergo a cell cycle (mitosis). Somatic cells in Go phase enter the GI phase and subsequently the S phase during which chromosome replication
occurs. Cell division is accomplished during the subsequent G2 and M phases. Subsequently, the two cell copies either enter a new Go phase or differentiate to a mature non-dividing cell. Cancer cells are unable to differentiate and hence they undergo a new cell cycle and divide repeatedly. Tumor cells can regulate their growth autonomously by secreting hormone-like proteins such as bombesin. Bombesin is a tetradecapeptide produced by most of the small-cell lung carcinomas. It is a potent mitogen which stimulates growth of small-cell lung cancer (SCLC) cells in serum cultures (MULLER, 1986). Several of the classical antitumor agents were shown to f affect individual phases o the cell cycle in a specific manner, e.g., daunorubicin and cisplatinum act on GI phase cells; arabinosyl cytosine, thioguanine, and doxorubicin act on Sphase cells; bleomycin, and cis-diamminedichloro-platinum act on GZphase cells; vincristine, doxorubicin, colchicine, etoposid, and bleomycin act on M phase cells. This suggested that they could change specific signaling pathways regulating cell division. The recent discovery of the cycline family of eukaryotic regulatory proteins unraveled specific signaling factors which trigger the coordinated events of cell division in yeast (SCHWOB NASMYTH, and 1993). Identical copies of the parental cell are formed in mammals in response to outer signals such as growth hormones (Fig. 1) (KAHN and GRAF,1986; BURCKet al., 1988). These exogenous signals are recognized by mem-
Fig. 1 Signal trans. duction pathway in cellular growth regulation (KAHNand

GRAF,1986).
2 Prescreening for Antineoplastic Agents
645
brane receptors which transmit the message via GTP-dependent membrane proteins (G proteins) to proteins located at the inner side of the cytoplasmic membrane. In response, the latter become activated and modify (e.g., phosphorylate, farnesylate, etc.) proteins which by switching on subsequent events of the cellular growth-regulating pathway, act as a kind of an operational amplifier (YARDIN and ULLRICH, 1988). A normal cell only divides in response to a given signal, but a cancer cell has lost all growth control and repeatedly enters a new cell cycle. Hence, a deregulated expression of potential cancer genes (protooncogenes) encoding growth factors, membrane receptors, transcriptional factors, and regulatory enzymes has been suggested. In turn, autonomous secretion of growth factors, deregulated responses to growth factors, and continuous cell division will occur. The existence of tumor suppressor genes has been proven, the products of which prevent the outbreak of malignant growth. Moreover, it became evident that carcinogenesis can be caused by chemical DNA strand-breaking and/or alkylating agents, and transforming oncogenic viruses (BRADSHAW and PRENTIS, 1987; PIMENTEL,1987). Insertion of the latter into the DNA causes particular genes (proto-oncogenes) involved in growth control to become cancer genes (oncogenes). Usually, oncogene-transformed cells can be distinguished from the ancestral type by the overproduction of growth factors (autocrine secretion of growth hormones), - altered structures of membrane receptors rendering them permanently activated even in the absence of exogenous growth factors, - altered structures of growth factors giving rise to permanent (irreversible) receptor activation, and - altered production rates, structures, and functions of intracellular signaling proteins such as protein kinases, phospholipases, inositol kinase.
-
tinent, deregulated cellular growth function. A series of receptor proteins and growth-regulating enzymes has been well characterized and in vitro screening assays have been developed (Tab. 2). Since the 1980s, many new structures which interfere with cellular signal transduction have been discovered. These are regarded as soft anticancer drugs. Their particular mode of action renders them invaluable biochemical tools in detailed investigations of malignant growth. The first part of this article surveys microbial products which are used as classical, cytotoxic antitumor agents. They interfere with DNA replication, transcription, and mitosis. In the second part, non-classical approaches with new and lowtoxicity anticancer agents are discussed (UMEZAWA, 1989; WILMAN, 1990). Some of these new drugs amplify the cytotoxicity of classical anticancer agents, even in resistant cell lines, and inhibit metastasis and angiogenesis of tumors. In addition, inhibitors of growth regulatory enzymes such as protein kinases, inositol kinases, and phospholipases will be reviewed.
2 Prescreening for Antineoplastic Agents

In order to screen microbial cultures for antitumor drugs, predictive prescreens are required because animal antitumor assays have poor sensitivity. For DNA-damaging and cytotoxic agents a series of specific high capacity assays has been developed (FOX, 1991). Promising results have been obtained with microorganisms which express special genes (such as recA) in response to DNA damage. Moreover, enhanced activity against repairdeficient microbial mutants is another indication for cytotoxic drugs interfering with DNA. Studies on DNA-replication in a cellfree system have led to the detection of covalently modifying agents (GREENSTEINand MAIESE,1984). Neoplastic cell lines have been widely used in many laboratories to screen microbial cul-
Different types of oncogene-expressing (transformed) cells are now available as tools in the search for specific inhibitors of the per-
646
Tab. 2. Inhibitors of Enzymes of the Cellular Growth-Regulating Signal Transmission Pathway and Drugs Exerting a Non-Classical Action on Tumors
EnzymelScreening Feature Protein kinase C Protein tyrosin kinase Protein phosphatase Calmodulin and CAMPPhosphodiesterase Phospholipase D Inositol-specific phospholipase C Inositol phosphate receptor agonist Inositol kinase Diacylglycerol kinase Inositol monophosphatase DNA topoisomerase I DNA topoisomerase I1 Poly(ADP-ribose)polymerase Glyoxylase I Induction of differentiation of cancer cell lines Metastasis inhibitors and inhibitors of angiogenesis Glutathion S-transferase Enhancer of the cytotoxicity of antitumor agents Reduction of side effects of highly toxic anticancer agents Ras-farnesyltransferase Immunostimulators Activity against tumor cells, resistant to other antitumor drugs Immunosuppressants Adenosine deaminase 5 -Nucleotidase Inhibitors of the eukaryotic cell cycle Melanogenesis Inhibitor Calphostin, staurosporin, RK-286c, K-252a, UCN-01, BE13793c, Sch45752, balanol, MS-282a BE-23372M, emodin, lavendustin, erbstatin, genistein, epiderstatin Tautomycin, okadaic acid, dephostatin, microcystin, calyculin KS-505a Sch49210, Sch53514, Sch53517 Hispidospermidin, psi-tectorigen, Q12713 Adenophostins Herbimycin, isoflavones, echiguanin, inostamycin, piericidin B1-N-oxide, echigramins, piericidin, benzaldehyde Cochlioquinone, temphone L-671-776 Dotriacolide, camptothecins (from plants), rebeccamycin UCT4B, saintopin, BE-10988, streptonigrin, anthracyclines, BE-22179, cyclothialidin 2-Methylquinazolines, benzamides (synth.) Glyo I and I1 Hygrolidine, cycloheximide, trichostatins, reveromycin, differanisol, differenol A, microcystilide A, herbimycin, indolecarbazoles (staurosporin), lavanducyanin, oxanosin, kasuzamycin TAN-1120, TAN-1323, U-77863, U-77864, matlystatins, siastatin B, WF-16775, erbstatin, herbimycin, lactoquinomycin TA-3037A, benastatins, rishirilid, cysfluoretin, bequinostatin Piperafizins, bisucaberin, verapamil (synth.), BE-l3793c, cyclosporin A, resorthiomycin, rubiginon, polyethers (laidlomycin), hatomarubigin Conagenin Pepticinnamine, streptonigrin, indolocarbazoles Forphenicinol, FK-156, bestatin Kazusamycin, lactoquinomycin Spergualin, 15-deoxyspergualin Adecypenol, pentostatin Nucleoticidin Reveromycins OH-3981
3 Classical Anticancer Drugs
647
ture or plant extracts; however, better purified fractions and components are required for results which are more reliable. Thus, the test program of the National Cancer Institute of the USA operates in vitro disease-oriented screens which are directed specifically to cell type specific agents. New compounds are tested against a panel of approximately 60 cell lines derived from human solid tumors such as colon tumors, melanoma, kidney tumors, ovarian tumors, brain tumors, and leukemia (GREVERet al., 1992; KAO and COLLINS, 1989). A more recent approach is the usage of cell cultures in the screening for inducers of tumor cell apoptosis (YAMAZAKI al., et 1995). Increasing use is been made of transformed cells (such as those transformed by retroviruses, e.g., the Rous sarcoma virus) to search for agents which display a selective toxicity or which alter tumor cell morphology. In addition, these cell lines are useful as tools in screening for inhibitors of special oncogeneencoded cellular functions (UMEZAWA, 1989). Another more recent, promising approach concerns the protein targets of drugs interfering with the regulation of the cell cycle. High-throughput screening assays have been developed for a series of receptors and enzymes of the cellular signaling cascade which are crucial in the development of malignancy. Despite these promising in vitro approaches trials with animals will finally be required to reduce the number of prescreeningpositive compounds to an amount which justifies continuing investigations, including clinical trials. In this context, the use of human tumor xenograft models represents a new approach to the evaluation of putative anticancer agents. An effective screening program for new anticancer agents will need new sources of drugs such as rare microorganisms, plants, and animals. During the last ten years, marine organisms such as tunicates, molluscs, dinoflagellates, and bryozoa have become the subject of an expanding field of research in the search for new antitumor agents (see, e.g., dolastatins, spongistatins, bryostatins, okadaic acid, and other cytotoxic metabolites) (JENSEN and FENICAL, 1994).

3.1 Nucleoside Antibiotics and Analogs of Nucleobases
The structural class of the naturally occurring nucleoside antbiotics comprises approximately 150 representatives (ISONO,1988; AsZALOS and BERDY,1978). Numerous analogs have been obtained by chemical synthesis. They exert their antimicrobial and cytotoxic activity as pseudo nucleobases or nucleosides which are activated to pseudo nucleotides. Thereafter, they act via a negative feedback mechanism on nucleotide-forming biosynthetic pathways or inhibit DNA- and RNA-dependent polymerases. However, the incorporation of pseudo nucleotides may also diminish the template function of DNA. Antagonists of folic acid, purines, and pyrimidines have been used in anticancer chemotherapy. Nucleobase analogs such as 5-fluorouracil (synthetic), 6-mercaptopurine (synthetic), thioguanine, thiouridine (NISHIKORI al., et 1992), 7-hydroxy-guanine (Streptomyces sp.) (KITAHARA al., 1985), and pseudo nuet cleosides such as arabinosyl cytosine (synthetic), cadeguomycin (TSUCHIYA al., 1992), et (Streptomyces hygroscopicus) oxanosin (YUAN al., 1985), neplanocin (Ampullarielet la reguluns) (YAGINUMA al., 1981), spicaet mycin, its derivative SPM VIII (KAMISHOHARA et al., 1994) and others (see, e.g., ISONO, 1988) display antitumor and antiviral activities against various cells and viruses. The general therapeutic problem associated with these agents is their poor selectivity and relatively high toxicity. However, several examples of a more selective action of nucleosides against viruses are known; synthetic drugs such as acyclovir and its derivatives are powerful inhibitors of herpes virus thymidylate kinase, and hence invaluable therapeutic agents (Fig. 2, see also the therapeutic effect of 2,3dideoxynucleosides on HIV virus infections). A more specific antitumor activity has been demonstrated for inhibitors of enzymes asso-
648
H2N
7~mxy#wnutn
H
T h ~ ~ i l
*
HO-CH, OH
&y
NH OH
FH2
y 2
KLy
va.oXO(lin
COOH
Fig. 2. Structures of neoplasm inhibitory nucleobase analogs, nucleosides, and antifolates.
ciated with special kinds of malignancies, such as adenosine deaminase. The importance of purine metabolism for the immune system was recognized due to the observable associaton of deficiencies in purine salvaging enzymes, such as adenosine deaminase and purine nucleoside phosphorylase, with heritable immunodeficiency. Coadministration of adenosine deaminase inhibitors in the treatment of leukemia by arabinosyl adenosine caused syndromes similar to those observed in immunodeficient patients.
Inhibitors of this enzyme such as pentostatin (2-deoxy-coformycin; Streptomyces antibioticus) (HOLLIS-SHOEWALTERal., 1992), and et adechlorin (Actinornadura sp. OMR-37) (OMURA al., 1985) (Fig. 2) were aimed to et prevent the degradation of arabinosyl adenosine by this enzyme (AGARWAL al., 1983; et OMURA al., 1986). et In eukaryotic cells, 5 '-nucleotidase is necessary for the formation of the 5'cap structure of mRNA. Hence, inhibitors of this enzyme such as the polysaccharide nucleoticidin
649
(Pseudomonus sp.) (OGAWARA al., 1985) (MURATA,et al., 1987). Diazaquinomycin A et inhibit malignant cell growth. Moreover, fo- (MURATAet al., 1985) and vanoxonin (KAlate metabolism was suggested as an addition- MA1 et al., 1985) are microbial inhibitors of al promising target for antitumor drugs (see, thymidylate synthase which have cytotoxic efe.g., the role of methotrexate as an inhibitor fects on cell cultures (Fig. 2). Moreover, a series of nucleotide analogs of tumor cell dihydrofolate reductase; Roshas been isolated from microorganisms, OWSKI et al., 1992) (Fig. 2). More recently, thymidylate synthase (VANDER WILT et al., which are active as antitumor and antimetas1994; KALMAN, 1989) and serine hydroxyme- tatic agents due to their interaction with sugar thyltransferase (RAo, 1991) have been sug- nucleotide glycosyltransferase (see below) gested as targets for new anticancer drug de- (KHANand MALTA,1992). velopment. The search for inhibitors of folic acid metabolism from microbial cultures (e.g., by the 3.2 Drugs Binding Non-Covalently use of Entercoccus facium as an assay organ- to the DNA ism) led to the discovery of analogs and inhibNetropsin and distamycin (pyrrolamidine itors of thymidylate synthase as antifolates. Thus, 7-hydro-8-methylpteroylglutami- antbiotics from Streptomyces netropsis and S. nylglutamic acid (HMPGG) was isolated as a dimtallus, respectively) are invaluable biofolic acid analog from an Actinomyces culture chemical tools in recent DNA research
Nelropsin
Distamycin
HO
Fig. 3. Structures of antibiotics binding non-covalently to the DNA.
Hm:
0
Charbeusin
650
(Fig. 3). Both antibiotics bind to the minor groove preferably at regions rich in adenosine and thymidine. The complex formation of netropsin and the duplex DNA (d(5CGCGAATTCGCG3 )) suggested that at the 5AATT sequence in the center the three NH groups of netropsin form hydrogen bonds with 0-2 of thymine and N-3 of adenine (KOPKAet al., 1985) (see Fig. 3). Distamycin which is a closely related antibiotic forms van der Waals bonds between its 5 NH-groups and 0 - 2 of thymidine and the N-3 of adenine in the DNA minor groove (COLLet al., 1987; DERVAN al., 1987) (see Fig. 3). et As a consequence of binding, the unwinding of DNA during replication and transcription as well as the access of the pertinent enzymes and proteins to specific DNA sites are prevented (ZIMMER al., 1990). Chemical et modifications of distamycin were done to improve the antiviral properties of its structures (ARCAMONE, 1994). Non-covalent binding to DNA at different sequences was also shown for a series of polycyclic aromatic antbiotics such as chartreusin (KRUGERet al., 1986) (Fig. 3), chromomycin (KOAMURA al., 1988; UCHIDA al., 1985), et et angucyclins (saquayamycin), and saframycin (see Fig. 8a). However, none of these compounds are used in anticancer therapy. Both chromomycin A3 and olivomycin inhibit the synthesis of DNA and RNA in vitro and the function of RNA polymerase and DNA polymerase I in vivo. Binding of chromomycin to DNA is magnesium-dependent and occurs preferably at regions with a high G/C content. The complex formed is stable to nuclease digestion. Similar to chromomycin A3, an antibiotic from Streptomyces plicatus, mithramycin binds preferably to d(5 ATGCAT3)2 regions (BEAUVILLE al., 1990). et Olivomycin interferes with the elongation step of RNA polymerase and inhibits DNA polymerase I by binding to guanine- and cytosine-containing DNA regions.
tions with DNA have been studied extensively by physicochemical methods (WANGet al., 1987; WANG,1992). The term intercalation refers to the insertion of flat, planar, aromatic molecules (quinolines, acridines, phenanthridines, phenoxazines, anthraquinones, fluorenes) into the minor groove of DNA, between two adjacent base pairs (WANG,1992; WARING, 1975). Polar substituents such as amino acids or sugars (see the structures of actinomycin D and doxorubicin, Fig. 4) stabilize the intercalation complex by the formation of hydrogen bonds with the phosphodiester groups of the deoxyribose moieties of DNA (MANGER,1980; KNUGHet al., 1980). Intercalation results in the distortion of the DNA and steric stress whereby neighboring bonds are weakened. As a consequence, the template function of the DNA is reduced, and single and double strand breaks as well as frameshift mutations may occur. In addition to some synthetic and plant-derived drugs (flavines, acridine dyes, ellipticin), the actinomycins (KNUGHet al., 1980; ADAMSON al., 1979) and the anthraet cyclines (EL KHADEM,1982; CASSIDY and DOUROS, 1988; LOWN,1988) are well known intercalators of microbial origin. The latter type of antbiotics differs from the intercalating drugs by its radical-forming properties and its inhibitory effect on DNA topoisomerase I1 (see below).
3.3.1 Ellipticin
Ellipticin and its 9-methoxy derivative were isolated from plants such as Excavatia coccinea and Chrosia moorie. They inhibit DNA synthesis by binding to the minor grove of double-stranded DNA.
3.3.2 Actinomycins
Already in 1955 the antitumor activity of actinomycin D (Fig. 4) (from Streptomyces antibioticus) was observed (TSCHAGOSHI et al., 1986). Representatives of the actinomycin chromopeptide family are extremely potent inhibitors of RNA polymerase, and they have widely been used as a biochemical tool to
3.3 Intercalating Antibiotics

Intercalating drugs have been important both as tools of biochemical research and as anticancer agents. Their molecular interac-

0 OH
651
0
Mel-Val-0
MA-ValMeGly
L+
D-Val
4
MGJ
OH Me Me
Doxorubicin Actinomycin D
Echinomycin
UK 63.052
CH.
Fig. 4. Structures of intercalating drugs.
Mitoxantron(X=oH) Ametantron (X=H)
Bisanthrene
NHyN'3
NH
652
study RNA synthesis (GALE et al., 1981; FORNICA, 1977). Actinomycins are composed of a phenoxazinone chromophore, which is the intercalating unit, and two linked pentapeptidolactone moieties (MAYERand KATZ, 1978; KATAGIRI,1975). The homologs are distinguishable by their amino acid composition. The compositon of the actinomycin complex depends on the available amino acid precursors (FORNICA, 1977). Intercalation of the actinomycins requires the 2-NH2 group of the phenoxazinone chromophore and the quinoid DNA binding occurs at (d(CAT2-C 02; GAT))2 sequences (ZHOUet al., 1989). Intercalation of actinomycin D and its analogs in DNA was extensively studied by spectroscopic methods (KNUGHet al., 1980). Computational techniques have been employed to calculate the substituent effect on free energy of binding to DNA (LYBRAND, 1988). In the past, actinomycin D (dactinomycin, Fig. 4) was used in clinical anticancer trials but due to its high toxicity its therapeutic use in Wilms tumor, choriocarcinoma, testis carcinoma, neuroblastoma, and sarcoma is disputed. In order to reduce toxic side effects and to improve the efficacy against certain tumor cells, a series of analogs of actinomycin D was obtained by semisynthesis, but none of them appeared to be more promising than the parent compound (ADAMSON al., 1979; FORet NICA, 1977).
sandramycin, a related 3-hydroxy-quinoxalinic acid cyclopeptide antibiotic was isolated, which lacks a sulphide bridge (MATSON al., et 1993). The compound UK-63.052 (Sfreptomyces bruegensis) (Fig. 4) is a new representative of the quinomycin group of antibiotics (RANCE al., 1989). et In every case, the alanine residue of the octapeptide ring of echinomycin type antibiotics is a critical determinant of the observed specificity for CpG dinucleotide sequences (WARING 1979, 1990). This is due to hydrogen binding interactions which involve NH and CO groups of the alanines together with the 2-amino group and N-3 of guanine in the minor groove of the helix. The binding of echinomycin type antibiotics to DNA is very tight and remains stable even during gel electrophoresis of DNA fragments in footprinting analysis (Fox, 1990).
3.3.4 Anthracyclines and Related p-Quinones

Both, daunomycin and doxorubicin (Fig. 4) (a semisynthetic derivative of daunomycin) bind to DNA sites which contain either only (G:C) or (A:T) base pairs (ASZALOSand BERDY,1978; SALUZand WIEBAUER, 1995; NEIDLY and WARING, 1983; WILMAN, 1990). In general, a three base pair binding site is required for the intercalation. The aminosugar at C-7-0 of the benzo[a]naphthacene quinoid aglycone is needed to replace water from the minor groove and to anchor the molecule to the phosphodiester linkages between the adjacent deoxyribose moieties (WANG et al., 1987). Due to these properties some representatives of the anthracyclines induce a small degree of helical unwinding. Doxorubicin and less frequently daunorubicin are used in the treatment of acute lymphatic leukemia, acute myeloid leukemia, lymphoma, sarcoma, Wilms tumor, and cancer of the breast, lung, bladder, thyroid, and prostate. After recognizing the anthracyclines as lead structures a series of intercalating anthraquinone type anticancer agents such as ) mitoxantron, ametantron (Fig. 4 , anthropyrazol, and bisanthrene (Fig. 4) was developed as anticancer drugs and used therapeutically,
3.3.3 Quinoxaline Antibiotics

The anticancer activity of echinomycin (Fig. 4) was rediscovered due to investigations of its mode of action at the molecular level. It was the first DNA bis-intercalator to be identified (WARING, 1992). Bis-intercalation involves the binding of a drug to DNA via the quinoxaline rings, with the peptide moieties binding to the DNA minor groove (ADDESet al., 1992). Representatives of this group of antibiotics display different sequence specificity (ADDESS et al., 1992), whereby CpG sequences are preferred. Triostin and luzopeptin are related antibiotics which display similar bis-intercalating properties (WANG,1992) as detected by X-ray diffraction and NMR spectroscopy. Recently,
653
e.g., in the treatment of breast cancer (ASZA- SHEN, 1994; CHAKRABORTY al., 1994; et LOS,1988; HOLLIS-SHOEWALTER, 1988). CHENand LEROY,1994; HSIEH,1992), either by interfering with ATP hydrolysis or with the DNA cleaving subunit A. 3.4 Inhibitors of Enzymes of DNA Drugs introducing lesions in the DNA trap the transient topoisomerase 11-DNA interReplication and Transcription: mediate which is held together by two covalDNA Topoisomerases ent bonds. Anthracyclines (see Fig. 7), ellipticin, and epipodophyllotoxins (etoposid and Dramatic advances have been achieved in tenoposid) (Fig. 5b) are the major antitumor the field of DNA topoisomerases. These in- drugs that form cleavable complexes in eukaclude the cloning of new topoisomerase genes ryotes. involved in DNA recombination and segregaIn a search for new inhibitors of topoisomtion of replicated DNA (WATT and HICK- erase I1 (HECHTet al., 1992) a benzoanthraSON, 1994; POOT and HOEHN, 1993; GIAC- quinone (UCE 1022) has recently been isoCONE, 1994). Hence, topoisomerases have be- lated from Puecilomyces sp. (FIJI] et al., come promising targets of modern anticancer 1994). Other inhibitors of microbial origin are agents (PAOLETIT,1993). Antitumor agents dotriacolide (FIJII et al., 1994) (inhibitor of such as epipodopyllotoxins (etoposid, tenipo- topoisomerase I and DNAse), TAN-1496A (a sid), acridine dyes, anthracyclines, and ellipti- diketopiperazine, Strepfomyces sp.; Fig. 5a) et cins were found to be specific inhibitors of to- (FUNABASHI al., 1994), and UCE6 (a nonpoisomerase 11, and the plant alkaloid camp- glycosylated anthraq uinone, Streptomyces sp.; tothecin was recognized as a specific inhibitor Fig. 5a) (FIJII et al., 1993). Anthracyclines of mammalian topoisomerase I (Fig. 5a) were also shown to inhibit toposomerase I in addition to topoisomerase I1 (CROW and (FOSTELand SHEN,1994). 1994). Topoisomerases as targets of antibiotics CROTHERS, Recent screening approaches in the search and antitumor drugs, their relation to the topological stage of intracellular DNA, and for microbial inhibitors of type I1 topoisomertheir function in replication have been dis- ases (PAOLETTI,1993) revealed new inhibicussed in a series of reviews (ZIMMER al., tors, such as BE-22179 (a peptide from Strepet and 1990 FOSTEL SHEN,1994; CHENand LE- fomyces) (OKADAet al., 1994), saintopin (a ROY, 1994). The enzymes catalyze the con- benzo[a]anthraquinone from Paecilomyces et certed breakage and rejoining of the DNA sp.; Fig. 5b) (YAMASHITA al., 1990a), BEbackbone, and hence they are indispensable 10988 (Sfrepfomyces xanthocidicus; Fig. 5b) for DNA replication, transcription, chromo- (OKAet al., 1991), streptonigrin (TOLSTIKOV et some segregation, and recombination. Mam- et al., 1992; YAMASHITA al., 1990b) and its malian type I topoisomerases mediate the re- derivatives, and UCT4B (terpentecine type, laxation of negatively or positively super- Strepfomyces sp.; Fig. 6b) (UOSAKI et al., et coiled DNA by transiently breaking and re- 1993; KAWADA al., 1992a). The majority of leasing one DNA strand in such a manner these inhibitors exhibit potent antitumor acthat the linking number changes by steps of tivity. Some also affect bacterial topoisomerone. Type I1 topoisomerases break and reli- ase I1 (gyrase) which is the target of several gate phosphodiester bonds in double- important antibacterial agents. A cytotoxic stranded DNA passing another duplex region quinoid system, popolahuanone E (Fig. Sb), through the break, thereby altering DNA to- was isolated from a Pohnpei sponge. It is sepology. In addition, they are structural pro- lectively toxic for the A549 non-small cell huteins involved in the spatial organization of man lung cancer cell line (CARNEY and 1993). In a series of comparative chromatin (ZUNINO and CAPRANICO, 1990). SCHEUER, One group of agents such as coumarin deriva- studies it was shown that the interaction with tives (novobiocin, coumermycin) and the syn- mammalian topoisomerase I1 is one of the thetic quinolone antibacterial agents act at major causes of cytotoxicity of the anthracyand the level of the enzyme subunits (FOSTELand cline type antibiotics (CAPRANICO ZUNI-
654
H % o
Me
OH
UCT 48 Sainbpin
OH
OH
"$+,pCoN4
N
0
Me
Popolohuanone E
BE409BB
MeO
Pdophyllotoxin(Rl: H; R2 Me)
Teniposid (R1:X ;R2: H); Etoposid (Rl: Y; R2: Me)
0 4
e
OH
Me'CH-O
0%
O \
Fig. 5. Structures of topoisomerase inhibitors. a Inhibitors of topoisomerase I, b inhibitors of topoisomerase 11.
655
1990 ZUNINO and CAPRANICO, 1990; GREVER al., 1992). Although interference et with the functions of DNA topoisomerases I and I1 holds great promise for the treatment of cancer, these drugs may themselves be dangerous due to their mutagenic and carcinogenic potential (ANDERSON BERGER, and 1994). DNA helicase has recently been proposed as a target involved in DNA replication, repair, recombination and transcription by unwinding of double-stranded DNA. Heliquinomycin as a new spirocyclic aromatic compound from Streptomyces sp. effectively inhibits the enzyme (CHINO al., 1996). et
NO,
are active against a broad spectrum of human tumors (SLICHENMYER al., 1993). They et slow down the religation step of topoisomerase I and stabilize the covalent adduct between the enzyme and DNA. In S phase cells double-stranded breaks occur at the position of the topoisomerase I-DNA adducts which is probably the reason for the high cytotoxicity of these agents. Both drugs are now under clinical investigation (SLICHENMYER al., et 1993; CURRAU al., 1993). et
3.5 Agents Forming Covalent Bonds with DNA

Covalent modification of DNA by interaction of reactive molecules with bases or sugars blocks its template function and induces the onset of repair processes. As an initial step, radical species are generated by the drug molecule due to its interaction with metabolic enzymes such as cytochrome-dependent and other oxidoreductases. The result is interstrand crosslinking and single- or doublestranded DNA cleavages. Some of the prominent representatives of this type of commercially available agent will be discussed (PINEDO, 1980).
Podophyllotoxins as inhibitors of topoisomerase 11: The antitumor drugs podophyllotoxin (Fig. 5b), peltatins, and other lignane type compounds have been extracted from Podophyllum peltatum L. These compounds interact non-covalently with tubulin and inhibit mitosis during metaphase (see Sect. 3.6). Due to their high toxicity for mammalian organisms they cannot be used for therapeutic purposes. Semisynthetic derivatives, such as tenoposid, etoposid (Fig. 5b), and mitopodosid appear to be more promising. They poorly interfere with tubulin and mitosis but instead, inhibit the transport of nucleosides in cells and the incorporation of thymine and uridine into DNA and RNA due to interaction with topoisomerase I (FRANZ, 1990). Camptothecin type compounds as inhibitors of topoisomerase 11: Camptothecins are plant dimeric indole alkaloids which were discovered as selective growth regulators for several species of mono- and dicotyledonous plants (BUTAand KALINSKI, 1988; WALLand WANI,1993; SLICHENMYER et al., 1993; CURRAU, 1993) (Fig. 5a). Chemical studies on inhibitors of DNA topoisomerase I which lead to the production of semisynthetic derivatives, such as CPT-11 and topotecan suggested a therapeutical potential for these types of compounds (ANDOH al., 1993). Topotecan and CPT-11 et
3.5.1 Mitomycin C
Mitomycin C (Streptomyces caespitosus) (Fig. 6) is used in anticancer chemotherapy by virtue of its inhibitory effect on DNA synthesis and its ability to induce strand cleavages (TOMASZ,1994). It is known as one of the most potent antitumor antibiotics. The molecular mechanism of action of this highly toxic antibiotic involves the reduction of the quinone and subsequent activation of C-1 (PINEDO, 1992; TOMASZ, 1994; SARTORELLI al., et 1993). Mitomycin C and its analogs, such as porfiromycin and mitiromycin (KASAIet al., 1991; ARAIet al., 1994) are used as an alternative to radiation during the treatment of hypoxic solid tumors which have acquired enhanced sensitivity to this bioreductive alkylating agent. It has been suggested that in hypoxic cells, one-electron transfer enzymes, such as diaphorase, control bioreductive alkyl-
656
ation. At least six different enzymes are capable of activating mitomycin C and other similar drugs. The nature of the activated molecular species and the resultant biological lesions can vary with the activating enzyme. This, in turn, causes variations in toxicity for different cell types. The process of activation is accompanied by the opening of the conformationally constrained aziridine ring to yield free radicals as intermediates. The double-bonded structure formed interacts with guanine bases in opposite DNA strands, thereby crosslinking them in a bifunctional manner (RAUTH and RAYMOND, 1993; TOMASZ, 1994; ROCKWELL et al., 1993; WARDMAN, 1990; BUTLER and HAEY, 1987). Frameshift mutations appear to be another consequence of DNA repair and cause of potent mutagenic and teratogenic activity of the mitomycins. In order to obtain less toxic and more potent mitomycin C analogs, both congeners of mitomycin fermentations (see, e.g., albomitomycin KW2149) and synthetic derivatives have been isolated (KONOet al., 1991, 1995; KASAIet al., 1991). DNA damage appears to be critical for the cytotoxicity of mitomycins. Both interstrand and intrastrand crosslinking occur in mitomycin-treated cells.
3.5.3 Cyclopropane, Aziridines, and Epoxide Compounds
The compound CC-1065 (Sfreptomyceszelet ensis; Fig. 6) (MARTIN al., 1980) is a heterocyclic antibiotic which forms a DNA adduct via the initial interaction of the DNA guanine residues with a cyclopropane ring attached to a cryptic p-quinoid structure. This is followed by molecular rearrangements and cleavage of the phosphodiester bond from deoxyribose. The compound CC-1065 displays sequence specificity in that it binds to regions rich in guanine (SCAHILL al., 1990). et A series of microbial products containing aziridine and epoxide structures, such as azinomycins (Strepfomyces griseofuscus), WF3405 (Amauroascus aureus), and duocarmycins A and SA (S. zelensis) (Fig. 6; ICHIMURA et al., 1991) was suggested to form covalent bonds with DNA. The exhibit extremely potent cytotoxic activity (ZCso=lo-'* M10-9M for Hela S3 cells). The constrained three-membered ring structure in addition to particular steric and electronic conditions permits a nucleophilic attack on DNA bases. Duocarmycin SA (ICHIMURA al., 1990) et was found to be the most stable and most potent cytotoxin of these agents. They alkylate 3.5.2 Anthramycins DNA in mechanistically similar manner to CC-1065 (ICHIMURA al., 1990,1991). DNA et Anthramycins (pyrrolo(l,4)-benzodiaze- alkylation occurs via addition of adenine N3 pines) bind tightly via their C-11 atom to the to the least substituted carbon of the acti2-amino group of guanine bases with the vated cyclopropane within AT-rich minor elimination of water (Fig. 6). The amino groove sites (BOGERand JOHNSON, 1996). In groups of these antibiotics play an important this context, scirpene type mycotoxins such as role in their molecular mechanism of action trichothecin (Fig. 6), trichodermole, anguid(BARKLEY al., 1986 MORRIS al., 1990). ine, nivalenol, crotocin, and their analogs et et Representatives of these agents include an- were tentatively used as antitumor agents. thramycin (Fig. 6), sibiromycin, tomaymycin, They are frequently occurring products of neothramycin, and porothramycin (TSUNA- Fungi imperfecti (Trichothecium, Fusarium, Myrothecium, Trichoderma)and inhibit DNA KAWA et al., 1988). Due to the high risk of hazardous side effects, the anthramycins offer and protein synthesis. no advantage in cancer treatment compared to other cytotoxic antitumor drugs. Some anthramycins were evaluated in clinical trials 3.5.4 Streptonigrin but they have not come to general use. Streptonigrin (bruneomycin) (Fig. 6) and its natural analogs (LIN et al., 1992) are cytotoxic radical-forming anticancer drugs which modify DNA covalently and cause strand
657
H2n*
,o y ,c
NH
H$cyJ&
OH H 0
Anthramycin
hlt
0
MitomyeinC
0
0
Pynolo~,l-cl[lnbcnz~i~epine DNA adduct (41
H.CO Duocarmycin A
p: :
CONH, OCH,
I
NH
'0
Streptonigrin
Me0
cc-lo65
hwo
0
crotonyl
Triehothccin
Fig. 6. Structures of drugs binding covalently to the DNA.
breaks. Interactions with topoisomerase I1 35.5 Anthracyclines appear to be involved in this activity (YAMASHITA et al., 1990b). Streptonigrin and a seThe anthracyclines (Fig. 7a) belong to a ries of chemical derivatives were shown to in- class of antitumor drugs which exhibit activity hibit HIV reverse transcriptase (TAKEet al., against a spectrum of human cancers. Only a few tumors, such as colon cancer, melanoma, 1989).
658

O-(sugar)i-3
HO
w 2 p F R
COOMe
OH
H CH3
\
OH
O
OH
0(sugar)l-3
OMe
@ ; &
Osugar
\
&~HzoH \
OH
OMe
OH
DounoruMcin
OMe
OH
Doxorubicin
c 3OH H& 0
c 3 0 OH H@
cH3@ OH
Aclacinomycin A (AcbruMcin)
Nogalamycin
IL-rhodospminylL-fucosylLcinerulose
b
Fig. 7. Structures of anthracycline type antibiotics. a General substitution pattern of anthracyclinone aglycones of anthracyclines, b anthracycline structures, c recently discovered new anthracycline structures.
659
M @NMe, e
Yellamycin B
OH
Alldimycin OH
COOCH,
OH
OH
Fig. 7c
Epelmycin A
0 L-rhodosaminL-rhodosaminylLcinerulose A
chronic leukemias, and renal cancer are unresponsive to them. The first clinically effective anthracycline, daunorubicin (DNR), was discovered independently in 1963 at Farmitalia (daunomycin) and RhGne-Poulonc (rubidomycin). Later on, adriamycin (doxorubicin, DOX) was discovered as the product of a mutant strain and was also obtained semisynthetically. In 1969, it was demonstrated that DOX was less toxic and more active against a much broader spectrum of tumors than DNR. The former rapidly replaced the older drug in clinical applications. In general, cumulative cardiotoxicity limits treatment with DOX to approximately nine months at usual dosages, and most cancers develop resistance to this agent. DOX differs from DNR only by an additional hydroxyl group. This fact has encouraged researchers worldwide to search for analogs of DOX that display lower acute toxicity or cardiomyopathy, that can be administered orally, and that have different or greater antitumor efficacy. The aim of this research was three-fold:
(1) Synthesis of new anthracycline analogs: This process has continued from the 1960s up to the present and might also continue in the future (LOWN,1993). No one has kept a count of the anthracycline analogs synthesized over the past 25 years, but their number probably exceeds 2000, and more are being reported every month. (2) Coadministration of other agents with DOX Researchers have attempted worldwide to administer doxorubicin in conjunction with other substances that will mitigate cardiotoxicity or overcome drug resistand ance of the cancer cells (STEINHERZ et STEINHERZ, 1991; VAN KALKEN al., 1991). Moreover, immunoconjugates of DOX and different anthracyclines incorporated into liposomes are being evaluated in clinical trials to minimize heart exposure to the drug while maintaining antitumor efficacy (PEREZ-SOLER al., 1995). et (3) Screening for new anthracycline compounds of microbial origin:
660
DNR and DOX as well as their biosynthetic congeners are all derived from actinomycetes, particularly from the genus Streptomyces. They are obtained by using the well known fermentation and recovery techniques generally used in antibiotic technology. Many of the more than 400 anthracyclines isolated and characterized to date have been isolated from blocked mutants of a variety of strains. The main anthracyclines selected for clinical evaluation are shown in Fig. 7b. Alldimycins (JOHDOet al., 1991a), yellamycins (JoHDO et al., 1991b), epelmycins (Fig. 7c) (JOHDO et al., 1991c), respinomycin (UBUKATA et al., 1993), cororubicin (ISHIGAMIet al., 1994), mutactimycin (MAEDA et al., 1992), betaclamycin B (YOSHIMOTO al., 1992), cinerubin R et (NAKATA al., 1992), and rubomycins F et and H (FOMICHOWA al., 1992) are exet amples of the previously discovered new anthracycline type structures.
Anthracyclines consist of a tetracyclic aglycone (anthracyclinone, tetrahydro-naphthacene-quinone) which is linked with up to ten sugars usually attached at positions C-7 or C-10 C-Cglycosylated derivatives have also been reported. The general substitution pattern of the anthracyclinones is shown in New anthracycline analogs: Fig. 7a. The nogalamycin family is distingui- the next generation: shable by the unusual substitution pattern of the left-side aromatic ring by a In the 1970s and the early 1980s the main bis-oxa-bicyclo[3.3.l.]nonane ring system 'aims were to expand the antitumor spectrum (DUMITRIU, 1996). of DOX, to reduce its cardiotoxicity, and to Fig. 7c shows a series of anthracycline develop an orally administrable compound. structures representing various building pat- The outcome of this research was a few anaterns. The aglycones differ in the hydroxyla- logs which display moderate clinical advantion of the aromatic nucleus and in the sub- tages (WEISS, 1992). In the 1980s developstituents at positions 7 and 10. Modifications mental efforts in this field also turned to the of the side chain (e.g., by oxidations, reduc- problem of multi-drug resistance of cancer tions) andor methylations are a consequence cells. For instance, F860191, AD-198, of numerous enzymatic and non-enzymatic FCE23762, ME2303, and MX-2 were rereactions occurring during biosynthesis. The ported to be active against DOX resistant and anthracyclinone part of the anthracyclines is multi-drug resistant cell lines and even to exproduced via a polyketide synthase system by ceed DOX activity sometimes by a factor of a series of reactions similar to that of fatty up to ten (for a survey, see WEISS,1992). acid biosynthesis (VANEK al., 1977). The et intermediate undergoes hydroxylations, methylations, glycosylations, and other modifications. Aklanonic acid and aklaviketone are
the earliest detectable intermediates of anthracycline biosynthesis (WAGNER et al., and 1988). More 1991; ECKARDT WAGNER, recently, the biosynthesis of anthracycline aglycones was studied in detail by a genomic analysis (STROHL al., 1989; BARTELet al., et 1990). The biosynthetic origin of many of the known anthracycline aglycones, such as Eand ppyrromycinones, E-, p, a-, and &,-rhodomycinones, -iso-rhodomycinones, daunomycinone, and adriamycinone could be ascribed to aklanonic acid as a single intermediate. The pigmented antibiotics are formed in the mycelium of microorganisms grown in shake flasks or in stirred aerated fermenters on media containing the usual organic nutrients and inorganic salts. For the biotechnical production of daunorubicin, the fermenters are operated usually for up to 10 days. Because anthracyclines are hazardous chemicals, due to their cardiotoxicity and/or mutagenicity, special care is needed when isolating and purifying these agents from the fermentation broth (UMEZAWA al., 1987). Recovery et and purification of the anthracyclines from large-scale fermentations is described elsewhere (WHITEand STROSHANE, 1984).
661
feeding of amino acids has been reported (TAKITA,1984). Peplomycin and liblomycin (A, B, C) (KURAMOCHI al., 1988; KURAet As a general feature, p-quinoid structures MOCHI-MOTEGI al., 1991; TAKAHASHI et et such as mitomycin C, streptonigrin, tetraceno- al., 1987c) are semisynthetic derivatives of mycin type antibiotics, angucyclines, and the bleomycins which contain a more space-filling representatives of the anthracycline family side chain. The aim of their synthesis was to (BUTLER and HAEY,1987; KONOet al., 1991; improve the stability against hydrolysis and to KASAI et al., 1991; BARKLEYet al., 1986; increase lipophilicity (OTSUKAet al., 1988; et FEIGand LIPPARD, 1994) can be reduced en- SEBTI and LAZO, 1988; NISHIMURA al., zymatically by cytochrome-dependent en- 1987; OHNO,1989). A general characteristic zymes to yield semiquinone radicals (Fig. 8). of these agents is that they form heavy metal These will combine with molecular oxygen to complexes with divalent cations such as form superoxide anion and hydroxyl ion radi- Cu2+, Co2+, Zn2+, Fe2+, and can even be cals. Thereafter, the free radical species react isolated as heavy metal (Cu) chelates from preferably with the deoxyribose moieties of the aqueous solution (Fig. 9). The mode of antitumor action of bleomycin DNA splitting single bonds between neighboring sugars. Although the anthracyclines has been subject to detailed investigations 1988; STUBBEand KOZARICH, could be considered as intercalating agents (MATSHURA, 1992). The antibiotic and and inhibitors of topoisomerase I1 (see 1987; STREKOWSKI, above) they are also potent producers of free its derivatves are commonly thought to exert radicals. The anthracycline semiquinones are their biological effects as metal-drug comsubsequently stabilized by deglycosylation of plexes which bind to the DNA. Thereafter, the C-7-0-linked sugar. As a consequence of the adduct causes oxidative strand cleavage, free radical formation, frameshift mutations and hence may be regarded as a low-molecuare induced. Thus the anthracyclines, bleomy- lar weight DNAse (HECHT,1994). The dicin (see below), mitomycin C, and similar an- saccharide moiety appears to be necessary to ticancer agents are hazardous chemicals enable penetration of the antibiotic through the cytoplasmic membrane barrier, while the (UMEZAWA al., 1987). et thiazolopeptide side chain is essential for the contact to the DNA (POVIRKand AUSTIN, 3.5.6 Bleomycin 1991). The cytotoxicity of bleomycin for some kind of tumor cells results from the reductive Bleomycin is a glycopeptide antbiotic pro- activation of dioxygen by metallobleomycins. duced by Streptomyces verticillatus as a com- Iron-I1 ions are able to transfer electrons to plex of not less than sixteen related antitumor molecular oxygen to form reactive and damantibiotics (TAKITA, 1984; MURADA,1988) aging intermediates of oxygen, such as super(Fig. 9). The main component bleomycin A2 oxide anion radical. Radical generation (02) was originally used clinically in the treatment is initiated by the binding of iron-I1 ions to of lung carcinoma, plate epithelium carcino- the paminoalanine-pyrimidine core of bleoma, Hodgkins lymphoma, glioma, skin carci- mycins. Interaction of Fe-I1 with molecular noma, and testis carcinoma (GIRIand WANG, oxygen generates the superoxide anion radi1989; SOLAIMAN, 1988; POVIRK and AUSTIN, cal to form Fe(II1)-bound bleomycin. The 1991; TAKITA al., 1989). Pulmonary toxici- formation of radicals probably occurs in close et ty is dose-limiting. More than 200 semisyn- proximity to the DNA, since there is intercalthetic derivatives have been obtained in order ative binding of the bisthiazole moiety of to improve activity and reduce toxic side ef- bleomycin to special minor groove DNA sefects (MURAOKA al., 1988; TAKAHASHI quences. It was also shown that the bleomyet et al., 1987a, b; TAKITA OGINO,1987). The cin-Fe(II)-02 complex splits double-stranded and naturally occurring bleomycins are distin- DNA specifically at the GC(5-3) and guishable by the amino side chain. Directed GT(5-3) sequences. It has been proposed biosynthesis of individual components by the that upon cleavage of the DNA the bleomy-
Formation of oxygen radicals by anthracyclines:
662
a
0 OMe
0 Me 0 Menoxymycin A R=NO(CHd2
Medermycin (R= N(CHd2

OH
d'
I I
'c,H,
Safnmycin A
Fig. 8. Drugs forming oxy-
gen radicals in the cells. a Suggested mode of radical formation by anthracycline drugs (see GOORMAGHTIGH and RUYSSCHAERT, 1984), b selected structures of drugs forming oxygen radicals (MYERSet al., 1986).
&:
\
Me
OH Atramycin H o
Hydromycin
663
Fig. 9. Structure of bleomycin and the bleomycin-Fe(II)-O2 complex (STUBBE and KOZARICH,1987; MATSUHARA, 1988).
cin-Fe( 111) complex dissociates from the target and is reduced to the (Fe-11) complex. Enzymatic modification of the antibiotic occurs by serum proteases which hydrolyze the amide structure (SEBTI and LAZO, 1988). Hence, inhibitors of proteases can be used to improve the efficacy of bleomycin. Moreover, bleomycin-Fe(I1)-0, and bleomycinFe( 111)-O2 complexes catalyze lipid peroxidation concomitantly with singlet oxygen evolution. Thus, oxidative damage of the cellular membranes may be one of the mechanisms determining the cytotoxic activity of this antibiotic (KIKUCHI and TETSUKA, 1992).
3.5.7 Macromolecular Antitumor Antibiotics and the Enedyine Family of Cytotoxic Drugs
The most recent discovery in the field of radical-forming, cytotoxic antitumor antibiotics concerned the extremely potent structural group of the enedyines. In general, these antibiotics are highly toxic for tumor cells, such as P388 leukemia and B16 melanoma. Enedyine chromophores are frequent constituents of macromolecular antitumor antbiotics, such as neocarzinostatin and C-1027 (HOFSTEADet al., 1992; ISHIDAet al., 1965; OTANI,1993). The enedyine subunits display antitumor activity even in the absence of the pertinent
664
apoprotein. Other members of this protein group are auromomycin (YAMASHITA al., et 1979), macromomycin (CHIMURAet al., 1968), kedarcidin (LEET, 1992; HOFSTEAD et al., 1992), largomycin (YAMAGUCHI al., et 1970; MURAMATSU al., 1991), maduropepet tin (HANADA al., 1991a), and actinoxanet thin (KHOKHLOV al., 1969), but the chemiet cal nature of their chromophores has not always been explored in detail. Similar to neocarzinostatin, madurapeptin is composed of a large peptide backbone and an aromatic chromophore which is responsible for the biological activity (HANADA al., 1991a, b). et The agent C-1027 from Streptomyces globisporus is the most recently discovered representative of the high molecular-weight antitumor antibiotics. It is composed of a 110 amino acid backbone and an unstable enedyine chromophore (OTANI, 1993) (Fig. 10a). A synthetic route to C-1027 was recently elaborated (IIDAet al., 1993). The apoprotein of C-1027 displays some amino peptidase activity which is inhibited by amastatin and bestatin (SAKATA al., 1992). et More recent research in the field of high molecular weight antitumor antibiotics has focussed on the nucleotide sequences of apoprotein genes (SAKATAet al., 1989, 1992). Advances in physicochemical analysis permitted the structural elucidation of the enedyine chromophore of neocarzinostatin from Streptomyces carzinostaticus (EDO et al., 1988). Several non-protein-bound representatives of the unique enedyine structure such as esperamicins (MAGNUS and BENNETT, 1989), calicheamicin (LEE, 1992), and dynemicins (KONISH al., 1991) have been isolated from et Actinomyces strains (Fig. 10a).
mechanism of neocarzinostatin action was shown to involve the formation of bisradicals and subsequent damage at C-5 of deoxyribose of thymidylate in the DNA. The neocarzinostatin chromophore binds to DNA in a two-step process. The first is an external binding and the second involves the intercalation of the naphthoate moiety between adjacent DNA base pairs in the minor groove and the electrostatic interaction of the amino sugar with the charged phosphate backbone of the DNA. Upon addition of thiol agents (R-SH) free radicals are produced which attack the DNA to cause double-stranded breaks (GOLDBERG,1991; HENSENS and GOLDBERG, 1989; DUMITRIU, 1996). The esperamicins (LAM et al., 1993) (Fig. 10a) as the main representatives of the non-protein group are composed of a bicyclic core (an enedyine, an allylic trisulfide, and an enone) which is bound to a trisaccharide and a substituted 2-deoxy-~-fucose. Esperamicin Al (Actinomadura verrucosospora) was produced on a large scale for clinical trials in cancer treatment (GOLDBERG al., 1989). Bioet synthetic studies on esperamicin Al (BENTLER et al., 1994) (Fig. lob) investigated the incorporation of the single and double I3C-labeled acetates ~-(methyl-I~C) methionine and NaS04. The CI5 bicyclic enedyine core is derived by head-to-tail condensation of seven acetate units. The S-methyl groups of the trisulfide, the thiosugar, and 0-methyl sugars are incorporated from S-adenosyl-methionine. A minor congener of esperamicin A], esperamicin P, was isolated from the fermentation broth of Actinomadura verrucospora (GOLDBERG al., 1989). It differs frorri eset peramicin Al in that it contains a methyl tetrasulfide moiety instead of a methyl trisulfide. Mode of action of the enedyines: The dynemicins (Micromonospora chersinu) lack the apoprotein constituent of the enNeocarzinostatin (NCS) was the first natu- edyine chromophore. Dynemycin A contains ral enedyine described which directly attacks the bicycl0[7.3.1]-1,5-diyn-3-eneand 1,4,6-trithe sugar moiety of DNA residues (GOLD- hydroxy-anthraquinone functions (Fig. 10a). BERG, 1986; GOLDBERG al., 1989). In Con- It exhibits a potent antibacterial and antituet trast to ionizing radiation or bleomycin which mor activity against a wide range of bacteria damage deoxyribose through reactive oxygen and cells, respectively. Satellite compounds, species, NCS produces DNA sugar lesions in such as L, M, N, 0, P, and Q have recently a sequence-specific manner at closely corre- been isolated as shunt metabolites in fermensponding sites of the two DNA strands. The tations of Micromonospora chersina and from
665
mutants of this strain (KONISHIet al., 1991; KAMEIet al., 1991; MIYOSHI-SAITOH al., et 1991). The calicheamicins (Fig. 10a) belong to a family of seven glycosylated compounds from Micromonospora echinaspora ssp. calichensis which demonstrate potent activity in vivo against the murine tumors P388 and B16 (LEE et al., 1989, 1992). Enedyines, such as esperimicin have previously been synthesized via several routes (LAM et al., 1993; LEE, 1992).
3.6 Inhibitors of Mitosis and the Microtubular System

The early discovery of the antimitotic activity of plant products such as colchicine (Fig. 11) led to some clinical trials for cancer treatment, but due to their high toxicity and relatively poor efficacy they have not found therapeutic application. The complex indole type vinca alkaloids (Catharanthus roseus) (GUNDAet al., 1994; HEINSTEIN and CHANG,1994; KINGSTON, 1994), such as vincristine and its derivatives (POITIERet al., 1994; JOEL, 1994), vinblastine, and the taxol type agents (from Taxus brevifolia; ATTA-UR-RACHMAN al., 1994; et NOBLE, 1990) are plant-derived antitumor drugs which prevent mitotic cell division (Fig. 11). While the vinca alkaloids inhibit tubulin polymerization, the latter (taxoids) appear as particularly promising due to their opposite effect of promoting the polymerization of tubulin. In folk medicine, leaf extracts of the subtropical plant Catharanthus roseus were known to be effective in the treatment of diabetes. Attempts to isolate the active hypoglycemic principle led, instead, to the discovery and isolation of cytotoxic vinblastine and vincristine. These complex indole alkaloids are now used in the clinical treatment of a variety of cancers (NOBLE,1990) and in therapeutic tumor cell synchronization (CAMPLEJOHN, 1980). A major problem in cancer treatment by vincristine is the development of resistance due to P glycoprotein-mediated drug efflux (HILL, 1986). 500 kg of Catharantus roseus (Vinca rosea) have to be extracted and the extract fractionated to obtain 1 g of vincristine. Hence, partial synthesis of vincristine, vinblastine and vindesin is done using vindolin. Taxol (paclitaxel) (Fig. 11) was discovered in 1964 during a large screening for antitumor drugs at the National Cancer Institute of the USA. It is an exciting new anticancer drug, exhibiting clinical activity in the treatment of ovary and breast cancer (KINGSTON, 1994; MARTYet al., 1994; ROTHENBERG,1993). Taxol was first isolated from the pacific yew tree (Taxus brevifolia). To overcome the problems of raw material supply,
3.5.8 Other Agents Forming Active Oxygen Radicals in the Cells

All normal cells have various defense systems against active oxygen species. However, several tumor cell types have lost a part of these defense mechanisms. Hence, substances generating active oxygen radicals such as quinoid structures have been suggested to exhibit selective cytotoxicity against such tumor cells (CURRAU al., 1993). In screening for et such antitumor agents Sfrepfomycessp. KBlO was found to produce the new menoxymycins A and B (related to medermycin) which actively generate superoxide radicals in N18RE-105 cell lysates (HAYAKAWA al., 1994) et (Fig. 8b). Numerous p-quinoid structures such as dioxamycin (SAWAet al., 1991), hydramycin (HANADA al., 1991b), the glycoet sylated saptomycins (ABE et al., 1993a), tetracenomycins, and angucycline type aromatic polycycles, such as saframycins (KANEDAet al., 1987) and atramycins (ABE et al., 1991) possibly owe their cytotoxic and cytostatic effects to the formation of active oxygen radicals during aerobic metabolism (Fig. 8b). Saframycin A was also reported to bind covalently to duplex DNA (KANEDA al., 1987). et In addition, various polycyclic aromatic compounds with quinone structures or an epoxy side chain display cytotoxic activities. This feature was also ascribed to the formation of free radicals or other reactive structures. Representatives of this kind of agents are the saet purimycins (Streptomyces sp.) (UOSAKI al., 1991).
666
OMe
+
OH
OMe M
on a
OMe
Callcheamlcln a1Br
C-1027 Chromophore
NH2
SSSMe
"YL+d OMe
Esperamicin A,
y p
OH
OMe
on
R = CH(Me)2 Al b
AlC
M MeO
Dynemicin A
R=Et
R=Me
.&me
CH2
Fig. 10. Structures of enedyine drugs. a Chromophore structures of esperamicin, calicheamicin, dynemicin, and C-1027, b enrichment pattern of esperamicin A1 supplemented with 13C-labeledsingle and double-labeled acetate. 13C labeling is indicated by 0, W, and WO.
667
0
a) CH3COOH
Ri
CH,O,CNH
0
OH
rz,
I
b) C!H3COOH
CM,S,
CH,O,CNH
Fig. lob
paclitaxel and docetaxel (taxotere) (Fig. 11) were developed (POITIER al., 1994; MARTY et et al., 1994). In particular, the semisynthesis from the related structures of the baccatins appear to be a feasible way of taxane production (HEINSTEIN and CHANG,1994; KINGSTON, 1994). Baccatins are produced by the leaves of the European yew and can be chemically transformed to taxotere (docetaxel) (POITIER al., 1994; JOEL,1994). Recently, et an endophytic hyphomyces fungus of Taxus
brevifoliu was reported to form low amounts of taxol, but so far a biotechnical procedure for the fermentative production of taxol has still not been developed (STIERLE al., et 1994). Taxoids are particularly promising antitumor agents due to their novel cancerostatic mechanism of action (ROTHENBERG, 1993). These compounds interfere with essential cellular processes such as mitosis, cellular motility, transport, and maintenance of cellular structures. The mechanism of action of
R1
OH OH
0
668
"q
Me HCOCH, Me0
.
CH,
OMe
Colchicin
R1 R2
R3
Vinblaslin CH3 OCH3 C W H 3 Wncrlstln CHO OCH3 COCH3 CI Vindesin CH3 NH2 H
CH,
OCH,
Maytansine R= COCH(CH3)N(CH3)COCH3 Ansamitocin P-3 R= COCH(CH3)*
Paclitaxel (TaxolR)
Wo
Fig. 1 . Inhibitors 1 of the microtubular system.
taxanes is to promote the formaton of extremely stable non-functional microtubular aggregates. The cells are thus prevented from entering the G, or M phase of the cell cycle. Both paclitaxel and docetaxel are more PO-
tent against proliferating cells than non-proliferating cells in v i m (ROTHENBERG, 1993). Myelosuppression is a dose-limiting side effect. Current research is centered now around improving drug extraction, the semisynthetic
669
production of the parent taxane structures 1992). To overcome myelosuppression as one (PAQUE-ITE, 1993; GUERITTE-VOEGELEIN of the limiting factors in cancer chemotherapy et al., 1994), the enhancement of water solubili- by anthracyclines, a search was initiated for ty, and the identification of taxane analogs immunomodulatory stimulators of leukocyte which are active against resistant cancer cells and platelet formation (KAWATSUet al., (ROTHENBERG, 1993; VYAS, 1993). Due to 1993,1994). As a result of this screening, conthe impressive clinical activity of paclitaxel agenin was detected in cultures of Streptoand docetaxel, the taxanes will be subject to myces roseosporus. It binds exclusively to T preclinical and clinical development in the fu- cells activated by concanavalin A or cytokines et al., 1994). and enhances both T cell proliferation and ture (GUERI-ITE-VOEGELEIN Moreover, structure-activity relationships lymphokine formation. Due to the induction (SAR) of drug-tubulin interactions are under of cytokines, the production of leukocytes et al., and platelets is subsequently stimulated. investigation (GUERI-ITE-VOEGELEIN 1994). Chemical modifications such as C-2 deoxygenation result in a total loss of tubulin function. A relatively small number of micro- 3.8 Cytotoxic Compounds with bial drugs inhibits mitosis in tumor cells. Ex- Poorly Characterized Mode of amples are rhizoxin (Rhizopus chinensis), a macrolide antibiotic and tubulin inhibitor of Activity Aspergillus nidulans (KATOet al., 1991), anThere are continuously references in the samitocin from Acfinomyces sp. (similar to maytansine from plant Maytanus sp.) literature to new cytotoxic structures (see, (Fig. ll), and curacin A from the marine cya- e.g., reports on structures such as russuphe1993), heptelidic acid (KInobacterium Lyngbya mujuscula (PIRRUNG lins (TAKAHASHI, and NAUHANS, 1994). Rhizoxin exhibits po- WASHIMA et al., 1994), and stubomycin tent antimitotic activity by binding to Ptubu- (KOMIYAMA al., 1983). Cytotoxicity apet lin, and is active against most eukaryotic cells. pears to be a very frequent property of new It shares the same binding site with maytan- natural products. In a personal literature sine and ansamitocin P-3 (Fig. 11) on porcine, compilation, out of 4,000 microbial drugs, 600 brain, and fungal tubulin which is different were indicated to suppress cell growth. For from that of colchicine and vinblastine (IWAS- the majority of these agents the mechanism of action has not been investigated. In the past, AKI, 1989). much interest has been focussed on new metabolites which might inhibit tumor cell lines that are intrinsically resistant to the known 3.7 Reduction of the Side Effects set of anticancer drugs or that have acquired of Highly Toxic Anticancer Agents multi-drug resistance. In the course of screening using adriamycin-resistant HL-60 cells, Highly effective cytotoxic anticancer homooligomycins A and B (Sfrepfomycesbotagents, such as the anthracyclines, vincristine, tropensis) were detected as a result of their and bleomycin, usually have severe side ef- particular effect on colon-26 carcinoma (YAfects such as cardiotoxicity, myelosuppression MAZAKI et al., 1992). A further focus of interest have been the (FURKet al., 1989), release of interleukin 2, and others which are dosage limiting. Prein- numerous neoplasm inhibitors which are seduction of metallothein synthesis, e.g., by bis- lectively toxic against a special type of tumor muth salt was reported to exert a protective cell. For instance, the homooligomycins A effect against various anticancer agents (e.g., and B (Sfreptomyces boffropensis) were debleomycin, cis-diammine-dichloro-platinum) tected due to their particular effect against (NAGAMURA al., 1993). et et colon-26 carcinoma (MAGAE al., 1993). In Iron chelators and antioxidants have also a screening program for immunomodulators, been studied with regard to brain protection the melastins (Sfrepfomycessp.) (MAGAEet al., against neoplasm inhibitors (GU~TERIDGE, 1993) were similarly found to selectively
670
inhibit the growth of leukemia cells and the lipopolysaccharide-inducedblastogenesis of T cells.
4.1 Potentiators of Cytotoxic Antitumor Agents
DNA repair mechanisms serve as a useful target for modulating the cytotoxic and chemotherapeutic effects of those agents. Their mechanism of action causes the induction of DNA damage. Poly(ADP-ribose) polymerase responds to DNA breaks by cleaving the substrate NAD and using the resultant ADPribose moieties to synthesize homopolymers Much interest has been centered on new of ADP-ribose. Inhibitors of this enzyme such anticancer agents which exert their antitumor as benzamide derivatives and benadrostin activity via a hitherto unknown interaction (Fig. 12) (YOSHIDAet al., 1988) prevent with cellular regulation and cellular transduc- DNA repair and potentiate the tumoricidal tion pathways. Among these new agents, inhib- effects of DNA strand-breaking agents such et itors of tumor cell resistance to anticancer as bleomycin (BERGER al., 1987; YOSHIDA agents, tumor metastasis, and angiogenesis of et al., 1988; GAAL and PEARSON, 1986; Sutumors appear promising. ZUKI et al., 1990). The simple compound, 2Otherwise, the discovery of the oncogenes methyl-4[3H]-quinazoline, was discovered in and their role in malignant cell growth has Bacillus cereus cultures and found to be an inopened new horizons for target-directed hibitor of poly(ADP-ribose) synthetase from et screening of new anticancer agents. This de- calf thymus (YOSHIDA al., 1988). In addivelopment was initated by the exploration of tion, calcium channel antagonists such as particular oncogene-encoded functions such verapamil and fendiline are known to potenas those of erb, sis, and ras in tumor cells. On- tiate the activity and toxicity of mitomycin C cogenes are modified forms of normal genes in mammalian and bacterial cells (SCHEIDet (proto-oncogenes) which have been altered al., 1991). The mechanism of this effect has either by chemical carcinogenesis or retrovi- not yet been explored in detail; however, it ruses in such a manner that they escape the seems reasonable to suggest that P glycopronormal pattern of control (FERDINAND, tein-mediated efflux of the drug is prevented. 1989). These alterations may result in the Similarly, a series of other agents were reoverexpression of the genes, the loss of down- ported to increase the cytotoxic effect of neoregulation of the genes, and the formation of plasm inhibitors. A major obstacle in cancer altered proteins. Proteins encoded by these chemotherapy is the appearance of resistant genes, such as the epidermal growth factor re- tumor cells which are non-treatable by cytoceptor (EGF receptor) and membrane-asso- static drugs such as vincristine and doxorubiciated Rus-proteins are abnormal derivatives cin. Several reasons for this are known (HILL, of the normal proteins involved in cellular 1986; BIEDLERet al., 1993). Overexpression signaling and cell cycle regulation. The pres- of metallothein in cancer cell lines, e.g., conently available information on cellular regula- fers resistance to anticancer drugs such as plation appears to be the tip of the iceberg. tinum complexes (KELLEYet al., 1988). DeRecent additions to our knowledge were pro- toxification of cytotoxic drugs may also occur vided by the discovery of the cyclins, the tu- by glucuronidation (BURCHELL al., 1991). et mor suppressor genes and their proteins. Other cancer cell lines become resistant to a These findings will doubtlessly promote the series of anticancer drugs such as taxol, coldiscovery of new agents which interfere with chicine, doxorubicin, actinomycin, and vinthese targets and module their activity. The cristine (HILL, 1986). This so-called multifollowing survey summarizes the results of re- drug resistance (MDR) is frequently assocent screening approaches to these non-classi- ciated with increased drug efflux and decreased accumulation within the cells. The cal anticancer drugs.
4 Non-Classical Approaches to Antitumor Drugs
b
OMe Me
a
? !
Me Me On Me Me Me
M ~ O on
on
Smbomycin
FD895
HnbXlUNblglIlA
cocn.
Rubiginon A,
Me
672
I4 Biotechnical Drugs as Antitumor Agents
phenomenon is due to membrane proteins, such as P 170 glycoprotein, which is capable of conveying drugs out of the cells. This glycoprotein was detected in the mouse macrophage-like cell line J 774.2 and in other cells, and its role as mediator of the active efflux of cytotoxic drugs was studied (GUNICKE and HOFMANN, 1992; BORREL et al., 1994; REICHLE al., 1991). Mobile ionophoric caret riers, such as valinomycin, lasalocid, monensin, calcimycin, and others inhibit the efflux of anthracyclines, whereas channel-forming ionophores (e.g., gramicidin A) do not. Cyclosporine, a calcium chelator, also interfered with /3-glycoprotein-mediated efflux of these drugs (BORRELet al., 1994; DIETEL,1991). Directed screening for drugs which inhibit the growth of multiresistant tumor cells (MDR) or which could restore and potentiate the antitumor activity of known agents thus appeared to yield promising substances. Examples of these agents (Fig. 12) are the hatomarubigins (Streptomyces sp., inhibitors of multi-drug resistant tumor cell lines) (HAYAKAWA et al., 1991), BE 137932~ (Streptornyces sp., inhibits the growth of doxorubicin-resistant cancer cells) (KEFRI et al., 1991), the piperaficins (Streptoverticillium aspergilloides, potentiate vincristine activity against resistant murine P388 cell lines) (KAMEI et al., 1990), bestatin (Streptoverticillium olivoreticuli, an immunostimulator, increases the in vivo vincristine cytotoxicity against colorectal K562 carcinoma cells) (UMEZAWA, 1989), sekothrixid (an antitumor drug) (KIM et al., 1991), rubiginon (Sfrepfomycesgriseorubigenosus, restores colchicine sensitvity to resistant tumor cells) (OKAet al., 1990,1991), resorthiomycin (potentiates the antitumor activity of vincristine) (TAKARA al., 1990), et laidlomycin (a polyether antibiotic from Streptomyces sp., potentiates anticancer drug activity against MDR carcinoma KB-4 cells) (KAWADA al., 1992b), bisucaberin (Alteroet monus haloplunctis, sensitizes tumor cells to cytolysis mediated by macrophages) (TAKAHASHI et al., 1987a), kazusamycin (see et Fig. 18) (Sfrepfomycessp.) (YOSHIDA al., 1987), lactoquinomycin (Sfrepfomyces funashiriensis, active against resistant L 51178~ lymphoblastoma) (TANAKA al., 1989), and et semisynthetic derivatives of staurosporin (re-
version of the multi-drug resistance of cancer cells) (WAKUSAWA al., 1993), 5-N-actylaret deemin (Fig. 12) (Aspergillus fischeri, reversion of multi-drug resistance in tumor cells) (HOCHLOWSKI al., 1993) and BE-12406 A et and B (Sfreptomyces sp., inhibition of vincristine- and doxorubicin-resistant P388 murine leukemia cells) (KOGIRIet al., 1991). A particular feature of the potentation of antitumor drug activity is the potentiating effect of inhibitors of bleomycin hydrolases (leupeptin, pepstatin) on bleomycin activity (NISHIMURA et al., 1987). Some tumor cells resistant to inhibitors of topoisomerase I1 display a type of MDR that differs from P glycoprotein-associated MDR and is restricted to drugs that inhibit topoisomerase I1 by stabilizing cleavable DNA-protein complexes (BECK et al., 1993).
4.2 Inhibitors of Glutathione S Transferase as Enhancers of Antitumor Activity of Drugs

Glutathione conjugate formation may constitute an important drug detoxification and bioactivation mechanism for several classes of mutagenic and carcinogenic compounds (SATO,1989). For instance, the metabolism of nephrotoxic and nephrocarcinogenic haloalkenes involves glutathione S conjugate formation. Glutathione was suggested also to play a role in the toxicity or action of bleomycin, cyclophosphamide, and neocarcinostatin (ANDERS,1991; SATO, 1989). An increased glutathione S transferase activity may create resistance to some anticancer drugs such as alkylating agents, doxorubicin, and cis-diammine-dichloro-platinum. Consequently, inhibitors of glutathione S transferase may improve the activity of antitumor drugs. Moreover, they could also act as antiinflammatory and antiallergic agents due to their effect on the prostaglandin and eicosanoid metabolism. In a worldwide screening approach several new inhibitors were discovered, such as benastatins C and D (Streptomyces sp.) (AOYAM A et al., 1993a), cysfluoretins (Streptomyces sp.) (AOYAMAet al., 1993b), TA-3037 A (KOMAGATA al., 1992a), rishirilide B (KOet
673
magata et al., 1992b), and bequinostatins C and D (YAMAZAKI al., 1993) (Fig. 13). et
4.3 Antimetastasis Drugs and Inhibitors of Angiogenesis

Prevention of metastasis of cancer cells and angiogenesis is a major aim of cancer chemotherapy. Angiogenesis is a tissue differentiation process which connects a growing tumor with the blood vessels. Recently, phospholipase D was proposed as a target for tumor invasion inhibitors. Drugs, such as Sch 49210, Sch 53514 and Sch 53517 from the fungus Nattrasia mangifera displayed potent activity in the antitumor invasion chamber assay (CHU, 1994). Among the rnetalloproteinases involved in tumor invasion, angiogenesis, and rheumatoid arthritis, the type N enzymes may play a particular role in the degradation of basement membranes. Searching for inhibitors of this enzyme, such as the matlystatins from Actinomadura atramentaria, has been proposed as a promising approach towards new antitumor drugs (HARUYAMA al., et 1994). Antimetastasis activity was reported for U-77863 (TROLLet al., 1987) and U-77864 (methylphenyl-propenyl-carboxamides) from Streptomyces griseoluteus (HARPER and WELCH, 1992). These compounds were shown to inhibit the growth of tumor xenograft models (Fig. 14). Nitrogen-containing pseudo-sugars such as siastatin B, nagstatin,
nojirimycins and their chemically derived derivatives have recently been reported to prevent metastasis by the inhibition of p-glucuronidase, heparanase, and P-D-mannosidase of tumor cells (SATOHet al., 1996 TSURUOKA et al., 1996; KAWASE al., 1996; TATSUet DA et al., 1996). Using the chicken embryo chorioallantoic membrane assay system, a series of angiogenesis inhibitors were disclosed, such as herbimycin A (Streptomyces sp.) (YAMASHITA et al., 1989; SAKAI et al., 1989), irsogladin (SATO et al., 1993), erbstatin (Streptomyces sp.) (OIKAWA al., 1993), TAN-1120 (an anet et thracycline, Streptomyces sp.) (NOZAKI al., 1993), TAN-1323 (Streptomyces sp. S-45628) (MUROI,1990), analogs of siastatin B, WF16775 A,, A2 (derivatives of pyridazine from Chaetosbolisia erysiophoides) (OTSUKA al., et 1992), 15-deoxyspergualin (NISHIKAWA al., et 1991a,b), and staurosporin (OIKAWA al., et 1992). These drugs are expected to prevent neovascularization and thus seem to be promising not only in the prevention of metastasis but also in the treatment of retinopathy and rheumatoid arthritis (OTSUKA al., 1992). et
4.4 Antitumor Effects of Immunomodulators

In the healthy human, the immune system recognizes both microbes and malignant cells and destroys them. The outbreak of cancer
6 H 0 OH Benadrostin
Cysfluoretin
Rishirilide
COOH
H
TDD
H OH0
OH
Bequinostatin
Fig. 13. Inhibitors of glutathione S transferase.
674

0
on
NHCHO
6H
Erbstatin
U-77863
Me
TAN-I1
Me
NMC
,+OH 0
i i
Matlystatin
lrsogladin
WF-I 6775 A1
I I Me -ClKONH-CH-COD L
OH
Me
D
NH- CHCOOH
I
(CH,),CONHCHCONHCH,COOH FK-606
NH, Forphenicin
NH, Forphenicinol
I '
Spemidin (=OH) Deoxyspetmidin(R=H)
Conagenin
Fig. 14. Selected structures of antimetastatic inhibitors and anti-angiogenetic and immunomodulatory drugs.
may be promoted by insufficient immunological defense and poor recognition of abnormal cells. Hence, immunostimulating and immunoregulatory agents have been considered as potential anticancer therapeutic agents. The immunostimulatory and antineoplastic properties of bacterial wall preparations, especially from Mycobacteriurn spp. were discovered already in the 1970s. High molecular weight
glucans secreted into the medium, such as shizophyllan and lentinan, belong to a group of immunostimulatory compounds which exhibit antitumor activity. They are derived from higher fungi and mushrooms. Their antitumor activity has been attributed to the stimulatory effect on the immune sysem (LAATSCH, 1992). Subsequently, the chemical structures of other naturally occurring immunomodula-
675
HerbimycinA
Herbimycin B
OH 0
Me
&NH2 \
N OH"
Me
M .
Fig. 14
U-7786
W-16775
tors were elucidated. These compounds may provide the pharmaceutical scientist with an armamentarium that can be used to mount a rational treatment of immunopathologies including cancer. For several immunomodulators of microbial origin, protective effects against experimental tumors have been established in mice (KLEGEMANN, 1993) (Fig. 14). Best atin (Strepto verticillium olivoreticuli) which was discovered by H. UMEZAWA'S group in Japan (ABE et al., 1985) enhances the activity of antitumor drugs and is used in clinical applications. The drug was initially found to be an inhibitor of leucine aminopeptidase. Immunostimulatory and antitumor effects have been reported also for some other microbial protease inhibitors (BILLINGS, 1993). The compound FK-156 (Streptomyces et olivaceogriseus) (IZUMI al., 1983) and conagenin (Streptomyces roseosporus) (YAMASHITA et al., 1993; KAWATSU al., 1993) are et additional examples of microbial metabolites which may exert anticancer activity due to their stimulatory effect on cytokine production by T cells and macrophage activity. Forphenicin (Streptomyces sp.) and its biotransformation product forphenicinol inhibit chicken alkaline phosphatase. Forphenicinol induces yinterferon production in mice which were sensitized by the BCG vaccine, and enhances macrophage activity. It also displays an antitumor effect on MetA fibrosarcoma and adenocarcinoma (OKURAet al., 1986). For a series of immunosuppressing
agents such as spergualin (Bacillus laterosporus) and its semisynthetic 15-deoxy derivative, antitumor activities were demonstrated, e.g., against the murine leukemia P 388 (NISHIKAWA et al., 1991a, b). Retardation of malignant cell growth was ascribed to the cytotoxic and cytostatic effects of these drugs, but the mode of action of these and other immunosuppressants (see, e.g., cyclosporine A and FK-506) has not been explored in detail.
4.5 Inhibitors of the Cellular Mitogenic Signal Transduction Pathway

Fig. 15 shows the role of protein kinases, phosphoprotein phosphatases (NEER and CHAPHAM, 1988), phospholipases, and farnesyltransferases in the transduction of an extracellular signal, such as a growth hormone, into an intracellular response of DNA replication and mitogenesis. In oncogene-transformed cells these enzymes are often involved in neoplasmic cell growth. Much effort in the search for new anticancer agents is now focussed on specifically acting inhibitors of the cell cycle such as acetophthalidin from a marine Penicillium strain which arrests the mammalian cell cycle in the G2/M phase (CUI et al., 1996a, b). So far, more than 40 distinguishable oncogenes have been identified. They can be clas-
676
sified into four main groups according to their function:
4.5.1 Inhibitors of Protein Kinases and Protein Phosphatases
(1) oncogenes encoding tyrosine protein kinases such as src (similar to insulin receptor and catalytic chain of mammalian CAMP-dependent protein kinase from Rous sarcoma virus), erbB, fgr, and fes (see also Fig. l), (2) oncogenes encoding proteins involved in the metabolic regulation of GTP-binding proteins as essential parts of G proteindependent membrane receptors (K-Ras, H-Ras, N-Ras), (3) oncogenes encoding proteins which act at the level of gene regulation as transcription factors such as Myc and Myb, and (4) oncogenes encoding growth factor-like proteins such as Sis (similar to plateletderived growth factor).
Protein kinases are ubiquitous regulatory proteins in microbial, plant, and animal cells which exert their activity by the ATP-dependent phosphorylation of serine, threonine, and tyrosine residues (Fig. 16). The phosphorylated proteins display an altered function (activated or inactivated) (BASU and LAZO, 1994; LEVITZKI, 1994). Phosphorylation and dephosphorylation of proteins can result in the intracellular amplification of a mitotic signal, generated by interaction of a growth hormone with a membrane receptor. The protein kinases share at least some common features in their secondary and tertiary structures 1989; TOWBRIDGE, 1991). Myosin (TAYLOR, light chain kinase, a cytoplasmic enzyme, has a Ca2+kalmodulin-binding domain carboxy To escape the normal pattern of control terminal to the catalytic core, and binding of proto-oncogenes have to be transformed ligands activates the kinase. Membrane-assointo oncogenes either by structural altera- ciated protein kinase C is activated by Ca2+, tions, due to chemical carcinogenesis (GUEN- diacylglycerol or phorbol ester type comGRICH,1988) or by integration into retrovi- pounds, and phospholipids, and the recogniruses and reintegration into the host genome tion sites for these ligands lie amino terminal (HART and TURTURRO,1988 BERTRAM, to the catalytic core. Tyrosine-protein kinase activity was first detected as a characteristic 1990 WEINBERG, 1985).
Fig. 15. Key steps in the cellular mitotic signal transduction pathway.
677
of the transforming protein from Rous sarcoma virus, pp60. Removal of a phosphorylation site of this protein converts the protooncoprotein into a transforming protein. Following myristylation (farnesylation) pp60- moves to the plasma membrane. Receptors of growth factors, such as the epidermal growth factor (EGF) receptor, span the membrane via a single membrane-spanning element. Binding of EGF to the outer domain activates the cytoplasmic tyrosine protein kinase activity. Another prototype of a tyrosineprotein kinase transmembrane protein is CD45 (T200or leukocyte common antigen) (TROWBRIDGE, 1991). Another type of protein kinase is activated by cyclic 3,5-adenosine monophosphate (CAMP). The activating ligand (CAMP)binds to a distinct regulatory subunit thereby inducing conformational changes that provoke dissociation of the holoenzyme (TAYLOR, 1989). Reversible phosphorylation/dephosphorylation of proteins is involved in many cellular activities (KAHN and GRAF,1986). For example, the M phase promoting factor (MPF) in mammalian cells is composed of a catalytical ~ 3 4 protein~ which has protein kinase ac~ ~ tivity and a regulatory protein (cyclin B). The activity of MPF is regulated by reversible phosphorylation/dephosphorylation.Recently, tautomycin (Streptomyces sp.) was found to inhibit dephosphorylation of MPF thereby preventing mitosis (MAGAE et al., 1988, 1992). In the search for inhibitors of protein kinase C, the morphogenic effect of phorbol esters on mammalian cells provided a useful screening feature (TAKAHASHI al., 1987, et 1989). Phorbol esters and indolactams (teleocidin and blastmycetin from Streptoverticilium sp.) (HAGIWARA al., 1988) are known as et tumor promoters. They stimulate growth of tumors but are not carcinogenic. The discovery of tumor promoters from Euphorbiaceae, such as phorbol esters contributed to a better understanding of the cellular mechanism of growth regulation. In addition to tumor promotion, other processes such as inflammation, mitogenesis of lymphocytes, platelet activation, etc. were affected by structural analogs of the second messenger diacylglycerol.
The phorbol esters and similar structures (Fig. 16) activate protein kinase C. This effect has been visualized as bleb formation on the cell surface of K562 chronic myeloic leukemia cells which is induced in the presence of phorbol dibutyrate. Bleb induction can be prevented by specific inhibitors of protein kinase C and thus a practicable high-throughput screening assay was developed (OSADA al., et 1988). The same bleb formation assay has also been used to search for inhibitors of protein phosphatases. These enzymes provoke the same type of stimulaton of bleb formation which is a characteristic of the phorbol esters (MAGAE al., 1988, 1992; BLUMBERG al., et et 1989). On the other hand, differences in the effect on tetrazolium blue reduction of HL-60 cells of phorbol esters and tautomycin suggested that the latter has a different mode of action (MAGAE al., 1992). For the sake of et completeness, the bryostatins (from marine Bryozoa) should be mentioned since they are similar tumor promotors (Fig. 16). They are antitumor and antileukemic compounds isolated from marine animal organisms as a type of cyclic polyether which also mimics the structure of 1,2-diacylglycerol (RAMSDELL and PETTIT, 1986; WOLF and BAGGIOLINI, 1988; TAKAHASHI al., 1987b). Similarly, et novel inhibitor structures of protein kinase C and phosphoprotein phosphatases were discovered, such as staurosporin, UCN-1, UCN2 (TAKASHI al., 1989), tautomycin (Streptoet myces spiroverticillatus) (MAGAE et al., 1992), calyculin, microcystin (Fig. 16), and okadaic acid (from a marine organism) (SuGANUMA et al., 1988). Calyculin A affects phosphoprotein phosphatase 1 (PP1) 30 to 250fold more than okadaic acid, and also inhibits protein phosphatase 2A (PP2A). Tautomycin inhibits several protein phosphatases, even those of smooth muscles. No activity was found on myosin light chain kinase or protein kinase C. Okadaic acid, a polyetherlike shellfish toxin, inhibits serine-threonine specific protein phosphatases, particularly protein phosphatase 2A. Moreover, isopalimurin, a mild protein phosphatase inhibitor, has recently been isolated from a sponge (MURRAY al., 1993) (Fig. 16). et Presently, much interest is focussed on phosphotyrosyl protein phosphatases which
618

0
II
Phorbol 12,13dibutyraI
Teieocidin 84
BRYOSTATIN 1
&
Cocn, QCQ
I 2
Epiderstatin
isoflavonoids
Dephostatin
on
Me
0
O H Q H 0
lsopalimurin
BE33372M
4 Non -Classical Approaches to Antitumor Drugs
679
4 Fig. 16. Selected structures of inhibitors of proteinkinase C, tyrosine protein kinase, and protein phosphatases.
product) (NAKANO al., 1987; CAI et al., et 1996). The antitumor antibiotic calphostin (UCN1028) (Fig. 16) consists of five components (A, B, C, D, I) (IIDAetal., 1989). It is proare a particular family of transmembrane en- duced by the fungus Cladosporium cladosporzymes (LAN et al., 1989; KRUEGER al., oides and strongly inhibits protein kinase C. et et 1990; FISCHER al., 1991). Another inhibitory structure is balanol (azepinostatin) (Fig. 16) isolated from Verficillium balanoides and Fusarium merismoides (BOR0s et al., 1994; OHSHIMA al., 1994). In adet 4.5.1.1 Inhibitors of Protein dition nucleoside antibiotics, such as sangivaKinase C mycin (Fig. 16), and membrane-active compounds, such as polymyxin B, were reported Members of the indolocarbazole family of to inhibit protein kinases including protein antibiotics such as staurosporin (Sfreptomyces kinase C (LOOMISand BELL, 1988; ALLsp.), UCN-01, UCN-02 (Sfrepfomyces sp.) GAIER et al., 1986). (AKINAGA al., 1993), K-252a (Nocardiopet sis sp.) (NAKANISHI al., 1986), RK-286C et (OSADA et al., 1990), KT 6006 and 4.5.1.2 Inhibitors of Other Protein CGP41251, are potent inhibitors of protein Kinases kinase C (Fig. 16). These unique structures The pamamycin-type agents MS-282a and are produced by actinomycetes; however, staurosporin-like compounds have recently MS-282b, are macrodiolide-type inhibitors of been discovered in extracts from marine tuni- calmodulin-activated myosin light chain kincates. Derivatives of staurosporin such as ase from Strepfomyces fauricus ATCC 27470 1994). Other protein kinases MLR-52, display immunosuppressive activity. (NAKANISHI, This effect is comparable to that of other im- are not inhibited by these agents. The interacmunosuppressors, such as cyclosporine A and tion of MS-282 compounds with the enzyme FK-506, which affect protein phosphoryla- occurs at the calmodulin-binding site. Calmotions and phosphoprotein dephosphoryla- dulin-dependent nucleotide phosphodiesterase is also affected. The compound Sch45752 tions (MCALPINE al., 1994). et Moreover, indolocarbazoles display re- (Fig. 16) from the fungus SCF-125 is an inhibmarkable antimicrobial activities against bac- itor of various protein kinases (IIDA et al., teria and fungi including Candida albicans 1989); however, it also inhibits calmodulinand Botryfis cinerea, but no correlation was sensitive cyclic nucleotide phosphodiesterfound between protein kinase inhibitory po- ase. tencies and inhibition of Strepfomyces sporulation or growth inhibition (SANCELME al., et 1994). The indolocarbazole compounds dis- 4.5.1.3 Inhibitors of Tyrosine play antitumor activity against human and Protein Kinases murine tumor cell lines both in vitro and in vivo (AKINAGA al., 1993; TAKAHASHI et et Erbstatin (Fig. 14), which was isolated from al., 1987b; WOLF et al., 1988). This type of the culture broth of Strepfomyces sp. MH 435structure (SANCELME al., 1994) inhibits UF3, is a potent inhibitor of the EGF assoet either protein kinase C (see above) or topo- ciated protein tyrosine kinase (IMOTOet al., isomerase I (rebeccamycin, AT 2433, and der- 1987; SONODA al., 1989). It specifically inet ivatives) (OSADA al., 1990 KANEKO al., hibits the autophosphorylation of EGF recepet et 1990). AT 2433 has an inhibitory effect on tor but does not inhibit either cyclic AMP-deprotein kinase C. Staurosporin and its analogs pendent protein kinases or protein kinase C also inhibit the activity of the Rous sarcoma (IMOTOet al., 1987; UMEZAWA al., 1986). et virus-transformed protein p60 (scr oncogene Also, synthetic peptide sequences were found
680
to be good substrates for the EGF tyrosine kinase and have been used in kinetic studies. According to these investigations, erbstatin competes with the peptide substrate but not with ATP (UMEZAWA al., 1986). 4',7,8-triet hydroxyisoflavone, 3 ' ,4 ',7-trihydroxyisoflavone, 8-chloro-3',4 ',5,7-tetrahydroxyisoflavone, and orobol were isolated from the culture broth of Sfrepfomyces sp. OH-1049 are isoflavones (Fig. 17) which also inhibit EGF receptor tyrosine-protein kinase. This effect coincided with their increasing effect on the life span of tumor-bearing mice (S180, P388) (KOMIYAMA al., 1989; OGAWARA al., et et 1989a). This compounds do compete with ATP. The interaction of flavonoids with mammalian protein kinases was later found to be non-specific (END, 1987). Although they occur in microbial cultures, isoflavone glycosides such as genistein and daidzin appear to stem from the plant-derived nutrients. Microbial biotransformation results in the formation of substituted or even glycosylated products (ANGANWUTAKUal., 1992). et More recently discovered structures with inhibitory activity against EGF receptor mediated protein-tyrosine kinase are, e.g., dephostatin (Sfreptomyces sp.) (KAKEYA al., et 1993), BE-13793C (TANAKAet al., 1992), et BE-23372M (Rhizoctoniu sp., TANAKA al., 1994), tyrphostins (synthetic) (Fig. 16), and pquinoid structures like paeciloquinones A-F (FREDENHAGEN al., 1995). Epiderstatin (a et substituted 2,6-piperidine-dione) (Fig. 16) was isolated from Sfreptomyces pulveruceus ssp. epidersfugenes (SONODAet al., 1989). The compound inhibits incorporation of [3H]thymidine into quiescent cells stimulated by EGF and also reverts the morphology of "src-transformed cells to normal. Contact inhibition of growth by neighboring cells is an essential characteristic of normal cells. However, tumor cells lack this behavior. Herbimycin A (Fig. 14) (Sfrepfomyces sp.) is a benzoquinoid ansamacrolide which was shown to prevent the interaction of v-src oncogene-expressed cells due to the inhibition of pp60""" tyrosyl protein kinase and phosphatidylinositol kinase (IWAI al., 1980; et SUZUKAKE-TSUCHIYA 1989). This antiet al., biotic was originally screened for its herbicidal activity.
4.5.2 Inhibitors Affecting the Metabolism of Phosphoinositols

Phosphatidylinositol turnover appears to be correlated with cell transformation by some types of oncogenes (rus) and also with cellular responses to growth factors (EGF, PDGF) (FLEISCHMAN al., 1986; JACHOWSet KI et al., 1986; BERRIDGE and IRVINE, 1984, 1989). The signal generation induced by the enzymatic breakdown of phosphatidylinositol-4,5-diphosphate involves phospholipase C. This key enzyme is capable of forming two second messenger molecules, inositol-1,4,5triphosphate and 1,2-diacylglycerol. Inositol1,4,5-triphosphate mobilizes the intracellular calcium pool and 1,2-diacylglycerol activates protein kinase C (Fig. 15). Consequently, phospholipase C, the phosphatidylinositol turnover, the phosphatidylinositol receptor, protein kinase C (see above), and diacyglycerol kinase are potential targets in the search of cellular growth regulators. Q 12713 (Acfinomuduru sp.) (OGAWARA al., 1992), (hiset pidospermidins (Chuefomiu sp.) (YANAGISAWA, 1994a, b), and caloporoside (Culoporus dichrous) (WEBERet al., 1994; TATSUDA and YASUDA, 1996) which are inhibitors of phospholipase C have recently been discovered (Fig. 17). The search for inhibitors of the inositol-triphosphate turnover provided a series of different chemical structures, such as psitectorigen (an isoflavone from Actinomyces cultures and plants, Fig. 17) (IMOTOet al., 1988, 1991), inostamycin A (a polyether from a Sfreptomyces strain) (IMOTOet al., 1990), echiguanins (TAKEUCHI, 1992), and piericidins B1, B5 and their "-oxides (NISHIOKA et al., 1991). Recently, automated screens have been developed for inhibitors of myo-inositol monophosphatase such as ATCC 20928 A-C (diterpene types; STEFANELLI al., 1996). et Piericidin B1 N-oxide (Fig. 17) was isolated from a Sfrepfomyces strain as an inhibitor of phosphatdylinositol phosphate turnover (NrSHIOKA, 1994). The drug did not inhibit the synthesis of DNA, RNA, or proteins. However, it reversibly reduced growth of A431 cells and Ehrlich carcinoma (NISHIOKA,1994). Echiguanin (TAKEUCHI, 1992) was shown to inhibit phosphatidylinositol kinase. In con-
681
Me
w -
Me
-0
pni-lectorigen R1=OMe R2=
Hiapidospennin
Womolonln R,=H Rt-
Orobol
Trkhostann A R- NHofl; C: R= NKO-bDglucose
PLerMdrO B, N-oxide
Me
OH Adenospbomn
L671.776
Fig. 17. Inhibitors of the phosphoinositol metabolism.
trast to inostamycin, the inostamycins B and stance altering the transformed morphology C displayed no inhibitory effect on phosphati- of v-src-expressed cells to normal morpholodylinositol turnover despite their structural gy. These changes were accompanied by alsimilarity (ODAI et al., 1994; IMOTOet al., terations in cytoskeletal organization, synthesis of fibronectin, and colony-forming ability 1990). Herbimycin A (Fig. 14), a benzoquinoid on distinct media. The compound was shown antibiotic (IWAIet al., 1980; SUZUKAKE-TSU- to inhibit phosphoinositol kinase (SuzuKAKE-TSUCHIYA 1989). The trichostaet al., CHIYA et al., 1990) was reisolated as a sub-
682
tins (Streptomyces toyocaensis) (Fig. 17) are antifungal antibiotics but also potent inducers of erythroid differentiaton in mouse Friend leukemia cells. In low concentration trichostatin A arrested the cell cycle of normal fibroblast cells in both GI and GS, and induced the formation of proliferative tetraploid cells after release from the G2 arrest. Inhibition of phosphoinositol kinase was demonstrated for this agent (YOSHIDA al., 1990). et The opposite effect of inositol phosphokinase inhibitors, i.e., an increase in cellular phosphatidylinositol concentration can be achieved by inhibitors of inositol phosphate phosphatases. This idea led to the discovery of L-671,776 from a hyphomycete (Memnoniella echinata). It was shown to inhibit both the myo-inositol-l,4-disphosphatephosphatase and the 1,4,5-triphosphate 5-phosphatase (LAMet al., 1992). The adenophostins A and B (Fig. 17) occur as phosphorylated adenosine analogs in cultures of Penicillium brevicompactum (TAKAHASHI al., 1994 ). They et are potent agonists of the inositol-1,4,5-triphosphate receptor and mimic second messenger activity. Diacylglycerol kinase phosphorylates diacylglycerol to form phosphatidic acid. This enzyme is affected by extracellular stimulators and is involved in the regulation of protein kinase C by decreasing the intracellular concentration of diacylglycerol. In a search for inhibitors of this enzyme cochlioquinone and stemphone from Drechslera sacchari have recently been discovered in fungal strains (OGAWARA al., 1994) (Fig. 17). et
stance, Sch52900 and Sch52901 are new dimeric diketopiperazine structures from Gliocladium sp. which inhibit c-fos protooncogene induction as an early event in the transition of cells from the quiescent to the growing stage (CHUet al., 1995). Other fungal products, tryprostatins A and B (diketopiperazins from Aspergillus fumigatus), inhibit the cell cycle of murine tsFT210 cells by inhibition of Cdc2-kinase (CUI et al., 1996a, b). Moreover, indolocarbazoles inhibited the cell cycle progression of ras-transformed rat fibroblasts (AKINAGA al., 1993). et Ras-related proteins control a wide variety of cellular processes. They are associated with the cell membrane, and their function and activity is modified by prenylation, proteolysis, and carboxymethylation of the carboxyl terminus (KHOSRAVI-FAR al., 1992). The doret rigocins A and B (Streptomycesplatensis) and depudecins (Fig. 18) have recently been reported as new antifungal antibiotics that change the morphology of ras-transformed N/H/3T3 cells to that of normal cells by inhibiting protein carboxymethylation (KARWOWSKI et al., 1994; MATSUMOTO et al. 1992). Differanisol A (Fig. 18), from a Chaetomium strain induces the differentiation of Friend leukemic cells in mice. Interestingly, the chemical structure of this substance is very similar to that of the stalk cell differentiation inducing factor of the cellular slime mold Dictyostelium discoideum (KUBOHARA et al., 1993; MORI, 1989). The reveromycins A-D (Fig. 18) interrupt eukaryotic cell growth due to their antagonistic effect on the mitogenic activity of the growth hormone, EGF (KOSHINO al., 1992). They cause moret 4.5.3 Substances Changing the phological reversion of src+'-NRK cells and exhibit antiproliferative activity for human tuMorphology of mor cell lines. Rhizopodin (Fig. 18) is a cytoOncogene-Transformed Cells or static compound from Myxococcus stipitatus Showing Selective Toxicity which inhibits growth of various animal cultures without killing the cells. Fibroblast cells Oncogene-transformed (ras, myc, fos, erb, become larger and form long branches in an etc.) cell lines are now available. They are irreversible manner. Protein phosphorylation used to select for specific inhibitors of signal has been suggested to be the initial effect of transduction pathways. The selection of these this compound (SASSEet al., 1993). Redifferentiation of cancer cells was also inhibitors is based on their capacity to restore normal cell morphology or selectively kill on- reported for the anthracycline type cosmomycogene-expressing cells (SUZUKAKE-TSU- cins (Streptomyces cosmosus, redifferentiaton CHIYA et al., 1990; UMEZAWA, 1989). For in- inducer of Friend leukemia cells) (KHOS-
683
RAVI-FARal., 1992), latosillan (Alculigenes et lutus, inducer of differentiation of mouse myeloic leukemia cells; high-molecular weight polysaccharide) (HAYAKAWA al., 1982), et spicamycin (Streptomyces ulunosinicus, differentiation inducer of myeloic leukemia cells; Fig. 18) (HAYAKAWA al., 1983), citrinin et (Monosporuscus carpounus, differentiation inducer of myeloic leukemia cells; Fig. 18) (KAWASHIMA al., 1983), canacelunin et (Streptomyces sp., inhibitor of cancer cell agglutination, inducer of lymphoblastoid transformation of mouse brain cells) (IKEKAWA et al., 1980), microcystilide A (Microcystis seruginosu) (TSUKAMOTO, 1993), hygrolidin (SuZUKAKE-TSUCHIYA 1991), differenol A et al., (Streotomyces sp., redifferentiation inducer of murine leukemia cells) (ASAKIet al., 1981), lavanducyanin (Streptomyces ulriufer, stimulation of Hela cell proliferation; Fig. 18) (MATSUMOTO SETO,1991), kasuzamycin and (Streptomyces sp., antitumor antibiotic, affects L1210 cell cycle; Fig. 18) (TAKAMIYA et al., 1988), trichostatins (Streptomyces sp., similar to leptomycin, inducer of erythroid differentiation in murine Friend leukemia cells, inhibition of the cell phase transfer G1 to G2; Fig. 18) (YOSHIDA al., 1990), and hygrolidet in (SUZUKAKE-TSUCHIYA 1991). et al., Modulation of the actin filament network and/or protein phosphorylation is involved in tautomycin-induced bleb formation. Protein phosphorylation is also involved in recycling cell surface receptors for transferrin and EGF (KURISAKI al., 1992). et Redifferentiation of rus-transformed cells to normal morphology is also a characteristic of flavones and some glutarimide antibiotics such as acetoxycycloheximide (Fig. 18) and cycloheximide (OGAWARA al., 1989a, b). et Herbimycin (Fig. 14) and sparoxomycins reverse the transformed morphology of temperature-sensitve Rous sarcoma virus-infected rat cells (ts/NRK) to the normal morphology concomitant with a drastic reduction in intracellular p60"" kinase activity. Semisynthetic derivatives of herbimycin prolonged the life span of tumor-bearing mice (UEHARA al., et 1988; UBUKATA al., 1996). Some macrolide et antibiotics such as FD-891 (Fig. 18) and FD892, which possess polyhydroxy-alkane-substituted side chains, induce morphological
changes of HL-60 cells at low concentrations and display cytocidal activity at higher dosages (SEKI-ASANO al., 1994). Antitumor et activities are also a characteristic of macrolactams, such as leinamycin and hitachimycin (Fig. 18) (HARAet al., 1990; SHIBATA al., et 1988). The antibiotic FR 901228 from Chromobucterium violuceum (Fig. 18) also reverted the transformed morphology of a rus transformant cell line to normal and exhibited antitumor activity against murine and human tumors (UEDA et al., 1994). Melastin (mol. wt. 5 OOO f 3 000) suppresses lipopolysaccharide-induced blastogenesis of B cells more effectively than concanavalin A and selectively inhibits growth of several leukemia cells (MAEDA,1993). Leptomycin B and leptolstatin from Streptomyces strains (Fig. 18) have recently been found to inhibit proliferation of cultured animal cells by blocking progression of the cell cycle in G1 and G2 phases (ABEet al., 1993b). Phosmidosin C nucleosides from Streptomyces sp. display morphology reversion activity against src transformed NRD cells (MATSUURA al., 1996). Last but not et least, reductioleptomycin A (Streptomyces sp.) (Fig. 18) was shown to flatten the moret phology of v-rusts NRK cells (HOSAKAWA al., 1993).
4.5.4 Inhibitors of ras-Farnesyltransferase

The rus genes are mutated in 50% of colon and 90% of pancreatic carcinoma. Farnesylprotein transferase catalyzes farnesylation of a protein encoded by rus (Ras or p21). This is essential for the associaton of Ras proteins with the cell membrane and rus-mediated transformation. Inhibition of the prenylating enzyme, rus-farnesyltransferase, was therefore proposed as a specific target of "soft" anticancer agents. In screening for representatives of such drugs, streptonigrin (Streptomyces albus) and its 10 '-desmethyl derivative (VAN DER PYL et al., 1992), representatives of the indolcarbazole type inhibitors of protein kinase C (CGP 412511, staurosporin, UCN-01, K-252a), the pepticinnamins (lipopeptide, Streptomyces sp.) (OMURAet al.,
684

OH 0
Me
0
Me
Me
me
MW
no
o
0
on
a
CY
Leinamycin
CY
Dorrigocin A
Differanisol A
OR
FD-891
Hichimycin
Reveromycin A
MPW
M;
Me Me
NHOH
Trichostatin
_.
Fr 901228
OH AH
Splcamycln
tnrlnln
Kazuramycln B
OH OH 0 OH
Me
Me Me
Me Me Me
Me Me
Leptolatath
bpudain
LIV.II~UCY.II~I
Fig. 18. Examples of drugs changing the morphology of oncogene-transformed cells or showing selective toxicity against oncogene-expressedcells.
685
1993; SHIOMI al., 1993), saquayamicins (SEet et al., 1996), and andrastins (Penicillium sp., OMURA al., 1996) were found to et be new inhibitors of this enzyme and blockers of the cell cycle progression of ras-transformed cells (Fig. 19). More recently, a series of other structures from Actinoplanes sp. such as actinoplanic acid A and its derivatives have been reported to inhibit ras-farnesyltransferase (SINGH al., 1994a, b, c). In this context, et fungal structures such as fusidienol (Fusidium griseum) (SINGHet al., 1994a): kurasoins A and B (Paecilomyces sp. FO-3648, UCHIDA et al., 1996), preussomerins, deoxypreussomerins, Sch 49210, Sch 53514-53517 (Preusseria isomera, Harmonema demativides) (SINGHet al., 1994c), chaetomellic acid (Chaetomella acuseta) (SINGHet al., 1993) (Fig. 19), gliotoxin, and acetylgliotoxin (VAN DER PYL et al., 1992) provided non-conventional leads in the search for inhibitors of this enzyme.
KIZAWA
similar to pyocyanin, WS-9659A and B, were also obtained from Streptomyces cultures (NAKAYAMA al., 1989) (Fig. 20). et
4.5.6 Miscellaneous Drugs with Potential Antitumor Activity

The discovery of new cytotoxic compounds will continue. New structures such as duacins (HIDAet al., 1994), leptosins (TAKAHASHI et al., 1994), cochleamycin (SHINDO al., 1996), et and himastatin (LEET et al., 1996) may be suitable for studying structure-activity relationships and modes of action. A therapeutic potential was suggested for inhibitors of melanin synthesis such as OH-3984K1 and -K2 (Streptoymces sp.), members of the albocycline family of antibiotics (KOMIYAMA al., et 1993) (Fig. 21) and melanoxazal from the hemolymphe of the silkworm Bombyx mori (TAKAHASHI al., 1996) as well as for inet hibitors of the CAMP-phosphodiesterase (HEDGEet al., 1993), in addition to glyoxylase inhibitors such as glyo I and I1 (ALLEN et al., 1993; TORNALLAY, 1990). Marine organisms have contributed many new and promising antitumor agents (Fig. 21). The spongistatins from the Eastern Indian Ocean sponge species are exceptionally potent growth inhibitors for human cancer cells. They comprise a series of homologous compounds distinguishable by chlorine substitutions (PETTITet al., 1993a, b). Similarly, Dolabella auricularia from the Indian Ocean contains the cytotoxic dolastatins (see the dolastatin D) and its congeners (SAE et al., 1993). Other recent examples of cytotoxic compounds from marine organisms are phloeodictines Al-A7 and Cl-C2 which are guanidine alkaloids from the New Caledonian sponge Phloeodictyon sp. (KOURANY-LEFOLL et al., 1994), phakellistatins from a Comoros marine sponge (PETTIT,1994), agelasphins (sponge Agelas mauritianus) (NATORI et al., 1994; HARADA al., 1993), amphidiet nolide F (KOBAYASHI al., 1993) and bryoset tatins 16-18 from the marine bryozoan Bugula neritina (PETTITet al., 1996). Cyanobacteria also supply a rich source of cytotoxic metabolites. However, due to their general toxicity against human cells a therapeutic application
4.5.5 Inhibitors of Sexual Hormone Production and Hormone-Receptor Interactions

Suppression of the production of sexual hormones (estrogens and androgens) and of receptor interaction of these hormones were proposed as a therapeutical strategy in the treatment of hormone-dependent tumors. The therapeutic efficacy of synthetic aromatase inhibitors such as aminoglutethimides, imidazoles, and triazoles (VANDEN BOSCHEet al., 1994; BRAND al., 1988) for breast canet cer and tetrazoles motivated a search for nonsteroidal microbial compounds which could reduce the effects of estrogen by antagonizing its receptor in a manner similar to the synthetic drug tamoxifen. The compound R1128, an estrogen receptor antagonist, is an alkylated trihydroxyanthraquinone from Streptomyces sp. 1128 (HORI et al., 1993a, b). Related polycyclic phenolic structures are the napyradiomycins A and B (HORI et al., 1993c) which have also recently been isolated from a Streptomyces culture. A promising approach to the treatment of prostata carcinoma is the use of inhibitors of testosterone5areductase. Non-steroidal inhibitor structures
686

a
yQ .
- $ Q 0
M * ,
Ml
0 COT 41251
UWMr
m
CO,H
OH
0 Fu~ldkul
AdkphkwidA
Fig. 19. Examples of inhibitors of ras-farnesyltransferase.

o
on
on
n
on
Me
on
a#+
R 1129 (R= Pmp,Butpent)
lMe
Napyradlmycin A WS-9859A
Fig. 20. Estrogen receptor antagonists and testosterone Sa-reductase inhibitors.
of these compounds not is foreseeable at present.
5 Closing Remarks
During the past ten years many new drug structures have been discovered from new sources, such as new microbes, plants, and
marine animals, by using new target-oriented screening assays. Both in the field of classical cytotoxic drugs and in the recognition of specific inhibitors of the mammalian cell cycle, major advances have been made. Investigations with enedyines and taxol have led to new types of neoplasm inhibitors which are now being introduced to clinical therapy. Various novel effectors of the cell cycle and the signaling cascade are now used as invaluable biochemical tools which may help to gain
5 Closing Remarks
681
2
Me
hlldnA
Me
AphldlnolfdeF
Lephla
Fig. 21. Selected structures of cytotoxic metabolites of marine organisms.
Y., H., ABE, N., ENOKI, NAKAKITA, UCHIDA, N., NAKAMURA, MUREKATA, (1993a), Novel T., N. antitumor antibiotics septomycins. 11. Isolation, physico-chemical properties and structure elucidation, J. Antibiot. 46, 1536-1549. M., H., ABE, K., YOSHIDA, NAOKI, HORINOUCHI, S., BEPPU, (1993b), Leptolstatin from StreptoT. myces sp. SAM 1595, a new GAP phase-specific Acknowledgement inhibitor of the mammalian cell cycle. 11. Physico-chemical properties and structure, J. Antibiot. We are deeply indebted to Mrs. HILTRUD 46,735-740. KLOSE for typing and preparing the manu- ADAMSON, H., SIEBER, M., DOUROS, D. R. S. J. script. (1979), Development of actinomycin analogs. Adv. Med. Oncol. Res. Educ. Proc. Int. Cancer Congr. 12th (1979), Meeting Date 1978, Vol. 5, pp. 93-99. Oxford: Edn. B.W. Fox, Pergamon. K. D. J. ADDESS, J., GILBERT, E., FEIGOU, (1992), Sequence specificity of quinoxaline antibiotics: a comparison of the binding of [N-MeCyS3] [NMeCys71 tandem and echinomycin to DNA ABE, F., SHIBUYA, ASHIZAWA, TAKAHAK., Y., studied by proton NMR spectroscopy. Struct. SHI, K., HORINISHI, MATSUDA, ISHIZUH., A., Funct. Proc. Conversation Discip. Biomol. SteH. KA,M., TAKEUCHI, UMEZAWA, (1985), T., reodyn. 7th (1992), Meeting Date 1991, Vol. 1, Enhancement of antitumor effect of cytotoxic pp. 147-164. (SARMA, H., SARMA,M. H., R. agents by bestatin, J. Antibiot. 38, 411-414. Eds.). New York: Adenine Press. ABE, N., ENOKI, NAKAKITA, UCHIDA, N., Y., H., NAKAMURA, (1991), Isolation and characteri- AGARWAL, P., BELL, R., LILLQUIST,A., T. R. zation of atramycin B, new isotetracenone type MCCAFFEY, (1983), Purine metabolism in R. antitumor antbiotic, J. Antibiot. 44,1025-1027. leukemia, Ann. Natl. Acad. Sci. 2, 160-168.
a deeper insight into the mechanism of malignant cell growth. Moreover, their therapeutic potential is presently under investigation. It is likely that new antitumor drugs will be discovered in the future when new screening assays and natural sources become available.
6 References
688
A, a differentiation inducing substance against AKINAGA, NOMURA, GOMI, OKABE, S., K., K., M. mouse leukemia cells, J. Antibiot. '34, 919-920. (1993), Diverse effects of indolecarbazole comA. pounds on the cell cycle progression of ras-trans- ASZALOS, (Ed.) (1988), Antitumor compounds of natural origin: chemistry and biology, Vol. 11. formed rat fibroblast cells, J. Antibiot. 46, 1767Boca Raton, FL: CRC Press. 1770. ALLEN,R. E., Lo, T. W. C., THOMALLY, J. ATTA-UR-RACHMAN, IQBAL,Z. NASIR,H. P. A., (1993), Inhibitors of glyoxylase. I. Design, syn(1994), Synthetic approaches to vinblastine and thesis, inhibitory characteristcs and biological vincristine - anticancer alkaloids of Catharanevaluation, Biochem. SOC. Trans. 21, 535-540. thus roseus. Stud. Nut. Prod. Chem. 14 (StereoG. ALLGAIER, HERTTING, (1986), Polymyxin C., selective Synthesis) (Pt. I.), 805-877. B, a selective inhibitor of protein kinase C, dim- BARKLEY, D., CHEATHAM, THURSTON, S., D. M. inishes the release of noradrenaline and the enE., HURLEY, H. (1986), Pyrrolo[l,4]benzodiaL. hancement of release caused by phorbol 12,13zepine antitumor antbiotics: evidence for two dibutyrate, Naunyn-Schmiedeberg's Arch. Pharforms of tomaymycin bound to DNA, Biochemmacol. 333,218-221. istry 25, 3021-3031. ANDERS, W. (1991), Glutathione-dependent BARTEL, L., ZHU, C.-B., CAMPEL, s.,DOSCH, M. P. Z. bioactivation of xenobiotics: implications for D. C., CONNORS, C., STROHL, R., BEALE, N. W. mutagenicity and carcinogenicity, Proc. Int. J. M., FLOSS,H. G. (1990), Biosynthesis of anSyrnp. Princess Takamatsu Cancer Res. Fund thraquinones by interspecies cloning of actino(1991) Vol. Date 190,21 (Xenobiot. Cancer) 89rhodin biosynthesis genes in streptomycetes: 99. clarification of actinorhodin gene functions, J. ANDERSON, D., BERGER, A. (1994), MutagR. N. Bact. 172,48164826. enicity and carcinogenicity of topoisomerase in- BASU,A., LAZO,J. S. (1994), Protein kinase C, in: teractive agents, Mutat. Res. 309,109-142. New Molecular Targets in Cancer Chemotherapy. ANDOH, KJELDSEN, BOUVEN, J., WESTT., E., B. (KERR,D. J., WORKMAN, Eds.), pp. 121-14. P., O., ERGAARD, ISHII, K., OKADA,K. (1993), Boca Raton, FL CRC Press. Camptothecin-resistant DNA topoisomerase I, BEAUVILLE, L., KENIRY,M. A., SHAFER, H. R. D. in: DNA Topoisomerase in Cancer (POTMESIL, (1990), NMR investigation of mithramycin A M., KOHN, W., Eds.), pp. 249-295. New York: K. binding to d(ATGCAT)2: a comparative study Oxford Univ. Press. with chromomycin Al, Biochemistry 29, 9294E., J. ANGANWUTAKU, ZIRBES, ROSAZZA, P. I. O., 9304. N. (1992), Isoflavonoids from streptomycetes: BECK,W. T., DANKS, K., WOLVERTON, S., M. J. origins of genistein, 8-chlorogenistein, and 6,8KIM,R., CHEN, (1993) Drug resistance assoM. dichlorogenistein, J. Nut. Prod. 55, 1498-1504. ciated with altered DNA topoisomerase 11, Adv. AOYAMA, KOMIYAMA, YAMAZAKI, TAT., F., T., Enzyme Regul. 33, 113-127. Y., TEE, T., ABE, F., MURAOKA, NAGANAWA, J. P., A. T., T. H., AOYAGI, TAKEUCHI, (1993a), Bena- BENTLER, A., CLARK, ALVAREDO, B. (1994), Esperamicin P, the tetrasulfide analog of statins C and D, new inhibitors of glutathione Sesperamicin Al, J. Nut. Prod. 57,629-633. transferase, produced by Streptomyces sp. N. S. S. MI384-DF12: production, isolation, structure de- BERGER, A., BERGER, J., GERSON, L. (1987), DNA repair, ADP-ribosylation and pyrtermination and biological activities, J. Antibiot. idine nucleotide metabolism as targets for cancer 46,712-718. chemotherapy, Anti-Cancer Drug Res. 2, 203AOYAMA, ZHAO,W., KOJIMA, MURAOKA, T., F., 210. H., T., T. Y., NAGANAWA, TAKEUCHI, AOYAGI, M. R. (1993b), Cysfluoretin, a new inhibitor of gluta- BERRIDGE, J., IRVINE, F. (1984), Inositol trisphosphate, a novel second messenger in celthione S-transferase produced by Streptomyces lular signal transduction, Nature 312, 315-321. sp. MI384-DF-12, J. Antibiot. 46, 1471-1474. M. R. ARAI, KANDA, ASHIZAWA, MORIMOTO, BERRIDGE, J., IRVINE, F. (1989), Inositol H., Y., T., phosphates and cell signalling, Nature 341 M., GOMI, KONO,M., KASAI, M. (1994), MiK., (6239), 197-205. tomycin derivatives having unique condensedBERTRAM, S. (1990), The chemoprevention of J. ring structures, J. Antibiot. 47, 1312-1321. human cancer: an overview, in: Effects of TheraARCAMONE, (1994), Design and synthesis of anF. p y on Biology and Kinetics of the Residual Tuthracycline and distamycin derivatives as new, mor. Part A: Preclinical Aspects, pp. 345-360. sequence-specific DNA-binding pharmacologiNew York, London: Wiley-Alan Liss. cal agents, Gene 149,57-61. J. M. H., ASAKI,K., ONO, I., KUSUKABE, NAKAMURA, BIEDLER, L., MYERS, B. (1993), Multidrug resistance (vinca alkaloids, actinomycin D, and anK. G., ISONO, (1981), Studies on differentiationthracycline antibiotics), in: Drug Resistance of inducing substances of animal cells. I. Differenol
6 References
689
Mammalian Cells, Vol. 2 (GUPTA,R. S., Ed.), pp. 57-88. Boca Raton, FL: CRC Press. BILLINGS, P. (1993), Approaches to studying the C. target enzyme of anticarcinogenic protease inhibitors, in: Protease Inhibitors as Cancer Chemotherapeutic Agents (TROLL,w., KENNEDY, A. R., Eds.), pp. 191-198. New York Plenum Press. BLUMBERG, M., PETTIT,G. R., WARREN, S., P. B. SZALLASI, SCHUMAN, D., SHARKEY, A., L. N. A., NAKAMURA, DELLAQUILA, L., DE H., M. ORIES,D. J. (1989), The protein kinase C pathway in tumor promotion, in: Brain Carcinogenesis: Mechanisms and Human Relevance, pp. 201212. New York: Alan Liss. D. BOGER, D. L., JOHNSON, S. (1996), CC-1065 and the duocarmycins: understanding their biological function through mechanistic studies, Angew. Chem. (Int. Edn. Engl.) 35, 1438-1474. BOROS,C., HAMILTON, M., KATZ,B., KULANS. THAIVEL, P. (1994), Comparison of balanol from Verticilliurn balanoides and ophidiocordin from Cordyceps ophioglassoides, J. Antibiot. 47,10101016. BORREL, N., PEREIRA, FIALLO, (1994), M. E., M. Analysis of multidrug transporters in living cells, Met.-Based Drugs 1,175-182. BRADSHAW, A., PRENTIS, J. (1987), OncoR. S. genes and Growth Factors. Amsterdam: Elsevier. R., BRAND, E., PUETT, D., GAROLA, FENOLL, M. K., COVEY,D. F., ZIMNISKI, J. (1988), AroD. matase and aromatase inhibitors: from enzymology to selective chemotherapy, in: Hormones, Cell Biology and Cancer: Perspectives and Potentials, pp. 65-84. New York Alan Liss. BURCHELL, BAIRD, COUGHTRIE, W. H. B., S., M. (1991), The role of xenobiotic glucuronidating enzymes in drug resistance of tumor cells and tissues, Proc. Int. Symp. Princess Takamatsu Cancer Res. Found., Volume Date 1990, 21 (Xenobiot. Cancer) 263-275. BURCK, B., LIN, E. T., LARRICK, W. (1988), K. J. Oncogenes. New York, Heidelberg: SpringerVerlag. BUTA,J. G., KALINSKI, (1988), Camptothecin A. and other plant growth regulators in higher plants with antitumor activity, ACS Symp. Ser. 380,294-304. BUTLER, HAEY,B. M. (1987), Are reduced quiJ., nones necessarily involved in the antitumor activity of quinone drugs? Br. J. Cancer (Suppl.) 55, 53-59. CAI, Y., FREDENHAGEN, HUG,P., MEYER,T., A., PETER,H. H. (1996), Further minor metabolites of staurosporin produced by a Streptomyces longisporojlavus strain, J. Antibiot. 49, 519-526.
CAMPLEJOHN, S. (1980), A critical review of the R. use of vincristin, Cell Tissue Kinet. 13(3), 327335. CAPRANICO, ZUNINO, (1990), Structural reG., F. quirements for DNA topoisomerase I1 inhibition by anthracyclines, Jerusalem Symp. Quantum Chem. Biochem. 23 (Mol. Basis. Specif. Nucleic Acid Drug Interact.), 167-176. CARNEY, C., SCHEUER, J. (1993), PolohuaJ. P. none E, a topoisomerase-I1 inhibitor with selective lung tumor cytotoxicity from the Pohnpei sponge Dysidea sp., Tetrahedron Lett. 34, 37273730. CASSIDY, M., DOUROS, D. (1988), Anticancer J. J. Agents Based on Natural Product Modes. New York: Academic Press. CHAKRABORTY, K., MAJUMDER, A. HEMATA, K., HODGSON, P. (1994), DNA topoisomerase inC. hibitors: potential uses in molecular medicine, Expert Opin. Ther. Pat. 4, 655-668. CHEN,A. Y., LEROY,L. F. (1994), DNA topoisomerases: essential enzymes and lethal targets, Ann. Rev. Pharmacol. Toxicol. 34,191-218. CHIMURA, ISHIZUKA, HAMADA, HORI, H., M., H., S., KIMURA, IWANAGA, TAKEUCHI, K., J., T., UMEZAWA, (1968), A new antibiotic, macroH. momycin, exhibiting antitumor and antimicrobial activity, J. Antibiot. 21, 4449. CHINO, NISHIKAWA, UMEKITA, HAYAM., K., M., SHI, C., YAMAZAKI, TSUCHIDA, SAWA, T., T., T. T., HAMADA,M., TAKEUCHI,(1996), Heliquinomycin, a new inhibitor of DNA helicase, produced by Streptomyces sp. MJ929-SF2. I. Taxonomy, production, isolation, physico-chemical properties and biological activities, J. Antibiot. 49, 752-757. CHU,M. (1994), A novel class of antitumor metabolites from the fungus Nattrasia mangifera, Tetrahedron Lett. 35, 1343-1346. CHU, M., TRUUMEIS, ROTHOFSKY, L., PAI., M. TEL, M. G., GENTILE, DAS, R. P., PUAR,S. F., M., UN, S. L. (1995), Inhibition of c-fos protooncogene induction by Sch 52900 and Sch 52901, novel diketopiperazines produced by Gliocladium sp., J. Antibiot. 48, 1440-1445. COLL,M., FREDERICH, A., WONG,A., RICH,A. C. (1987), A bifurcated hydrogen-bonded conformation in the d(A,T) base pairs of the DNA dodecamer d (C6CA3T3GCG) and its complex with distamycin, Proc. Natl. Acad. Sci. USA 84, 8385-8389. CROW,R. T., CROTHERS, M. (1994), Inhibition D. of topoisomerase I by anthracycline antbiotics. Evidence for general inhibition of topoisomerase I by DNA-binding agents, J. Med. Chem. 37, 3191-3194.
690
Cur, C., UBUKATA, KAKEYA, ONOSE, M., H., R., OKADA, TAKAHASHI, ISONO,K., OSADA, G., I., H. (1996a), Acetophtalidin, a novel inhibitor of mammalian cell cycle, produced by a fungus isolated from a sea sediment, J. Antibiot. 49, 216219. Cur, C., KAKEYA, OSATA,H. (1996b), Novel H., mammalian cell cycle inhibitors, tryprostatins A, B, and other diketopiperazines produced by Aspergillus fumigatus. 11. Physico-chemical properties and structures, J. Antibiot. 49, 534-540. CURRAU, P. (1993), The camptothecins: a reD. born family of antitumor agents, J. Clin. Chem. SOC. (Tapei) 40(1), 1-6. CURRAU, P., SISKO,J., YESKE,P. E., LIN, H. D. (1993), Recent applications of radical reactions in natural product synthesis, Pure Appl. Chem. 65(6), 1153-1159. DERVAN, D., SCOTT YOUNGQUIST, SLUKA, P. R., J. P. (1987), in: Stereochemistry of Organic and W., Bioorganic Transformations (BARTHMANN, SHARPLESS, B., Eds.), pp. 221-234. WeinK. heim: VCH. DIETEL,M. (1991), Whats new in cytostatic drug resistance and pathology, Pathol. Res. Pract. 187, 892-905. DUMITRIU, (1996), Polysaccharides in Medicinal S. Applications. New York, Basel, Hongkong: Marcel Dekker. ECKARDT, WAGNER, (1988), Biosynthesis of K., C. anthracyclinones, J. Basic Microbiol. 28, 137144. EDO, K., SAITO,K., AKIYAMA-MURAI, MrzuY., M., N. GAKI, KOIDE,Y., ~ S H I D A , (1988), An antitumor polypeptide antibiotic neocarzinostatin: the mode of apo-protein - chromophore interaction, J. Antibiot. 41, 554-562. EL KHADEM, S. (Ed.) (1982), Anthracycline AnH. tibiotics. New York Academic Press. END, D. W. (1987), Non-selective inhibition of mammalian protein kinases by flavinoids in vitro, Res. Commun. Chem. Pathol. Pharmacol. 56, 76-86. FEIG, A. L., LIPPARD, J. (1994), Reactions of S. non-hemic iron (11) centers with dioxygen in biology and chemistry, Chem. Rev. (Washington D.D.) 94,7594305. FERDINAND, J. (1989), Mechanism of tumor inF. duction by human retroviruses, Arzneim. Forsch. 39(6), 735-740. FIJII, YAMASHITA, CHIBA, UOSAKI, N., Y., S., Y., SAITOH, TUJI,Y., NAKANO, (1993), UCE Y., H. 6, a new antitumor antibiotic with topoisomerase I medaited DNA cleavage activity from Actinomycetes, J. Antibiot. 46, 1173-1174. FIJII, YAMASHITA, CHIBATA, SAITOH, F., Y., S., H. Y., NAKANO, (1994), UCE 1022, a new anti-
tumor antibiotic with topoisomerase I mediated DNA cleavage activity from Paecilomyces, J. Antibiot. 47, 949-951. FISCHER, H., CHARBONNEAN, TACHS, K. E. H., N. (1991), Protein tyrosine phosphatases: a diverse family of intracellular and transmembrane enzymes, Science 253,401-406. FLEISCHMANN, F., CHAHWALA, B., COUNTF. S. LEY, L. (1986), Ras transformed cells: altered levels of phosphatidylinositol 4,s-bisphosphate and catabolites, Science 232,407-410. FOMICHOVA, V. (1992), Design of enzyme inE. hibitors as drugs, J. Antibiot. 45, 1185-1186. FORNICA, (1977), The biogenesis of actinomyJ. cins, Actinomycetes Relat. Org. l2(4), 12-20. FOSTEL, SHEN,L. L. (1994), DNA TopoisomerJ., ases, Design of Enzyme Inhibitors as Drugs, Vol. 2 (SANDLER, SMITH,H. J., Eds.), pp. 564M., 624. Oxford: Oxford University Press. FOX,K. R. (1990), Footprinting studies of the interaction of quinomycin antibotic UK 63052 with DNA: comparison with echinomycin,J. Antibiot. 43, 1307-1315. FOX, B. W. (1991), Natural products in cancer treatment from bench to the clinic, Trans. R. SOC. Trop. Med. Hyg. 85,22-25. FRANZ, (1990), Biogene Cytostatica, PharmuZ G. 6,257-262. FREDENHAGEN, HUG,P., SAUTER, PETER, A., H., H. H. (1995), Paeciloquinones A, B, C, D, E and F New potent inhibitors of protein tyrosine kinase produced by Paecilomyces carneus. 11. Characterization and structure determination, J. Antibiot. 48, 199-204. FUNABASHI, HORIGUCHI, IINUMA, TANY., T., S., IDA, S., HORADA, (1994), TAN-l496A, C and S. E, diketopiperazine antibiotics with inhibitory activity against mammalian DNA topoisomerase I, J. Antibiot. 47, 1202-1218. FURK, L., ABDULHAMIED, A., PARKER, J. T. D. (1989), Potentiation of interleukin-2 release by anticancer drugs, Agents Actions 26(1-2), 156157. GAAL, J. C., PEARSON,C. K. (1986), Covalent modification of proteins by ADP-ribosylation, TIBS 11, 171-175. GALE, E. F., CUNDLIFFE, REYNOLDS, E., E., P. RICHMOND, H., WARING, J. (1981), The M. M. Molecular Basis of Antibiotic Action. New York, Sydney, Toronto: Wiley. GIACCONE, (1994), DNA topoisomerases and G. topoisomerase inhibitors, Pathol. Biol, 42, 346352. GIRI,S., WANG,Q. (1989), Mechanism of bleomycin-induced lung injury, Comments Toxicol. 3, 145-179.
6 References
691
GOLDBERG, I. (1986), Molecular mechanisms of H. DNA sugar damage by antitumor antibiotics, Ponti$ Acad. Sci. Scr. Varia 70, 425-462. GOLDBERG, I. (1991), Mechanism of neocarziH. nostatin action: role of DNA microstructure in determinaton of chemistry of bistranded oxidative damage, Acc. Chem. Res. 24, 192-196. GOLDBERG, I., KAPPA, S., CHIN, H., LEE, H. L. D. S. H. (1989), Bistranded oxidative DNA sugar damage by targeted antibiotic diradicals, New Methods Drug Res. 3, 27-42. GOORMAGHTIGH, RUYSSCHAERT, (1984), E., J. M. Anthracycline glycoside-membrane interactions, Biochim. Biophys. Acta 779,271-288. GREENSTEIN, MAIESE,W. M. (1984), PreM., screens for novel antineoplastic agents, Dev. Ind. Microbiol. 25, 267-275. GREVER, R., SCHEPARTZ, A., CHAPNER, M. S. B. A. (1992), The national cancer institute: cancer drug development and development program, Semin. Oncol. 19,622-638. GUENGRICH, P. (1988), Roles of cytochrome F. P-450 enzymes in chemical carcinogenesis and cancer chemotherapy, Cancer Res. 48, 29462954. GUERIITE-VOEGELEIN, GUENARD, DuF., D., P. BOIS, WAHL,A., POITIER, (1994), ChemiJ., cal and biological studies in the taxol (pacitaxel) and taxotere (docetaxel) series of new antitumor agents, J. Pharm. Belg. 49, 193-205. GUNDA, I., ALI, S. M., ZYGMUNT, JAGAG. J., SINGHE, L. R. (1994), Taxol: a novel antitumor agent, Expert Opin. Ther. Pat. 4, 109-120. GUNICKE, HOFMANN, (1992), Cytotoxic and H., J. cytostatic effects of antitumor agents induced at the plasma membrane level, Pharmacol. Ther. 55, 1-30. GUITERIDGE, M. C. (1992), Iron and oxygen raJ. dicals in brain, Ann. Neurol. 32 (Suppl.), 516521. HAGIWARA, IRIE,K., FUNAKI, HAYASHI, N., A., K. H., ARAI,M., KOSHIMIZU, (1988), Structure and tumor-promoting activity of new teleocidinrelated metabolites (blastmycetins) from Streptoverticillium blastmyceticum, Agric. Biol. Chem. 52, 641-648. HANADA, OKKUMA, YONEMOTO, TONM., H., T., M., H., T., ITA,T, OHBAYASHI, KAMEI, HIYAKI, KONISHI,M., KAWAGUCHI, (1991a), MaduH. ropeptin, a complex of new macromolecular antitumor antibiotic, J. Antibiot. 44,403-414. HANADA, KANEDA, NISHIYAMA, HoM., K., Y., SHINO,Y., KONISHI, OKI,T.(1991b), HydraM., mycin, a new antitumor antbiotic. Taxonomy, isolation, physico-chemical properties, structure and biological activity, J. Antibiot. 44,824-830.
HARA,M., ASANO,K., KAWAMOTO, I., TAKIGUCHI, T., KATSUMATA, TAKAHASHI, NAS., K., KANO, H. (1990), Leinamycin, a new antitumor antibiotic from Streptomyces. Producing organism, fermentation and isolation, J. Antibiot. 42, 1768-1774. HARADA, MAGUMI, SHIMADA, SUZUKI, K., T., T., M. (1993), Occurrence of four depsipeptides aerugicopeptins, together with microcystins from toxic cyanobacteria, Tetrahedron Lett. 34, 60916094. HARPER, E., WELCH,D. R. (1992), Isolation, D. purification, synthetics and antiinvasivelantimetastatic activity of U-77863 and U-77864 from Streptomyces griseoluteus strain WS6724, J. Antibiot. 45, 1827-1836. HART, R. W., TURTURRO, (1988), Current A. views of the biology of cancer, in: Carcinogen Risk Assessment (TRAVIS,C. C., Ed.), New York: Plenum Publishing Corp. HARUYAMA, OHKUMA, NAGAKI, OGIH., Y., H., TA,T., TANAKI, KINOSHITA,T. (1994), MatK., lystatins, new inhibitors of type IV collagenases from Actinomadura atramentaria. 111. Structure elucidation of matlystatins A to F, J. Antibiot. 47, 1473-1480. HAYAKAWA, NAKAGAWA, ANDO, SHIY., M., T., MAZU, A., SETO,H., OTAKE, (1982), Studies N. on the differentiation inducers of myceloid leukemic cells. I. Latosillan, a new inducer of the differentiation of M1 cells, J. Antibiot. 35, 12521254. HAYAKAWA, NAKAGAWA, KAWAI, H., Y., M., TANABE, NAKAYAMA, SHIMAZU, K., H., A., N. SETO,H., OSAKE, (1983), Studies on the differentiation of inducers of myeloid leukemic cells. 111. Spicamycin, a new inducer of differentiation of HL-60 human promyeolocytic leukemia cells, J. Antibiot. 36, 934-937. HAYAKAWA, HA, S., KIM,Y. J., FURIKATA, Y., K., SETO,H. (1991), Hamatorubicins A, B, C and D, new isotetracenone antibiotics effective against multidrug-resistant tumor cells, J. Antibiot. 44, 1179-1 186. HAYAKAWA, ISHIGAMI, SHIN-YA, SETO, Y., K., K., H. (1994), Menoxymycins A and B, antitumor antibiotics generating active oxygen in tumor cells, J. Antibiot. 47, 1344-1347. HECHT,S. M. (1994), RNA degradation by bleomycin, a naturally occurring bioconjugate, Bioconjugate Chem. 5,513-526. HECHT,S . M., BERRY, E., MACKENZIE, J., D. L. DUSOY, W., NASUTI, A. (1992), A strategy K. C. for identifying novel mechanistically unique inhibitors of topoisomerase I, J. Nut. Prod. 55, 401-413.
692
HEDGE,V. R., MILLER, R., PATEL,M. G., KING, J. A., A. H., PUAR, M. S., HORAN, HART, R., YARBOROUGH,R., GULLO, V. (1993), Sch45752 - an inhibitor of calmodulin-sensitive cyclic nucleotide phosphodiesterase activity, J. Antibiot. 46, 207-213. HEINSTEIN, F., CHANG.C. J. (1994), Annu. Rev. P. Plant. Physiol. Plant. Mo1.-Biol. 45,663-674. HENSENS, D., GOLDBERG, H. (1989), Mecha0. I. nism of activation of the antitumor antibiotic neocarzinostatin by mercaptan and sodium borohydride, J. Antibiot. 42, 761-767. HIDA,T., MERROI,M., TANIDA,S., HARADA,S. (1994), Structures of duacin Al and B1 new naphthyridinomycin-type antitumor antibiotics, J. Antibiot. 47, 917-921. HILL,B. T. (1986), Resistance of mammalians tumor cells to anticancer drugs: mechanisms and concepts relating specificity to methotrexate and vincristine, J. Antimicrob. Chernother. 18 (Suppl. B), 61-73. HOCHLOWSKI, E., MULLALLY, M., SPANTON, J. M. S. G., WHITTERN, N., HILL, P., MCALPINE, D. J. B. (1993), Reversion of multi-drug resistance to antitumor antibiotics was also reported for 5-Nacetylardeeminfrom Aspergillus fischeri, J. Antibiot. 46,380-386. HOFSTEAD, S . J., MATSON,J. A., MALACKO,A. R., MARQUARDT, (1992), Kedarcidin, a new H. chromoprotein antitumor antibiotic. 11. Isolation, purification and physico-chemical properties, J. Antibiot. 45, 1250-1292. HOLLIS-SHOEWALTER, D. (1988), Design, tuH. mor biology, and biochemical pharmacology of anthrapyrazoles, in: Anthracycline and Anthracenedione-Based Anticancer Agents (LOWN, J. W., Ed.). Amsterdam, New York: Elsevier. HOLLIS-SHOEWALTER, D., BUNGE, R. H., H. FRENCH,J. C., HURLEY,T. R., LEEDS, R. L., P. C. LEJA,B., MCDONNELL, D., EDMUNDS, R. (1992), Improved production of pentostatin and identification of fermentation metabolites, J. Antibiot. 45, 1914-1918. HORI, ABE, Y., EZAKI,M., GOTO, T., OKUNAY., MA, M., KOHSAKA,M. (1993a), R1128 substances, novel non-steroidal estrogen-receptor antagonists produced by a Streptornyces. I. Taxonomy, fermentation, isolation and biological properties, J. Antibiot. 46, 1055-1062. HORI, Y., ABE, Y., NISHIMURA, M., GOTO, T., OKUHARA, KOHSAKA,M. (1993b), R1128 M., substances, novel non-steroidal estrogen-receptor antagonists produced by a Streptomyces. 111. Pharmacological properties and antitumor activities, J. Antibiot. 46, 1069-1075. HORI, Y., ABE, Y., SHIGEMATSU, GOTO, T., N., OKUNARA, KOHSAKA, (1993c), NapyraM., M.
diomycins A and B: non-steroidal estrogen receptor antagonists produced by a Streptornyces, J. Antibiot. 46, 1890-1893. HOSOKAWA,N., IIMURA, NAGANAWA, H., H., HAMADA, TAKEUCHI, (1993), A new anM., T. tibiotic, structurally related to leptomycin A, flattens the morphology of r-ras" NRK cells, J. Antibiot. 46, 676-679. HSIEH,T. S. (1992), DNA topoisomerases, Curr. Opin. Cell Biol. 4, 396-400. ICHIMURA, OGAWA, TAKAHASHI, KoM., T., K., E., I., T., BAYASHI, KAWAMOTA, YASUZAWA, I., TAKAHASHI, NAKANO,H. (1990), Duocarmycin SA, a new antitumor antibiotic from Streptornyces sp., J. Antibiot. 43, 1037-1038. ICHIMURA, OGAWA,T., KATSUMATA, TAM., S., KAHASHI, K., TAKAHASHI, NAKANO,H. I., (1991), Duocarmycin S, new antitumor antibiotics produced by streptomycetes: producing organism and improved production, J. Antibiot. 44, 1045-1053. IIDA, KOBAYASHI, YOSHIDA, SANO,H. T., E., M., (1989), Calphostins, novel and specific inhibitors of protein kinase. 11. Chemical structures, J. Antibiot. 42, 1475-1481. IIDA, K., ISHII, T., HIRAMA,M., OTANI,T., MINAMI, Y., YOSHIDA,K. (1993), Synthesis and absolute stereochemistry of the aminosugar moiety of antibiotic C-1027 chromophore, Tetrahedron Lett. 34,40794082. IKEKAWA, ASARI,T., MANABE, UMEJI,M., T., T., YANOMA,S. (1980), Canacelunin, a cancer cell agglutinin from Streptornyces spec., J. Antibiot. 33,776-777. IMOTO,M., UMEZAWA,K., ISSHIKI,K., KUNIMOTO, S., SAWA, TACKEUCHI, UMEZAWA, T., T., H. (1987), Kinetic studies of tyrosine kinase inhibition by erbstatin, J. Antibiot. 40,1471-1473. IMOTO,M., YAMASHITA, SAWA,T., KURASAT., WA, S., NAGANAWA, H., TAKEUCHI, BAOT., QUAN,Z., UMEZAWA,K. (1988), Inhibition of cellular phosphatidylinositol turnover by psi-tectorigen, FEBS Lett. 230, 43-46. IMOTO,M., UMEZAWA,K., TAKAHASHI, NAY., GANAWA, JITAKA, NAKAMURA, KAH., Y., H., M., ZUMI,Y., SASAKI,Y., HAMADA, SAWA,T., TAKEUCHI, (1990), Isolation and structure T. determination of inostamycin, a novel inhibitor of phosphatidylinositol turnover, J. Nat. Prod. 53,825-829. IMOTO, SHIMURA, UMEZAWA, (1991), M., N., K. Inhibition of epidermal growth factor-induced activation of phospholipase C by psi-tectorigen, J. Antibiot. 44,915-917. ISHIDA, MIYAZAKI, KUMAGAI, RIKIMAN., K., K., RU, K. (1965), Neocarzinostatin, an antitumor antibiotic of high molecular weight. Isolation,
6 References
693
physicochemical properties and biological activi- KAHN, GRAF, (1986), Oncogenes and Growth P., T. ties, J. Antibiot. Ser. A 18, 68-76. Control. Berlin, Heidelberg: Springer-Verlag. M., Y., ISHIGAMI, HAYAKAWA, SETO,H. (1994), KAKEYA,H., IMOTO, TAKAHASHI, NAGAK., Y., NAWA,H., TAKEUCHI, UMEZAWA, T., K. Cororubicin, a new anthracycline antibiotic gen(1993), Dephostatin, a novel protein tyrosine erating active oxygen in tumor cells, J. Antibiot. phosphatase inhibitor produced by Strepto47, 1219-1225. myces. 11. Structure determination, J. Antibiot. ISONO,K. (1988), Nucleoside antibiotics: Structure, 46, 17161719. biological activity and biosynthesis, J. Antibiot. KALKENVAN, K., HOEVEN C. VAN DER, J. J. M., 41, 1711-1752. JONGDE, J. et al. (1991), Bepridil in combinaIWAI, NAKAGAWA, SADAKANE, OMUY., A., N., tion with anthracyclines to reverse anthracycline RA, S . (1980), Herbimycin B, a new benzoquinresistance in cancer patients, Eur. J. Cancer 27, oid ansamycin with anti-TMV and herbicidal ac739-744. tivities, J. Antibiot. 38, 1114-1119. T. IWASAKI, (1989), Rhizoxin, an Inhibitor of Tu- KALMAN, I. (1989), Inhibition of thymidylate S. synthase cycle of leukemia cells by antifolates: bulin. Novel Microbial Products for Medicine the usefulness of cellular tritium assay, in: Chem. and Agriculture (DEMAIN, L., SAMKUTI, A. G. Biol. Pteridines Proc. Int. Symp. Pteridines Folic A., HUNTER-CEVERA, ROSSMORE, W., J. C., H. Acid Deriv. 9th (1980) Meeting Date 1989 (CUREds.). Washington, D. C.: Society for Industrial S., N., TINS, H., GHISLA, BLAN, Eds.), pp. 1192Microbiology. 1197. Berlin: DeGruyter. IZUMI, NAKAHARA, GOTOH, HASHIMO- KAMAI, ISSHIKI, UMEZAWA, MORISHIS., K., T., F., K., Y., TO, S., KINO, OKUHARA, ADRI, IMAT., M., V., H., T., MA,H., NAGANAWA, TAKITA, TAKEUNAKA, H. (1983), Antitumor effects of novel imCHI, T., UMEZAWA, (1985), Vanoxonin, a H. munoactive peptides, FK-156, and its synthetic new inhibitor of thymidylate synthetase. 11. derivatives, J. Antibiot. 36, 566-574. Structure determination and total synthesis, J. JACHOWSKI, REITENMEIR, W., SHERR, J., S., C. C. Antibiot. 38, 31-38. ROCK,C. 0. (1986), A guanine nucleotide-de- KAMEI, OKA,M., HAMAGISHI, TOMITA, H., Y., K., pendent phosphatidylinositol 4d-diphosphate KONISHI, M., OKI, T. (1990), Piperafizines A phospholipase C in cells transformed by the Vand B, potentiators of cytotoxocity of vincristine, fms and V-fes oncogenes, J. Biol. Chem. 261, J. Antibiot. 43, 1918-1920. 4978-4985. KAMEI, NISHIYAMA, TAKAHASHI, OBI, H., Y., A., JENSEN, R., FENICAL, (1994), Strategies for P. W. Y., OKI,T. (1991), Dynemicins, new antibiotics the discovery of secondary metabolites from mawith the 15-diyn-3-ene and anthraquinone subrine bacteria: ecological perspectives, Annu. unit. 11. Antitumor activity of dynemicin A and Rev. Microbiol. 48, 559-584. its triacetyl derivative, J. Antibiot. 44, 1306JOEL,P. S. (1994), Taxol and taxotere: from yew 1311. tree to tumor cell, Chem. Ind. (London) ( S ) , KAMISHOHARA, KAWAI, ODAGAWA, M., H., A,, 172-175. ISOE, MOCHIZUKI, UCHIDA, HAYAT., J.4, T., JOHDO, TONE,H., OKAMOTO, YOSHIMOO., R., KAWA, Y., SETO,H., TSURUO, OTAKE, T., N. H., T., TO, A., NAKAGAWA, SAWA, TAKEUCHI, (1994), Antitumor activity of SPM VIII, a deriT. (1991a), Anthracycline metabolites from vative of the nucleoside antibiotic spicamycin, Streptomyces violaceus A262. V. New anthracyagainst human tumor xenografts, J. Antibiot. 47, cline alldimycin A: a minor component isolated 1305-1312. from obelmycin beer, J. Antibiot. 44, 1160- KANEDA, YOUNG, c., YAZAWA, TAKAHAs., K., 1164. SHI, K., MIKAMI, ARAI, (1987), Biological Y., T. activities of newly prepared saframycins, J. AntiJOHDO, WATANABE, ISHIKURA, YOSHIO., Y., T., biot. 40,1640-1642. MOTO, A., NAGANAWA, H., SAWA, TAKEUT., CHI, (1991b), Anthracycline metabolites from KANEKO, WONG, UTRIG, SCHURIG, T. T., H., J., J., Streptomyces violaceus A262. 11. New anthracyDOYLE, W. (1990), Water soluble derivatives T. cline epelmycins produced by a blocked mutant of rebeccamycin, J. Antibiot. 43, 125-127. strain SU2-730, J. Antibiot. 44,1121-11. KAO,J. W. Y., COLLINS, L. (1989), A rapid in J. vitro screening system for the identification and JOHDO, TONE,H., OKAMOTO, YOSHIMOO., R., evaluation of anticancer drugs, Cancer Invest. 7, TO, A., NAGANAWA, SAWA, TAKEUCHI, H., T., 303-31 1. T. (1991c), Anthracycline metabolites from Streptomyces violaceus A262. IV. New anthracy- KARWOWSKI, P., JACKSON, SUNGA,G., J. M., cline yellamycins produced by a variant strain SHELDON, PODDIG,J. B., KOHL, W. L., P., SC-7, J. Antibiot. 44,1155-1159. KADAM, (1994), Dorrigocins: novel antifungal S.
694
B. antibiotics that change the morphology of ras- KELLEY,S. L., BASU,A., TEICHER, A., HACKtransformed N/H/3T3 cells to that of nomal cells. ER, M. P., HAMER,P. H., LAZO, J. S. (1988), I. Taxonomy of the producing organism, J. AntiOverexpression of metallothionein confers rebiot. 45, 862-689. sistance to anticancer drugs, Science 241(4874), 1813-1815. KASAI,M., KONO,M., SHIRAHATA, (1991), The K. derivation of a novel mitomycin skeleton: 3a-al- KHAN, S. H., MALTA, K. L. (1992), Recent Adkoxymitomycin, J. Antibiot. 44,301-308. vances in the Development of Potential Inhibitors of Glycosyltransferase. Glycoconjugates (ALLEN, KATAGIRI, (1975), Quinoxaline, in: Antibiotics, K. E. H. J., KISAILUS, C., Eds.), pp. 361-378. New Vol. 3, Mechanism of Action of Antimicrobial York: Marcel Dekker. and Antitumor Agents (CORCORAN, J. W., B. A. HAHN, F. E., Eds.). Berlin, Heidelberg, New KHOKHLOV, S., CHERCHES, Z., RESHETOV, York: Springer-Verlag. P. D., SMIRNOVA, M., SOROKINA, B., PROG. I. KOPTZEWA, T. A., KOLODITSKAYA, T. A., Y., T., S., KATO, Y., OSAWA, IMADA, IWASAKI, S. I. SMIRNOV, V., NAVASHIN, M., FOMINA, V. SHIMAZAKI, KOBAYASHI, KOMAI,T. N., T., P. (1969), Physico-chemical and biological stud(1991), Studies on macrocyclic lactone antibioties on actinoxanthin, an antibiotic from Actinoics. XIII. Anti-tubulin activity and cytotoxicity of myces globisporus 1131, J. Antibiot. 22, 541rhizoxin derivates: synthesis of a photoaffinity 544. derivate, J. Antibiot. 44,66-75. R., KAWADA, YAMASHIDA, UOSAKI,Y., GOMI, KHOSRAVI-FAR, Lox, A. D., KATO, K., DER, S., Y., C. J. (1992), Protein prenylation: key to ras funcT., H. K., IWASAKI, TAKIGUCHI, NAKANO, T., tion and cancer intervention, Cell Growth Differ. (1992a), UCT4B, a new antitumor antibiotic 3, 461469. with topoisomerase I1 metiated DNA cleavage H., T. activity, from Streptomyces sp., J. Antibiot. 45, KIKUCHI, TETSUKA, (1992), On the mecha1182-1184. nism of lipoxygenase-like action of bleomyciniron complexes, J. Antibiot. 45, 548-555. K., KAWADA, SUMI,S., UMEZAWA, INONYE, M., S., SAWA,T., SETO, H. (1992b), Circumvention KIM,Y. J., FURIKATA, SHIMAZU, FURIKAK., A., of multidrug resistance in human carcinoma KB TA, K., SETO,H. (1991), Isolation and structural cells by polyether antibiotics, J. Antibiot. 45, elucidation of sekothrixide, a new macrolide ef556-562. fective to overcome drug-resistance of cancer cells, J. Antibiot. 44,1280-1282. KAWASE, Y., TAKAHASHI,M., TAKATSU, T., ARAI, M., NAKAJIMA,M., TANZAWA,K. KINGSTON, G. I. (1994), Taxol: the chemistry D. (1996), A-72363 A-1, A-2, and C, novel heparaand structure-activity relationship of a novel annase inhibitors from Streptomyces nobilis SANK ticancer agent, Trends Biotechnol. 12,222-227. 60192. 11. Biological activities, J. Antibiot. 49, KITAHARA, ISHII, K., KUMADA, SHIRAIY., M., 61-64. SHI,T., FURUTA,T., MIWA,T., KAWAHANADA, KAWASHIMA, NAKAGAWA, HAYAKAWA, H., WATANABE, (1985), 7-hydroxyguanine, a A., M., K. Y., KAWAI,H., SETO, H., OTAKE,N. (1983), novel antimetabolite from a strain of StreptoStudies on differentiation inducers of myeloid myces purpurasceus. I. Taxonomy of the producleukemia cells. 11. Citrinin, a new inducer of the ing organism, fermentation, isolation and biologdifferentiation of M1 cells, J. Antibiot. 36, 173ical activity, J. Antibiot. 38, 972-978. 174. KIWASHIMA, ITO, F., KATO, T., NIWANO,M., J., KAWATSU, YAMASHITA, OSONO,M., ISHIM., T., H., M. KOSHINO, URAMOTO, (1994), Antitumor ZUKA, M., TAKEUCHI, (1993), T cell activaT. activity of heptelidic acid chlorohydrin, J. Antition by conagenin in mice, J. Antibiot. 46, 1687biot. 47, 1562-1563. 1691. KLEGEMANN, E. (1993), Use of bacterial immuM. KAWATSU,M., YAMASHITA, ISHIZUKA,M., T., nomodulators for the treatment of cancer, in: TAKEUCHI, (1994), Effect of conagenin of T. Recent Adv. Pharm. Ind. Biotechnol. Minutes thrombocytopenia induced by antitumor agents Int, Pharm. Technol. Symp., 6th (1993), Meeting in mice, J. Antibiot. 47, 1123-1129. Date 1992 (HINCAL,A. A., KAS, S. H., Eds.), pp. 179-195. Paris: Sante. KEFRI, K., KONDO, H., YOSHINORI, ARAKAT., WA, H., NAKAGAMA, SATOH,F., KAWAMU- KNUGH,T. R., HOOK, J. W. III., BALAKRISHNAN, S., RA, K., OKURA,A., SUDA,H., OKANISHI, M. M. S., CHEN.F. M. (1980), Spectroscopic studies (1991), A new antitumor substance, BE-13793c of actinomycin and ethidium-complexes with produced by a streptomycete. Taxonomy, ferdeoxyribonucleic acids, in: Nucleic Acids Geommentation, isolation structure determination and etry and Dynamics (SARMA,R., Ed.), pp. 351biological activity, J. Antibiot. 44,723-728. 366. Elmsford, N Y Pergamon Press.
6 References
695
KOAMURA, YOSHIMURA, MATSUMOTO, M., Y., K., Y. TERUI, (1988), New amedic acid antibiotics. 11. Structure elucidation, J. Antibiof. 41, 68-72. KOBAYASHI, TSUDA, ISHIBASHI, SHIGEJ., M., M., MORI, H., YAMASHI, HIROTA, T., H., SAZAKI, T. (1993), Amphidinolide F, a new cytotoxic macrolide from the marine dinoflagellate Amphidinum spec., J. Org. Chem. 58, 2645-2646. KOGIRI, ARAKAWA, SATOH, KAWAMUK., H., F., RA, K., OKURA, SUDA, OKANISHI, A., H., M. (1991), New antitumor substances, BE-12406 A and BE-12406 B produced by a streptomycete. I. Taxonomy, fermentation, isolation, physicochemical and biological properties, J. Antibiot. 44,1054-1060. KOMAGATA,D., SAWA, T., MURAOKA,Y., T. IMADA, OKAMI, TAKEUCHI, (1992a), C., Y., TA-3037A, a new inhibitor of glutathione Stransferase, produced by actinomycetes. I. Production, isolation, physio-chemical properties and biological activities, J. Antibiot. 45, 11171122. D., SAWA, R., KINOSHITA, N., KOMAGATA, IMADA,C., SAWA,T., NAGANAWA, HAH., MADA, OKAMI, TAKEUCHI, (1992b), M., Y., T. Isolation of glutathione transferase inhibitors, J. Antibiot. 45, 1681-1683. KOMIYAMA, EDANAMI, TANOH, YAMAK., K., A., I. MOTO, UMEZAWA, (1983), Studies on the H., biological activity of stubomycin, J. Antibiot. 36, 301-311. KOMIYAMA, FUNAYAMA, ANRAKU, K., S., Y., Y., S., MITO,A., TAKAHASHI, OMURA, SHIMASAKI, H. (1989), Isolation of isoflavonoids possessing antioxidant activity from the fermentation broth of Streptoymces sp., J. Antibiof. 42, 1344-1349, 1350-1355. KOMIYAMA, TAKAMATSU, TAKAHASHI, K., S., Y., SHINOSE, HAYASHI, TANAKA, IWAI, M., M., H., S. Y., OMURA, (1993), New inhibitors of melanogenesis OH-3984 K1 and K2. I. Taxonomy, fermentation, isolation, and biological activities. 11. Physico-chemical properties and structural elucidation, J. Anfibiof. 46, 1520-1525, 15261529. KONISHI, OHKUMA, MATSUMOTO, SAIM., H., K., TOH, K., MIGAKI, O K I , T., KAWAGUCHI, T., H. (1991). Dynemicins, new antibiotics with the 1,5diyn-3-ene and anthraquinone subunit. I. Production, isolation and physico-chemical properties, J. Antibiot. 44, 1300-1305. M., K., KONO,M., KASAI, SHIRAHATA, HIRAYAMA, N. (1991), The configuration of mitiromycin and its derivation from mitomycin B, J. Antibiot. 44, 309-315. KONO,M., KASAI, SHIRAHATA, HIRAYAM., K., MA, N. (1995), Isolation of albomitomycins from
a solution from a solution of 7-amino substituted mitomycins; mitomycin C and KW-2149, J. Antibiot. 48, 179-181. KOPKA, L., YOON, C., GOODSELL, PIMA, M. D., P., DICKERSON, E. (1985), The molecular origin R. of DNA-drug specificity in netropsin and distamycin, Proc. Natl. Acad. Sci. USA 82, 1 3 7 6 1380. KOSHINO, TAKAHASHI, OSADA, ISONO, H., H., H., K. (1992), Reveromycins, new inhibitors of eukaryotic cell growth. 111. Structures of reveromycins A, B, C and D. Inhibition of the mitogenic activity of epidermal growth factor, J. Anfibiot. 45, 1420-1427. o., KOURANY-LEFOLL, CAPREROTE, SEVEE., NET, T., MONTAGNAC, PAIS,M. (1994), PulA., veodictines Al-A7, and Cl-C2, antibiotic and cytotoxic guanidine alkaloids from the new caledonian sponge Sphleodictyon sp., Tetrahedron 11,3415-3426. KRUEGER, X., STREULI, SAITO,H. (1990), N. M., Structural diversity and evolution of human receptor-like protein tyrosine phosphatases, E M B O J. 9,3241-3252. KRUGER, C., PSCHIGODA, M., MOSCOWITZ, W. L. A. (1986), The binding of the antitumor antibiotic chartreusin to poly(dA-dT)-poly(dA-dT), poly(dG-dC)-poly(dG-dC) calf thymus DNA, transfer RNA and ribosomal RNA, J. Antibiot. 39, 1298-1303. KUBOHARA, OKAMOTO, TANAKA, AsY., K., Y., AHI,K., SAKURAI, TAKAHASHI, (1993), A., N. Differanisole A, an inducer of the differentiation of Friend leukemic cells, induces stalk cell differentiation in Dictyostelium discoideum, FEBS Lett. 322, 73-75. K., KURAMOCHI, MOTEGI,A., TAKAHASHI, H., TAKEUCHI, (1988), DNA cleavage activity of T. ablomycin (NK313), a novel analog of bleomycin, J. Antibiot. 41, 1846-1853. KURAMOCHI-MOTEGI, KURAMOCHI, TAA., H., KAHASHI, TAKEUCHI, (1991), Cell killing K., T. mode of liblomycin (NK313), a novel dose-survival relationship different from bleomycins, J. Antibiot. 44, 429-434. KURISAKI, MAGAE, ISONO, NAGAI, T., J., K., K., YANASAKI, (1992), Effects of tautomycin, a M. protein phosphatase inhibitor, on recycling of mammalian cell surface molecules, J. Antibiot. 45,252-257. LAATSCH, (1992), A fungal polysaccharide exhiH. biting antitumor activity, PharmuZ 21, 159-166. LAM,Y. K. T., WICHMANN, F., HEINZ,M. S., C. GUARIGLIA, GIACOBBE, A., MOCHALES, L., R. S., KONG, L., HONEYCUTT, S., ZINK, D., S. BILLS, G. F., HUANG, BURG,R. W., MONL., AGHAN, L., JACKSON, REID, G., MACR. R.,
696
14 Biotechnical Drugs as Antitumor Agents MAEDA, J. (1993), Melastin, a novel product of Streptomyces that selectively inhibits leukemia cell growth, Biosci. Biotech. Biochem. 57, 969972. MAEDA,J., YAZAWA, MIKAMI, ISHIBASHI, K., Y., J. M., KOBAYASHI, (1992), The producer and biological activities of SO-075R1, a new mutactimycin group antibiotic, J. Antibiot. 45, 14481852. MAGAE,J., WATANABE, OSADA,H., CHENG, C., X., ISONO,K. (1988), Induction of morphological changes of human myeloid leukemia and activation of protein kinase by a novel antibiotic, tautomycin, J. Antibiot. 41, 932-937. MAGAE, J., HINO, A., ISONO, K., NAGAI, K. (1992), Respiratory burst induced by phorbol ester in the presence of tautomycin, a novel inhibitor of protein phosphatases, J. Antibiot. 45,246251. MAGAE,J., TSUJI,R. F., WANG,Z., KATAOKA, T., LEE, M.-H., HANADA, KURISAKI, URAT., T., MOTO, M., YAMASAKI, NAGAI,K. (1993). M., Melastin, a novel product of Streptomyces that selectively inhibits leukemia cell growth, Biosci. Biotech. Biochem. 57, 969-972. MAGNUS, BENNEIT,F. (1989), Synthetic studies P., on the esperamicin/calicheamicin antitumor antibiotics, Tetrahedron Lett. 30,3637-3641. MANGER,A. B. (1980), The actinomycins, Top. Antibiot. Chem. 5,223-306. MARTIN,D. G., CHICHESTER, G., DUCHAMP, C. D. J., MISZAK,S. A. (1980), Structure of CC1065 (NSC-298223), a new antitumor antibiotic, J. Antibiot. 33, 902-903. MARTY, M., EXTRA,J. M., GIACCHETTI, CuS., VIER,C., ESPIE, M. (1994), Taxoids: a new class of cytotoxic agents, Nouv. Rev. Fr. Hematol. 36 (Suppl. l), 525-528. MATSHURA, (1988), Photochemical and oxidaT. tive strand cleavage of DNA, Stud. Org. Chem. Amsterdam 88,353-366. MATSON, A., COLSON,K. L., BELOFSKY, N., J. G. BLUMBERG, B. (1993), Sandramycin, a novel B. antitumor antibiotic produced by a Nocardioides sp. 11. Structure determination, J. Antibiot. 46, 162-166. MATSUMOTO, SETO, H. (1991), Stimulation of M., mammalian cell proliferation by lavanducyanin, J. Antibiot. 44,1471-1473. MATSUMOTO,M., MATSUTANI,S., SUGITA, K., YOSHIDA,H., HAYASHI, TERN], Y., NAKAI, F., H., NOTANI, KAWAMURA, MATSUMOTO, N., Y., K., SHOJI,J., YOSHIDA, (1992), Depudecin: a T. novel inhibitor of NIH3T3 cells doubly transformed by ras- and src-oncogene, produced by Alternaria brassicola, J. Antibiot. 45, 879-885. MATSUURA,N., ONOSE, R., OSADA, H. (1996), Morphology reversion activity of phosmidosine
GUIRE, J. J., MCKNIGHT,A. T., RAGAN,C. I. (1992), A novel phosphoinositol monophosphatase inhibitor from Memnoniella echinata, J. Antibiot. 45, 1397-1404. LAM, K. S., VEITCH,J. A., GOLIK,J., KRISHNAN, S., B., KLOHR, S. E., VOLK, K. J., FORENZA, DOYLE,T. W. (1993), Biosynthesis of esperamicin A,, an enedyine antitumor antibiotic, J. Am. Chem. SOC. 115, 12340-12345. LAN, K. H. W., FARLEY,J. R., BAYLINK, J. D. (1989), Phosphotyrosyl protein phosphatase, Biochem. J. 257,23-36. LEE, M. D. (1992), Calicheamicins, a novel family of antitumor antbiotics. 4. Structure elucidation of calicheamicins @Br, ylBr, J. Am. Chem. SOC. 114,985-997. LEE, M. D., MANNING, K., WILLIAMS, R., J. D. KUCK, N. A., TESTA, R. T., BORDERS,D. B. (1989), Calicheamicins, a novel family of antitumor antibiotics. 3. Isolation, purification, and characterization of calicheamicins, J. Antibiot. 42,1070-1087. LEET, J. E. (1992), Kedarcidin, a new chromoprotein antibiotic: structural elucidation of the kedarcidin chromophore, J. Am. Chem. SOC. 114, 7946-7948. LEET, J. E., SCHROEDER, R., GOLIK,J., MATD. SON, J. A., DOYLE,T. W., LAM, K. S., HILL, S. E., LEE, M. S., WHITNEY, L., KRISHNAN, S. J. B. (1996), Himastatin, a new antitumor antibiotic from Streptomyces hygroscopicus. 111. Structural elucidation, J. Antibiot. 49, 299-311. LEVITZKI, (1994), Protein tyrosine kinase inhibA. itors, in: New Molecular Targets in Cancer CheP., motherapy (KERR, D. J., WORKMAN, Eds.), pp. 67-79. Boca Raton, FL: CRC Press. LIU, W. C., BARBACID, BULGAR, CLARK, M., M., J. M., ROSSWELL, R., DEAN,L., DOYLE, T. A. W., FERNANDES, B., HUANG, MANNE, P. S., V., PIRNITZ, M., WELLS, J. S., MYERS,E. (1992), D. 10-Desmethoxystreptonigrin,a novel analog of streptonigrin, J. Antibiot. 45, 454-457. LOOMIS,C. R., BELL, R. M. (1988), Sangivamycin, a nucleoside analogue, is a potent inhibitor of protein kinase C, J. Biol. Chem. 263, 16821692. LOWN,J. W. (Ed.) (1988), Anthracyclines and Anthracene-Dione-Based Anticancer Agents. Amsterdam, New York Elsevier. LOWN,J. W. (1993), Anthracycline and anthraquinone anticancer agents: current status and recent developments, Pharmacol Ther. 60,185-214. LYBRAND, P. (1988), Interaction of peptide antiT. biotics with DNA. Peptides. Chemistry and biology, Proc. Amer. Pept. Symp. loth, meeting date 1987 (MARSHALL,G. R., Ed.), pp. 416419. Leiden: ESCOM Sci. Publ.
6 References and phosmidoxine B, a newly isolated derivative, on src-transformed NRK cells, J. Antibiot. 49,361-365. MAYER,A. B., KATZ, E. (1978), Actinomycins, J. Chromatogr. Libr. 15 (Antibiotics: isolation, separation, purification), 1-33. J. M., MCALPINE, B., KARWOWSKI, P., JACKSON, J. J. E., MULLALY, M., HOCHLOWSKI, PREMAM. CHANDRAN, BURRES, S. (1994), MLR-52, U., N. (4' - demethylamino - 4 ',5'- dihydroxy - staurosporine), a new inhibitor of protein kinase C with immunosuppressive activity, J. Antibiot. 47,281286. MIYOSHI-SAITOH, MORISAKI, TOKIWA, M., N., Y., IWASAKI, KONISHI, SAITOH,K., OKI,T. S., M., (1991), Dynemicins 0, P and Q: novel antibiotics related to dynemicin A. Isolation, characterization and biological activity, J. Antibiot. 44, 10371044. MORI,K. (1989), Synthesis of differanisol A, an inducer of differentiation, Liebigs Ann. Chem. 3, 303-305. D. MORRIS,S. J., THURSTON, E., NEVELL,T. G. (1990), Evaluation of the electrophilicity of pyrrolo[2,1-c][1,4]benzodiozeDNA-binding pines by HPLC, J. Antibiot. 43, 1286-1294. MULLER,R. (1986), Proto-oncogenes and differentiaton, TZBS 11,129-132. MURADA, (1988). Semisynthetic bleomycins, in: Y. Horizons on Antibiotic Research (DAVIES,B. D., ISHIKAWA, MAEDA, K., MITSCHER,L. T., A., Eds.), pp. 88-101. Toyko: Japan Antibiotics Association. MURAMATSU, ABE, S., HAYASHI, YAMAR., H., GUCHI,K., JINDA, K., SAKANO, NAKAMUK., RA, S. (1991), Complete amino acid sequence of phenomycin, an antitumor polypeptide antibiotic, J. Antibiot. 44, 1222-1227. MURAOKA, SAITO, S., NOGAMI,T., UMEZAY., WA, K., TAKITA, TAKEUCHI, UMEZAWA, T., T., H., SAKATA,N., HORI, M., TAKAHASHI, K., EKIMOTO, MINAMIDES, NISHIKAWA, H., S., K., KURAMOCHI, MOTEGI, A., FUKUOKA, H., T., A., A. NAKATANI, FUJII, MATSUDA, (1988), T., Bleomycin: structure-activity relationship in semisynthetic bleomycin, in: Horizons on AntiT., biotic Research (DAVIES,B. D., ICHIKAWA, MAEDA, K., MITSCHER,L. A., Eds.), pp. 88101. Tokyo: Japan Antibiotics Research Association. H., MURATA,M., MIYASAKA, TANAKA, OMUT., RA, S. (1985), Diazoquinomycin A, a new antifolate antibiotic inhibits thymidylate synthase, J. Antibiot. 38, 1025-1033. MURATA, TANAKA, OMURA, (1987), 7M., H., S. hydro-8-methyl-pteroylglutaminyl-glutamic acid, a new antifolate from an actinomycete, J. Antibiot. 40,251-257.
697
MUROI,M. (1990), Angiogenesis inhibitors TAN1323 C and D manufacture with Streptomyces, Jpn. Patent J P 03290193. MURRAY, SIM,A. T. R., POSTAS,J. A. P., CAL., PON, R. J. (1993), Isopalimurin: a mild protein phosphatase inhibitor from a Southern Australian marine sponge Dysidea sp., Aust. J. Chem. 46,1291-1294. NAGAMURA, SATOH, M., IMURA,N. (1993), A., Metallothionein 111 (International Conference of Metallothionein, 3rd) Meeting Date 1992 (SUZUKI, K. T., IMURA, KIMURA, Eds.). Basel: N., M., Birkhauser. NAKANISHI, (1994), MS-282a and MS-282b, new S. inhibitors of calmodulin-activated myosin light chain kinase from Streptomyces tauricus ATCC 27470, J. Antibiot. 47, 855-861. NAKANISHI,S., MATSUDA, Y., IWAHASHI, K. KOSE, H. (1986), K-252b, c and d, potent inhibitors of protein kinase C from microbial origin, J. Antibiot. 39, 1067-1071. NAKANO, KOBAYASHI, TAKAHASHI, H., E., I., TAMAOKI, K u z u u , Y., IBA, H. (1987), StauroT., sporine inhibits tyrosine-specific protein kinase activity of Rous sarcoma virus transforming protein p60, J. Antibiot. 40,706-708. NAKATA,M., SAITO, M., INOUYE,Y., NAKAMURA, S., HAYAKAWA, SETO, H. (1992), A new Y., anthracycline antibiotic, cinerubin R. Taxonomy, structural elucidation and biological activity, J. Antibiot. 45, 1599-1608. O., N., NAKAYAMA, SHIGEMATSU, KATAYAMA, M., A., TAKASE,S., KIYOTO,S., HASHIMOTO, KOHSAKA, (1989), WS-9659A and B, novel M. testosteron 5a reductase inhibitors isolated from a Streptomyces. 11. Structural elucidation of WS9659 A and B, J. Antibiot. 42, 1230-1234. NATORI,T., MORITA,M., AKIMOTO, KOEZUK., KA,Y. (1994), Agelasphins, novel antitumor and immunostimulatory cerebrosides from the marine sponge Angelus induritianus, Tetrahedron 50,2771-2784. NEER, E. J., CHAPHAM, E. (1988), Roles of G D. protein subunits in transmembrane signalling, Nature 333, 129-134. M. NEIDLY,S., WARING, J. (Eds.) (1983), Molecular Aspects of Anticancer Drug Action. Weinheim: Verlag Chemie. M., NISHIKAWA, SHIBASAKI, HIRATSUKA, K., C., ARAKAWA, TAKAHASHI, TAKEUCHI, M., K., T. (1991a), Antitumor spectrum of deoxyspergualin and its lack of cross-resistance to other antitumor antibiotics, J. Antibiot. 44, 1201-1209. NISHIKAWA, SHIBASAKI, UCHIDA,T., TAK., C., KAHASHI, TAKEUCHI, (1991b), The naK., T. ture of in vivo cell-killing of deoxyspergualin and its replication in combination with other antitumor agents, J. Antibiot. 44,1237-1246.
698
S., M., NISHIKORI, HIRUMA, KUROHAWA, SAI- OHSHIMA, YANAGISAWA, KATOH, A., FuT., S., T., JII, T., SANO, T., MATSUKAMA, FURUMAI, S., N. TO, S., SHIMADA, (1992), Microbial producK., K., T., FUJIU,M., WATANABE, YAHOSE, tion of 4-thymidine, J. Antibiot. 45, 1376-1377. ARISAWA, OKUDA, (1994), Fusarium M., T. NISHIMURA, NISHIMURA, TANAKA, SuC., T., N., merismoides Corola Nr. 6356, the source of the ZUKI, (1987), Potentiation of the cytotoxicity H. protein kinase C inhibitor, azepinostatin. Taxonof peplomycin against Ehrlich Ascites Carcinoomy, yield improvement, fermentation and bioma by bleomycin hydrolase inhibitor, J. Antibiot. logical activity, J. Antibiot. 47, 639-647. 40,1794-1795. T., M., H., NISHIOKA, (1994), Antitumor effect of piericid- OIKAWA, SHIMAMURA, ASHINO, NAH. KAMURA, 0. (1992), Inhibition of angiogenesis in B, N-oxide inhibitors of phosphatidylinositolby staurosporine, a potent protein kinase inhibiphosphate turnover, J. Antibiot. 47, 441452. NISHIOKA,H., SAWA, ISSHIKI, TAKAHASHI, tor, J. Antibiot. 45, 1155-1160. T., K., M., T., H., Y., NAGANAWA, MATSUDA, HATTORI, OIKAWA, ASHINO, SHIMAMURA,HASEH., N., I., S., GAWA, MORITA, MUROTA, ISHIZUNM., S., HAMADA, TAKEUCHI, UMEZAWA, M., T., H. KA,M., TAKEUCHI, (1993), Inhibition of anT. (1991), Isolation and structure determinaton of a giogenesis by erbstatin, an inhibitor of tyrosine novel phosphatidylinositol turnover inhibitor, kinase, J. Antibiot. 46, 785-790. piericidin B1 N-oxide, J. Antibiot. 44, 1283Y., K., OKA,M., KAMEI,H., HAMAGISHI, TANITA, 1285. MIGAKI, KONISHI, OKI, (1990), ChemT., M., T. NOBLE, L. (1990), The discovery of the vinca alR. ical and biological properties of rubiginone, a kaloids - chemotherapeutic agents against cancomplex of new antibiotics with vincristine-cytocer, Biochem. Cell. Biol. 68, 1344-1351. toxicity potentiating activity, J. Antibiot. 43,967NOZAKI,Y., HIDA, IINUMA, ISHII, SUDO, T., S., T., 976. M., T. K., MUROI, KANAMARU, (1993), TANOKA, YOSHINARI, MURAI, KAWAMURA, H., T., T., 1120, a new anthracycline with potent angiostatK., SATOH,F., FUNAISHI, OKURA,A., K., ic activity, J. Antibiot. 46, 569-579. SUDA, OKANISHI, SHIZURI, (1991), A H., M., Y. ODAI, SHINDO, ODAGAWA, MOCHIZUH., K., A., new topoisomerase-I1 inhibitor, BE-10988, proJ., M., T. KI, HAMADA, TAKEUCHI, (1994), Inosduced by a streptomycete. I. Taxonomy, fermentamycins B and C, new polyether antibiotics, J. tation, isolation and characterization,J. Antibiot. Antibiot. 47, 939-941. 44,486491. OGAWARA, UCHINO, AKIYAMA, WA- OKADA, SUZUKI, YOSHINARI, ARAKAH., K., T., H., H., T., TANABE, (1985), A new 5-nucleotidase inS. H., A., WA,H., OKURA, SUDA, YAMADA, A., hibitor, nucleoticidin. I. Taxonomy, fermentaD. UEMURA, (1994), A new topoisomerase I1 intion, isolation and biological properties, J. Antihibitor, B E 22179, produced by a streptomycete biot. 38, 153-156. I producing strain, fermentation, isolation and OGAWARA, AKIYAMA, WATANABE, H., T., S., biological activity, J. Antibiot. 47, 129-135. Iso, N., KOBORI, SEODA, (1989a), Inhibi- OKURA, NAKADAIRA, NAITO, ISHIZUM., Y. A., M., K., tion of tyrosine protein kinase activity by synKA,M., TAKEUCHI, UMEZAWA, (1986), T., H. thetic isoflavones and flavones, J. Antibiot. 42, Effect of forphenicinol on y-interferon produc340-343. tion in mice sentisized with BCG, J. Antibiot. 39, OGAWARA, HASUMI, HIGASHI, ISHII, H., Y., K., 569-580. Y., SAITO, WATANABE, SUZUKI, Ko- OMURA, IMAMURA, KUGA, ISHIKAWA, T., S., K., S., N., H., BORI, M., TANAKA, (1989b), AcetoxycycloK. H., YAMAZAKI, OKANO, KIMURA, Y., K., K., heximide and cycloheximide convert transTAKAHASHI, TANAKA, (1985), AdechloY., H. formed morphology of ras-transformed cells to rin, a new adenosine deaminase inhibitor connormal morphology, J. Antibiot. 42, 1530-1533. taining chlorine. Production, isolation, properOGAWARA, HIGASHI, MANITA,S., TANAH., K., ties, J. Antibiot. 38, 1008-1015. KA, K., SHIMIZU, SHUFANG, (1992), An Y., L. OMURA, TANAKA, KUGA, IMAMURA, S., H., H., N. inhibitor for inositol-specific phospholipase C (1986), Adecypenol, a unique adenosine defrom Actinomadura sp., J. Antibiot. 45, 1365aminase inhibitor containing homopurine and 1367. cyclopentene rings, J. Antibiot. 39, 309-310, OGAWARA, HIGASHI, MACHIDA, TAH., K., T., 1219-1224. J., N., T. KASHIMA,CHIBA, MIKAWA, (1994), In- OMURA, VANDER PYL,D., INOKASHI, TAS., J., hibitors of diacylglycerol kinase from Drechslera KAKASHI, Y., TAKESHIMA, (1993), PepticinH. sacchari, J. Antibiot. 47,499-501. namins, new farnesyl-protein transferase inhibitors produced by an actinomycete. I. Producing OHNO,M. (1989), From natural bleomycins to strain, fermentation, isolation and biological acman-designed bleomycin, Pure Appl. Chem. 61, tivity, J. Antibiot. 46, 222-228. 581-584.
6 References
699
OMURA, INOKOSHI, UCHIDA, SHIOMI, S., J., R., K., MASUMA,R., KAWAKUBO, TANAKA, T., H., S. IWAI, Y., KOSEMURA,S., YAMAMURA, (1996), Andrastins A-C, new protein farnesyltransferase inhibitors produced by Penicillium sp. FO-3929. I. Producing strain, fermentation, isolation, and biological activities, J. Antibiot. 49, 414-417. OSADA, MAGAE, WATANABE, ISONO, H., J., C., K. (1988), Rapid screening method for inhibitors of protein kinase C, J. Antibiot. 41,925-931. H., K., OSADA, TAKAHASHI,TSUNODA, KUSAH., KABE, H., ISONO, (1990), A new inhibitor of K. protein kinase C, RK-286C (4 -Demethylamino4-hydroxystaurosporin). I. Screening, taxonomy, fermentation and biological activity, J. Antibiot. 43, 163-167. OTANI, (1993), Conformation studies and assessT. ment by spectral analysis of the protein-chromophore interaction of the macromolecular antitumor antibiotic C-1027, J. Antibiot. 46, 791-802. T., S., H., OTSUKA, TAKASE, TERANO, OKUHARA, M. (1992), New angiogenesis inhibitors WF16775 A, and A2, J. Antibiot. 45, 1970-1973. PAOLETTI, (1993), The localization of topoisoC. merase I1 cleavage sites on DNA in the presence of antitumor drugs, Pharmacol. Ther. 60, 381387. PAQUETTE, A. (1993), Synthetic progress toward L. powerfully cytotoxic taxol and related taxane diterpenes, in: Organic Chemistry: Its State of Art (KISAHUEREK, V., Ed.), pp. 103-115. Basel: M. Helvetica Chimica Acta Publishers. S., PEREZ-SOLER,R., SUGARMAN, Zou, Y., PRIEBE, (1995), Use of drug carries to ameW. liorate the therapeutic index of anthracycline antibiotics, ACS Symp. Ser. 574 (Anthracycline Antibiotics), 300-319. PETTIT, G. R., CHICHACZ, A., GAO, F., HERZ. ALD,c. L., BOYD,M. R. (1993a), Isolation and structure of the human cancer cell growth inhibitors. Spongistatins 2 and 3 from an Eastern Ocean Spongia sp., J. Chem. SOC. Chem. Commun. 1993, 1166-1193. PETTIT,G. R., HERALD, L., CHICHACZ, A., C. Z. GAO,F., SCHMIDT, M., BOYD,M. R. (1993b), Y. Isolation and structure of the powerful human cancer cell growth inhibitors spongistatins 4 and 5 from an African Spirastrella spinispinilifera (Porifera), J. Chem. SOC. Chem. Commun. 24, 1805-1807. PETTIT, R., TAN,R., HERALD, L., CERNY, G. D. R. C., WILLIAMS, D. (1994), Antineoplastic M. agents 277. Isolation and structure of phakellistatin 3 and isophakellistatin 3 from a Republic of Comoras marine sponge, J. Org. Chem. 59, 1593-1595.
PETTIT,G. R., GAO, F., BLUMBERG, M., HERP. ALD, c. L., COLL, J. O., KAMANO, LEWIN, Y., J. J. N. E., SCHMIDT, M., CAAPIUS, C. (1996), Antineoplastic agents 340. Isolation and structural elucidation of bryostatins 16-18, J. Nat. Prod. 59,286-289. PIMENTEL, (1987), Oncogenes. Boca Raton, FL: E. CRC Press. PINEDO, M. (Ed.) (1980), Cancer ChemotheraH. py: The EORT Cancer Chemotherapy. Amsterdam, Oxford: Annual Excerpta Medica. PIRRUNG, C., NAUHANS, K. (1994), Structure M. S. of curacin A, a novel antimitotic, antiproliferative, and bine shrimp toxic natural product from the marine cyanobacterium Lyngbya majuscula, Chemtracts. Org. Chem. 7, 128-129. POITIER, P., GUERITTE-VOEGELEIN, GUENC. F., ARD, D. (1994), Taxoids, a new class of antitumor agents of plant origin: recent results, Nouv. Rev. Fr. Hematol. 36 (Suppl. l), S21-S23. POOT, M., HOEHN,H. (1993), DNA topoisomerases and the DNA lesion in human genetic instability, Toxicol. Lett. 61, 297-308. POVIRK, AUSTIN, J. F. (1991), Genotoxicity C., M. of bleomycin, Mutat. Res. 257(2), 127-143. RAMSDELL, S., PEITIT, R. G. (1986), Three actiJ. vators of protein kinase C, bryostatins, calyculin and phorbol esters show differing specificities of action of CH4 pituitary cells, J. Biol. Chem. 261, 17073-17 080. RANCE, M. J., RUDDOCK,J. C., PACEY, S., M. CULLEN, P., HUANG, H., JEFFERSON, W. L. M. T., MAEDA,H., TONE, J. (1989), UK-63052 complex, new quinomycin antibiotics from Streptomyces bruegensis subsp. taponicus. Taxonomy, fermentation, isolation, characterization and antimicrobial activity, J. Antibiot. 42, 206217. RAO, A. N. (1991), Serine hydroxymethyl transferase: a target for cancer chemotherapy, New T., Trends Biol. Chem. (OZAWA, Ed.), pp. 333340. Tokyo: Japan Sci. SOC.Press. RAUTH,M. A., RAYMOND, S. (1993), Cellular R. approaches to bioreductive drug mechanisms, Cancer Metastasis Rev. l ,153-164. 2 REICHLE, DIDDENS, RASTAITER, BERH., H., J., DEL,W. E. (1991), Resislenzmechanismen maligner Zellen gegenuber Zytostatika, Dtsch. Med. Wochenschr. 116,186-191. S., A. ROCKWELL, SARTOLINI, C., TOMASZ,M., K. KENNEDY, A. (1993), Cellular pharmacology of quinone bioreductive alkylating agents, Cancer Metastasis Rev. 12, 165-176. ROSOWSKY, FORSCH, A,, REICH, V. E., A,, R. FREISHEIM, H., MORAN, G. (1992), Side J. R. chain modified 5-deazafolate and 5-deazafolate analogues as mammalian folylpolyglutamate
700
synthetase and glycinamide ribonucleotide for- SAWA, MATSUDA, UCHIDA, IKEDA, R., N., T., T., myltransferase inhibitors. Synthesis and in vitro SAWA, NAGANAWA, HAMADA, TAT., H., M., biological evaluation, J. Med. Chem. 35, 1578KEUCHI, T. (1991), Dioxamycin, a new benz1588. [alanthraquinone antibiotic, J. Antibiot. 44,396402. ROTHENBERG, L. (1993), Taxol, taxotere and M. D. T. R. other new taxanes, Curr. Opin. Invest. Drugs 2, SCAHILL, A., JENSEN, M., SWENSON, H., 1269-1277. HAKENBUHLER, T., PEKOLD,G., WIERENN. SAE, H., NEMOTO,T., ISHIWATA, OJIKA, H., M., GA,W., BRAHME, D. (1990), An NMR study N. YAMADA, (1993), Isolation structure and synK. of the covalent and noncovalent interactions of thesis of dolastatin D, a cytotoxic cyclic depsiCC-1065 and DNA, Biochemistry 29, 2852peptide from the sea hore Dolabella auricularia, 2860. Tetrahedron Lett. 34, 8449-8452. SCHEID, WEBER, ROETTGERS, TRAUT, W., J., U., K. SAKAI, KAWAI, AONO,M., TAKAHASHI, M., Y., H. (1991), Enhancement of the mutagenicity of (1989), A new activity of herbimycin A: inhibianticancer drugs by the calcium agonists veration of angiogenesis, J. Antibiot. 42, 1015-1017. pamil and fendilihe, Arzneim. Forsch. 41, 901904. SAKATA, KARUBE, TANAKA, HAYASHI, N., T., M., M. E., H., HORI,M., HOTTA,K., HAMADA, (1989), SCHWOB, NASMYTH, K. (1993), CLBS and Nucleotide sequence of the macromomycin apoCLB6, a new pair of B cyclins involved in DNA protein gene and its expression in Streptomyces replication in Saccharomyces cerevisiae, Genes macromomyceticus, J. Antibiot. 42, 1704-1712. Dev. 7, 1160-1175. SAKATA, SUZUKAKE-TSUCHIYA, N., K., MORIYA, SEBTI,S. M., LAZO,J. S. (1988), Metabolic inactiY., HAYASHI, HORI,M., OTANI, NAKAI, H., T., vation of bleomycin analogs by bleomycin hyM., AOYAGI, (1992), Aminopeptidase activity T. drolase, Pharmacol. Ther. 38,321-329. of an antitumor antibiotic, C-1027, J. Antibiot. SEKI-ASANO, OKAZAKI, YAMAGICHI, M., M., T., 45, 113-117. SAKAI, HAMADA, MIZORIE,K. (1994), N., K., SALUZ, P., WIEBAUER, (Eds.) (1995), NatuH. K. Isolation and characterization of new 18-memral products for studying the structure of nucleic bered macrolides FD-891 and FD-892, J. Antiacids and DNA biomolecular complexes, Chapbiot. 47, 1226-1233. R., H., H., ter 3 in: DNA and Nucleoprotein Structure in SEKIZAWA, IINUMA, NAGANAWA, HAvivo. Investigations on ProreidDNA Interactions MADA, M., TAKEUCHI, YAMAZUMI, UMET., J., and Nucleic Acid Structures. Austin, T X Landes ZAWA, (1996), Isolation of novel saquayamiK. Company. cins as inhibitors of farnesyl-protein transferase, M. J. Antibiot. 49,487-490. SANCELME, M., FABRE, S., PRUDHOMME, (1994), Antimicrobial activities of indolocarba- SHIBATA, SATSUMABAYASHI, SANO, H., K., S., KOMIYAMA, K., NAKAGAWA, OMURA, A., S. zole and bis-indole protein kinase C inhibitors, J. Antibiot. 47, 792-798. (1988), Chemical modification of hitachimycin. S. SARTORELLI, C., TOMASZ, ROCKWELL, A. M., Synthesis, antibacterial, cytocidal, and in vivo (1993), Studies on the mechanism of the cytoantitumor activities of hitachimycin derivatives, toxic action of the mitomycin antibiotics in hyJ. Antibiot. 41, 614-623. poxic and oxygenated EMT6 cells, Adv. Enzyme SHINDO, MATSUOKA, KAWAI, (1996a), K., M., H. Regul. 33, 3-17. Studies on cochleamycins, novel antitumor antiSASSE,F., STEINMETZ, H ~ L L E , REICHENH., G., biotics. I. Taxonomy, production, isolation and BACH, H. (1993), Rhizopodin, a new compound biological activities, J. Antibiot. 49, 241-244. from Myxococcus stipitatus (Myxobacteria) SHIOMI, YANG, Y., INOKOSHI, VAN DER K., J., causes formation of rhizopodia-like structures in PYL, D., NAKAGAWA, TAKESHIMA, A., H., animal cell cultures, J. Antibiot. 46, 741-748. S. OMURA, (1993), Pepticinnamins, new farneSATO,K. (1989), Glutathione transferases as marksyl-protein transferase inhibitors produced by an ers of preneoplasia and neoplasia, Adv. Cancer actinomycete. 11. Structural elucidation of peptiRes. 52, 205-255. cinnamin E, J. Antibiot. 46, 229-234. S. J. SATO,Y., MORIMOTO, KINE,A., OKAMURA, SINGH, B., ZINK,D. L., LIESCH, M., GOETZ, A., R. M., M. A., JENKINS, J., NALLIN-OMSTEAD, K., HAMANAKA, KOHNO, KUWANO, R., K., M., SAKATA, (1993), Irsogladine is a potent inhibT. G. R. SILVERMAN, C., BILLS, F., MOSLEY, T., K. G., GIBBS,J. B., ALBERS-SCH~NBERG,LINGitor of angiogenesis, FEBS Lett. 322, 155-158. SATOH,T., NISHIMURA, KONDO,S., TAKEUY., HAM,R. B. (1993), Isolation and structure of chaetomellic acids A and B from Chaetomella CHI,T. (1996), Synthesis and antimetastatic activity of 6-trichloroacetamido and 6-guanidino acutiseta: farnesyl pyrophosphate mimic inhibianalogues of siastatin B (1996), J. Antibiot. 49, tors of ras farnesyl-protein transferase, Tetrahe321-325. dron 49,5917-5926.
6 References
701
SINGH, B., JONES,E. T, GOETZ,M. A., BILES, S. R. G. F., NALLIN-OUSTEAD, JENKINS, G., M., LINGHAM, B., SILVERMAN, C., GIBBS,J. R. K. B. (1994a), Fusidienol: a new inhibitor of ras farnesyl-protein transferase from Fusidium griseum, Tetrahedron. Lett. 35, 46934696. R. SINGH,S. B., NESCH, J. M., LINGHAM, B., GOETZ,M. A., GIBBS,J. B. (1994b), Actinoplanic acid A a macrocyclic polycarboxylic acid which is a potent inhibitor of ras farnesyl-protein transferase, J. Am. Chem. SOC. 116, 1160611607. SINGH, B., ZINK,D. C., NESCH, M., BALL,R. S. J. G., GOETZ,M. A., BOLESSA, A., GIACOBBE, E. R. A., SILVERMAN, C., BILLS, F., PELAEZ, K. G. F., CASCALES, GIBBS, B., LINGHAM, B. C., J. R. (1994c), Preussomerins and deoxypreussomerins: novel inhibitors of ras farnesyl-protein transferase, J. Org. Chem. 59, 6296-6302. E. SLICHENMYER, J., ROWINSKY, K., DONEW. KOWER, R. c., KAUFMANN, H. (1993), The s. current status of camptothecin analogs as antitumor agents, J. Nut. Cancer Inst. 85(4), 271-291. SOLAIMAN, (1988), Bleomycin and its Metal D. Complexes. Metal-Based Anti-Tumour Drugs (GIELEN,M. F., Ed.), pp. 235-256. London: Publisher Freund. SONODA, OSADA, URAMOTO, UZAWA, T., H., M., J., ISONO, (1989), Epiderstatin, a new inhibiK. tor of the mitogenic activity induced by epidermal growth factor. 11. Structure elucidation, J. Antibiot. 42, 1607-1609. s., F., P., STEFANELLI, SPONGA, FERRARI, SOTC., TANI,C., CORTI,E., BRUNATI, ISLAM,K. (1996), Inhibition of myo-inositol monophosphatase ATCC20928 factors A and C. Isolation, physico-chemical characterization and biological properties, J. Antibiot. 49, 611-616. STEINHERZ, STEINHERZ, (1991), Delayed L., P. cardiac toxicity from anthracycline therapy, Pediatrician 18, 49-52. STIERLE, STIERLE, STROBEL, BIGNAMI, A., D., G., G., GROTHANS, (1994), Endophytic fungi of P. Pacific yew (Taxus brevifolia) as a source of tax01, taxanes and other pharmacophores, ACS Symp. Ser. 557 (Bioregulators for Crop Protection and Pest Control), 6477. STREKOWSKI, (1992), Molecular basis for the enL. hancement and inhibition of bleomycin-mediated degradation of DNA and DNA-binding compounds,. Adv. Detailed React. Mech. 2 (Mech. Biol. Importance), 61-109. W. P. N. STROHL, R., BARTEL, L., CONNORS, C., ZHU, C.-B., DOSCH,D. C., BEALE,J. M., JR., FLOSS, G., STUTZMAN-ENGWALL, OTH. K., TEN, s. L., HUTCHINSON, R. (1989), Biosync. thesis of natural and hybrid polyketides by an-
thracycline producing Streptomycetes, in: Genetics and Molecular Biology of Industrial MiC. croorganisms (HERSHBERGER,L., QUEENER, S. W., HEGEMAN, Eds.), pp. 68-84. WashingG., ton: American Society for Microbiology. J. STUBBE, KOZARICH, W. (1987), Mechanism J., of bleomycin-induced DNA degradation, Chem. Rev. 87,1107-1136. S., SUGANUMA, YOSHIZAWA, HIROTA,M., M., H., NAKOYASU, OJIKA, FUJUKI, SUGOM., M., RI, H., WAKAMATSU, YAMADA, K., K., SIGUMURA, (1988), Okadaic acid an additional T. non-phorbol-12-tetradecanoiete-13-acetate-type tumor pormoter, Proc. Natl. Acad. Sci. USA 85, 1768-1771. SUZUKAKE-TSUCHIYA, K., MORIGA, HORI,M. Y., (1989), Induction by herbimycin A of contact inhibition in v-src-expressed cells, J. Antibiot. 42, 1831-1834. SUZUKAKE-TSUCHIYA, K., MORIYA, YAMAZAY., N., KI, K., HORI, M., HOSOKAWA, SAWA,T., C., IINUMA, NAGANAWA, IMADA, HAH., H., MADA, (1990), Screening for antibiotics preM. ferentially active against ras oncogene-expressed cells, J. Antibiot. 43, 1489-1495. K., MORIYA,Y., KAWAI, SUZUKAKE-TSUCHIYA, H., HORI,M., UEKARA, IINUMA, NAGAY., H., NAWA, TAKEUCHI, (1991), Inhibition of H., T. pinocytosis by hygrolidin family antibiotics: Possible correlation with their selective effects on oncogene-expressed cells, J. Antibiot. 4 , 3444 348. M., CARCERERI, D., SUZUKI,H., MENEGAZZI, PRATI, GEROSA, TAMMASI, SCARPA, A., F., M., S., R., A., SORRENTINO, DE PRISCO, LIBONATI, M. (1990), Studies on potential markers in human prostatic carcinoma and on a possible role of poly(ADP4bose) polymerase in carcinogenesis, in: Pathology of Gene Expression (FRATI, L., AARONSON, A., Eds.). New York: Raven S. Press. TAKAHASHI, (1993), Russuphelins B, C, D. E A. and F, new cytotoxic substances from the mushroom Russula subniglicans HONGO, Chem. Pharm. Bull. 41, 17261729. TAKAHASHI, (1994), Adenophostins A and B: M. potent agonists of inositol1,4,5-trisphosphatereceptor produced by Penicillium brevicompaccum. Taxonomy, fermentation, isolation, physico-chemical and biological properties, J. Antibiot. 47, 1643-1647. TAKAHASHI, NAKAMURA, KAMEYAMA, A., H., T., KURASAWA, NAGANAWA, OKAMI, S., H., Y., TAKEUCHI, UMEZAWA, (1987a), BisucaT., H. berin, a new siderophore sensitizing tumor cells to macrophage-mediated cytolysis. 11. Physicochemical properties and structure elucidation, J. Antibiot. 40,1671-1676.
702
TAKAHASHI, KOBAYASHI, ASANO,K., YoI., E., SHIDA, NAKANO, (1987b), UCN-01, a seM., H. lective inhibitor of protein kinase C from Streptomyces, J. Antibiot. 40,1782-1784. TAKAHASHI, KATSUTOSHI, HISAO, SEIK., E., M., KI, N. (1987c), Liblomycin, a new analogue of bleomycin, Cancer Treat. Rev. 14, 169-177. I., I., TAKAHASHI, ASONO,K., KAWAMOTO, TAMAOKI, NAKANO, (1989), UCN-01 and T., H. UCN-02, new selective inhibitors of protein kinase C. I. Screening, producing organism and fermentation, J. Antibiot. 42, 564-570. TAKAHASHI, NUMATA, MATSUMURA, C., A., E., K., T., MJNOURA, ETO, H., SHINGU, ITO, T., HASEGAWA, (1994), Leptosins I and J, cytoT. toxic substances produced by a Leprosphaeria sp. Physico-chemicalproperties and structures, J. Antibiot. 47, 1242-1249. S., P., K., TAKAHASHI, HASHIMOTO, HAMANO, SUZUKI, NAKAGAWA, (1996), MelanoxaT., A. zal, new melanin biosynthesis inhibitors discovered by using the larval haemolymph of the silkworm Bombyx mori. Production, isolation, structural elucidation and biological properties, J. Antibiot. 49, 513-518. TAKAMIYA, YOSHIDA,E., TAKAHASHI, K., T., OKURA,A., OKANISHI, KOMIYAMA, M., K., UMEZAWA, (1988), The effect of kazusamycin I. B on the cell cycle and morphology of cultured L1210 cells, J. Antibiot. 41, 1854-1861. TAKARA, TOMIDA, NISHIMORA, YAM., A., T., MAGUCHI, SUZUKI, H. (1990), ResorthiomyH., cin, a novel antitumor antibiotic. 111. Potentiation of antitumor drugs and its mechanism of action, J. Antibiot. 43, 138-142. E., Y., TAKE,Y., KUBO,T., TAKEMORI, INOYE, NAKAMURA, NISHIMURA, SUZUKI, S., T., H., YAMASUCHI, (1989), Biological properties of H. streptonigrin derivatives. 111. In vitro and in vivo antitumor activities, J. Antibiot. 42, 968-1976. TAKEUCHI, (1992), Novel physiologically active T. substance, Jpn. Patent JP 04217681. TAKITA,T. (1984), The bleomycins: properties, biosynthesis and fermentation, in: Biotechnology of Industrial Antibiotics (VANDAMME, J., E. Ed.), pp. 595-621. New York Marcel Dekker. TAKITA, MURAOKA, (1990), Biosynthesis T., Y. and Chemical Synthesis of Bleomycin. Biochemistry of Peptide Antibiotics (KLEINKAUF, H., VON DOEHREN, H., Eds.), pp. 289-309. Berlin: DeGruyter. TAKITA, OGINO, (1987), Peplomycin and liT., T. blomycin, new analogs of bleomycin, Biomed. Pharmacother. 41,219-226. K. TAKITA,T., MURAOKA,Y., TAKAHASHI, (1989), Bleomycins: basic research, Gann Monogr. Cancer Res. 36 (Antitumor. Nat. Prod.), 59-70.
TANAKA, OKABE, ISONO, KASHIWAGI, N., T., F., M.,NOMOTO,K., TAKAHASHI, SHIMAZU, M., T. A., NISHIMURA, (1989), Lactoquinomycin, a novel anticancer antibiotic. I. Taxonomy, isolation and biological activity, J. Antibiot. 38, 13271332. S., M., TANAKA, OHKUBO, KOJIRI,K., SUDA,H. (1992), A new indolocarbazole antitumor substance ED-110, a derivative of BE-l3793c, J. Antibior. 45, 1797-1798. TANAKA, OKABE, NAKAJIMA, YOSHIDA, S., T., S., E., SUDA,H. (1994), BE-23372M, a novel protein tyrosine kinase inhibitor. 1. Producing organism, fermentation, isolation, and biological activities, J. Antibiot. 47,289-293. S. TATSUDA, YASUDA, (1996), Synthesis and K., biological evaluation of caloporoside analogs, J. Antibiot. 49, 713-715. TATSUDA, IKEDA, MIURA, (1996), SynK., Y., S. thesis and glycosidase inhibitory activities of nagstatin triazole analogs, J. Antibiot. 49, 836837. G. TAYLOR, S. (1989), CAMP-dependent protein kinases, J. Biol. Chem. 264, 8443-8446. THORNALLEY,J. (1990), The glyoxylase system: P. new developments towards functional characterization of a metabolic pathway fundamental to biological life, Biochem. J. 269, 1-11. TOLSTIKOV, V., HOLPNE V. KOZLOVA, Y., ORN. ESHKINA, T. D., OSIPOVA, V., PREOBRAZT. HENSKAYA, M. N., SZTARICSKAI, F., BALZARINI,J., DE CLERCQ, E. (1992), Amides of antibiotic streptonigrin and amino dicarboxylic acids or amino-sugar. Synthesis and biological evaluation, J. Antibiot. 45, 1020-1025. TOMASZ, (1992), Mitomycin C: DNA sequence M. specificity of a natural DNA cross-linking agent, Adv. DNA sequence specific Agents 1,247-261. TOMASZ, (1994), DNA-adducts of mitomycin. M. ZARC Sci. Publ. 125 (DNA-Adducts: Identification and Biological Significance), 349-357. TROLL,W., WIESNER,R., FRENKEL, (1987), K. Anticarcinogenic action of protease inhibitors, Adv. Cancer Res. 49,265-283. TROWBRIDGE,S . (1991), CD45, J. Biol. Chem. I. 266,23 517-23 520. TSCHAGOSHI, TAKENDO, UMEZAWA, S., T. H. (1986), Antitumor substances, in: Biotechnology, Vol. 4 (REHM,H.-J., REED,G., Eds.), pp. 509530. Weinheim: VCH. Y., TSUCHIYA, S., MORIYA, HON,M., ITOH,O., K. TAKEUCHI, EKIMOTO, HIRATSUKA, T., H., M. (1992), Synergism between 5-fluorouracil and oxanosine in inhibition of growth of ras-expressed cells in vitro and in vivo, J. Antibiot. 45, 283-285. TSUKAMOTO, (1993), Microcystilide A. A novel S. cell-differentiation-promotingdepsipeptide from
6 References
703
Microcystis seruginosa NO-15-1840, J. Amer. UMEZAWA, HARESAKU, MURAMATSU, K., M., M., T. Chem. SOC. 115, 11046. MATSUSHIMA, (1987), Mutagenicity of anM., H., M., TSUNAKAWA, KAMEI, KONISHI, MIGAthracycline glycosides and bleomycins in a SalKI, T., OKI,T., KAWAGUCHI, (1988), PoroH. monella assay system, Biomed. Pharmacother. 41,214-218. thramycin, a new antibiotic of the anthramycin group: production, isolation, structure and bio- UOSAKI,Y., YASUZAWA, HARA, SAITOH, M., T., logical activity, J. Antibiot. 41, 1366-1373. Y., SANO,H. (1991), Sapurimycin, new antituTSURUOKA, FUKUYASHI,ISHI,M., USUI,T., T., H., mor antibiotic produced by streptomyces strucSHIBAHARA, INOUYE, (1996), Inhibition of S., S. ture determinaton, J. Antibiot. 44, 40-44. mouse tumor metastasis with nojirimycin-related UOSAKI,Y., KAWADA, NAKANO, SAITOH, S., H., compounds, J. Antibiot. 49, 155-161. Y., SANO,H. (1993), UCT4B, a new antitumor M., H., UBUKATA, UZAWA,J., OSADA, ISONO,K. antibiotic with topoisomerase I1 mediated DNA (1993), Respinomycins A l , A2, B, C, and D, a cleavage activity. Structure determination, J. Annovel group of anthracycline antibiotics. 11. Phytibiot. 46, 235-240. sicochemical properties and structure elucidaVAN DER PYL, D., INOKOSHI,J., SHIOMI, K., tion, J. Antibiot. 46, 942-951. H., S. YANG,H., TAKESHIMA, OMURA, (1992), UBUKATA, M., MORITA, T., URAMOTO, M., Inhibition of farnesyl protein transferase by glioOSADA, (1996), Sparoxomycins A1 and A2, H. toxin and acetylglyotoxin, J. Antibiot. 45, 1803new inducers of the flat reversion of NRK cells 1805. transformed by temperature sensitive Rons sarVAN DER WILT, C. L., PETERS,G. J. (1994), New koma virus. 11. Isolation, physicochemical proptargets for pyrimidine antimetabolites in the erties and structure elucidation, J. Antibiot. 49, treatment of solid tumors. I. Thymidylate 65-70. synthase, Pharm. World Sci. 16, 84-103. UCHIDA,T., IMOTO,M., WATANABE, MIURA, Y., BOSCHE,H., MOLREELS, KOYMANS, H., T., N., T., K., DOBASHI, MATSUDA, SAWA, NA- VANDEN L. M. H. (1994), Aromatase inhibitors - mechaGANAWA, H., HAMADA,M., TAKEUCHI, T., nisms for nonsteroidal inhibitors, Breast Cancer UMEZAWA, H. (1985), Saquayamycins, new Res. Treat. 30,43-55. aquayamycin-group antibiotics, J. Antibiot. 38, VANEK,Z., TAX,J., KOMERSOVA, SEDMERA, 1171-1181. I., UCHIDA,R., SHIOMI, INOKOSHI, MASUMA, K., P., VOKOUN,J. (1977), Anthracyclines, Folia J., T., H., IWAI, Y., R., KAWAKUBO, TANAKA, Microbiol. 22, 139-159. OMURA, (1996), Kurasoins A and B, new pro- VYAS, D. M. (1993), Recent advances in medical S. tein farnesyltransferase inhibitors produced by chemistry of taxol, Pharmacochem. Libr. 20 Paecilomyces sp. FO-3684, J. Antibiot. 49, 932(Trends in Drug Research), 261-270. 934. WAGNER, ECKARDT, IHN,W., SCHUMANN, C., K., UEDA, H., NAKAJAMA, HORI,Y., FUJITA,T., H., W. D. G., STENGEL,C., FLECK, F., TRESSELT, NISHIMURA,M., GOTO, T., OKUHARA, M. (1991), Biosynthese der Anthracycline: eine (1994), FR 901228, a novel antitumor bicyclic Neuinterpretation der Ergebnisse zur Daunomydepsipeptide produced by Chromobacterium cin-Biosynthese, J. Basic Microbiol. 31, 223violaceum, No. 968, I. Taxonomy, fermentation, 240. isolation, physico-chemical and biological prop- WAKUSAWA, S., INOKO, K., MIGAMOTO, K. erties, and antitumor activities, J. Antibiot. 47, (1993), Staurosporine derivatives reverse multi 310. drug resistance without correlation with their UEHARA, Y., MURAKAMI, SUZUKAKE-TSUY., protein kinase inhibitory activities, J. Antibiot. CHIYA, K., MORIYA,Y., SANO, H., SHIKATA, 46,353-355. K., OMURA, (1988), Effects of herbimycin S. derivatives on src oncogene function in relation WALL,M. E., WANI,M. C. (1993), Camptothecin and analogs: synthesis, biological in vitro and in to antitumor activity, J. Antibio?. 41, 831-834. vivo activities, and clinical possibilities, ACS UMEZAWA, (1989), Screening of oncogene funcK. Symp. Ser. 534 (Human Medical Agents from tion inhibitors, possible antitumor agents. AntiPlants), 146-169. cancer Drugs, Colloques INSERM, Vol. 191, WANG,H. H. J. (1992), Intercalative drug binding H., (TAPIERC, Ed.), pp. 13-21. to DNA, Curr. Opin. Struc?. Biol. 2, 361-368. H., UMEZAWA, IMOTO,M., SAWA,T., ISSHIKI,K., H., M., MATSUDA,N., IINUMA, HAMADA, TA- WANG,A. H. J., LEGHETTO, G., QUIGLEY,G. J., KEUCHI, T. (1986), Studies on a new epidermal RICH, A. (1987), Interactions between an angrowth factor-receptor kinase inhibitor, erbstathracycline antibiotic and DNA: molecular structin, produced by MH435-hF3, J. Antibiot. 39, ture of daunomycin complexed to d( CGTACG) at 1,2A-resolution, Biochemistry 26, 1152-1163. 170-1 73.
704
WARDMAN, (1990), Bioreductive activation of P. quinones: redox properties and thiol reactivity, Free Radicals Res. Commun. 8,219-229. WARING,M. E. (1975), Ethidium and propidium, in: Antibiotics, Vol. 3, Mechanisms of Action of Antimicrobial and Antitumor Agents (CORCORAN, J . W., HAHN, F. E., Eds.), pp. 141-166. Berlin, Heidelberg, New York: Springer-Verlag. WARING,M. J. (1979), Echinomycin, triostin and related antibiotics, in: Antibiotics, Vol. 512, Mechanisms of Action of Antibacterial Agents (HAHN,F. E., Ed.), pp. 173-194. Berlin, Heidelberg, New York: Springer-Verlag. WARING, J. (1990), The molecular basis of speM. cific recognition between echinomycin and DNA. Jerusalem Symp. Quantum Chem. Biol. 23 (1990) (Molecular Basis of Specific Nucleic Acid-Drug Interactions) 225-24. WARING, J. (1992), Echinomycin, Puthof. Biof. M. 40,1022-1024. WAIT, P. M., HICKSON, D. (1994), Structure and I. function of type I1 DNA topoisomerases, Biochem. J. 303,681-695. WEBER, H., SCHU, P., ANKE, T., VELTEN, R., STEGLICH, W. (1994), Caloporoside, a new inhibitor of phospholipases C from Caloporus dichrous (Fr.) Ryr., J. Antibiot. 47, 1188-1194. WEINBERG, A. (1985), The action of oncogenes R. in the cytoplasm and nucleus, Science 230, 770776. WEISS, R. B. (1992), The anthracyclines will ever find a better doxorubicin? Semin Oncol. 19,670686. WHITE,J. R., STROSHANE, M. (1984), DaunoruR. bicin and adriamycin: properties, biosynthesis, and fermentation, Chapter 19, in: Biotechnology of Industrial Antbiotics (VANDAME, J., Ed.), E. pp. 569-594. New York, Basel: Marcel Dekker. WILMAN, E. V. (Ed.) (1990), The chemistry of D. antitumor agents. New York: Chapman and Hall. WOLF, M., BAGGIOLINI, (1988), The protein M. kinase inhibitor staurosporin, like phorbol esters, induces the association of protein kinase C with membranes, Biochem. Biophys. Res. Commun. 154,1273-1279. YAGINUMA, MOTO, N., TSUJINO, SUDATE, S., M., Y., HAGASHI,M., OTANI,M. (1981), Neplanocin A, a new antitumor antibiotic. I. Producing organism, isolation and characterization, J. Antibiot. 34,359-366. YAMAGUCHI, FURUMAI, SATO,M., OKUDA, T., T., T., ISHIDA,N. (1970), Studies on a new antitumor antibiotic, largomycin. I. Taxonomy of the largomycin-producing strain and production of the antbiotic, J. Antibiot. 23, 369-372.
YAMASHITA, NAOI,N., HIDAKA, HORI, S., T., T., KIMURA,K., TAKEUCHI, UMEZAWA,H. T., (1979), Studies on auromomycin, J. Antibiot. 32, 330-339. YAMASHITA, SAKAI,M., KAWAI,Y., AONO, T., M., TAKAHASHI, (1989), A new activity of K. herbimycin A: inhibition of angiogenesis, J. Antibiot. 42, 1015-1017. YAMASHITA, SAITOH, ANDO,K., TAKAHAY., Y., H., H. SHI,K., OHNO, NAKANO, (1990a), Saintopin, a new antitumor antibiotic with topoisomerase I1 dependent DNA cleavage activity, from Paecilomyces, J. Antibiot. 43, 1344-1346. YAMASHITA, KAWADA, FUJII, NAKANO, Y., S., N., H. (1990b), Induction of mammalian DNA topoisomerase I1 dependent DNA cleavage by antitumor antibiotic streptonigrin, Cancer Res. 50, 5841-5844. YAMASHITA, OSONO, MASUDA, ISHIZUT., M., T., KA, M., TAKEUCHI, KAWATSU,M. (1993), T., Effect of conagenin in tumor bearing mice. Antitumor activity, generation of effector cells and cytokine production, J. Antibiot. 46, 1692-1698. YAMAZAKI, YAMASHITA, HARADA, NIM., T., T., N., SHIKORI, SAITO,S., SHIMADA, FUGII,A. T., (1992), 44-Homooligomycins A and B, new antitumor antibiotics from Streptomyces bottropensis, J. Antibiot. 45, 171-179. YAMAZAKI, TATEE,T., HOYAMA, KOJINA, T., T., F., TAKEUCHI, (1993), Bequinostatins C and T. D, new inhibitors of glutathione S-transferase produced by a Streptomyes spec. MB84-DF12, J. Antibiot. 46, 1309-1311. YAMAZAKI, AMEMIAY, ISHIZUKA, TAK., M., M., KEUCHI, T. (1995), Screening for apoptosis inducers in microbial products and induction of apoptosis by cytostatin, J. Antibiot. 48, 11381140. YANAGISAWA, (1994a), Hispidospermidin, a M. novel phospholipase C inhibitor produced by Chaetomia hispidulum (Cda) Moesz NR 7127. I. Screening, taxonomy, and fermentation, J. Antibiot. 47, 1-5; 11. Isolation, characterization and structural elucidation, J. Antibiot. 47, 6-17. YANAGISAWA, (1994b), Hispidospermidin, a M. novel phospholipase C inhibitor produced by Chaetomia hispidulum (Cda) Moesz NR 7127. 11. Isolation, characterization and structural elucidation, J. Antibiot. 47, 6-17. YARDIN,Y., ULLRICH,A. (1988), Molecular targets of signal transduction by growth factors, Biochemistry 27, 3113-31 19. YOSHIDA, KOMIYAMA, NORITO,K., WATAE., K., NABE, Y., TAKAMIYA, OKURA, A., FUNAIK., SHI, K., KAWAMURA, FUNAYAMA, UMEK., S., ZAWA, I. (1987), Antitumor effect of kazusamy0 cin B on experimental tumors, J. Antibiot. 4 , 1596-1604.
6 References YOSHIDA,S., NAGAWARA, AOYAGI,T., TAH., KEUCHI, T., UMEZAWA, (1988), Benadrostin, A. new inhibitor of poly(ADP-ribose) synthetase, produced by actinomycetes. 11. Structure elucidation, J. Antibiot. 41, 1015-1018. YOSHIDA, HOSHIKAWA, KOSEKI, HORI, M., Y., K., K., BEPPU,T. (1990), Structural specificity for biological activity of trichostatin A, a specific inhibitor of mammalian cell cycle with potent differentiation-inducing activity in Friend leukemia cells, J. Antibiot. 43, 1101-1106. O., Y., YOSHIMOTO, JOHDO, WATANABE, NIA., R. SHIDA,H., OKAMOTO, (1992), Production of a new anthracycline antibiotic betaclamycin B by microbial conversion with a specific aclacinomycin-negative mutant, J. antibiot. 45, 1005-1007.
705
T., YUAN,B. D., Wu, R. T., SATO, I., OKABE, SuT., N. ZUKI,H., NISHIMURA, TANAKA, (1985), Biological activity of cadeguomycin. Inhibition of tumor growth and metastasis. Immunostimulation, and potentiation of 1-P-D-arabino-furanosylcytosine, J. Antibiot. 38, 642-648. ZHOU, N., JAMES,T. L., SHAFER, R. H. (1989), Binding of actinomycin D to (d(ATCGAT)),: NMR evidence of multiple complexes, Biochernistry 28, 5231-5239. C., ZIMMER, STORL, K., STORL, J. (1990), Microbial DNA topoisomerases and their inhibition by antibiotics, J. Basic Microbiol. 30,209-224. G. ZUNINO, CAPRANICO, (1990), DNA topoisoF., merase I1 as the primary target of antitumor anthracyclines, Anti-Cancer Drug Res. 5, 307-317.
Biotechnology
Index
A acarbose 438f acarbose family, structure of - 472 Acinetobacter baumannii, acetinobactin 219 Acinetobacter huemolyticus, acetinoferrin 221f acinetobactin, siderophore of Acinetobacter baumannii 219 - structure 219 acetylcholine receptor agonist 123 acetylenes, antifungal activity 503 acinetoferrin, dihydroxamate siderophore of Acinetobacter haemolyticus 221f actagardine, action on cell wall of susceptible bacteria 357 - primary structure 337f actinoidin A 375 - structure 373 Actinomadura madurae, maduraferrin 219f actinomycetes, as sources of nonribosomal peptides 304ff - lateral gene transfer to Bacillaceae 457 - oligosaccharide structures 35 actinomycin D 650ff - antitumor agent 651 - structure 651 actinomycins, antitumor activity 650ff ACVS see &(L-a-aminoadipy1)-L-cysteinyl-D-Valine synthetase acyl-coenzyme A :isopenicillin N acyltransferase (AT) 257ff adechlorin, inhibition of adenosine deaminase 648 adenomycin, cyclitol moieties 454 adenosine deaminase inhibitors, treatment of leukemia 648 Aerobacter aerogenes, aerobactin 222f aerobactin, siderophore of Aerobacter aerogenes 222f Aeromonas hydrophila, amonabactins 206 A-factor 27 - of Streptomyces griseus, cascade 87f - structure 80
Agaricus bisporus, endogenous siderophore 228 Agrobacterium rhizogenes, initiation of plant hairy root cultures 619 Agrobacterium tumefaciens, agrobactin 204 agrobactin, catecholate siderophore of Agrobacterium tumefaciens 204 - structure 204 ajmalicine 607ff - from plant cell cultures 608f - - yields 608 - structure 607 D-alanine, biosynthesis 330 alanine racemase, characterization 580 - role in cyclosporin biosynthesis 580f albomycins 211f - hydroxamate siderophores of Streptomyces, ferrioxamine type 211f - structure 211 Alcaligenes eutrophus, poly-phydroxybutyrate 178 alcaligin, hydroxamate siderophore of Bordetella pertussis 210f - structure 211 aleurodiscal 502f - antifungal activity 503 - structure 502 algae see oleaginous microalgae alkaloids, from plant cell cultures 603ff alldimycin, antitumor agent 659 - structure 659 alliacols, cytotoxic activity 507 - structure 505 allosamidin, chitinase inhibitor 440 allosamidins, structure 472 alterobactin, peptide siderophore of Alteromonas luteoviolacea 212 - structure 212 Alteromonas haloplanctis, bisucaberin 210f Alteromonas luteoviolacea, alterobactin 212 ametantron, antitumor agent, structure 651 amikacin, semisynthetic aminoglycosides 461 amino acids, found in lantibiotics 327ff. 347
108
Index
- - D-alanine 330 - - didehydroalanine 328

- - didehydrobutyrine 328 - - lanthionine 328 - - methyllanthionine 328 - - structures 327 - uncommon -, in glycopeptide antibiotics 384ff
- semisynthetic 461
sorbistins 439f streptomycin-related - 427ff - boholmycin 428 - hygromycin A 428f - kasugamycins 427 - minosaminomycin 428 - - myomycin 428 - - spectinomycins 427 - streptomycins 415ff - target site interactions 449 - trehalosamines 437f - trehazolin 439 - validamycins 438f aminopeptidase inhibitors, from basidiomycetes 518ff - - structure 519 amonabactins, catecholate siderophore of Aeromonas hydrophila 206 - structure 206 ampicillin, effect in mouse septicemia 380 Amycolatopsis sp., production of dalbaheptides 380 amylostatin family, structure of - 472 anabasine, biosynthesis 625 - structure 624 Anchusu officinalis, accumulation of rosmarinic acid 599f ancovenin, primary structure 336f androgen receptor antagonists 116 androgens, inhibition of - 685 angiogenesis inhibitors 673ff - structures 674 anguibactin, catecholate siderophore of Vibrio unguillarum 206f - structure 207 animal fats, commercial markets 135 - fatty acid composition 136 ansamitocin, inhibition of the microtubular system 668 - structure 668 anthocyanin, structure 610 anthocyanins, from plant cell cultures 609f anthracycline analogs 660 anthracyclines, antitumor agents 652, 657ff - - structures 657ff - formation of oxygen radicals 661f - pigmented antibiotics, as antitumor agents 660 - structures 34 anthramycins, antitumor agents 656 anthraquinones, anticancer drugs 652 - biosynthetic pathway 602f - from plant cell cultures 601ff - - yields 601 - structure 601 antibacterial agents, from basidiomycetes 501ff antibiotic export, regulatory cassettes, in streptomycetes 92
-
amino hexitol components, structures 406 amino pentitol components, structures 406 &(L-cY-aminoadipyl)-L-cysteinyl-D-valinesynthetase (ACVS) 254ff aminochelin, catecholate siderophore of Azotobacter vinelandii 203, 203 - structure 203 aminocyclitol aminoglycosides, producing bacterial genera 401 aminoglycoside-like compounds, structures 402 aminoglycoside pathways, overview 448f aminoglycoside producers, resistance in - 44Off - - genes 440 aminoglycoside production, common gene pool 462 aminoglycosides 397ff - acarbose 438f - adverse effects in humans 451 - allosamidin 440 - as translation inhibitors 450 - bactericidal effects 450 - biochemical classification 400 - biosynthesis of - 409ff - biosynthetic genes, in Streptomyces 447 - bluensomycin 415f - carbohydrate precursors, from primary metabolism 459 - compilation of 400 - decoding site of bacterial ribosomes 450 - deoxystreptamine-containing - 432ff - - apramycins 436 - - biosynthetic pathways 433 - - destomycins 436 - - 4,5-disubstituted - 434 - - 4,6-disubstituted - 434f - ecology 401 - evolutionary aspects 457f - fortimicins 429 - functions of - 4opff - genetics of 409ff - glycosidase inhibitors 436f - hexitol components, structures 406 - istamycins 429ff - oligosaccharidic - 438f - pathway engineering 459ff - pathway formula 400 - pentitol components, structures 406 - producing bacteria 401 - regulation of production, in streptomycetes 443ff - resistance mechanisms in 44Off
Index antibiotic formation, in strepomycetes, factors determining onset of - 86 - pathway-specific activator genes 89f antibiotic production, effect of a two-component regulatory system 83 - effect of metabolic imbalances 74 - gene clusters 81 - genes affecting also morphological differentiation 84f - genetics 79ff - in streptomycetes, metabolite interference 72ff - multicopy inhibition of - 84 - onset of -, role of protein phosphorylation 81f - organization of antibiotic biosynthetic genes 79ff - regulation in Streptomyces, physiological aspects 71ff - regulation of - 57ff - - plactams 65ff - - effect of growth conditions 59f - - in streptomycetes 70ff - - in unicellular bacteria 61ff - - peptide antibiotics 67ff - role of small diffusible signaling compounds 78f - stimulation by ppGpp 61ff, 75ff antibiotic regulatory genes, pleiotropic -, interactions 85ff antibiotic resistance, in streptomycetes, induction of - 92ff antibiotics see also p-lactam antibiotics, see also dalbaheptides, see also lantibiotics, see also peptide antibiotics - aminoglycosides, target site interactions 449 - anthracycline type - 658 - binding non-covalently to the DNA, structures 649 - evolutionary significance 60 - intercalating - 650ff - plactam biosynthesis 247ff - macromolecular antitumor - 663ff - resistance mechanisms against -, producing microbes 442 anticancer drugs see also antitumor agents - angiogenesis inhibitors 673ff - - structures 674 - antifolates 648 - antimetastatic inhibitors 673f - - structures 674 - baccatins, from the European yew tree 669 - binding non-covalently to the DNA 649f - - structures 649 - DNA topoisomerase inhibitors, camptothecins 655 - - podophyllotoxins 655 - - structures 653f - immunomodulators 675f
709
inhibitors of, cellular mitogenic signal transduction pathway 675f - glutathione S transferase 672 - mitosis 665 - protein kinases 676ff - protein phosphatases 676ff - ras-farnesyltransferase 683 - sexual hormone production 685 - the microtubular system 665 - tyrosine protein kinases 679 intercalating antibiotics, actinomycins 65Off - anthracyclines 652 - ellipticin 650 - p-quinones 652 - quinoxaline antibiotics 652 - - structures 651f - non-classical approaches 670 - nucleobase analogs 647f - nucleoside antibiotics 647f - radical generation 667 - screening assays, regulation of the cell cycle 647 - - use of transformed cells 647 - side effects, reduction of - 669 antifolates 648f - structures 648 antifungal agents, from basidiomycestes 50lff antihelminthic compounds, from secondary metabolites 10 anti-HIV activity, of cyclosporins 540ff, 550f antimetastasis drugs 673f antimicrobial agents, compilation of 4ff antineoplastic agents see antitumor agents antitumor agents 641ff, see also anticancer drugs - biotechnical drugs 643 - classical - 647ff - compilation of - 6ff - forming covalent bonds with DNA 656ff - - anthracyclines 657 - - aziridines 655f - - bleomycin 661, 663 - - cyclopropane 655 - - epoxide compounds 655f - - streptonigrin 656 - - structures 657 - from basidiomycetes 503ff - from plant cell cultures 614ff - non-classical action on tumors 646 - of marine organisms 685 - plant-derived drugs 643 - potentiators of cytotoxicity 670ff - - structures 671 - prescreening for - 645ff - radical generation, by anthracyclines 661 - - by bleomycin 663 - - by enedyines 664 - showing selective toxicity 682ff antitumor polysaccharides 508
710
Index
antiviral agents, from basidiomycetes 509ff, 520

- - structures 510f
- - surfactin 67ff
Bacillus spp., peptide antibiotic production 67ff Bacillus subtilis, stationary-phase functions 67
antramycin, antitumor agent 662

- structure 662
apramycin family, structure of - 469 apramycins 436 arachidonic acid (ARA), formation in, algae 161 - - microalgae 167f - - phycomycetes 161 arbekacin, semisynthetic aminoglycosides 461 arbutin, production by biotransformation 622 Archaea, aminoglycoside-likeglycolipid structure 402 arginine-vasopressin receptor antagonists 121 Arthrobacter sp., arthrobactin 222 arthrobactin, siderophore of Arthrobacter sp. 222 ascidians, as sources of nonribosomal peptides 313 aspergillic acid, iron-binding compounds 233 - structure 233 Aspergillus nidulans 25Off - clustering of p-lactam biosynthetic genes 259ff - gene expression, regulation of - 266 - penicillin biosynthetic cluster, regulation of 21 - penicillin biosynthetic enzymes 255ff asperlicin 117 - activities 122 - structure 119, 122 astaxanthin, structure 174 - use as feed for pen-reared salmonids 175 AT see acyl-coenzyme A: isopenicillin N acyltransferase ATPase inhibitors 519f atramycin, antitumor agent, structure 662 atrial natriuretic peptide receptor antagonists 120f atromentic acid, iron-binding compounds 232 - structure 232 atromentin, iron-binding pigments 231 - structure 231 Atropa belladonna, transgenic -, enhanced scopolamine production 625f autoimmune diseases, treatment with cyclosporin A 540 aziridines, antitumor agents 6561 Azotobacter vinelandii, catecholate siderophores 203 Azotobacteriaceae, azotobactin 216 azotobactin, chromophore 216 - peptide siderophore of Azotobacteriaceae 216 azotochelin, catecholate siderophore of Azotobacter vinelandii 203 - structure 203
- subtilin production 331 - surfactin production 67ff - surfactin synthesis, scheme 68
bacteria, accumulation of PHB 178

- affection by dalbaheptides 377ff - affection by lantibiotics 354ff - aminocyclitol aminoglycoside-producing - 401 - as producers of lantibiotics 326 - as sources of nonribosomal peptides 304ff, 308f - as sources of ribosomal peptides 299 - Plactam producing -, actinomycetes 250 - - eubacteria 250 - production of PUFA 147ff - - eicosapentaenoic acid 147f - regulation of antibiotic production 57ff - - in unicellular - 61ff - resistance mechanisms in - 440ff - signaling molecules, structures 14
bacterial siderophores, acinetobactin 219

- carboxylate siderophores 223ff
- - cepabactin 224 - - rhizobactin DM4 223
B baccatin 615 baccatins, antitumor agents 667
- - staphyloferrin A 223 - - staphyloferrin B 224 - - vibrioferrin 224 - catecholate siderophores 201ff - - agrobactin 204 - - aminochelin 203 - - amonabactins 206 - - anguibactin 206f - - azotochelin 203 - - chrysobactin 202f - - desferrithiocin 207 - - enterobactin 201f - - ferrorosamine A 208 - - fulvibactin 205f - - myxochelin A 203 - - parabactin 204f - - protochelin 203 - - pyochelin 207f - - siderochelin A 208 - - vibriobactin 205f - - vulnibactin 205f - - yersiniabactin 207f - citrate hydroxamates 220ff - - actinoferrin 221f - - aerobactin 222f - - arthrobactin 222 - - rhizobactin 1021 221 - - schizokinen 220f - frankobactin 219 - hydroxamate siderophores 209ff - - albomycins 211f - - alcaligin 210f - - bisucaberin 210f
Index
- ferrimycins 211 - ferrioxamines 209f
711
keto hydroxy bidentates 225 maduraferrin 219f mycobactins 217ff peptide siderophores 212ff - alterobacin 212 - ferribactin 214 - ferrocins 212f - - pseudobactins 213ff - - pyoverdins 213ff bacteriocinkytolysin, structural gene 344 basidiomycetes, aminopeptidase inhibitors 518ff - antibacterial metabolites 501 - antitumor compounds 503ff - antitumor polysaccharides 508 - antiviral compounds 509ff - ATPase inhibiting metabolites 520f - cultivation of - 490f - cytotoxic compounds 503ff - herbicidal compounds 513ff - immunosuppressive compounds 509 - inhibitors of cholesterol biosynthesis 521f - inhibitors of leukotriene biosynthesis 522f - insecticidal compounds 515f - life cycle 490 - metabolic inhibitors of signal transduction 524 - metabolites from - 489ff, 496ff - - history 490 - - oudemansis 496ff - - pleuromutilin 496 - - screening of - 496 - - strobilurins 496ff - metabolites inducing differentiation of leukemia cells 524 - nematicidal compounds 515f - phospholipase inhibiting metabolites 518f - platelet aggregation inhibitors 512f - primary and secondary metabolism, interrelationship 492 - producing, ondemansins 499 - - strobilurins 499 - reverse transcriptase inhibitors 509ff - uses 490 beef tallow, fatty acid composition 136 benastatins, glutathione S transferase inhibitors 672 benzophenanthridine alkaloids, from plant cell cultures 603ff - - yields 605 - structures 604 berberine, structure 604 betacyanin, structure 610 betacyanins, from plant cell cultures 611 betalains, from plant cell cultures 611 betaxanthins, from plant cell cultures 611 bialaphos production, in Streptomyces hygroscopicus, regulatory genes 91
bicyclic plactams see plactam antibiotics biopreservatives, nisin 358 biosurfactants 180f biosynthetic pathways, of secondary metabolites 31ff biosynthetic strategies, formation of nucleoside antibiotics 32 biotechnical drugs, as antitumor agents 641ff biotransformations, of dalbaheptides, deacylation 389 - of dalbaheptides 389f - - deglycosylation 389 - - glycosylation 389 - with cultured plant cells 621ff bisanthrene, antitumor agent, structure 651 bisucaberin, hydroxamate siderophore of Alteromonas haloplanctis 21Of - structure 211 blennin, leukotriene biosynthesis inhibitor, structure 523 bleomycin 661 - mode of antitumor action 663 - structure 663 blood group markers 403 bluensidine, biosynthetic pathway 417,421 bluensomycin 415f - biosynthetic pathway 416 boholmycin, streptomycin-related aminoglycosides 428 - structure 464 Bordetella pertussis, alcaligin 210f botryococcene 174ff - structure 174 Botryococcus braunii, hydrocarbon production 175f bryostatins, inhibition of protein kinase C 677 butanolides, of Streptomyces virginiae, structure 80 butirosin family, structure of - 467 butirosins 434 butterfat, fatty acid composition 136 butyl-methyl threonine, biosynthetic pathway 581 - polyketide synthase 581 - role of transaminases 581 y-butyrolactone autoregulators, synthesis by Streptomyces sp. 446 ybutyrolactones, as regulatory signals 27ff - role in antibiotic production 78ff - structures 80f
C
C-1027 chromophore, antitumor agent, structure 666 cadeguomycin, antitumor activity 647 caffeic acid, from plant cell cultures 598ff caffeoyl putrescine, from plant cell cultures, yields 600
712
Index parabactin, of Micrococcus denitrijicans 204f protochelin 203 pyochelin 207f siderochelin A 208 vibriobactin 205f - vulnibactin 205f - yersiniabactin 207f Catharanthus roseus, accumulation of monoterpene indole alkaloids 607ff - - vinblastine 614f - - vincristine 614 - callus cultures 607 - indole alkaloids 665 - suspension cultures 608 CBE see cocoa butter equivalent cell cultures see plant cell cultures cell divison, phases 644 cellular growth regulation, signal transduction pathway 644 cepabactin, carboxylate-type siderophore of Pseudomonas cepacia 224f - structure 224 cephaloridine, effect in mouse septicemia 380 cephalosporin biosynthesis 251f - genes 251 - - cloning of - 259 - - clustering of - 259ff - pathway 252 cephalosporins, semi-synthetic - 264 Cephalosporium acremonium 249ff - gene expression, regulation of - 267 - penicillin biosynthetic enzymes 254ff - use of recombinant DNA technology, in p-lactam biosynthesis 2641 cephamycin biosynthesis 252f - genes 251 - - cloning of - 259 - - clustering of - 259ff ceramides, use in skin care formulations 184 chalciporon, antimicrobial activity 495 chartreusin, structure 649 Chlorella minutissima, formation of eicosapentaenoic acid 169 chloropeptin I, inhibition of HIV replication 125 chloropeptins 124 - structures 126 cholecalciferol, structure 171 cholecystokinin receptor antagonists 116 cholesterol, structure 173 cholesterol biosynthesis, inhibitors of -, from basidiomycetes 518f chrysobactin 202f - catechol-type siderophore 202f - structure 202 cinnamic acid derivatives, from plant cell cultures 598ff cinnamycin, antibiotic activity 357f - prepeptide structure 345
-
- structure 599
calcineurin 558f, 561f - phosphatase activity 5492 - role in T cell activation 549 calicheamicin PIBr, antitumor agent, structure 666 caloporoside 518f - inhibition of, phosphoinositol metabolism 680 - phospholipase, inhibition of - 518 - structure 520 calphostin, antitumor activity 679 calyculin, inhibition of protein kinase C 677 camptothecin-type compounds, as inhibitors of topoisomerase ll 655 cancer cells see also tumor cells - cell cyle 644 cancer genes see oncogenes candicidin production, in Streptomyces griseus 71f - - growth-phase dependency 71 - - phosphate repression 74 Candida bombicola, sophorolipid 180 cantharanthine 607f - structure 607 carbapenem synthesis, by Erwinia carotovora, regulation of - 65f - quorum regulation of - 66 carbon catabolite repression, global regulation 59 carboxylate siderophores 223ff - cepabactin 224 - rhizobactin DM4 223 - staphyloferrin A 223f - staphyloferrin B 224 - vibrioferrin 224 cardiac glycosides 617 - production by biotransformation 622 carnocin UI49, structure 336 pcarotene, formation in halophilic algae 175 - in fungi 174f - in microalgae 172ff - structure 173 carotenoids, occurrence in microalgae 172ff - structures 173f caryophyllanes, structure 506 catabolite regulation, in microbial drug synthesis
11
catabolite repression 59 catecholate siderophores, agrobactin, of Agrobacterium tumefaciens 204 - aminochelin 203 - anguibactin 206f - azotochelin 203 - chrysobactin 202f - desferrithiocin 207 - enterobactin 201f - ferrosamine A 208 - fluvibactin 205f - myxochelin A 203 - of Azotobacter vinelandii 203
Index
- primary structure 336f
713
- solution structure 340

- structural gene 344 -
citrate hydroxamates, acetinoferrin 221f aerobactin 222f arthrobactin 222 rhizobactin 1021 221 schizokinen 220f structure 220 clavicoronic acid 509f - structure 510 clitocine 509f - structure 510 cocoa butter, fatty acid composition 136 cocoa butter equivalent (CBE) 150 - costs of yeast processs 155 - formation, from Cryprococcus mutants 153 - growth of Cryptococcus, oxygen limitation 154 - price 148 Coleus blumei, accumulation of rosmarinic acid 598ff collybial, antiviral activity 520 - structure 522 complement C5a receptor antagonists 121 coprogens, structure 227 - trihydroxamate siderophores 227 Coptis japonica, accumulation of protoberberines 604f coriolin, ATPase inhibiting metabolite 521 - structure 521 cosmomycins, redifferentiation of cancer cells 682 crinipellins, cytotoxic activity 508 - structure 506 Cryptococcus curvatus 152ff - cocoa butter fatty acids 153 - fatty acyl composition 153 CsA see cyclosporin A cyanobacteria, as sources of nonribosomal proteins 307f cyclitols 454f - basic pathways 402ff - benzoic acid derivatives, structures 407 - biogenesis 402ff - components of nucleoside-type antibiotics 454 - dehydroquinate pathway 403ff - myo-inositol pathway 403ff - sugar components, attachment of - 405ff cycloleucomelone, iron-binding compounds 232 - structure 232 cyclophilin A, complexed with cyclosporin A, 3Dstructure 566 - complexed with cyclosporin derivatives 568ff - - X-ray structures 568 - 3D-structure, complexed with a tetrapeptide 563f - peptidyl-prolyl isomerase active site 564ff - target of cyclosporin A 560
cyclophilin B, complexed with a cyclosporin derivative 570f - - X-ray structure 570 cyclophilin C, complexed with cyclosporin A, Xray structure 571 cyclophilins 5475 560ff - amino acid residues 560f cyclopropane, antitumor agent 655 cyclosporin A 538ff - 3D-structure, in apolar environment 561 - - in polar environment 562 - effector site 548 - immunosuppressive properties 539ff - indications 539 - mode of immunosuppressive action 546ff - - dual domain concept 546,548 - structure 538 - T cell selective immunosuppressant 539 - X-ray structure 553 cyclosporin A analogs, non-immunosuppressive -, antiviral activity 541, 551 - - effect on multidrug resistant tumor cells 540, 551 - SDZ NIM 811, structure 541 - SDZ PSC 833,541 - SDZ214-103 580 - - structure 538 - - X-ray structure 562 cyclosporin biosynthesis 572ff - building units 573 - cofactors 573 - cyclosporin synthetase 572ff - mechanistic aspects 577ff - modified thiotemplate mechanism 578 cyclosporin C, cleavage of peptide bonds 558 cyclosporin<yclophilin complexes 549f - crystal structures 565ff cyclosporin receptors 547f cyclosporin synthetase, characterization 573f - enzymatic activities 572f - gene 573f - - manipulation by integrative transformation 574ff - - structure 575 - theoretical reprogramming 576 cyclosporin variants, biosynthesis 580 cyclosporins 535ff - activity against tumor multidrug resistance 540, 551 - analogs 294f - anti-HIV activity 540ff, 550f - antiinflammatory effects 550 - antimalarial activity 551 - binding to cyclophilins 547ff, 558f - - effects of valine for leucine substitutions 560 - chemical synthesis 552ff - constituent amino acids 551f
714
Index
- lipoglycopeptides, naturally occurring - 375
- cyclic peptide backbone, chemical transformations of - 555 - fermentation 552ff - immunosuppressive activity 539ff - - mechanism 546ff - - relation to structure 556ff - incorporating non-proteinogenic amino acids 556 - macrocycle, selective ring opening 555 - mode of immunosuppressive action 542ff - resistance to systemic proteases 551 - structural investigations 561ff - structure-activity relationships 557ff - structures in polar and non-polar solutions 552 - treatment of autoimmune diseases 540 - use in transplant rejection therapy 539f - variety of conformations 571 cyclovariegatin, iron-binding compounds 232 - structure 232 cytodifferentiation, in Streptomyces, A-factor 28 - - signal cascade 27 - role of secondary metabolism 16 cytokine receptors 545 cytokines, release upon T cell activation 545f cytolysinlbacteriocin 335f - primary structure 335 cytotoxic agents, from basidiomycetes 503ff cytotoxic metabolites, marine organisms 687 D dalbaheptide fermentation, feedback inhibition of - 382f - media 381f - - carbon sources 381f - - effect of phosphates 382 - - nitrogen sources 382 - product recovery 383 - purification 383 dalbaheptides 369ff - actinoidin type, naturally occurring - 375 - biological activity 377ff - - in vitro antibacterial activity 377f - - in vivo efficacy 379f - - mechanism of action 378f - - pharmacology 379f - biosynthesis 384ff - biotransformation of - 389f - - deacylation 389 - - deglycosylation 389 - - glycosylation 389 - chemical modifications of - 387f - definition 371 - descriptive chemistry 371ff - effect in mouse septicemia 380 - effect on pathogens 378 - glycolysation of - 387 - glycopeptide skeleton 372
- physicochemical properties 376f
producing organisms 372,374f, 380 resistance to - 379 ristocetin type, naturally occurring - 372 screening methods 381 uncommon amino acids in - 384ff vancomycin type, naturally occurring - 374 daunomycin, antitumor agents 652 daunorubicin, antitumor agent, structure 658 daunorubicin production, in Streptomyces peuceticus, regulatory genes 91 deacylation, of dalbaheptides 389 deglycosylation, of dalbaheptides 389 7-dehydrocholesterol, structure 171 dehydroquinate pathway 403ff dehydroquinate synthase, substrates 405 6-deoxyhexoses 410ff - as moieties of antibiotics 456 - biosynthesis, general pathway 408 - biosynthetic pathways, gene clusters 455 - structural variants, in secondary metabolites 410ff - - Occurrence in natural products 414 deoxyilludin M,antitumor activity 503 - structure 504 desferrithiocin, orally available iron chelator 207 destomycin family, structure of - 469 destomycins 436 devazepide, activities 122 - structure 122 devazepide receptor antagonist 121 DHA see docosahexaenoic acid DHGLA see dihomo-ylinolenic acid diazaquinomycin B, thymidylate synthase inhibition 648 2,3-didehydroalanine 328ff 2,3 didehydrobutyrine 328ff 3,6-dideoxyhexosepathway, of gram-negative bacteria 414 Digitalis lanata 621ff - tissue culture process 621 digitoxin, production by biotransformation 622 digoxin, production by biotransformation 622 dihomo-ylinolenic acid (DHGLA), as precursor of prostaglandins 160 - formation by deletion of desaturase 160 dihydrostreptose, biosynthetic pathway 422f dioxamycin, antitumor agent, structure 662 Dipodascopsis uninucleata, formation of prostanoid-type lipids 184f distamycin, antitumor agent 650 - structure 649 diterpenoids, Occurrence in basidiomycetes 493 DNA replication, inhibition of topoisomerases 653
- isoelectric points 376 - reverse-phase HPLC retention times 377
Index
715
DNA synthesis, inhibition by antitumor agents 655 DNA topoisomerase inhibitors 654ff - camptothecins 655 - podophyllotoxins 655 - structures 654f DNA topoisomerases, as targets of antitumor drugs 653ff docetaxel, inhibition of the microtubular system 668 - structure 668 docosahexaenoic acid (DHA), in bacteria 147f - in microalgae 16% - in phytoplankton species 169f - nutritional role 162 - occurrence in marine fungi 163 dopamine receptor antagonists 114 doxorubicin 651f - antitumor agent 651f, 658f - structure 651, 658 drugs see microbial drugs Dunaliella salina, &carotene content 173 duocarmycins, antitumor agents 657f duramycins, antibiotic activity 357f - primary structure 336f - solution structure, stereorepresentation of 340 dynemicin A 664ff - antitumor agent 665 - structure 666 - antitumor activity 664
echinomycin 651f - antitumor agent 651f - structure 651 eicosapentaenoic acid (EPA), in bacteria 147f - in microalgae 168f - physiological effects 162 - precursor of prostaglandins 162 eicosatrienoic acid (ETA), occurrence in marine fungi 163 ellipticin 650 endothelin receptor antagonists 116ff enedyines, cytotoxic antitumor agents 664ff - macromolecular antitumor antibiotics 663ff - mode of action 664ff - structures 666 enterobactin 201f - catecholate siderophore 201f - of enteric bacteria 202 - structure 201 enzyme inhibitors, of the growth-regulating pathway 646 EPA see eicosapentaenoic acid epelmycin A, antitumor agent 659 - structure 659
epidermin, C-terminus, aminovinyl cysteine residue 349 - medical uses 359 - prepeptide structure 345 - primary structure 332ff epidermin biosynthesis, from pre-epidermin, gene cluster 342 - - scheme 342 - gene cluster 348 - regulation of - 351 epilancin K7, primary structure 332ff epirubicin, antitumor agent, structure 658 epoxide compounds, antitumor agents 656f ergocalciferol, structure 173 ergosterol 171f - structure 171 ergotamine, as receptor antagonist 112 Erwinia carotovora, carbapenem production, regulation of - 65 Erwinia rhapontici, ferrorosamine A 208 erythromycin, effect in mouse septicemia 380 erythromycin A 122ff - activities 124 - structure 123 Escherichia coli, carbon catabolite-repressible operons 60 - microcin C7 synthesis 61ff - ppGpp 61ff - - synthesis 63 - RNA polymerase sigma factor 61ff - - regulation of - 64 esperamicin A, 664ff - antitumor agent 666 - enrichment pattern 667 - structure 666 estrogen receptor agonists 112, 123 estrogen receptor antagonists llSf, 686 estrogens, inhibition of -, treatment of hormonedependent tumors 685 ETA see eicosatrienoic acid ethanol, conversion to lipid 142 ether lipids, archaebacterial - 181ff - - structures 182 eukaryotes, DNA transfer to prokaryotes 263 Euphorbia milli, anthocyanin production 610 exochelins 217f - structure 218 F
u factors 61ff - initiation of transcription 62 - regulation of - 64 - structure 62
fasciculols, plant growth inhibitors, structures 515 fatty acids 146ff - esterification 146 - in bacteria, polyunsaturated fatty acids 147f
716
Index ferrichromes 225f fusarinines 228f rhizoferrins 229f rhodotorulic adid 228 fungi see also oleaginous molds - as sources of nonribosomal peptides 309 - p-lactam producing -, actinomycetes 250 - - eubacteria 250 - formation of PUFA 151,158 - method of resistance to their own fungicide
-
- in fungi, polyunsaturated fatty acids 151
- in lipoglycopeptides 386f
- in microalgae, arachidonic acid (ARA) 167f - - docosahexaenoic acid (DHA) 169f - - eicosapentaenoic acid (EPA) 168f
- - ylinolenic acid (GLA) 167

-
- in molds, arachidonic acid (ARA) 161f
- - dihomo-y-linolenic acid (DHGLA) 160

docosahexaenoic acid (DHA) 162f eicosapentaenoic acid (EPA) 162 eicosatrienoic acid (ETA) 163 ylinolenic acid (GLA) 159f polyunsaturated fatty acids 158 structures 139 ferrosamine A, catecholate siderophore of Envinia rhapontici 208 ferribactin, peptide siderophore of Pseudomonas fluorescens, non-fluorescent - 214 ferrichromes, fungal siderophores 225f - structure 225 ferric uptake regulation proteins 200 ferrimycins 211 ferrioxamines, hydroxamate siderophores of Streptomyces pilosus 209f - structures 209 ferrocins 212f - peptide siderophore of Pseudomonas fluorescens 213 - structure 212 fibrinogen receptor antagonists 115 fimicolon 502f - cytotoxic activity 503 - structure 502 FK506 546ff - immunosuppressive properties 546f - structure 546 FK506 binding proteins 547f flavidulols, structures 509 fluvibactin 205f - catecholate siderophore of Vibrio fluvialis 206 - structure 205 folic acid metabolism, inhibition by thimidylate synthase 649 fomannosin, plant growth inhibitor 514 - structure 514 formycin production, by Streptomyces lavendulae, effect of ppGpp 77 fortimicin family, structure of - 465 fortimicins 429ff - biosynthetic pathway 429 - enzymes, involved in production of - 431 - gene products, involved in production of - 431 frankobactin, structure 219 frogs, as sources of ribosomal peptides 301 fucoxanthin, structure 173 fulvoferruginin, structure 505 fungal siderophores 225ff - coprogens 227
occurrence of $-carotene 174f oil product containing y-linolenic acid 158 producing, oudemansins 499 - strobilurins 499 signaling molecules 14 transformation systems, for $-lactam production 253 fursarinines 228f - structure 228
-
501
gallidermin, medical uses 359
- solution structure, stereorepresentation of 339 gearbox promoters, expression at low growth rates 64 gene clusters, biosynthetic - 17ff - - analysis of - 20ff - - listing 18f - - organization of - 21 - in streptomycin production 420 gene products, in streptomycin production 418f genes, biosynthetic -, identification of - 20 genetic exchanges, between microbial species 15 gentamicin family, structure of - 468 gentamicins, biosynthetic pathway 433 ginseng, from plant cell cultures 611 GLA see y-linolenic acid glutarimde antibiotics, redifferentiation of cancer cells 683 glutathione S transferase inhibitors, structures 673 glycopeptide antibiotics see dalbaheptides glycosyl phophatidylinositols, resemblance to aminoglycosides 402 - structure 402 glycosylation, of dalbaheptides 389 glyoxylate cycle, key enzymes, inhibitors 513 gp120-CD4 binding inhibitors, structures 125ff groundnut oil, genetic engineering of - 138
- primary structure 333
H hairy root cultures, fermentation techniques 620 - production of tropane alkaloids 619
index
- scopolamine secretion 620 - secondary product formation 619f - transformation with Agrobacterium rhizogenes - antitumor activity 503 - biosynthesis 492 - toxic principle 493
717
719 Hansenula ciferri, formation of ceramides 184 hatomarubigins, inhibitors of multi-drug resistant tumor cells 672 hemimycin 502f - cytotoxic activity 503 - structure 502 heptapeptide structure, of dalbaheptides 372 herbicidal compounds, from basidiomycetes 513ff - - structures 514f herbimycin, immunostimulation 675 - structure 686 herbimycin A, angiogenesis inhibitor 674 - contact inhibition of tumor cells 680 - phosphoinositol kinase inhibiton 681 hexosamine pathway 423 hexosamines, as moieties of antibiotics 457 hexoses, amination of - 415 hispidospermidins 680 HIV see human immunodeficiency virus homoserine lactones 62f horizonal transfer hypothesis, @-lactam biosynthetic pathway 262f hormone-dependent tumors, treatment of - 685 human immunodeficiency virus (HIV), anti-HIV activity assay 124 - blocking of HIV entry 124 - HIV disease, use of cyclosporins 54Off, 55Of - inhibition of replication 125 hydromycin, antitumor agent, structure 662 hydroxamate siderophores 209ff - albomycins 211f - alcaligin 21Of - bisucaberin 210f - ferrimycins 211 - ferrioxamines 209f hydroxycinnamoyl putrescines 600f 7-hydroxyguanine, antitumor activity 647 p-hydroxyphenylglycine, biosynthesis 385 hygromycin, structure 464 hygromycin A, streptomycin-related aminoglycosides 428 hygromycin family, structure of - 469 hyoscyamine, from plant cell cultures 613 hyphodontal 511f - structure 511 hypnophilin, cytotoxic activity 507 - structure 505 I ibotenic acid, insecticidal compound 515f - structure 516 idarubicin, antitumor agent, structure 658 illudins 491ff
immunomodulators, antitumor effects of - 673f

- structures 674
immunophilins 547%559ff
- 3D-structure 561f
immunosuppressant macrolides, model of the mode of action 548 immunosuppressants, of microbial origin, dual domain concept 546 immunosuppression, by cyclosporins 539ff - - mode of action 542ff - - side effects 540 - - use in transplantation surgery 539f - by FK506 546ff - by rapamycin 546ff - drug-immunophilin complexes 549f immunosuppressive agents, from basidiomycetes, structures 509 immunosuppressors, from secondary metabolites 7 indolocarbazole compounds, antitumor activity 679 - bactericidal activity 679 - fungicidal activity 679 insecticidal compounds, from basidiomycetes 515f - - structures 516 insects, as sources of ribosomal peptides 300 intercalating drugs 650ff - actinomycins 65Off - ellipticin 650 - structures 651 interspecific communication 13, 15 IPNS see isopenicillin N synthase iron nutrition, role of siderophores 235 iron transport, siderophore-mediated systems 233ff - - in enterobacteria 234 iron uptake 2OOff isepamicin, semisynthetic aminoglycosides 461 isochalciporon, antimicrobial activity 495 isochromophilones 124ff - inhibition of HIV replication 125 - structures 125ff isopenicillin N synthase (IPNS) 256f isoprenoid lipids, structures 182 isovelleral, antimicrobial activity 495 isovellerol, antimicrobial activity 495 istamycin family, structure of - 465 istamycins 429ff - biosynthetic pathway 429
K kanamycin family, structure of - 467 kanamycins, biosynthetic pathway 433 kasugamycin, structure 464
718
Index
kasugamycins, streptomycin-related aminoglycosides 427 keto hydroxy bidentates 225 krestin, antitumor polysaccharide 509 kuehneromycins 511f - structure 511
- - veterinary - 359 - biological activities 353ff - chemistry of - 325ff - cinnamycin 336f, 340, 357 - compilation of - 326f - cytolysinlbacteriocin 335f - definition 325 - disruption of energy transduction 355 - duramycins 336f, 340, 357 - effect on cytoplasmic membrane of susceptible - epidermin 332ff - epilancin K7 332ff - gallidermin 333, 339 - lacticin 481 332, 335 - lactocin S 332, 335 - leader peptidases 348ff - mersacidin 3371 357 - modified amino acids in - 327ff, 347f - - D-alanine 330 - - didehydroalanine 328 - - didehydrobutyrine 328 - - lanthionine 328 - - methyllantionine 328 - - structures 327
bacteria 354
- carnocin U149 336 - chemical characteristics 326
Plactam antibiotics 247ff - biosynthesis of carbapenem 65 - biosynthetic clusters 21 - biosynthetic enzymes 253ff - - compartmentalization of - 261f - biosynthetic genes 251 - - cloning of - 259 - - clustering of - 259ff - - DNA sequences 260 - - orientation of transcription 261 - - regulation of gene expression 265ff - biosynthetic pathway, cephalosporin C 252 - - evolution of - 262f - - penicillin G 252 - genetic transformation systems, for fungi 253 - - for prokaryotes 253 - history 249f - penicillin biosynthesis, enzymes 253ff - producing organisms 250,253 - use of recombinant DNA technology 264f lacticin 481, primary structure 332 - structure 335 lactocin S, prepeptide structure 345 - primary structure 332,335 - structural gene 344 lactococcin, prepeptide structure 345 - structural gene 344 Lactococcus lactis, nisin production 330 lanosterol, structure 171 lanthionine, biosynthesis 328f - biosynthetic pathway 329 - non-protein amino acid 347 - structure 327 lantibiotic biosynthesis, epidermin 342 - genetic regulation 341ff - leader peptide, role of - 346f - modification events, sequence of - 353 - nisin, gene clusters 343 - regulation, by proteins 350f - structural genes 341ff lantibiotic prepeptides, leader peptide 345ff - primary structure 345 - structural genes, transcription of - 341ff ]antibiotics 323ff see also antibiotics - actagardine 337f, 357 - ancovenin 336f - applications of - 358f - - as foodlbeverage preservatives 358 - - medicaUparamedica1- 359
- mutacin
- nisin 330ff, 339 - non-disulfide cyclic peptides 281 - Pep 5 332ff, 340 - presursor peptide 328 - producing organisms 326 - ribosomal biosynthesis 325 - salivaricin A 332, 335 - streptococcin A-FF22 334 - subtilin 331ff, 339 - transport proteins, genes 348 - type A, induced pore formation 356 - - induction of autolysis 356 - - modes of action 353ff - - primary structures 330ff - - solution structures 339f - type B, mode of action 357f - - primary structures 336ff - - role in human immune and blood pressure - - solution structures 340f
lantibiotic synthesizing cells, self-protection mechanisms 351ff leader peptidases 348ff leader peptides, in lantibiotic biosynthesis 345ff - - conserved motifs 346f - - roles 346f leaianafulvene, cytotoxic activity 508 - structure 506 lentinan 508 regulatory systems 336
336
- physical characteristics
326
Index
719
- structure 502
lentinellic acid, antibacterial activity 501
- oligopeptide siderophore of Actinomadura ma-
durae 21%
- structure 220
leucomelone, iron-binding pigments 231 - structure 231 leukotriene B4 receptor antagonists 114 leukotriene biosynthesis, inhibitors of -, from basidiomycetes 522f ligand receptors, antagonists 117ff - - producers 117 ligands, radio-labeled, radioimmunoassay 110 - - screening of receptor-active compounds 110 light emission, by the autoinducer of Vi6rio jischeri 66 lincomycin A, biosynthetic gene, sugar subcluster 453 - hypothetical biosynthetic pathway 452 lincosamides 451ff - structure 473 - sugar containing moieties 451 ylinolenic acid (GLA), in microalgae 167 - medicinal uses 159 - production by phycomyceters 159 linseed oil, genetic engineering of - 138 lipid accumulation, biochemistry 143ff - culture conditions 140ff - effect of carbon-to-nitrogen ratio 141 - efficiency 142ff - - in molds 142f - - in yeasts 142f - in continuous culture 141 - patterns 139ff - - in molds 140 - - in yeasts 140ff - yields, on carbohydrates 142 - - on ethanol 142 lipids 133ff - biosurfactants 180f - classification 138f - ether lipids l8lff - fatty acid profile, in yeasts 148f - - in molds 157f - nomenclature 138f - phospholipids 183f - prostanoid-type - 184f - sphingolipids 183 lipoglycopeptides, fatty acids, origin 386 - naturally occurring - 375 - producing organisms 380 Lithospermum erythrorhizon, accumulation of shikonins 601f
macrolides, immunosuppressant -, model of the mode of action 548 - structures 34 maduraferrin 219f
Magnus test, screening of receptor-active compounds 109f mammals, as sources of ribosomal peptides 302f marasmic acid 503f - antitumor activity 503 - structure 504 maytansine, inhibition of the microtubular system 668 - structure 668 Mead acid see eicosatrienoic acid medermycin, antitumor agent, structure 662 menoxymycin, antitumor agent, structure 662 mersacidin, action on cell wall of susceptible bacteria 357 - prepeptide structure 345 - primary structure 337f - structural gene 344 merulidial 504f - structure 504 metabolic engineering, of secondary pathways, nicotine biosynthesis 624f - - scopolamine production 625f - - serotonin biosynthesis 624 - - use of plant cell cultures 623ff metabolism, secondary see secondary metabolism metastasis, prevention of - 673f methylenomycin biosynthesis, by Streptomyces coelicolor 74f methylenomycin production, in Streptomyces coelicolor, negative regulation of - 91f N-methyl-L-glucosamine, biosynthetic pathway 424 methyllantionine 328 microalgae see also oleaginous microalgae - carotenoid contents 172ff microbial drugs 2ff - catabolite regulation 11 - phase-dependency of biosynthesis 11 microbial lipids see lipids microbial siderophores see siderophores microcin C7, synthesis in E. coli 61ff microcins 61 microcystin, inhibition of protein kinase C 677 minosaminomycin, streptomycin-related aminoglycosides 428 - structure 464 mitogenic signal 644 mitomycin C, antitumor agent 655 mitosis, cellular signal transduction pathway 677 mitoxantron, antitumor agent, structure 651 mniopetals 510ff - structures 511 molds see oleaginous molds molluscs, as sources of nonribosomal peptides 314 - as sources of ribosomal peptides 301
720
Index Nicotiana tabacum, production of caffeoyl putrescine 599ff nicotine, biosynthesis 624f nisin, prepeptide structure 345 - primary structure 33Off - solution structure, in aqueous solution 339 - - in lipophilic solvents 339 - structural gene 342 - use in foods and beverages 358 nisin biosynthesis, gene clusters 343 NMDA receptor antagonists 114 Nocardia lactamdurans, p-lactam biosynthesis 251 - clustering of p-lactam biosynthetic genes 259ff Nocardia sp., siderochelin A 208 nogalamycin, antitumor agent, structure 658 nojirimycin, biosynthetic pathway 437 novobiocin resistance, in streptomycetes, induction of - 93 nucleobase analogs 647f - structure 648 nucleoside antibiotics 647f - biosynthetic strategies of formation 32 - structural class 40 - structure 648
monoterpene indole alkaloids, biosynthetic pathway 609 - from plant cell cultures 606ff Morinda citrifolia, accumulation of anthraquinones 603 morphinan alkaloids, from plant cell cultures 613 morphological differentiation, genes affecting also antibiotic production 84f Mortierella alpina, formation of arachidonic acid 161 - formation of eicosapentaenoic acid 162 motilides, receptor agonists 122f mucidin, biosynthesis 500 - structure 501 Mucor circinelloides, fatty acid profiles 158 multidrug resistance, of tumor cells 540, 551, 672 muscarine 112, 123 muscarinic acetylcholine receptor, agonists 112 - antagonists 112ff muscimol, insecticidal compound 515f - structure 516 mushrooms see basidiomycetes mutacin, structure 336 mycelianamide, iron-binding compounds 233 - structure 233 Mycobacterium sp., mycobactins 217 mycobactins 217ff - function as iron shuttle 217 - structure 218 myo-inositol pathway 403ff myo-inositol-~-3-phosphate synthase, substrates 405 myornycin, streptomycin-related aminoglycosides 428 myomycins, structure 464 myxochelin A, structure 203
nannochelins 223 naphthoquinones, from plant cell cultures 601f - - yields 601 neamine 434 nematicidal compounds, from basidiomycetes 515f. 521 - - structures 516 neomycin, biogenesis of components 435 neomycin family, structure of - 466 neomycins, biosynthetic pathway 433 neoplanocin, antitumor activity 647 neoplasm inhibitors see anticancer drugs neoplastic cell lines, use in screening for anticancer drugs 645ff netropsin, antitumor agents 650 - structure 649 neurokinin A, receptor antagonists 120 neuropeptides, receptor angatonists 120
0 OHHL, of Vibrio ficheri 65f - - structure 65 - regulators 65f okadaic acid, inhibition of protein kinase C 677 oleaginous microalgae 164ff - as sources of eicosapentaenoic acid 169 - cultivation of - 164 - fatty acid profiles 165f - fatty acids, arachidonic acid (ARA) 167f - - docosahexaenoic acid (DHA) 169f - - eicosapentaenoic acid (EPA) 168f - - y-linolenic acid (GLA) 167 - lipid contents 165 oleaginous microorganisms 140ff - pathway of triacylglycerol formation 143ff - - scheme 144 oleaginous molds 155ff - fatty acid profiles 156ff - fatty acids, arachidonic acid 161f - - dihomo-y-linolenic acid (DHGLA) 160 - - docosahexaenoic acid (DHA) 162f - - eicosapentaenoic acid (EPA) 162 - - eicosatrienoic acid (ETA) 163 - - y-linolenic acid (GLA) 159f - formation of PUFA 158 - lipid accumulation 140 - - efficiency 142f - lipid analyses 147 - lipid contents 156 - triacylglycerol storage 146
Index oleaginous yeasts, cocoa butter equivalent (CBE) 148. 150 - - process costs 155 - conversions of fatty acids 150 - fatty acid profile of lipids 148f - increase of stearic acid content 150ff - - by direct feeding 150 - - by inhibition of stearoyl desaturase 151f - - by metabolic manipulation 154 - - by mutation 152ff - lipid accumulation 14Off - - efficiency 142f - lipid analyses 147 - lipid contents 149 - triacylglycerol storage 146 oleandomycin formation, by Streptomyces antibioticus, growth-phase dependency 72 oligoglycosides, structural class 40 oligopeptides, biosynthesis of - 36 omphalotin, nematicidal activity 522 - structure 522 oncogenes 645 - classification 676 oncogene-transformed cell lines, availability 682 - substances changing the morphology of - 682ff ornibactins, structure 216 - tetrapeptide siderophores of Pseudomonas cepacia-like strains 216f oudemansins 496ff - blockage of respiration in eukaryotes 497 - fungistatic activity 497 - mode of action 497ff - possible functions 501 - producing organisms 499 - structures 498 oxanosin, antitumor activity 647 oxytocin receptor antagonists 121 penicillin biosynthesis 119f
- enzymes 253ff - - ACVS 254 - - AT 257ff - - IPNS 256 - gene dosage 264 - genes 251 - - cloning of - 259 - - clustering of - 259ff - pathway 252 - - compartmentalization of - 262 - regulation of gene expression 264 - side chain exchange 257
721
paclitaxel 665,667 - inhibition of the microtubular system 668 - structure 668 Panax ginseng, saponin aglycone, structure 611 panepoxydone, inhibition of signal transduction 524 panudial, antithrombotic compounds, structure 512 Papaver somniferum, accumulation of benzophenanthridine alkaloids 605f parabactin 204f - catecholate siderophore of Micrococcus denitrificans 204f - structure 204 Paracoccus denitrificans, parabactin 204f paromomycins, biosynthetic pathway 433 Peganum harmala, serotonin biosynthesis 624
penicillin fermentation 292 Penicillium chrysogenum 249ff - clustering of p-lactam biosynthetic genes 259ff - gene expression, regulation of - 266 - penicillin biosynthetic enzymes, AT 257ff pentosamines, as moieties of antibiotics 456 pentoses, amination of - 415 pentostatin, inhibition of adenosine deaminase 648 peptide antibiotics 67ff, 277ff - biosynthesis 284ff - biotechnology of - 284ff - directed biosynthesis 296 - future prospects 297f - nonribosomal origin 2791 - production by Bacillus sp. 67 - reprogramming, genetic approaches 297 - ribosomal origin 278f - side chain modifications 282 - structures 278ff peptide biosynthesis 284ff - biosynthetic modules 286ff - fermentation procedures 292,294 - gene clusters 286f - - compilation of - 287f - mechanisms 285 - multienzyme systems 292 - - domains 290 - nonribosomal system 285 - ribosomal system 285 - structural alterations 294 - tripeptide model system 291 peptide families, analogs 294f - - aureobasidin 295 - - cyclosporin 295 - - gramicidin 295 - - gramicidin S 295 peptide ligand receptors, antagonists of - 116ff - - structures 118ff peptide siderophores 212ff - alterobactin 212 - azotobactin 216 - ferribactin 214 - ferrocins 212f
722
Index
peptolide SDZ, 3D-structure, in apolar environment 562 peptolide SDZ 214-103 synthetase 580 Pep$ biosynthetic pathway 334 - prepeptide structure 345 - primary structure 332ff - structural gene 344 Phafia rhodozyma, astaxanthin production 175 PHB see poly-/3-hydroxybutyrate phellodonic acid, structure 505 phenoxazinone synthase, effect of glucose on activity 73 phopholipids, yeast lecithin, extraction of spent brewers yeast 183 phorbol, structure 678 phorbol esters, activation of protein kinase C 677 - tumor promoters 677 phosphilipids 183f phosphoinositol kinase inhibitors 681 phosphoinositol metabolism, inhibitors of - 680ff - - structures 681 phospholipase inhibitors, from basidiomycetes 518f phospholipids, medicinal uses 184 - structures 139 Phycomyces blakesleeanus, p-carotene production 175 pilatin 504f - structure 504 pinicoloform, inhibition of differentiation of leukemia cells 524 plant cell cultures 604 - biotechnological importance 596ff - biotransformations 621ff - commercial interest in - 595 - expression of pathways 603 - metabolic engineering 623ff - of transgenic plants 623f - production of secondary metabolites 593ff - regulation of secondary pathways 623 - technology 596 plant cell suspension cultures, secondary product formation 598ff anthocyanins 609f anthraquinones 601ff antitumor compounds 614ff benzophenanthridine alkaloids 603ff betalains 611 cardiac glycosides 617 cinnamic acid derivatives 598ff immunologically active polysaccharides 612 monoterpene indole alkaloids 606ff morphinan alkaloids 613 naphthoquinones 601f protoberberine alkaloids 603ff quinoline alkaloids 614 tropane alkaloids 613f vanillin 617f
- ornibactins 216f - pseudobactins 213ff - pyoverdins 213f

peptide synthetases 286ff
- carrier domains 289

-
- adenylate forming domains 286 - branched cyclopeptides 293
compilation of - 293 condensation domains 290 consensus sequences 289 cyclodepsipeptides 293 - cyclopeptides 293 - domain construction 288 - epimerization domains 290 - lactones 293 - linear peptides 293 - N-methylation domains 29Of - thioesterase domains 291 peptides, branched cyclic - 293 - compilation of - 299ff, 304ff - cyclic - 279, 293 - cyclodepsipeptides 293 - formed by in vitro synthesis 296f - gene-encoded - 283f - lactones 293 - linear 283,293 - monocyclic- 279f - multicyclic - 282f - of nonribosomal origin 279ff, 284 - - from actinomycetes 304ff - - from ascidians 313 - - from bacteria 304, 308f - - from cyanobacteria 307f - - from fungi 309f - - from molluscs 314 - - from plants 311f - - from scorpions 314 - - from sea hares 314 - - from spiders 314 - - from sponges 312f - - future prospects 298 - of ribosomal origin 278ff - - D-amino acids 281 - - features 284 - - fromfrogs 301f - - from insects 300 - - from mammals 302f - - from microbial sources 299 - - from molluscs 301 - - from plants 299 - - from scorpions 301 - - from snakes 302 - - from spiders 301 - - future prospects 297 - - production of - 283 - - structures 278ff peptidic drugs, structural class 39
Index plant growth regulators, from secondary metabolites 9 plant oils, commercial markets 135 - fatty acid composition 136 - fatty acid profiles 158 - modification by genetic engineering 138 plant tissue cultures, expression of pathways 597 - novel pharmacologically active compounds 621 - secondary metabolites 595 - technology 596 plants, as sources of nonribosomal peptides 311f - as sources of ribosomal peptides 299 - secondary metabolites, from cell cultures 595ff - - uses 595 platelet activating factor receptor antagonists 114f platelet aggregation, screening of receptor-active compounds 110 platelet aggregation inhibitors, from basidiomycetes 512f - - structures 512 pleuromutilin, antibiotic action 496 - production by fermentation of Pleurotus 496 - structure 495 pleurotellic acid, structure 505 pleurotellol, cytotoxic activity 507 - structure 505 Pleurotus sp., pleuromutilin production 496 podophyllotoxins 617 - as inhibitors of topoisomerase 11 655 podoscyphic acid, structure 510 poly-b-hydroxy alkanoates 177ff - Occurrence in Pseudomonas oleovorans 180 poly-phydroxybutyrate (PHB), accumulation in bacteria 178 - biosynthetic pathway 178 - structure 177 - synthesis by condensation of acetyl-CoA units 179 polyenes, structures 34 polyesters 177ff - poly-P-hydroxyalkanoates 177ff polyethers, structures 34 polyketide drugs, structural class 39 polyketide synthases 32ff - scheme 33 polyketides 32ff - structures 34 polyprene hydrocarbons, as components of the alga Botryococcus braunii 175f polyprenoids 175f polysaccharides, immunologically active -, from plant cell cultures 612 polysporic acid, iron-binding pigments 231 - structure 231 polyunsaturated fatty acids (PUFA) 138f - dietary uses 158 - in bacteria 147ff
- in fungi 151, 158
723
- in microalgae 165ff
- nutritional roles 164
Porphyridium curenturn, formation of polyunsaturated fatty acids 168 ppGpp 61ff, 75ff ppGpp synthesis, activation of - 63 pre-epidermin 342 pre-lantibiotics see lantibiotic prepeptides pre-nisin 341 proferrorosamine A, structure 208 prokaryotes, DNA transfer to eukarvotes 263 1 transformation systems, for plactam production 253 protease inhibitors, microbial -, antitumor effects 675 protein kinase inhibitors, antitumor agents 677ff - structures 678 protein phosphatase inhibitors, antitumor agents 677ff - structures 678 protein phosphorylation, role in antibiotic production 81f protoberberine alkaloids, from plant cell cultures 603ff - structures 604 protochelin, catecholate siderophore of Azotobacter vinelandii 203 - structure 203 protooncogene induction, inhibition of - 682 pseudobactins 213ff - peptide siderophore of Pseudomonas jluorescens 213 - structures 213, 215 Pseudomonas aeruginosa, pyochelin 207f - rhamnolipids 180f Pseudomonas cepacia, cepabactin 224 Pseudomonas cepaciu-like strains, ornibactins 216f Pseudomonas jluorescens, ferribactin 214 - ferrocins 213 - pseudobactins 213ff - pyoverdins 213f Pseudomonas oleovorans, poly-phydroxyalkanoates 180 PUFA see polyunsaturated fatty acids pulcherrimic acid, iron-binding compounds 233 - structure 233 putrebactin, structure 211 pyochelin, catecholate siderophore of Pseudomonus aeruginosa 207f - structure 207 pyoverdins 213ff - defective mutants 214 - peptide siderophore of Pseudomonas jluorescens 213 - structures 213, 215
724
Index
- of peptide ligands, screening o antagonists f
quinoline alkaloids, from plant cell cultures 614 p-quinones, antitumor agents 652 quinoxaline antibiotics, antitumor agents 652 quorum sensors 65
ranalexin, structural gene 280 rapamycin 546ff - biosynthetic clusters 21 - immunosuppressive properties 546f - structure 546 rapeseed oil, genetic engineering of - 138 ras-farnesyltransferaseinhibitors 685 - structures 686 receptor-active compounds 107ff see also receptor agonists, receptor antagonists, - new screening methods 107ff - screening assays 109ff - - Magnus test 109f - - using radio-labeled ligands 110 - - using recombinant cells l l O f - screening of, agonists 122ff - - antagonists l l l f f receptor agonists see also receptor-active compounds - erythromycin A 122ff - motilides 122f - muscarine 123 - zearalenone 123 receptor antagonists l l l f f see also receptor-active compounds, - androgen receptor 116 - ANP receptor 120f - arginine-vasopressin receptor 121 - C5a receptor 121 - CCK receptors 116, 121 - devazepide receptor 121 - dopamine receptor 114 - endothelin receptor 116ff - estrogen receptor 115f - fibrinogen receptor 115 - ligand receptors 112 - LTB4 receptor 114 - muscarinic acetylcholine receptor 112f - neurokinin receptors 120 - NMDA receptor 114 - oxytocin receptor 121 - PAF receptor 114f - peptide ligands 117ff - structures 113 - substance P receptor 120 receptor binding, of viruses, inhibitors 124ff receptor genes, of recombinant cells, use in screens of receptor-active compounds 110 receptors, of low molecular weight ligands, screening of antagonists 112ff
116ff resistance genes, in aminoglycoside producers 440 resistance mechanisms, in bacteria 44Off reveromycins, interruption of eukaryotic cell growth 682 reverse transcriptase inhibitors, from basidiomycetes 509ff - - structures 51Of rhamnolipids 180f - structure 181 - surfactant formation by Candida spp. 180 rhizobactin 1021, dihydroxamate siderophore of Rhizobium meliloti 221 rhizobactin DM4, carboxylate-type siderophore of Rhizobium meliloti 223 - structure 223 Rhizobium meliloti, rhizobactin 1021 221ff - rhizobactin DM4 223 rhizoferrin 229f - fungal siderophores 229 - structure 229 Rhodosporidium toruloides, fatty acyl composition 152 Rhodotorula glutinis, lipid accumulation patterns 140f rhodotorulic acid, fungal siderophores 228 - structure 228 ribostamin family, structure of - 467 ribostamycins 434 ristocetin A 372f - structure 373 RNA polymerase, of E. coli, d subunit 64 rosmarinic acid, from plant cell cultures 598ff - - yields 599 - structure 599 rotamases 547f S Saccharomyces cerevisiae, accumulation of sterols 172 - formation of ergosterol 172 - transformed -, screening of steroid receptors 111 Saccharopolyspora erythraea, polyketide synthase 33 saframycin A, antitumor agent, structure 662 salivaricin A, prepeptide structure 345 - primary structure 332 - structure 335 sanguinarine, from plant cell cultures, yields 606 saponin aglycone, from plant cultures 611f saponins, from plant tissue cultures 611f schizokinen 220f schizokinen A, structure 221 schizonellins, structure 504
Index
725
scopolamine, from hairy root cultures 620

- from plant cell cultures 613 - production in transgenic Atropa belladonna
- structure 624
scorodonin 502f
- antifungal activity 503 - structure 502
plants 625f
scorpions, as sources of nonribosomal peptides 314 - as sources of ribosomal peptides 301 screening, of natural products, compilation of 4ff - of bioactive metabolites, from basidiomycetes 496 - of dalbaheptides 381 - of natural products 3ff - of plant secondary metabolites 595 - of receptor-active compounds 107ff sea hares, as sources of nonribosomal peptides 314 secondary metabolic aminotransferases 422 secondary metabolism, evolution of - 16 - general aspects Iff - negative feedback control 30 - overproduction of enzymes, in transgenic plant cell cultures 627 - peptide bond formation 36 - regulation of - 17ff - - in eukaryotes 22f - regulatory molecules, induction of cytodifferentiation 16 secondary metabolite formation, effect, of nutrient depletion 24ff - - of stress conditions 24ff secondary metabolites, adaptation to metabolic imbalances 13 - agricultural uses 9ff - - plant growth regulators 9 - as a signal of interspecific communication 13, 15 - as drugs 2ff - - antimicrobial agents 4ff - - antitumor agents 6f - as endogenous signals 13 - as metabolic reserves 13, 15 - biological role 493 - biosynthesis, biosynthetic modifications 37f - - use of cell-free enzymes 37 - biosynthetic genes clusters 17ff - enhancement by metabolic engineering 623ff - - nicotine 624f - - scopolamine 625f - - serotonin 624 - expression of regulatory mechanisms 23ff - formation from a single precuror 31f - genetic instability 26f - immunosuppressors 7
incorporation into cellular structures 15 main biosynthetic pathways 31ff overproduction of - 29f pharmacological activities 7ff precursor pools of - 29f producing organisms 4ff production in plant cell cultures 593ff - properties 4ff - regulation in streptomycetes 443ff - roles in producing organisms l l f f - structural classification 38ff - structural types 4ff - sugar components in - 397ff secondary pathways, single site manipulations 626 seldomycin family, structure of - 469 seldomycins, biosynthetic pathway 433 septicemia, in mice, effect of dalbaheptides 380 serotonin, biosynthesis in plant cell cultures 624 - structure 624 serpentine, from plant cell cultures 608f - - yields 608 sexual hormone production, inhibitors of -, treatment of hormone-dependent tumors 685 shikimic acid, from plant cell cultures 603 shikonins, from plant cell cultures 601f - - yields 601 sideramines, ferric ion-chelating compounds 493 siderochelin A, catecholate siderophore of Nocardia sp. 208f - structure 208 siderophore-mediated iron transport 233ff siderophores 14, 199ff see also bacterial siderophores, fungal siderophores - biological properties 226 - carboxylate-type 223ff - catecholate-type 201ff - citrate hydroxamates 22Off - ferric uptake regulation proteins 200 - history 201 - hydroxamate-type 209ff - iron(II1) binding ligands 201 - metal-to-ligand charge transfer bands, coloration 200 - mycobactins 217ff - peptide-type 212ff - role in iron nutrition 235 signal cascades, in Bacillus subtilis, surfactin synthesis 68 signaling molecules, from bacteria and fungi 14f - - structures 14 - PPGPP 61 single cell oils 135ff - advantage over single cell protein 185 snakes, as sources of ribosomal peptides 302 solution structures, of lantibiotics 338ff sophorolipids, structure 180 - surfactant formation by Pseudomonas spp. 180
-
726
Index
sorbistins 439ff
- biogenesis of - 439 - structure 472
- oleandomycin production, growth-phase de- phenoxazinone synthase 73

Streptomyces clavuligerus, plactam biosynthesis 251 - clustering of p-lactam biosynthetic genes 259ff - penicillin biosynthetic enzymes 254ff Streptomyces coelicolor 71ff, 81, 83 - antibiotic regulatory genes 85ff - biosynthetic genes 447 - methylenomycin production 74f, 91f - - on minimal medium 75 - ppGpp production 75ff - regulatory cassettes for antibiotic export 92 Streptomyces fradiae, tylosin resistance 93 Streptomyces glaucescens, biosynthetic genes 447 - regulatory cassettes for antibiotic export 92 Streptomyces griseus, A-factor cascade 87f - A-factor-induced regulation 446 - candicidin production 71f - - phosphate repression 74 - decision phase 445 - decision-making growth phase 447 - exponential growth phase 445 - stationary growth phase 445 - streptomycin biosynthetic genes 91 - streptomycin production, regulation of - 445 Streptomyces hygroscopicus, bialaphos production 91 - rapamycin production 546f Streptomyces lavendulae, formycin production 77 Streptomyces peuceticus, daunorubicin production 91 Streptomyces pilosus, ferrioxamines 209f Streptomyces sp., formation of anticancer drugs 660 Streptomyces tsukubaensis, FK506 production 546 Streptomyces virginiae 78f streptomycetes, antibiotic formation, factors determining onset of - 86 - antibiotic production 7Off - - regulation of - 70 - ybutyrolactones, A-factor 78 - induction of antibiotic resistance 92ff - regulation of antibiotic production, interfering metabolites 73 - regulation of secondary metabolite production 443ff - regulatory genes 27 streptomycin, biosynthetic genes 417 - - in Streptomyces griseus 90 - pathway for its release 426 streptomycin biosynthesis, regulation of - 88 streptomycin production, enzymes, in biosynthetic pathway 418f - gene clusters 420 - gene products 418 - in Streptomyces griseus, regulation of - 445 pendency 72
soybean oil, fatty acid composition 136

- genetic engineering of - 138
sp&tinomychs, streptomycin-related aminoglycosides 427 - structure 464 spenolimycin, structure 464 sphingolipids, of yeasts, structure 183 spicamycin, antitumor activity 647 spiculisporic acid, structure 181 spiders, as sources of nonribosomal peptides 314 - as sources of ribosomal peptides 301 spiramycin formation, regulatory gene of in Streptomyces ambofaciens 90 Spirulina maxima, as source of polyunsaturated fatty acids 167 Spirulina platensis, as source of polyunsaturated fatty acids 167 - pcarotene content 172 sponges, as sources of nonribosomal peptides 312f Staphylococcus hyicus, staphyloferrin A 223 staphyloferrin, carboxylate-type siderophore of Staphylococcus hyicus 223f staphyloferrin A 2231 - structure 224 staphyloferrin B, structure 224 stationary growth phase, of Bacillus subtilis 67 staurosporin, inhibition of protein kinase C 677 stearic acid, increase in yeasts 15Off - - by direct feeding 15Of - - by inhibition of stearoyl desaturase 151f - - by metabolic manipulation 154 - - by mutation 152ff stearoyl desaturase, inhibition of - 151f stearoylphytosphingosine, structure 184 stearoylvelutinal, antimicrobial activity 495 sterigmatocystin, biosynthetic clusters 21 steroid receptors, screening with transformed Saccharomyces 111 steroidal compounds, from plant cell cultures 611f sterols 170ff - accumulation in yeasts 171f - commercial extraction of eukaryotes 170ff - structures 171 streptidine, biosynthetic pathway 417,421 streptococcin A-FF22, prepeptide structure 345 - primary structure 334f Streptomyces, cytodifferentiation, A-factor 27 - - signal cascade 27 - hydroxamate siderophores, ferrioxamine type 211f Streptomyces ambofaciens 92 - spiramycin production in - 90 Streptomyces antibioticus 72ff
Index streptomycins 415ff basic pathway formula 463 biosynthetic pathway 416 condensation of subunits 425 export of - 425 processing of subunits 425 streptonigrin, cytotoxic antitumor agent 656 striatals 493f - biosynthesis 494 - structure 504 striatins, cytotoxic activity 507 - structure 504 strobilurins 496ff - blockage o respiration in eukaryotes 497 f - fungistatic activity 497 - mode of action - 497ff - possible functions 501 - producing organisms 499 - structures 498 substance P receptor antagonists 120 subtilin, prepeptide structure 345 - primary structure 331ff - solution structure, in aqueous solution 339 - - in lipophilic solvents 339 - structural gene 343f sugar components, attachment to cyclitols 405ff - basic pathways 402ff - biogenesis 402ff - in secondary metabolites 397ff sugar derivatives, monomeric - 436f sunflower oil, genetic engineering of - 138 surfactin synthesis 67f - by Bacillus subtilis 67ff - - signal cascade 68 suspension cultures see plant cell suspension cultures Syringa vulgaris, production of verbascoside 599f
-
727
T cell activation, cytokine receptor expression 545 - gene 545 - inhibition o -, by rapamycin 547 f - - role of immunophilins 548f - release of cytokines 545f - role of calcineurin 549 - signal recognition 542ff - - binding of antigen 542f - - costimulatory signals 542ff - signal transduction 544f tautomycin, inhibition of protein kinase C 677 taxol 665ff - biosynthetic pathway 617 - from plant tissue cultures 614ff - - yields 616 - structure 615, 668
Taxus brevifolia 615 teicoplanin 375f - effect in mouse septicemia 380 - heptapeptide, amino acid composition 384 - structure 376 teicoplanin aglycone, stereo model 377 terpenes, biosynthesis 35 terpenoids, structural class 39 testosterone 5a-reductase inhibitors 686 tetracyclines, structures 34 tetrafibricin, fibrinogen binding antagonists 115 Thalictrum sp., accumulation of protoberberines 604f thelephoric acid, iron-binding compounds 232 - structure 232 thioguanine, antitumor activity 647 thiouridine, antitumor activity 647 tiamulin, antibiotic action 496 - structure 495 tissue culture technology see plant cell cultures, plant tissue cultures tolypocin, iron-binding compounds 232 - structure 232 topoisomerases, inhibition by anticancer drugs 653ff torulene 175 - structure 174 transduction of an extracellular signal, protein kinases, effect on neoplastic cell growth 675 transgenic plant cell cultures 623ff - overexpression of enzymes 627 transplant rejection therapy, use of cyclosporin A 539f trehalosamine family, structure of - 470 trehalosamines 437f - as glycosidase inhibitors 437 trehazolin, structure 472 - trehalase inhibitor 439 triacylglycerols 146ff - biosynthetic pathway 144 - glucose conversion, overall stoichiometry 145 - pathway of synthesis 143ff trichostatins, phosphoinositol kinase inhibition 681 triglycerides see triacylglycerols triptolide, structure 615 tropane alkaloids, from plant cell cultures 613f - from plant hairy root cultures 619 tumor cells see also cancer cells - efflux of cytotoxic drugs 672 - multidrug resistance 540,551 - redifferentiation of - 683 tylosin resistance, in streptomycetes, induction of - 93 tyromycin, structure 519 tyrosine protein kinase inhibitors 680
728
U
Index violaxanthin, structure 173 virus receptor binding, inhibition, by chloropeptins 124ff - - by isochromophilones 124f - - of HIV entry 124ff vitamin A, structure 174 formation in yeasts 172 vitamin DZ, vulnibactin 205f - catecholate siderophore of Vibrio vulnificus 206 - structure 205 vulpinic acid, iron-binding compounds 232 - structure 232
UK 63.052, antitumor agent, structure 651
validamycin family, biogenesis of - 438 - 471 validamycins 438f vancomycin 373f - biotransformation of - 390 - effect in mouse septicemia 380 - structure 373 vanilla flavor, from plant cell cultures 617f Vanilla planifolia, plant cell cultures 618 vanillin 617 variegatic acid, iron-binding compounds 232 - structure 232 variegatin, iron-binding pigments 231 - structure 231 verbascoside, from plant cell cultures, yields 600 - structure 599 Vibrio anguillarum, anguibactin 206f vibriobactin 205f - catecholate siderophore of Vibrio cholerae 205 - structure 205 Vibrio cholerae, vibriobactin 205f vibrioferrin, carboxylate-type siderophore of Vibrio parahaemolyticus 224 - structure 224 Vibrio fischeri, light emission 66 - OHHL 65f - - structure 65 Vibrio fluvialis, fluvibactin 206 Vibrio parahaemolyticus, vibroferrin 224 Vibrio vulnificus, vulnibactin 206 vinblastin, from plant cell cultures 614 - inhibition of the microtubular system 668 - structure 615, 668 vinca alkaloids 665 vincristin, from plant cell cultures 614 - inhibition of the microtubular system 668 - structure 668 vindesin, inhibition of the microtubular system 668 - structure 668
- structure of
8
wax esters 176f

- pathways to a diunsaturated wax ester 176
xerocomic acid, iron-binding compounds 232 - structure 232 xerulin, inhibitor of cholesterol biosynthesis 518 - structure 519 Y yeasts see also oleaginous yeasts - sphingolipids 183 - sterol contents 171f - yeast lecithin 183 yellamycin B, antitumor agent 659 - structure 659 yersiniabactin 207f - catecholate siderophore of Yersinia enterocolitica 208 - structure 207 Yersinia enterocolitica, yersiniabactin 208
2 zealarenone, as estrogen receptor agonist 112 zearalenone 123 zymosterol, structure 171

Products of Secondary Metabolism:wiley Encyclopedia

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Products of Secondary Metabolism:wiley Encyclopedia

Uploaded by

Copyright:

Available Formats

Biotechnology

Second Edition Volume 7

Products of Secondary Metabolism

A Multi-Volume Comprehensive Treatise

Products of Secondary Metabolism

Dr. G. R e e d 1914 N. Prospect Ave. #61 Milwaukee, WI 53202-1401

P.O. Box 100131 D-33501 Bielefeld FRG

Scientific Advisory Board

Pro$ Dr. M. J. Beker

Pro$ Dr. T. K. Ghose

Pro$ Dr. J. D. BuLock

Pro$ Dr. I. Goldberg

Pro$ Dr. C. L. Cooney

Pro$ Dr. G. Goma

Pro$ Dr. H. W. Doelle

Prof Dr. J. Drews

Pro$ Dr. E. H. Houwink

Pro$ Dr. A. Fiechter

Pro$ Dr. A. E. Humphrey

Scientific Advisory Board

Prof. Dr. I. Karube

Prof. Dr. K. Schiigerl

Prof. Dr. M . A. Lachance

Prof. Dr. P. Sensi

Prof. Dr. Y. Liu

Prof. Dr. Y. H. Tan

Prof. Dr. J. F. Martin

Prof. Dr. D. Thomas

Prof. Dr. B. Mattiasson

Prof. Dr. W . Verstraete

Prof. Dr. M . Roehr

Prof. Dr. E.-L. Winnacker

Prof. Dr. H. Sahm

H. von Dohren, H. Kleinkauf

Prof. Dr. Timm Anke

Prof. Keith Chater

Dr. Jochen Berlin

Dr. Mervin Bibb

Dr. Hans von Dohren

Dr. Bruno Cavalleri

Dr. Klausjiirgen Dornberger

Dr. Hartmut Drechsel

Prof. Dr. Gunter Jung

Dr. Gerhard Erkel

Dr. Jorg Kallen

Dr. Hans Fliri

Prof. Dr. Horst Kleinkauf

Dr. Friedrich Gotz

Dr. Giancarlo Lancini

Prof. Dr. Udo Grafe

Dr. Henk van Liempt

Dr. Ralph Jack

Dr. Vincent Mikol

Prof. Dr. Satoshi Omura

Dr. Kurt Schorgendorfer

Prof. Dr. Wolfgang Piepersberg

Dr. Torsten Schwecke

Dr. Valkrie F.J. Quesniaux

Dr. Paul L. Skatrud

Prof. Colin Ratledge

Dr. Haruo Tanaka

Prof. Dr. habil. Hans-Peter Saluz