Effet chez le porcelet d'une exposition Ã un rÃ©gime co-contaminÃ© en ...

AUTEUR : Bertrand GRENIERTITRE : Effet chez le porcelet d’une exposition à un régime co-contaminé en mycotoxines, etappréciation des stratégies de lutteDIRECTEURS DE THESE : Dr. Isabelle OSWALD et Pr. Martine KOLF-CLAUWRESUME :Les mycotoxines sont des métabolites secondaires des moisissures qui peuvent naturellementcontaminer de nombreuses denrées alimentaires, notamment les céréales. Dans les travaux dethèse, nous nous sommes intéressés à deux mycotoxines majeures produites par des champignonsdu genre Fusarium, le Déoxynivalénol (DON) et la Fumonisine (FB). Les objectifs de la thèse ont étéde déterminer les effets individuels et combinés d’une contamination en DON et FB chez le porc, uneespèce cible et sensible aux mycotoxines. Les effets sur les fonctions immunitaires lors d’unchallenge antigénique ainsi que sur les fonctions intestinales ont été évalués. Par ailleurs, dans lecadre d’un partenariat avec un industriel, nous avons évalué in vivo les effets de méthodes dedétoxification par biotransformation et ciblant spécifiquement ces deux toxines. Chez le porc,l’ingestion d’aliments contaminés avec de faibles doses de mycotoxines (DON, 3 mg/kg ; FB, 6 mg/kg)a provoqué des lésions tissulaires (foie, reins et poumons) et a fortement altéré la mise en placed’une réponse immunitaire spécifique de l’antigène (expression des cytokines, prolifération deslymphocytes et anticorps spécifiques). Les animaux ont été significativement plus affectés après laconsommation du régime co-contaminé, et l’interaction a pu être considérée comme additive. Deplus, les paramètres intestinaux examinés ont révélé des changements dans la morphologie, dans leprofil de sécrétion des cytokines et dans l’adhésion cellulaire. L’interaction des deux toxines a pu iciêtre caractérisée comme moins qu’additive. Les approches de détoxification biologique proposéespar l’industriel étaient basées sur la transformation par voie enzymatique du DON et des FB, à partird’un microorganisme entier et d’une enzyme respectivement. La stratégie d’élimination des FB asuscité un intérêt plus important étant donné que cette méthode est non commercialisée et en coursde développement. Ainsi, la toxicité du produit d’hydrolyse de la FB1 (mycotoxine principale de lafamille des FB) obtenu initialement par traitement enzymatique, a été comparée in vivo à celle de lamolécule mère la FB1. Les résultats ont montré que l’hydrolyse de la FB1 réduisait fortement latoxicité hépatique et intestinale chez les porcelets. L’expérimentation animale avec le DON et la FB,seuls ou en combinaison a ensuite été reproduite afin de déterminer l’efficacité d’hydrolyse de ceprocédé chez le porc après incorporation de l’enzyme dans les aliments contaminés. Dans cesaliments, le microorganisme entier ciblant le DON avait également été inclus. La nette diminution dumarqueur d’exposition des FB et la neutralisation partielle ou totale des effets ont suggéré que leprocédé avait fortement réduit la biodisponibilité des FB dans le tractus gastro-intestinal. Cetteobservation a aussi été en partie confirmée pour l’approche de dégradation du DON. Labiotransformation par voie enzymatique des mycotoxines représente ainsi une stratégiebiotechnologique prometteuse dans la lutte contre ces contaminants.MOT-CLES : Déoxynivalénol, Fumonisine, porc, co-contamination, réponse immunitaire, réponseintestinale, marqueur d’exposition, biotransformation enzymatiqueDISCIPLINE : Toxicologie alimentaireINTITULE ET ADRESSE DU LABORATOIRE :Institut National de la Recherche Agronomique – INRAUnité ToxAlim, Equipe d’Immuno-Mycotoxicologie180, chemin de Tournefeuille – BP 9317331027 TOULOUSE Cedex 3 – FRANCE

AUTHOR : Bertrand GRENIERTITLE : Effect in pigs of the exposure to a mycotoxins co-contaminated diet, and evaluation ofcontrol strategiesTHESIS SUPERVISORS : Dr. Isabelle OSWALD and Pr. Martine KOLF-CLAUWABSTRACT :Mycotoxins are secondary metabolites of fungi that are natural contaminants of severalcommodities, in particular cereals. In the present work, we focused on two major mycotoxinsproduced by the Fusarium genus, Deoxynivalenol (DON) and Fumonisin (FB). The main objectives ofthe thesis were to determine the toxic effects of individual and combined DON and FB contaminationin pig, a target species highly sensitive to mycotoxins. The effects on the immune functions followingan antigenic challenge and also on the intestinal functions were evaluated. Besides, within theframework of an industrial partnership, we evaluated in vivo the effects of detoxifying methods bybiotransformation and targeting specifically these two toxins. In pigs, ingestion of contaminatedfeeds with low doses of mycotoxins (DON, 3 mg/kg ; FB, 6 mg/kg) triggered tissular lesions (liver,kidneys and lungs) and strongly impaired the establishment of the antigenic immune response(cytokines expression, lymphocytes proliferation and specific antibodies). Animals consuming the cocontaminateddiet were more affected and the interaction could be considered as additive. Inaddition, changes in morphology, in profile of cytokines secretion and in cell adhesion were observedat intestinal level. The interaction here could be characterized as less than additive. The biologicaldetoxification approaches proposed by the industrial were based on the transformation by enzymaticway of DON and FB, from intact microorganism and enzyme respectively. We paid a particularattention to the strategy of FB removal as this method is not marketed and still in development.Therefore, the toxicity of the hydrolysis product of FB1 (major mycotoxin in the FB group) initiallyobtained by enzymatic way, was compared in vivo to the toxicity of the parent compound FB1.Results showed that the hydrolysis of FB1 strongly reduced the toxicity in piglets at intestinal andhepatic levels. The animal experiment with DON and FB, alone or in combination was then repeatedin order to determine in pigs the hydrolysis efficiency of this process when enzyme was incorporatedin contaminated feeds. In these feeds, the intact microorganism toward DON was also included. Themarked decrease of the biomarker of exposure to FB and the partial or total counteraction of theeffects suggested that the process had greatly reduced the FB bioavailability in the gastrointestinaltract. This observation was also in part confirmed for the method degrading DON. Thebiotransformation method of mycotoxins by enzymatic way represents therefore a promisingbiotechnological strategy in the control of these contaminants.KEY WORDS : Deoxynivalenol, Fumonisin, pig, co-contamination, immune response, intestinalresponse, biomarker of exposure, enzymatic biotransformationDISCIPLINE : Food ToxicologyHEADING AND ADRESS OF THE LABORATORY :Institut National de la Recherche Agronomique – INRAUnité ToxAlim, Equipe d’Immuno-Mycotoxicologie180, chemin de Tournefeuille – BP 9317331027 TOULOUSE Cedex 3 – FRANCE

REMERCIEMENTSLe travail ici présenté a été rendu possible grâce à un partenariat entre l’Association Nationale dela Recherche Technique (ANRT), le laboratoire d’Immuno-Mycotoxicologie de l’Institut National pourla Recherche Agronomique (INRA) et la société BIOMIN, formalisé dans une Convention Industriellede Formation par la Recherche (CIFRE n°065/2007).Je tiens à remercier toutes les personnes qui ont accepté de faire partie de ce jury :Le Dr. Nathalie Le Floc’h-Burban et le Pr. Yvan Larondelle, qui ont accepté d’évaluer ce travail enqualité de rapporteur.Le Pr. Cyril Feidt, qui a accepté de faire partie de ce jury en tant qu’examinateur.Le Dr. Wulf-Dieter Moll, le Dr. Gerd Schatzmayr et Mr. Christian Tenier (ces deux derniers étantnon présents dans le jury), pour m’avoir accordé votre confiance dans la réalisation de ce projet,m’avoir ouvert la porte sur le monde de l’entreprise. Merci Dieter d’être examinateur de ce travail.Le Pr. Martine Kolf-Clauw, co-directrice de thèse, pour sa disponibilité et ses conseils judicieuxtout au long de ces 3 années ½ de thèse. Je te remercie de ton implication dans les expérimentationsanimales et dans la rédaction.Le Dr. Isabelle Oswald, directrice de thèse, pour son accueil au sein de l’équipe d’immunomycotoxicologie.Je t’adresse tout particulièrement mes remerciements, la pertinence de tesconseils, ta capacité à gérer cette équipe et tous ces projets, et également ta capacité à répondreaussi vite à tes mails même à l’autre bout de la planète, m’impressionneront toujours. Merci pour taconfiance, ta disponibilité (même en vacances) et pour m’avoir permis de « voyager » et rencontrerde nombreuses personnes à travers tes collaborations ou lors des congrès.Je tiens à adresser tous mes remerciements, ma sympathie et mon affection aux personnes qui ontpartagé avec moi – de près ou de loin – ces 3 ans ½ de travail :- INRA, pôle de St Martin du Touch :Au personnel de l’équipe :Philippe, merci de ta générosité et de ta multi-compétence au labo. J’espère un jour pouvoir denouveau travailler avec une personne comme toi.Joëlle, je te remercie de ta gentillesse et de ton ouverture d’esprit. Les conversations que nousavons pu avoir sur la société, l’actualité politique ou encore la vie de tous les jours vont me manquer.Anne-Marie, sans qui les expérimentations animales auraient été impossibles. Merci pour tonprofessionnalisme et pour toutes ces pauses partagées à discuter avec toi.

Olivier, grâce à ce congrès en Autriche et cette superbe suite que nous avons partagé, j’ai pudécouvrir un chercheur fort sympathique et passionné des champignons ! J’espère que tu vas lagarder cette barbe !Je remercie tous les autres membres permanents de l’équipe pour leurs bonnes humeurs et leursdisponibilités : Laurence, Thierry, Soraya, Christiane.J’ai vu passé un nombre incroyable de stagiaires (entre 30 et 40), courtes ou longues durées, toutau long de ma thèse, et je tiens à vous remercier pour tous ces moments agréables passés au ou endehors du labo en votre compagnie. Je n’en citerai que quelques uns, Mél, Romain, Dima, Leticia. Etégalement bonne chance à tous ces nouveaux étudiants bien sympas et prêts à prendre la relève (oupas !) : Patricia, Julie, Selma, Joyce, German, Roberta.Au personnel du pôle ToxAlim :Marie-Jo, encore merci pour ta rapidité pour nous procurer toute la bibliographie et pour cetteaide très précieuse dans la reliure et l’envoi de ce manuscrit.Merci à tous ceux qui ont forcément contribué à faire de cette thèse une agréable expériencejusqu’au bout. Je pense en particulier, à Alice, Afi, Arnaud (bon courage pour ton petit bout), Fred,Hervé, Pascal, Nico, Cécile 1 et 2, Caro, Manue.Simon, je suis bien content d’avoir fait ta connaissance en thèse, je sais que nous resterons encontact, nos virées à la montagne et les soirées magrets de canard nous ont bien rapprochés ! Boncourage pour la suite.- BIOMIN :Je tiens à adresser mes remerciements à la société BIOMIN, que ce soit en France ou en Autriche.Par le biais de leurs investissements, leurs conseils mais aussi leurs confiances, ce projet de thèse apu se réaliser dans les meilleures conditions. Je remercie tout particulièrement Gerd Schatzmayr,Christian Tenier, Dieter Moll et Heidi Schwartz dans cette collaboration.Merci également aux autres collègues de chez BIOMIN, qui ont largement contribué à mesagréables séjours en Autriche ou lors des congrès : Daniel, Doris, Gertrude, Katia, Inès, Tamara, ….- Collaborateurs :Tout au long de ma thèse, j’ai pu rencontrer de nombreuses personnes et qui ont directement ouindirectement participées à la réalisation et/ou à la réussite de nos projets. Parmi ces personnes, jeciterai Ana-Paula Loureiro Bracarense, professeur à l’école vétérinaire de Londrina au Brésil et qui adepuis le début collaboré dans ce travail de thèse. Sa compétence en histopathologie a été trèsprécieuse tout au long de ce travail de thèse. Egalement un grand merci pour ton accueil lors de monséjour au Brésil.

J’ai une pensée particulière pour mes deux collègues roumaines, si gentilles et disponibles,Daniela et Ionelia. Merci de m’avoir donné l’opportunité de présenter mes travaux lors dusymposium que vous aviez organisé.Un grand merci à Georges Guillemois de l’INRA de St-Gilles pour son aide dans la fabrication denos aliments expérimentaux.- ENVT :Jean-Luc, si tu ne m’avais pas donné ma chance lors de ma première expérience professionnelle,je ne serai probablement pas là aujourd’hui. Merci pour ta confiance et ta bonne humeur.Je n’oublie pas toutes les personnes de cette période qui a précédée la thèse, avec qui je garde debons contacts : Kunta, Mag, Béa, Romain, Steph.- Activités sportives :Le sport a été mon compagnon fidèle tout au long de ces 3 ans ½ et ainsi j’adresse un grand mercià mon club de foot à 11 et à 7, au club de badminton de l’INRA, et aux stations de ski des Pyrénées.- Les amis :Ces 3 années ½ ne se seraient pas aussi bien passées si je n’avais pas été entouré de tous mesamis. D’abord toi Léon, mon pote qui m’a suivi de notre très cher Alsace jusqu’à Toulouse et quimaintenant fait le bonheur de la science à Los Angeles. Nos soirées et weeks ends avec les copains deToulouse et d’Alsace resteront des moments inoubliables. Merci donc les amis de la rue Guénot et larue Maran, merci pour toutes ces soirées League des Champions et ces week ends de détente, Nico,Yo, Sebbie, Ben, Aurel, Maud, Alicia, Nickauch c’était génial de vous avoir rencontré sur Toulouse !J’ai aussi une pensée pour mes amis d’enfance et de fac, les retours en Alsace ou les visites chezles uns et les autres sont toujours un plaisir. J’espère qu’on continuera à se voir régulièrement, merciles amis, Steph, Zgueg, Portos, Rik, Adrien, Eric, Yannick, José, Bastien.- La famille :Et enfin, je dédie tout spécialement mon travail de thèse à mes parents, Marie-Claude et Patrice,qui m’ont toujours soutenu et à qui je dois beaucoup. Merci la frangine, je suis fier de toi. Et bien sûrles liens familiaux étant très importants, je n’oublierai pas mes folles cousines, fafa et sa famille,Rémi, les tatas et tontons et mes grand-mères.Et enfin, merci à toi « Choucroute » pour ton soutien et ton amour qui m’ont tant aidé cesderniers mois.

PUBLICATIONS ET COMMUNICATIONSArticles :Grenier, B., Loureiro-Bracarense, A. P., Lucioli, J., Pacheco, G., Cossalter, A. M., Moll, W. D.,Schatzmayr, G., Oswald, I. P. 2011. Individual and combined effects of subclinical doses ofdeoxynivalenol and fumonisins in piglets. Molecular Nutrition & Food Research, DOI10.1002/mnfr.201000402, sous presse.Lucioli, J., Grenier, B., Pacheco, G., Cossalter, A. M., Moll, W. D., Schatzmayr, G., Oswald, I. PLoureiro-Bracarense, A. P. Chronic ingestion of deoxynivalenol and fumonisins induce changes inthe morphology and local immune response of the intestine of piglets. Manuscrit enpréparation.Grenier, B., Loureiro-Bracarense, A. P., Schwartz, H., Cossalter, A. M., Schatzmayr, G., Moll, W. D.,Oswald, I. P. Hydrolysis of fumonisin B1 strongly reduced toxicity in piglets at the intestinal andhepatic levels. Manuscrit en préparation.Revue et Chapitres :Grenier, B., Loureiro-Bracarense, A. P., Oswald, I. P. Physical and chemical methods for mycotoxinsdecontamination in maize. In Mycotoxin Reduction in Grain Chains: A Practical Guide. Edited byLogrieco, A. F., sous presse.Pedrosa, K., Schatzmayr, D., Jans, D., Bertin, G., Grenier, B. Mycotoxin reduction in animals –adsorption and biological detoxification. In Mycotoxin Reduction in Grain Chains: A PracticalGuide. Edited by Logrieco, A. F., sous presse.Grenier, B., Oswald, I. P. Mycotoxins co-contamination: meta-analysis of published data. WorldMycotoxin Journal, soumis.Communications orales à des congrès :Grenier, B., Loureiro-Bracarense, A. P., Lucioli, J., Pacheco, G., Cossalter, A. M., Moll, W. D.,Schatzmayr, G., Oswald, I. P. 2010. The issue of mycotoxins multi-contamination: example ofDeoxynivalenol and Fumonisins co-contamination. 6 th World Mycotoxin Forum. November 8-10.Noordwijkerhout, Netherlands.Grenier, B., Loureiro-Bracarense, A. P., Lucioli, J., Pacheco, G., Cossalter, A. M., Moll, W. D.,Schatzmayr, G., Oswald, I. P. 2010. Individual and combined effects of low doses ofdeoxynivalenol and fumonisins in piglets. 9 th International Symposium of Animal Biology andNutrition. September 23-24. Bucharest, Romania.Grenier, B., Loureiro-Bracarense, A. P., Lucioli, J., Pacheco, G., Cossalter, A. M., Moll, W. D.,Schatzmayr, G., Oswald, I. P. 2010. Effect of two Fusariotoxins, Deoxynivalenol and FumonisinB1, on the pig: general toxicity and the immune response. 32 nd Mycotoxin Workshop. June 14-16. Lyngby, Danemark.

Grenier, B., Loureiro-Bracarense, A. P., Schwartz, H., Cossalter, A. M., Schatzmayr, G., Moll, W. D.,Oswald, I. P. 2010. Differential toxicity of Fumonisin B1 and its hydrolyzed form on the pigletimmune response. 7 th meeting of Immunology of Domestic Animals. May 17-18. Paris, France.Grenier, B., Loureiro-Bracarense, A. P., Cossalter, A. M., Moll, W. D., Schatzmayr, G., Oswald, I. P.2009. Evaluation des effets combinés du Deoxynivalénol et de la Fumonisine B1 sur la réponseimmunitaire du porcelet. Journées d’Animation Scientifique du Département Santé Animale del’INRA. Mai 25-28. Port-Albret, France.Communications affichées à des congrès :Grenier, B., Loureiro-Bracarense, A. P., Schwartz, H., Cossalter, A. M., Schatzmayr, G., Moll, W. D.,Oswald, I. P. 2010. Hydrolysis of Fumonisin B1 strongly reduced toxicity for piglets at theintestinal and systemic levels. 6 th World Mycotoxin Forum. November 8-10. Noordwijkerhout,Netherlands.Grenier, B., Loureiro-Bracarense, A. P., Lucioli, J., Pacheco, G., Cossalter, A. M., Moll, W. D.,Schatzmayr, G., Oswald, I. P. 2010. Effects of subclinical doses of deoxynivalenol and fumonisins,alone or in combination on the systemic and intestinal responses of piglets. 4 th World NutritionForum. October 13-16. Salzburg, Austria.Grenier, B., Loureiro-Bracarense, A. P., Schwartz, H., Cossalter, A. M., Schatzmayr, G., Moll, W. D.,Oswald, I. P. 2010. Hydrolysis of Fumonisin B1 strongly reduced toxicity for piglets at theintestinal and systemic levels. 4 th World Nutrition Forum. October 13-16. Salzburg, Austria.Grenier, B., Loureiro-Bracarense, A. P., Lucioli, J., Pacheco, G., Cossalter, A. M., Moll, W. D.,Schatzmayr, G., Oswald, I. P. 2009. Combined effects of Deoxynivalenol and Fumonisin B1 on thepig immune response. Conference of International Society for Mycotoxicology. September 9-11.Tulln, Austria.Grenier, B., Loureiro-Bracarense, A. P., Lucioli, J., Pacheco, G., Cossalter, A. M., Moll, W. D.,Schatzmayr, G., Oswald, I. P. 2008. Evaluation of individual and combined effects ofDeoxynivalenol and Fumonisin B1 on the pig immune response. 3 rd World Nutrition Forum.September 18-19. Mayrhofen, Austria.

TABLE DES MATIERES1

TABLE DES MATIERES ........................................................................................................ 1LISTE DES ABREVIATIONS ..................................................................................................................... 4FIGURES ET TABLEAUX ......................................................................................................................... 6INTRODUCTION ................................................................................................................ 8CONTEXTE DE L’ETUDE ........................................................................................................................ 9ETUDE BIBLIOGRAPHIQUE ................................................................................................................. 131. Les mycotoxines : généralités, métabolisation et effets toxiques ............................................. 142. La co-contamination en mycotoxines : méta-analyse des données publiées (revue n°1) ......... 19Introduction....................................................................................................................................... 22Caractérisation de l’interaction entre les mycotoxines ...................................................................... 23Interaction entre les différentes mycotoxines ................................................................................... 24i. interactions entre les aflatoxines et les autres mycotoxines ........................................................... 24ii. interactions entre les fusariotoxines ............................................................................................... 35iii. interactions entre l’ochratoxine A et les autres mycotoxines ........................................................ 41Conclusion ......................................................................................................................................... 45Légende des tableaux ........................................................................................................................ 473. Procédés de décontamination des denrées contaminées et réduction des mycotoxines chezl’animal .......................................................................................................................................... 483.1. Méthodes physiques et chimiques pour la décontamination du maïs contaminé(revue n°2) ..................................................................................................................................... 48Introduction....................................................................................................................................... 52i. méthodes physiques ......................................................................................................................... 53ii. méthodes chimiques ........................................................................................................................ 61Conclusion ......................................................................................................................................... 683.2. Adsorption et détoxification biologique (revue n°3) ............................................................... 69Introduction....................................................................................................................................... 72i. adsorption des mycotoxines ............................................................................................................. 74ii. détoxification biologique ................................................................................................................. 84Conclusion ......................................................................................................................................... 88TRAVAIL EXPERIMENTAL ................................................................................................. 90OBJECTIFS DE LA THESE ..................................................................................................................... 91CHAPITRE 1 – TOXICITE IN VIVO DU DEOXYNIVALENOL ET DES FUMONISINES, SEULS OU EN COMBINAISON CHEZ LEPORCELET ............................................................................................................................................. 931. Toxicité in vivo du deoxynivalénol et des fumonisines, seuls ou en combinaison sur la réponseimmunitaire (article n°1) ............................................................................................................... 94Introduction....................................................................................................................................... 96Matériel & Méthodes ........................................................................................................................ 98Résultats .......................................................................................................................................... 101Discussion ........................................................................................................................................ 1042. Toxicité in vivo du deoxynivalénol et des fumonisines, seuls ou en combinaison sur lamorphologie et la réponse locale de l’intestin (article n°2) ......................................................... 107Introduction..................................................................................................................................... 110Matériel & Méthodes ...................................................................................................................... 112Résultats .......................................................................................................................................... 1152

Discussion ........................................................................................................................................ 117Légende des Figures ......................................................................................................................... 121CHAPITRE 2 – EVALUATION DE LA TOXICITE DU PRODUIT D’HYDROLYSE DE LA FUMONISINE B1 CHEZ LE PORCELET(ARTICLE N°3) ..................................................................................................................................... 123Introduction..................................................................................................................................... 128Matériel & Méthodes ...................................................................................................................... 130Résultats .......................................................................................................................................... 134Discussion ........................................................................................................................................ 137CHAPITRE 3 – EVALUATION DES EFFETS D’AGENTS DETOXIFIANTS LORS D’UNE EXPOSITION AU DEOXYNIVALENOL ETAUX FUMONISINES CHEZ LE PORCELET ...................................................................................................... 142Résumé de l’étude ........................................................................................................................... 143Matériel & Méthodes ...................................................................................................................... 144Résultats .......................................................................................................................................... 149Discussion ........................................................................................................................................ 156DISCUSSION GENERALE .................................................................................................. 1601. Critères expérimentaux............................................................................................................ 161Le modèle animal ............................................................................................................................ 161L’exposition aux mycotoxines et à leurs dérivés .............................................................................. 163Les suppléments alimentaires : micro-organisme/enzyme a propriétés détoxifiantes ..................... 1662. Les systèmes de défense de l’organisme ................................................................................. 168L’immunité systémique acquise et spécifique .................................................................................. 168Mécanisme cellulaire de modulation de l’immunité ........................................................................ 173L’immunité intestinale ..................................................................................................................... 1753. Stratégies de lutte pour réduire les effets toxiques du deoxynivalénol et des fumonisines –approches de biotransformation ................................................................................................. 179L’Eubacterium spécifique des trichothécènes .................................................................................. 179La carboxylesterase spécifique des fumonisines .............................................................................. 180La présence simultanée de la carboxylesterase et de l’Eubacterium ................................................ 183CONCLUSIONS ............................................................................................................... 184REFERENCES BIBLIOGRAPHIQUES ................................................................................... 1883

LISTE DES ABREVIATIONSAFAflatoxineDOM-1De-époxy-déoxynivalénolAFB1Aflatoxine B1DONDéoxynivalénolAFM1Aflatoxine M1EFSAEuropean Food Safety AuthorityAJAdherens Junctione.g.exempli gratia (par exemple)ANOVA Analysis of varianceANRTAPCA.U.b.w.BGYCASTCIFREAgence Nationale de la RechercheTechniqueAntigen Presenting CellsArbitrary Unitsbody weightBright Greenish YellowCouncil for Agricultural and SciencesTechnologyConvention Industrielle deFormation par la RechercheELISAERKFAOFBFB1FLDGALTGGTEnzyme Linked Immuno SorbentAssayExtracellular signal-RegulatedKinasesFood and Agriculture Organization ofthe United NationsFumonisineFumonisine B1Fluorescent detectionGut-Associated Lymphoid TissueGamma-Glutamyl TransferaseCITCitrinineHACCPHazard Analysis Critical Control PointCMHConACPAcpmComplexe Majeurd’HistocompatibilitéConcanavaline ACyclopiazonic acidcounts per minuteHEHFBHFB1HPLCHématoxyline-EosineHydrolyzed FumonisinHydrolyzed Fumonisin B1High Performance LiquidChromatographyDADeactivating AgentsHRPHorseradish peroxydaseDARTDASDCDDGSData Analysis for Real TimeDiacetoxyscirpenolDendritic CellsDried Distillers’ Grains and SolublesHSCASH 2 O 2IARCHydrated Sodium CalciumAluminosilicateHydrogen peroxydeInternational Agency for Research onCancer4

IFNInterféronppmpartie par million (e.g. mg/kg)IBDInflammatory Bowel Diseasepvpoids vifIgImmunoglobulineRPL32Ribosomal Protein L32ILInterleukineRTReverse TranscriptionINRAIPEC-1kGyLDMAPKInstitut National de la RechercheAgronomiqueIntestinal Porcine Epithelial Cells-1kilograyLimite de DétectionMitogen-Activated Protein KinaseSaSDSSEMSoTEERSphinganineSodium Dodecyl SulfateStandard Error MeanSphingosineRésistance électriquetransépithélialeMIPMacrophage Inflammatory ProteinTCATricarballylic acids side chainsMLNMesenteric Lymph NodesTCTTrichothécènesMONMoniliformineThT helper cellsMSMass SpectrometryTJTight JunctionNIVNivalénolTNFTumor Necrosis FactorNTCNon Template ControlT-2 Toxine T-2OTAOchratoxine AUE/EUUnion Européenne/European UnionOVAPAGEPASPCRppbOvalbuminePolyacrylamide gel electrophoresisPeriod Acid-SchiffPolymerase Chain Reactionpartie par milliard (e.g. µg/kg)USFDAZEAµCiUnited State Food and DrugAdministrationZéaralénoneµCurie3-aDON 3-acétyldéoxynivalénol5

FIGURES ET TABLEAUXFigure 1 : Mycotoxines majeures produites par les champignons et se retrouvant à l’état naturel dansdifférents produits de consommation………………………………………………………………………………………………… 9Figure 2 : Tentation de Saint Antoine peint entre 1512 et 1516 par Grünewald, et épis parasités parClaviceps purpurea, excroissance (sclérote) qui s’accroche aux épis de seigle……………………………………. 9Figure 3 : Diversité structurelle des mycotoxines majeures……………………………………………………………… 10Figure 4 : Caractérisation de l’interaction entre les mycotoxines………………………………………………………23Figure 5 : Effet individuel et combiné du DON et de la FB sur l’histologie du jéjunum et de l’iléon... 115Figure 6 : Effet individuel et combiné du DON et de la FB sur le nombre de cellules inflammatoires etde cellules caliciformes dans le jéjunum et l’iléon………………………………………………………………………….. 115Figure 7 : Effet individuel et combiné du DON et de la FB sur l’expression des ARNs codant descytokines dans le jéjunum……………………………………………………………………………………………………………… 116Figure 8 : Effet individuel et combiné du DON et de la FB sur l’expression des ARNs codant descytokines dans l’iléon…………………………………………………………………………………………………………………….. 116Figure 9 : Effet individuel et combiné du DON et de la FB sur l’expression intestinale de l’E-cadhérineet de l’occludine…………………………………………………………………………………………………………………………….. 116Figure 10 : Réaction de deestérification par la carboxylestérase sur la Fumonisine B1, résultant enFumonisine B1 totalement hydrolysée, et cinétique in vitro d’hydrolyse de la Fumonisine B1 par lacarboxylestérase……………………………………………………………………………………………………………………………. 124Figure 11 : Effets des traitements FB1 et HFB1 sur les lésions du foie……………………………………………. 134Figure 12 : Effets des traitements FB1 et HFB1 sur l’intestin grêle…………………………………………………. 135Figure 13 : Transformation par voie enzymatique de l’ochratoxine A et de la zéaralénone, vial’utilisation du microorganisme Trichosporon mycotoxinivorans, et transformation par voieenzymatique des trichothécènes de type A et B, via l’utilisation du microorganisme EubacteriumBBSH 797……………………………………………………………………………………………………………………………………….. 143Figure 14 : Illustrations de la procédure d’incorporation dans l’aliment de l’agent désactivateurciblant spécifiquement les fumonisines……………………………………………………………………………………….... 144Figure 15 : Plan expérimental de la phase animale avec le protocole de vaccination…………………….. 145Figure 16 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non enagents désactivateurs sur les bases sphingoïdes, sphinganine (Sa) et sphingosine (So)…………………. 150Figure 17 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non enagents désactivateurs sur le score lésionnel du foie………………………………………………………………………. 151Figure 18 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non enagents désactivateurs sur le score lésionnel des poumons…………………………………………………………….. 152Figure 19 : Effet de l’exposition au DON avec un régime alimentaire supplémenté ou non en agentsdésactivateurs sur la hauteur des villosités du jéjunum…………………………………………………………………. 152Figure 20 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non enagents désactivateurs sur la prolifération des lymphocytes après stimulation antigénique…………… 154Figure 21 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non enagents désactivateurs sur la concentration plasmatique des immunoglobulines A et G spécifiques del’ovalbumine………………………………………………………………………………………………………………………………….. 154Figure 22 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non enagents désactivateurs sur l’expression des ARNs spléniques codant pour des cytokines……………….. 155Figure 23 : Cheptel porcin des principaux pays producteurs en 2005, et évolution des échangesmondiaux de viande de porc………………………………………………………………………………………………………….. 161Figure 24 : Effets toxicologiques provoqués chez le porc par l’ingestion des mycotoxines majeures 1686

Figure 25 : Hypothèses proposées dans l’altération de la mise en place de la réponse spécifique,d’après les résultats obtenus dans le chapitre 1 sur l’effet du DON et de la FB, seuls ou encombinaison………………………………………………………………………………………………………………………………….. 169Figure 26 : Mécanisme d’inhibition de la céramide synthase par les fumonisines et implication dans lemétabolisme des sphingolipides……………………………………………………………………………………………………. 170Figure 27 : Vue globale de la muqueuse intestinale, et des compartiments et des acteurs clés dusystème de défense de l’intestin……………………………………………………………………………………………………. 175Figure 28 : Réseau qui constitue le système de perméabilité paracellulaire de l’intestin, impliquant lesprotéines de jonction…………………………………………………………………………………………………………………….. 176Figure 29 : Effets d’exposition à des mycotoxines chez l’homme et l’animal…………………………………. 185Figure 30 : Graphique sur l’analyse mondiale de denrées agricoles. Représentation du nombre demycotoxines détectées dans les échantillons, et par conséquent de la fréquence des multicontaminations……………………………………………………………………………………………………………………………...186Tableau 1 : Interaction entre les Aflatoxines et les Fumonisines……………………………………………………… 24Tableau 2 : Interaction entre les Aflatoxines et l’Ochratoxine A………………………………………………………. 27Tableau 3 : Interaction entre les Aflatoxines et les Trichothécènes…………………………………………………. 31Tableau 4 : Interaction entre les Aflatoxines et les autres mycotoxines…………………………………………… 33Tableau 5 : Interaction entre les fusariotoxines………………………………………………………………………………. 35Tableau 6 : Interaction entre l’Ochratoxine A et les autres mycotoxines…………………………………………. 41Tableau 7 : Evaluation toxicologique des procédés de décontamination du maïs……………………………. 58Tableau 8 : Résultats des effets des HSCAS dans les aliments contaminés en AFB1 chez différentesespèces………………………………………………………………………………………………………………………………………….... 75Tableau 9 : Résultats des effets des parois de levure dans les aliments contaminés en mycotoxineschez différentes espèces………………………………………………………………………………………………………………….. 80Tableau 10 : Origines et dilutions des anticorps primaires utilisés pour la détection des protéines dejonction et de la β-actine en Western Blot…………………………………………………………………………………….. 113Tableau 11 : Séquence nucléotidique des primers utilisés en PCR temps réel……………………………….. 114Tableau 12 : Séquence nucléotidique des primers utilisés en PCR temps réel……………………………….. 132Tableau 13 : Effets des traitements FB1 et HFB1 sur les paramètres biochimiques……………………….. 134Tableau 14 : Effets des traitements FB1 et HFB1 sur l’expression des cytokines du foie………………… 134Tableau 15 : Effets des traitements FB1 et HFB1 sur le ratio sphinganine/sphingosine dans leséchantillons biologiques………………………………………………………………………………………………………………… 135Tableau 16 : Effets des traitements FB1 et HFB1 sur la hauteur des villosités dans l’intestin grêle… 135Tableau 17 : Effets des traitements FB1 et HFB1 sur l’expression des cytokines dans l’intestin grêle etdans les ganglions mésentériques………………………………………………………………………………………………….. 135Tableau 18 : Régimes formulés et teneurs finales en mycotoxines………………………………………………… 144Tableau 19 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou nonen agents désactivateurs sur le gain de poids total………………………………………………………………………… 149Tableau 20 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou nonen agents désactivateurs sur certains paramètres hématologiques et biochimiques à J35…………….. 149Tableau 21 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou nonen agents désactivateurs sur l’index de prolifération des hépatocytes…………………………………………… 150Tableau 22 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou nonen agents désactivateurs sur la population cellulaire et la prolifération cellulaire du jéjunum et del’iléon…………………………………………………………………………………………………………………………………………….. 153Tableau 23 : Teneurs maximales recommandées en mg/kg pour un aliment pour animaux ayant untaux d’humidité de 12%....................................................................................................................... 161Tableau 24 : Analyse de denrées agricoles de janvier à décembre 2009 et détermination des niveauxde contaminations mondiales en mycotoxines………………………………………………………………………………. 1647

INTRODUCTION8

Figure 1 :Mycotoxines majeures produites par les champignons et se retrouvant à l’état naturel dansdifférents produits de consommation

INTRODUCTIONCONTEXTE DE L’ETUDELes mycotoxines sont des produits du métabolisme secondaire de moisissures pouvant sedévelopper sur la plante au champ ou en cours de stockage, et douées de potentialités toxiques àl’égard de l’homme et des animaux. Les moisissures toxinogènes peuvent se développer sous tous lesclimats, sur tous les supports solides ou liquides, dès l’instant qu’il y a des éléments nutritifs et del’humidité (activité en eau A w supérieure à 0,6).Les toxines produites se retrouvent ainsi à l’état de contaminants naturels dans de nombreusesdenrées d’origine végétale, notamment les céréales (maïs, blé, orge, soja) mais aussi les fruits, noix,amandes, grains, fourrages, et les aliments composés et manufacturés contenant ces matièrespremières destinés à l’alimentation humaine et animale (Figure 1).Il a été recensé que les mycotoxines contaminent 25 à 40 % des productions céréalièresmondiales, à des niveaux mesurables. Leurs effets toxiques sur l’homme et l’animal, leur stabilité lorsdes processus de transformation et de cuisson des aliments, et leur présence sur de nombreuxproduits agricoles justifient ainsi l’attention croissante qui leur est portée (nombreuses recherchesen cours, renforcement de la législation en Europe).Historiquement, de nombreuses pathologies liées à des épisodes de contamination enmycotoxines ont été rapportés. L’une des plus connues s’est produite au Moyen-Age, égalementappelée le « Mal des Ardents » ou « Feu de Saint-Antoine », et provoquée par les toxines deClaviceps élaborées par l’ergot de seigle (Figure 2).Figure 2 :Tentation de Saint Antoine peint entre 1512 et 1516 parGrünewald (photo à droite)Epis parasités par Claviceps purpurea, excroissance (sclérote)qui s’accroche aux épis de seigle (photo ci-dessous)9

Figure 3 :Diversité structurelle des mycotoxines majeuresFumonisine B1DéoxynivalénolZéaralénonePatulineAflatoxine B1Ochratoxine A

INTRODUCTIONElle se présentait sous la forme de délires, prostrations, douleurs violentes, abcès, gangrènes desextrémités aboutissant à des infirmités graves et incurables. Des épidémies ont sévi du 8ème au16ème siècle en raison des conditions d’alimentation misérables des populations, en particulier laconsommation de farines contaminées par les sclérotes de ces champignons. De la même manière,les fusariotoxines (toxine T-2 et zéaralenone) sont considérées comme responsables du déclin de lacivilisation Étrusque et de la crise athénienne cinq siècles avant J.-C. Certains tombeaux égyptiensont également été suspectés de renfermer des moisissures sécrétant une mycotoxine, l’ochratoxineA, qui aurait été responsable de la « malédiction des Pharaons ». La littérature vétérinaire rapportede nombreux cas de mycotoxicoses, notamment le syndrome de Turkey X ou « maladie de la dinde »,qui a décimé en Angleterre des milliers de dindes, de canetons et autres animaux domestiques dansles années 1960. Elle a permis de découvrir les aflatoxines, mycotoxines produites par Aspergillusflavus en quantités importantes dans la farine d’arachide importée d’Amérique latine dont senourrissaient les volailles.Mais de nos jours en Europe, il demeure exceptionnel d’être exposé à des doses toxiques en uneseule ingestion d’aliments contaminés, provoquant ainsi une « mycotoxicose » aiguë. Les effetschroniques (exposition répétée à de faibles voire très faibles doses) sont les plus redoutés en raisondes habitudes alimentaires et du pouvoir de rémanence de ces toxines.Les mycotoxines peuvent être classées en polycétoacides, terpènes, cyclopeptides et métabolitesazotés selon leur origine biologique et leur structure (Figure 3). On peut aussi classer les mycotoxinesplus simplement selon leurs principaux effets toxiques. On distingue parmi les groupes demycotoxines considérées comme importantes du point de vue agro-alimentaire et sanitaire lesaflatoxines, les ochratoxines et l’ochratoxine A en particulier, la patuline, les fumonisines, lazéaralènone et les trichothécènes et tout spécialement le déoxynivalénol (Figure 3).Un aspect important mais pourtant peu documenté concerne la présence simultanée de plusieursmycotoxines dans les denrées alimentaires. En effet, une même moisissure a la capacité de produiredifférentes mycotoxines. A l’inverse, une même mycotoxine pourra être produite par plusieursespèces et genres de moisissures. Ainsi, plusieurs toxines d’une même famille structurale ouprésentant des structures différentes peuvent se retrouver dans le même produit alimentaire et, afortiori, dans une ration composée de divers ingrédients alimentaires : on parle alors de multicontamination.Cette situation naturelle pose des interrogations sur les interactions toxiques quipeuvent s’opérer et ainsi pourrait se traduire par un effet antagoniste ou additif ou encoresynergique. De plus, les recommandations et régulations actuelles ne fixent des seuils que pour unemycotoxine et ne tiennent pas compte des contaminations avec plusieurs mycotoxines.10

INTRODUCTIONUne partie du travail de thèse a donc consisté à étudier l’interaction de deux mycotoxines, ledeoxynivalenol et la fumonisine, à des faibles doses. En effet, la co-contamination par ces deuxtoxines majeures produites par des champignons du genre Fusarium, a été rapportée lorsd’échantillonnage de denrées agricoles, et sont d’intérêt majeur en termes d’ubiquité et de toxicité.De plus, ces résultats s’inscrivent dans une étude bibliographique que nous avons réalisée etprésentée dans ce mémoire, sur les données toxicologiques actuelles, rapportant les résultatsd’expérience in vivo suite à l’exposition à des toxines seules ou en combinaison.L’autre partie du travail de thèse a consisté à évaluer l’efficacité d’agents détoxifiants demycotoxines, dans le cadre de notre collaboration avec l’industriel BIOMIN.En effet, le risque mycotoxique étant d’origine naturelle, l’homme n’en maîtrise pas la survenuequi est notamment liée aux conditions climatiques, et ainsi la contamination fongique estdifficilement contrôlable. Par conséquent, le contrôle du niveau de contamination des aliments exigel’emploi de stratégies variées et complémentaires. Des méthodes préventives telles que despratiques culturales adaptées existent pour diminuer le risque de prolifération des moisissures.Cependant, l’élimination totale du risque fongique et mycotoxique est impossible.Les stratégies impliquant le tri et/ou la destruction des denrées contaminées sont peu réalistes dufait de leur coût économique important. Par ailleurs, la dilution des aliments contaminés afind’abaisser le niveau de contamination en-dessous des normes réglementaires, a été interdite enEurope à partir de juillet 2003. Ces méthodes d’élimination ou de réduction des teneurs enmycotoxines, directement sur les matières brutes, sont détaillées dans le présent manuscrit, suite àla publication d’un chapitre sur les méthodes physiques et chimiques existantes.Une autre façon d’augmenter la qualité sanitaire des aliments est l’utilisation de ligands minérauxet organiques ou de micro-organismes/enzymes détoxifiant les mycotoxines afin de limiterl’absorption des mycotoxines dans l’organisme de l’animal. Cet aspect est également détaillé dans lemémoire de thèse, suite à un chapitre complémentaire sur les alternatives biologiques. Nousprésenterons également les résultats d’une des méthodes développées par la société BIOMIN,concernant l’action d’une enzyme sur la fumonisine B1, mycotoxine prédominante de la famille desfumonisines et une des toxines les plus difficiles à éliminer.Grâce à nos installations au sein du pôle ToxAlim, les phases animales expérimentales ont étéconduites sur le porc. D’un point de vue agronomique, les animaux monogastriques d’élevage telsque le porc et la volaille sont particulièrement exposés aux mycotoxines du fait de l’importance de lapart des céréales dans leur alimentation. De plus, ces animaux et en particulier le porc sont trèssensibles aux mycotoxines, du fait de l’absence de réservoir ruminal, connu pour contenir des microorganismescapables de dégrader les toxines avant leur absorption intestinale. Finalement,11

INTRODUCTIONl’utilisation de ce modèle animal permet aussi d’extrapoler nos données à l’homme, considérant lasimilarité de leurs systèmes immunitaire et digestif.12

INTRODUCTIONETUDE BIBLIOGRAPHIQUEL’étude bibliographique réalisée se présente sous la forme de deux grandes parties, précédéesd’une brève introduction sur les mycotoxines majeures, en termes d’occurrence et de toxicité. Lapremière grande partie est une revue exhaustive sur les expériences menées in vivo et quicaractérise les interactions des mycotoxines. La seconde partie présente les méthodes développéespour neutraliser, éliminer et décontaminer les denrées alimentaires contaminées en mycotoxines.Comme indiqué précédemment, la présence de plusieurs mycotoxines en même temps n’est pasun cas isolé mais plutôt une situation fréquente. Les relevés de terrain et les études toxicologiquessur les multi-contaminations sont encore peu nombreux contrairement aux données individuelles, etpar conséquent la connaissance du risque pour la santé humaine et animale est limitée. Nousprésentons donc une synthèse des résultats d’expériences in vivo, où des animaux ont été exposés àdeux toxines, seules ou en combinaison. A partir des données brutes, nous avons classé paramètrepar paramètre l’effet des mycotoxines en association, tel que synergique, additif, moins qu’additif ouantagoniste. De par son aspect exhaustif, cette revue résume l’état actuel des connaissances surl’interaction des mycotoxines chez l’animal.La contamination en mycotoxines étant peu maîtrisable, de nombreuses approches pour réduireou éliminer les mycotoxines ont été développées, et certaines intégrées dans les chaînes deproduction et/ou appliquées en aval dans les élevages. Ces stratégies sont présentées sous forme dedeux chapitres en cours de publication dans un ouvrage scientifique. Le premier rapporte lesméthodes physiques et chimiques qui agissent directement sur les denrées alimentairescontaminées. Nous présentons l’efficacité de ces techniques expérimentales ou appliquées, et aussila toxicité potentielle des denrées modifiées physiquement et chimiquement. Le second chapitretraite des méthodes biologiques qui agissent directement dans le tractus gastro-intestinal desanimaux, suite à l’ingestion d’aliments contaminés. Nous pouvons distinguer les approches baséessur la capacité d’adsorption des mycotoxines par des ligands minéraux ou organiques, et sur lacapacité de biotransformation des toxines par voie enzymatique.13

INTRODUCTION1. Les mycotoxines : généralités, métabolisation et effets toxiquesLes mycotoxines sont des molécules de faible masse moléculaire issues du métabolismesecondaire des moisissures. Plus de 300 métabolites secondaires ont été identifiés mais seule unetrentaine possède de réelles propriétés toxiques préoccupantes.Les effets toxiques sont de nature variée. Certaines toxines exercent un pouvoir hépatotoxique(aflatoxines), d’autres se révèlent oestrogéniques (zéaralénone), immuno/hématotoxiques (patuline,trichothécènes, fumonisines), dermonécrosantes (trichothécènes), néphrotoxiques (ochratoxine A)ou neurotoxiques (toxines trémorgènes). Certaines mycotoxines sont reconnues ou suspectéesd’être cancérogènes.Pour les consommateurs humains, un autre type de risque est indirect car induit par la présencepossible de résidus dans les productions issues des animaux de rente exposés à une alimentationcontaminée par les mycotoxines. Ces résidus correspondent à la toxine elle-même et/ou à desmétabolites bioformés conservant les propriétés toxiques du composé parental. Les espècesd’élevage peuvent donc constituer un vecteur de ces toxines ou de leurs métabolites dans desproductions telles que les abats, le lait ou le sang. C’est le cas notamment de l’aflatoxine B1, dont lemétabolite l’aflatoxine M1 est retrouvé dans le lait des mammifères lorsque ceux-ci ont ingéré desaliments contaminés par l’aflatoxine B1.Les mycotoxines sont généralement thermostables et ne sont pas détruites par les procédéshabituels de cuisson et de stérilisation. Leur capacité à se lier aux protéines plasmatiques et leurlipophilie en font des toxiques capables de persister dans l’organisme en cas d’expositions répétéeset rapprochées.Les animaux monogastriques d’élevage, porcs et volailles sont particulièrement exposés auxmycotoxicoses du fait de l’importance de la part de céréales dans leur alimentation et de l’absencede réservoir ruminal contenant des micro-organismes capables de dégrader les toxines avant leurabsorption intestinale.En France, en dehors de cas sporadiques correspondant à des accidents aigus observables dansdifférentes espèces animales, l’essentiel des problèmes est lié à une contamination chronique par lesfusariotoxines (trichothécènes, zéaralénone, fumonisines) des aliments produits en France ouimportés. Les problèmes ponctuels dus à l’importation de matières premières contaminées justifientdes procédures de surveillance et de contrôle.14

INTRODUCTIONa) Les aflatoxines (voir la revue de Meissonnier et al., 2005)Chez les animaux monogastriques, l’absorption pourrait représenter près de 90% de la doseadministrée. Après absorption, les aflatoxines (AF) sont véhiculées dans l’organisme après fixationsur les protéines plasmatiques, c’est le cas de l’Aflatoxine B1 (AFB1) liée à l’albumine. La présence decet adduit dans le sérum peut servir de bio-indicateur d’exposition.La plus toxique des quatre aflatoxines naturelles est l’AFB1, viennent ensuite par ordredécroissant de toxicité l’AFG1 puis les AFG2 et AFB2. Les effets des aflatoxines sont principalementliés à l’action toxique des époxydes formés par l’action du système enzymatique réactionnel desmono-oxygénases à cytochromes P450. Ainsi l’AFB1 8,9-époxyde conduit à la formation d’adduits àl’ADN en créant des liaisons avec les bases guanines, et pouvant entraîner des effets carcinogènes.En raison de ses capacités de bioactivation, le foie est la cible principale des aflatoxines.L’exposition à l’AFB1 peut aboutir à un cancer hépatique surtout lorsqu’elle est associée à uneinfection par le virus de l’hépatite B ou C. Chez l’animal, AFB1 exerce des propriétésimmunosuppressives affectant en particulier l’immunité à médiation cellulaire par inhibition de laphagocytose, diminution de la production de radicaux oxygénés et altération de la production decytokines. La réactivation d’infections parasitaires et la diminution de l’efficacité vaccinale ont étémises en évidence expérimentalement sur plusieurs modèles animaux après administration d’AFB1.b) L’ochratoxine A (voir revue de Pfohl-Leszkowicz & Manderville, 2007)L’ochratoxine A (OTA) est connue pour sa nephrotoxicité. Elle serait l’un des facteurs potentiels àl’origine de troubles rénaux chez l’homme, connus sous le nom de Néphropathie Endémique desBalkans. Elle s’avère également immunotoxique, tératogène et neurotoxique.L’OTA est d’abord absorbée dans l’estomac en raison de ses propriétés acides. Néanmoins, le sitemajeur d’absorption de l’OTA est l’intestin grêle avec une absorption maximale au niveau du jéjunumproximal. Elle est hydrolysée en OTα non toxique par la carboxylpeptidase A et la chymotrypsine ainsique par les flores microbiennes (rumen des polygastriques et gros intestin chez toutes les espèces).Au niveau hépatique, l’OTA est transformée en des métabolites mineurs qui permettent unedétoxification partielle.Au niveau immunitaire, l’un des effets les plus notables de l’OTA est la diminution de la taille desorganes lymphoïdes, probablement par un mécanisme nécrotique ou apoptotique.15

INTRODUCTIONc) La zéaralénone (voir rapport de l’AFSSA, 2009)Les propriétés physiques et chimiques de la zéaralénone (ZEA) et notamment son hydrophobicitésont des caractéristiques favorables à une large diffusion à l’intérieur des tissus. Les différentesétudes de pharmaco-cinétique montrent d’ailleurs que la ZEA est absorbée rapidement aprèsadministration orale et peut être métabolisée par le tissu intestinal. La ZEA et ses dérivés ont lacapacité de se fixer de façon compétitive sur les récepteurs oestrogéniques cellulaires. La ZEA estdonc un perturbateur endocrinien. Sa fixation est due à sa capacité à adopter une conformationsimilaire aux oestrogènes naturels tels que les 17β-estradiol.Chez l’homme, la ZEA est suspectée d’une vague de changements pubertaires chez des milliers dejeunes enfants à Porto-Rico. Des toxicoses aux Etats-Unis, en Chine, au Japon et en Australie ont étéliées à la présence de ZEA dans les denrées alimentaires.Les différentes propriétés d’absorption, de distribution, de biotransformations et d’excrétion de laZEA chez différents animaux ont été résumées par Gaumy et al. (2001). Plusieurs métabolites telsque les α et β zéaralénols et α et β zéaralanols peuvent être formés. Ils peuvent ensuite subir uneglucurono-conjugaison. L’α-zéaralénol est un métabolite 3 à 4 fois plus actif que la molécule initiale.Le porc est particulièrement sensible à la ZEA. Le taux de survie de l’embryon est notammentdiminué pour les femelles en gestation ayant ingéré de la ZEA.Plusieurs altérations des paramètres immunologiques ont été montrées in vitro après expositionde lymphocytes à la ZEA : inhibition de la prolifération lymphocytaire après stimulation par unmitogène, augmentation de la production d’IL-2 et d’IL-5. Par contre, aucune étude réalisée in vivone montre d’immunotoxicité de la ZEA.d) Les trichothécènes (voir rapport de l’AFSSA, 2009)‣ La toxine T-2Chez l’animal, la toxine T-2 est rapidement absorbée après ingestion et distribuée dansl’organisme sans accumulation dans un organe spécifique. La toxine T-2 est rapidement métaboliséepar déacétylation, hydroxylation, glucuronoconjugaison et dé-époxydation : la principale voie debiotransformation est une déacétylation qui aboutit à la formation de toxine HT-2.La toxine T-2 est probablement à l’origine de l’Aleucie Toxique Alimentaire, maladie qui a touchédes milliers de personnes en Sibérie pendant la seconde guerre mondiale. Elle provoque chezl’animal une perte de poids, des vomissements, des dermatoses sévères et des hémorragies pouvantentraîner la mort. La toxine T-2 possède des propriétés immunosuppressives intervenant à la fois sur16

INTRODUCTIONle nombre de macrophages, de lymphocytes et d’érythrocytes. Elle inhibe la synthèse protéique et lasynthèse d’ADN et d’ARN.‣ Le déoxynivalénolLa biodisponibilité du deoxynivalénol (DON) est extrémement variable selon les espèces animales,allant de 10% chez les ovins à plus de 50% chez le porc. Le DON est un inhibiteur de la synthèseprotéique, inhibant l’élongation de la chaîne protéique.Le DON, communément appelé vomitoxine, provoque des vomissements chez le porc induisantune réduction de la consommation alimentaire, une diminution du gain de poids et des perturbationsde certains paramètres sanguins. L’impact du DON sur le système immunitaire a été relativementétudié, montrant une augmentation de la sensibilité face à certains pathogènes ou encore unediminution de l’efficacité vaccinale. Par ailleurs, l’ingestion prolongée de DON provoque uneaugmentation de la concentration des IgA, pouvant induire une néphropatie à IgA.e) Les fumonisines (voir rapport de l’EFSA, 2005)Chez l’animal, après administration par voie orale, la fumonisine B1 (FB1), la plus toxique desfumonisines, est faiblement absorbée et se retrouve majoritairement dans les fecès. Labiodisponibilité est évaluée de 1 à 6% selon l’espèce. La majeure partie de la toxine absorbée seretrouve dans le foie et les reins.Les fumonisines agissent en particulier sur la synthèse des sphingolipides, en inhibant de façoncompétitive l’activité de la céramide synthase. Cette perturbation de la synthèse des sphingolipidesentraîne une accumulation de bases sphingoïdes (la sphinganine et la sphingosine) et une déplétionen céramide et en sphingolipides complexes. Ceci a de multiples conséquences sur la physiologiecellulaire, puisque les bases sphingoïdes et les céramides sont des second messagers impliqués dansde nombreuses fonctions telles que l’apoptose, la croissance et la différenciation cellulaire,l’inflammation ou encore la sécrétion protéique. Par ailleurs, les sphingolipides sont des constituantsstructuraux essentiels des membranes cellulaires et la FB1 peut provoquer une atteinte de l’intégritémembranaire.Pour toutes les espèces animales étudiées, le foie est la principale cible de la FB1. Les reins sontégalement affectés chez de nombreuses espèces, notamment les rongeurs. Mais la particularité decette toxine est la capacité d’induire lors d’intoxications aigües des troubles spécifiques de l’espècecible. Ainsi, à forte dose la FB1 est capable d’induire des oedèmes pulmonaires chez les porcins, des17

INTRODUCTIONleucoencéphalomalacies chez les équidés, des lésions rénales et hépatiques chez les rongeurs, ouencore d’induire des cancers de l’œsophage et des anomalies du tube neural chez l’homme.18

INTRODUCTION2. La co-contamination en mycotoxines : méta-analyse des données publiéesDue à leur grande diversité structurale, la toxicité induite par les mycotoxines est très différenteet varie en fonction de la cible cellulaire et tissulaire, mais aussi de l’espèce cible. Ainsi, il est trèscompliqué de prédire l’effet de leurs interactions, en se basant simplement sur les effets individuels.Néanmoins, lorsque certaines mycotoxines ou famille de mycotoxines ont des modes d’actionsimilaires ou partagent des propriétés toxiques communes, telles que des effetsimmunosuppresseurs, mutagènes ou tératogènes, un effet au moins additif est attendu lors de leurassociation chez l’homme ou l’animal. Cette revue reporte l’analyse de plus de 100 expériences invivo publiées dans la littérature, avec un plan factoriel 2 x 2, i.e. avec les effets individuels etcombinés des mycotoxines. Pour chaque paramètre analysé, nous avons reporté l’effet del’interaction tel que synergique, additif, moins qu’additif ou antagoniste. A noter que les interactionsimpliquant les aflatoxines représentent plus de la moitié des études reportées.Cette revue, écrite sous la direction d’Isabelle Oswald, a été soumise en janvier 2011 dans leWorld Mycotoxin Journal.19

INTRODUCTIONMycotoxins co-contamination : meta-analysis of published dataB. GRENIER 1,2 & I. P. OSWALD 11INRA, Unité ToxAlim, Toulouse, France.2BIOMIN Research Center, Technopark 1, Tulln, Austria.Address correspondence toDr Isabelle P. OswaldINRA-Unité ToxAlim180 chemin de Tournefeuille BP 9317331027 Toulouse Cedex 3Phone : +33561285480E-Mail : isabelle.oswald@toulouse.inra.fr20

INTRODUCTIONABSTRACTMost fungi are able to produce several mycotoxins simultaneously; moreover food and feed canbe contaminated by several fungi species at the same time. Thus, humans and animals are generallynot exposed to one mycotoxin but to several toxins at the same time. Most of the studies concerningthe toxicological effect of mycotoxins have been carried out taking into account only one mycotoxin.In the present review, we analyzed 112 reports where laboratory or farm animals were exposedto a combination of mycotoxins, and we determined for each parameter measured the type ofinteraction that was observed. Most of the published papers concern interactions with aflatoxins andother mycotoxins, especially fumonisins, ochratoxin A and trichothecenes. A few papers alsoinvestigated the interaction between ochratoxin A and citrinin, or between different toxins fromFusarium species. Only, experiments with a 2 x 2 factorial design with individual and combinedeffects of the mycotoxins were selected. Based on the raw published data, we classified theinteractions in four different categories: synergistic, additive, less than additive or antagonisticeffects.This review highlights the complexity of mycotoxins interactions which varies according to theanimal species, the dose of toxins, the length of exposure but also the parameters measured.Keywords : co-contamination, toxicological synergism, animal health, feeding trials, chronic toxicity.21

INTRODUCTIONINTRODUCTIONFood safety is a major issue throughout the world. In this respect, much attention needs to bepaid to the possible contamination of food and feed by fungi and the risk of mycotoxin production.Mycotoxins are secondary metabolites produced by fungi, mainly by species from the genusAspergillus, Fusarium and Penicillium. Mycotoxins are very common contaminants of cereals.Furthermore, most mycotoxins are resistant to milling, processing and heating and, therefore, readilyenter the food and feed chains (Bullerman and Bianchini, 2007). The toxicological syndromes causedby ingestion of such toxins range from acute mortality, to slow growth and reduced reproductiveefficiency. Consumption of fungal toxins may also result in impaired immunity and decreasedresistance to infectious diseases (Oswald and Comera, 1998).Most fungi are able to produce several mycotoxins simultaneously in separate feedstuffs, andconsidering, it is a common practice to use multiple grain sources in animals diets, the risk to beexposed to several mycotoxins at the same time increases. This is supported by global surveys thatindicate humans and animals are generally exposed to more than one mycotoxin (Bermudez et al.,1997; Boeira et al., 2000; Monbaliu et al., 2010; Rodrigues and Griessler, 2010; Speijers and Speijers,2004; Wangikar et al., 2004b). The toxicity of combinations of mycotoxins cannot always bepredicted based upon their individual toxicities. Interactions between concomitantly occurringmycotoxins can be antagonistic, additive, or synergistic. The data on the in vivo combined toxiceffects of mycotoxins are limited and therefore, the health risk from exposure to a combination ofmycotoxins is incomplete.The aim of this review is to summarize the published experiments, where laboratory and farmanimals were exposed to a combination of mycotoxins and to describe, parameter by parameter, thetype of interaction observed.22

Figure 4 : characterization of the interaction between mycotoxinsType 1Potentialization Type 2Type 3Synergisticinteraction1086108610861086444422220Cont Tox A Tox B Tox A+B000Cont Tox A Tox B Tox A+B Cont Tox A Tox B Tox A+B Cont Tox A Tox B Tox A+BAdditiveinteraction1086Less than additiveinteraction108644220Cont Tox A Tox B Tox A+B0Cont Tox A Tox B Tox A+BAntagonisticinteraction10Type 1861086Type 21086Type 24442220Cont Tox A Tox B Tox A+B0Cont Tox A Tox B Tox A+B0Cont Tox A Tox B Tox A+B

INTRODUCTIONCHARACTERIZATION OF THE DIFFERENT INTERACTIONS BETWEEN MYCOTOXINSIn the present review, in order to be consistent among the different papers, we went back to theoriginal published raw data, and for each parameter that was measured, we classified the interactionin four different categories: synergistic, additive, less than additive and antagonistic effect.- The “synergy” category was complex and we could differentiate 3 types of synergistic interaction(Figure 4) : Synergism type 1: contains experiments where the effect of the mycotoxins combination wasgreater than expected from the sum of the individual effects of the two toxins. We alsoincluded in this category, experiments where one toxin did not display any effect but wherethe effect of the co-contaminated treatment was greater than the effect of the other toxinalone (potentialization, P). Synergism type 2: the two mycotoxins induce opposite effects and the combined treatmentinduces an effect greater than the individual effect. So even though the toxins have oppositeindividual effects, the interaction led to an exacerbated effect. Synergism type 3: the two mycotoxins induce similar effects and the combined treatmentinduces an opposite effect than the individual effects.- The “additive” category includes experiments where the effect of the combination could becalculated as the sum of the individual effects of the two toxins (Figure 4).- The “less than additive” category includes experiments where the effect of the combined treatmentmainly reflected the effect of only one of the toxin without additional effect of the other toxin(Figure 4).- We also differentiated 2 types of antagonistic interaction (Figure 4) : Antagonism type 1: the two mycotoxins induce similar effects and the combined treatmentinduces a lower effect. So, the combination lowered the effect of the more potent toxin. Antagonism type 2: the two mycotoxins don’t induce the same effects and the combinedtreatment induces an effect intermediate between the two individual treatments. So, thecombination lowered the effect of one of the toxin.Results of the interaction were only reported, where significant differences were observedbetween the control group (no mycotoxin) and at least one of the mycotoxin-contaminated group(mycotoxin A, mycotoxin B or mycotoxin A+B), on parameters measured at the end of theexperiment. Regarding histopathological analysis, few studies scored the incidence and severity oflesions; so we could not have analyzed the results based on the raw data, and include them in thedifferent tables. Nevertheless, we included the conclusions as indicated by the authors in the text.23

Table 1 : Interaction between Aflatoxins (AF) and Fumonisins (FB)MycotoxinsSpecies(exposure)Doses0.1 – 3AF-FBFish0.1 – 23(322 d) 1AF-FBMouse(90 d)AF-FBChicken(21 d)AF-FBPig(28 d)0.1 – 1040.01 – 100.05 – 100.35 – 102.45 – 100.05 – 100.05 – 30AF-FBRat0.017 bw/d(56 d) 2 – 250AF-FBPig(35 d)AF-FBTurkey(21 d)AF-FBChicken(27 d)2.5 – 1000.75 – 2002.5 – 200AF-FB 0.05 – 10SYNERGISTIC INTERACTIONANTAGONISTIC INTERACTIONLESS THANADDITIVEADDITIVEType 1 Type 2 Type 3 INTERACTIONType 1 Type 2INTERACTION- tumors in AF-initiatedliver ↗ (P)- tumors in AF-initiatedliver ↗ (P)- AST ↗ (P)- intensification of- congestion, hemolysislesions in liver ↗in kidney ↗- enlarged thymus ↗- heterophils ↘ - lymphocytes ↗- tumors in AF-initiatedliver ↗- feed intake ↗ - cholesterol ↘- triglyceride →- ALP ↘- lymphocytes ↗- heterophils ↘- albumin ↘ - lymphocytes ↗- heterophils ↘- BWG ↘ (P)- feed intake ↘ (P)- GSTP + lesions/foci ↗- BWG, feed intake ↘- AST, CHL, ALP ↗- mortality ↗- glucose, inorganicphosphorus ↘- BWG ↘ (P)- RW-L, K ↗ (P)- feed intake ↘- egg weight ↘- RW-L ↘ - triglycerides ↗- RBC ↗- lymphocytesstimulation ↘- RW-L ↘ - BWG ↘- RW-Pc ↗- AST ↗- RW-L,H, Lg ↗- GGT ↗- BUN ↘- RW-G ↗- TP, cholesterol,triglycerides, calcium ↘- LDH ↗- lymphocytesstimulation ↘- TP ↘- RW-S ↗- egg production ↘- mortality ↗- BUN ↗- CK ↗- calcium ↘- congestion,hemolysis in spleen →- total iron ↗- RW-K ↗- albumin ↘- uric acid →- ALP ↘- hemoglobin ↘- albumin ↘- RW-G ↗REFCarlson etal., 2001Casado etal., 2001Del Bianchiet al., 2005Dilkin et al.,2003Gelderblomet al., 2002Harvey etal., 1995bKubena etal., 1995bMiazzo etal., 2005Ogido et al.,

Quail(140 d)AF-FBRabbit(21 d)AF-FBRat(21 d)AF-FBChicken(33 d)AF-FBChicken(33 d)AF-FBRat(90 d)AF-FBRat(90 d)AF-FBTurkey(21 d)0.2 – 100.03 bw/d –1.5 bw/d0.72 bw/d –5 bw/d- BW ↘ 2004- feed intake ↘- egg production ↘- egg weight ↘- BW ↘- BWG ↘ (P)- RW-K ↗ - urea ↗ - RW-L ↘- TP → - albumin ↘- mortality ↗- ALP ↗- Sa/So liver ↗Orsi et al.,- AST, GGT ↗- Sa/So urine ↗2007- creatinine ↗- Sa/So serum ↗ (P)- RW-L ↗- AST, ALT ↗1) Carlson et al. (2001) : 30 min initiation by immersion with AF, followed by 4 weeks of recovery, and then, 42 weeks FB feeding2) Gelderlom et al. (2002) : 14 days initiation with AF, followed by 3 weeks of recovery, and then 3 weeks FB feeding- BWG ↗- urea →- RW-K ↘- ALP, GGT ↗- TP, creatinine, albumin↗- RW-K ↘ - creatinine ↗ - urea ↗0.72 bw/d –15 bw/d- GGT ↗ - BWG ↘- RW-L ↘- AST, ALP, ALT ↗- TP, albumin ↗0.05 – 50 - AST ↗0.05 – 200 - AST ↗0.2 – 50 - AST ↗0.2 – 200 - TP ↗ - AST ↗0.05 – 500.05 – 2000.2 – 500.2 – 2000.04 – 1000.04 – 1000.2 – 75- BWG ↘- RW-H,BF ↗- Ab ND ↘ (P)- Ab ND ↘ - BWG ↘- RW-L ↗- RW-H ↗- RW-BF →- RW-BF ↗- BWG ↘ - RW-L ↗ - RW-H ↗- Ab ND ↘- Ab ND ↘ (P) - BWG ↘- RW-BF ↗ - RW-L ↗- RW-H ↗- SOD activity ↗ - DNA damage in SMC - DNA damage in SMC(ACA) ↗(MN) ↗- MDA, CAT activity ↗- BW ↘ (P)- Incidence, severity ofliver & kidney lesions ↗- Sa/So kidney, urine,serum, liver ↗ (P)- feed intake ↘- Incidence, severity oflung & intestine lesions↗- RW-Pc ↗ - Ab SRBC ↗ - BWG ↘- RW-S ↗- albumin, TP,cholesterol ↘- AST ↗- hemoglobin ↗- RW-BF →- RW-L ↗- Sa/So serum ↗Pozzi et al.,2001Tessari etal., 2010Tessari etal., 2006Theumer etal., 2010Theumer etal., 2008Weibking etal., 1994

INTRODUCTIONINTERACTION BETWEEN THE DIFFERENT MYCOTOXINSMost of the published papers, which are dealing with the effect of mycotoxin multi-contaminationon animals, concern aflatoxins (62/112 reports). The main mycotoxins investigated in associationwith aflatoxins are fumonisins (16 reports), ochratoxin A (19 reports), and trichothecenes (16reports), especially T-2 toxin (11 reports). The other important interactions involved ochratoxin A andcitrinin (11 reports), and, fumonisins and moniliformin (7 reports). Despite the high occurrence ofmycotoxins from Fusarium spp., only few studies have investigated the interaction betweenFusariotoxins (23 reports).I. INTERACTIONS BETWEEN AFLATOXINS AND OTHER MYCOTOXINS1) Interaction between Aflatoxins (AF) and Fumonisins (FB)a) effects of AF and FB on zootechnical parametersAs displayed in the Table 1, this association of mycotoxins resulted most of the time in asynergistic effect on body weight gain (Dilkin et al., 2003; Harvey et al., 1995b; Miazzo et al., 2005;Ogido et al., 2004; Orsi et al., 2007; Pozzi et al., 2001; Tessari et al., 2006; Theumer et al., 2008).Surprisingly, Pozzi et al. (2001) reported an increase and an antagonistic interaction on body weightgain when rats were fed both AF and FB.Reduced egg production and egg weight is one of the effects of aflatoxicosis in laying hens. Inquails, one study reported an antagonistic effect of AF and FB on egg production (Table 1).Surprisingly, egg production was reduced in animals fed FB-contaminated diet, and this effect waspartially spared when animals received AF+FB contaminated diet (Ogido et al., 2004).b) effects AF and FB on biochemical parameters and organs weightSeveral experiments have measured enzyme concentrations in the serum of animals fed muticontaminateddiets, in order to investigate tissue damage, especially the liver. The levels ofmycotoxins used in these reports were moderate to high, and therefore induced organ injuries,physiologically translated by increase of these biomarkers. The lack of response in some experimentsis due to the low doses of mycotoxins used (Del Bianchi et al., 2005). Overall, association of AF andFB induced a significant increase of the serum levels of AST, ALP, ALT and GGT (synergistic or additiveeffect, Table 1). Such increase in enzymatic activity can be attributed to cell necrosis, changes in cellmembrane permeability or impairment of biliary excretion. This strong hepato-biliary dysfunction24

INTRODUCTIONdue to the ingestion of co-contaminated feed is in accordance with the histopathologicalobservations (see below).Another parameter correlated with hepatotoxicity, is the relative weight of liver. Interestingly,either an increase or a decrease of the RW-L was observed in the experiments for the combinedtreatment (Table 1). This is attributed to the individual effect of each toxin, showing either anincrease or a decrease for this parameter, according to studies. A synergistic interaction type 2 or 3was recorded in pigs, turkeys, and rats exposed to both toxins (Harvey et al., 1995b; Kubena et al.,1995b; Pozzi et al., 2001). It would be possible that the mixture induced enough hepatic cell walldamage to have a major effect on total hepatic parenchymal mass. By contrast, other studies havereported an increase in the RW-L, but leading to a less than additive (Dilkin et al., 2003; Tessari et al.,2006) or an antagonistic interaction (Tessari et al., 2006; Weibking et al., 1994) in animals receivingmulti-contaminated feed.In two studies, ALP concentration was decreased when animals were fed with AF (Casado et al.,2001; Kubena et al., 1995b). The authors suggested that it was due to the inhibition of proteinsynthesis by AF. The combination of AF with FB led to a lesser reduction, suggesting an antagonisticinteraction (Table 1).c) effects of AF and FB on lipidsDisruption of sphingolipid biosynthesis is the main mechanism involved in FB toxicity, withinhibition of ceramide synthase leading to accumulation of sphingoid bases (sphinganine, Sa andsphingosine, So). This inhibition is well documented and the Sa/So ratio is considered as a goodindicator of FB exposure (Theumer et al., 2008).Three studies analyzed the effect of co-exposure to AF and FB on sphingoid bases in differentbiological samples, with the aim of investigating if AF could exacerbate the toxicity of FB (Orsi et al.,2007; Theumer et al., 2008; Weibking et al., 1994). These studies gave fairly different results (Table1). In contrast with the two other studies, Orsi et al. (2007) observed an increased Sa/So ratio inurine of animals exposed to AF alone. Both, Orsi et al. (2007) and Weibking et al. (1994)demonstrated an antagonistic interaction between AF and FB on the Sa/So ratio in the liver and inthe serum of exposed animals. Conversely, Theumer et al. (2008) observed a synergistic effect of thetwo toxins when looking at the Sa/So ratio in the kidney, urine, and to a lesser extent in the serumand the liver of exposed rats. Of note, this study also described a synergistic interaction in theincidence and the severity of kidney and liver lesions (Theumer et al. 2008).25

INTRODUCTIONd) effects of AF and FB on microscopic lesionsMicroscopic lesions following ingestion of AF and FB were mostly evaluated in the liver andkidneys, their respective target organs. In these studies, additive or synergistic interactions betweenthe two toxins were observed (Casado et al., 2001; Del Bianchi et al., 2005; Dilkin et al., 2003;Gelderblom et al., 2002; Harvey et al., 1995b; Orsi et al., 2007; Pozzi et al., 2001; Tessari et al., 2010;Tessari et al., 2006; Theumer et al., 2008; Weibking et al., 1994). Histopathological alterations in liverconsisted mainly in vacuolar degeneration of hepatocytes, apoptotic and mitotic figures, dysplasticnodules, megalocytosis and fibrosis. In animals receiving both AF and FB, cirrhotic livers were alsoobserved. Enlargement of gallbladder and duct proliferation were also reported (Orsi et al., 2007). Inkidney, apoptosis of tubular epithelial cells, glomerulonephritis and tubular epithelium with areas ofdegeneration and necrosis, as well as congestion and hemolysis were reported in animals fed AF andFB contaminated diet (Casado et al., 2001; Del Bianchi et al., 2005; Theumer et al., 2008). Theselesions were considered as moderate to severe, depending on the species and the doses used.Lymphocytic infiltrates in the small intestine, and thickening of the alveolar walls and lymphocyticinfliltrates, as well as apoptosis in the lungs, were observed in rats fed with diet containing AF(Theumer et al., 2008). Incidence and severity of these alterations were similar in rats fed the cocontaminateddiet, suggesting a less than additive interaction (Table 1).e) effects of AF and FB on genotoxicity and carcinogenicityBoth AF and FB have carcinogenic effects, and besides, co-contamination of maize with AF and FBwas related to a high-incidence area of human primary hepatocellular carcinoma in China (Li et al.,2001). At the molecular level, the toxicology of AF involves its metabolic conversion by thecytochrome P450 system to the highly electrophilic AF-exo-8,9-epoxyde, which in turn binds to theDNA guanines to form adducts. Therefore, AF has been reported to be a good initiator and a provencomplete carcinogen. On the other hand, FB has been described to be a potent tumor promoter buta weak initiator (Riley, 1998).Two long-term trials, performed in rats and in rainbow trouts, initiated hepatocarcinogenesis withAF, and then treated the animals with high doses of FB (Carlson et al., 2001; Gelderblom et al., 2002)(Table 1). In trouts, FB promoted liver cancer in AF-initiated animals, but did not promote tumors inother tissues (Carlson et al., 2001). Similarly, in rats, despite the fact that AF and FB wereadministered three weeks apart in a sequential model, they acted synergistically with respect tocancer initiation, as demonstrated by the number of hepatocytes nodules and foci (Gelderblom et al.,2002).26

Table 2 : Interaction between Aflatoxins (AF) and Ochratoxin A (OTA)MycotoxinsSpecies(exposure)Doses0.025 –AF-OTA 1.15Chicken egg (µg/egg)(10 d) 1 0.05 – 2.3(µg/egg)AF-OTA 0.025 – 1.0(µg/egg)Chicken egg(15 d) 1 0.05 – 2.0(µg/egg)AF-OTAPig(28 d)AF-OTAChicken(21 d)AF-OTAChicken(42 d)AF-OTAChicken(42 d)AF-OTAChicken(21 d)AF-OTAChicken(42 d)AF-OTACalve(87 d)2.0 – 2.02.5 – 2.02.5 – 2.02.5 – 2.03.5 – 2.00.2 – 0.20.013 –0.450SYNERGISTIC INTERACTIONANTAGONISTIC INTERACTIONLESS THANADDITIVEADDITIVEType 1 Type 2 Type 3 INTERACTIONType 1 Type 2INTERACTION- embryos mortality ↗ - abnormalities ↗- abnormalities ↗ - embryos mortality ↗- abnormalities ↗ - embryos mortality ↗- abnormalities ↗ - embryos mortality ↗- triglycerides ↗ (P) - TP ↗- RW-L ↗- BW ↘- mortality ↗- RW-L,K,G,S,Pv,Pc ↗- BW ↘- prothrombin times ↗- live, dressed,eviscerated weight ↘- carcasse yield, breast,drum, thigh, wing, backweights ↘- plasma carotenoids ↘- BW ↘- ALP ↗- inorganic phosphorus↘- incidence andseverity of breastbruises ↗- BW ↘- RW-L,K,H,Pv ↗- RW-S ↘- DTH reaction ↘- BUN, cholesterol,glucose ↘- calcium →- sodium ↘- incidence and severityof right thigh bruises ↗- albumin →- hemoglobin →- RW-K →- liver lipid levels →- breast yield ↘ - liver lipid levels ↗- hemoglobin ↘- albumin, cholesterol,triglycerides ↘- creatinine, uric acid ↗- ALP ↗- RW-L ↗- RW- BF,T ↘- albumin, TP,cholesterol ↘- Ab ND ↘- WBC ↘- uric acid ↗- AST ↘REFEdringtonet al., 1995Edringtonet al., 1995Harvey etal., 1989aHuff andDoerr, 1981Huff et al.,1983Huff et al.,1984Huff et al.,1992Kalorey etal., 2005Patterson etal., 1981

AF-OTAChicken(35 d)AF-OTAChicken(35 d)AF-OTAGuinea pig(28 d)AF-OTAChicken(42 d)AF-OTAPig(42 d)AF-OTALaying hen(50 d)AF-OTALaying hen(50-75 d)AF-OTAChicken0.3 – 2.00.3 – 2.00.01 mg/d –0.45 mg/d0.2 – 0.20.375 – 1.00.5 – 1.01.0 – 2.02.0 – 4.00.5 – 1.01.0 – 2.02.0 – 4.0- feed consumed forproduction of 1 dozeneggs ↗- metabolizable energy,protein retention ↘- maintenance energyrequirement ↗- maintenance energyrequirement ↗- RW-G,A ↗- GGT ↗- hemoglobin ↘- BW ↘- cholesterol ↘- RW-S,T ↘- Ab ND, DTHreaction ↘- BW, feed intake ↘- ALT ↘- TP, globulin ↘- complement titer ↘- triglyceride ↘- creatinine, uric acid ↗- mortality ↗- RW-L, K ↗- TP, cholesterol, BUN↘-coagulation time →- RW-BF,T ↘- Ab ND, IBD ↘- hemoglobin ↘- RW-L ↗, RW-BF ↘- WBC →Raju andDevegowda, 2000Raju andDevegowda, 2002Richard etal., 1975Sakhare etal., 2007- BWG ↘ - creatinine ↗ - BUN → Tapia andSeawright,1985- egg production ↘ - egg shape index ↘ - feed intake ↘- egg shell thickness →- egg production ↘- feed consumed forproduction of 1 dozeneggs ↗- egg shape index ↘- egg production ↘- feed consumed forproduction of 1 dozeneggs ↗- egg energydeposition ↘- metabolizableenergy, proteinretention, egg energydeposition ↘- metabolizableenergy, proteinretention ↘- feed intake ↘- egg shape index ↘- egg energy deposition↘- maintenance energyrequirement ↗- feed intake ↘ - egg shell thickness↘- egg shell thickness →- BWG ↘- RW-L ↗- Ab SRBC ↘0.5 – 1.0- feed intake ↘- DTH reaction ↘1.0 – 2.0 - feed intake ↘ - BWG ↘ - RW-BF ↘ - RW-K ↗Verma etal., 2003Verma etal., 2007Verma etal., 2004

(49 d) - RW-L ↗- Ab SRBC ↘- RW-K ↗ - BWG ↘2.0 – 4.0- feed intake ↘- Ab SRBC ↘1) Edrington et al. (1995) : one administration and measurements 10 or 15 d post-administration- DTH reaction ↘- RW-BF ↘ - RW-L ↗- DTH reaction ↘

INTRODUCTIONCellular oxidative stress was also proposed as a possible mechanism of cancer initiation. However,Theumer et al. (2010) only observed a less than additive and an antagonistic interaction between thetwo toxins when looking at DNA lesions using the alkaline comet assay (ACA) and the micronucleustechnique (MN), respectively (Table 1). Furthermore, in the same study, a less than additiveinteraction was recorded in the MDA and CAT activity, both are being biomarkers of oxidative stress.f) effects AF and FB on immunityFew studies investigated the combined effect of AF and FB on immunity (Table 1). After ingestionof the co-contaminated diet, the interaction on the reduced lymphocytes proliferation uponmitogenic stimulation, was described either as less than additive (Kubena et al., 1995b) or additive(Harvey et al., 1995b). Tessari et al. (2006) showed a synergistic decrease of the antibodies titersagainst Newcastle Disease. By contrast, Weibking et al. (1994) reported an unexpected increase andan additive effect of the two toxins when looking at the hemagglutination titers against SRBC inturkey poults.2) Interaction between Aflatoxins (AF) and Ochratoxin A (OTA)a) effects of AF and OTA on zootechnical parametersAs indicated in Table 2, the interaction between AF and OTA on animal performance has beenstudied in 11 publications. In most of these studies, the association induced a synergistic or anadditive effect on body weight gain (Harvey et al., 1989a; Huff and Doerr, 1981; Huff et al., 1983;Huff et al., 1992; Sakhare et al., 2007; Verma et al., 2004). In addition, Huff et al. (1983) observedthat even four weeks after exposure to the co-contaminated diet had stopped, the animals did notrecover their normal body weight. The same author also observed that in chicken, ingestion of feedco-contaminated with AF and OTA decreased carcass yield, breast weight and other processingparameters in a synergistic manner (Huff et al., 1984) (Table 2).Both AF and OTA affect egg production and hatchability, and one group investigated thecombined effect of these two toxins on laying hens (Verma et al., 2003; 2007) (Table 2). An additiveinteraction of AF and OTA was observed on egg production and on the feed efficiency (consumptionfor egg production) (Verma et al., 2003). Regarding egg quality, less than additive and antagonisticinteractions were observed on egg. Protein and energy utilization were also modulated by theconsumption of AF and OTA. However the interaction between the two toxins varied according totheir concentrations ranging from synergistic to less than additive effect (Verma et al., 2007).27

INTRODUCTIONb) effects AF and OTA on biochemical parameters and organs weightRegarding biochemical parameters, many studies reported a less than additive or an antagonisticinteraction between AF and OTA (Table 2). These types of response were observed for serumconcentrations of cholesterol, albumin, total proteins, creatinine, uric acid or blood urea nitrogen(Harvey et al., 1989a; Huff et al., 1992; Kalorey et al., 2005; Raju and Devegowda, 2000; Richard etal., 1975; Sakhare et al., 2007; Tapia and Seawright, 1985).Several studies have assessed the effect of the co-contaminated diet on relative weight of liverand on liver functions (Table 2). The type of interaction between the two toxins on the increasedrelative weight of the liver varied a lot from one experiment to another. Huff and Doerr (1981), andHarvey et al. (1989) observed a synergistic interaction, while Huff et al. (1992) and Verma et al.(2004) observed an additive interaction. Kalorey et al. (2005) noted a less than additive interaction,and with higher doses of toxins, Verma et al. (2004) described an antagonistic interaction.Antagonism was also observed in the experiments of Raju and Devegowda (2000) and Sakhare et al.(2007). Interestingly, OTA seems to inhibit or spare the effect of AF on the accumulation of lipid inthe liver, leading to an antagonistic interaction between the two mycotoxins (Huff and Doerr, 1981;Huff et al., 1984) (Table 2). Few data are available on the biomarkers of hepatotoxicity, but anadditive or less than additive effect was observed for the increased concentrations of ALP (Harvey etal., 1989a; Kalorey et al., 2005) and GGT (Raju and Devegowda, 2000), and an antagonistic or lessthan additive effect was reported for the decreased concentrations of AST (Huff et al., 1992) and ALT(Raju and Devegowda, 2000) (Table 2).Concerning the increased relative weight of kidney, as already mentioned for the liver, the type ofinteraction of the two toxins on this organ range from synergism to antagonism (Harvey et al., 1989a;Huff and Doerr, 1981; Huff et al., 1992; Raju and Devegowda, 2000; Verma et al., 2004) (Table 2).Based on the relative weights of gizzard, proventriculus and pancreas, it seems that the upperalimentary tract is more sensitive to the combination than mycotoxins alone (Huff and Doerr, 1981;Huff et al., 1992) (Table 2).Of note, a lack of response for most of the variables studied was observed in calves fed anycontaminated diets (Patterson et al., 1981). This failure was probably attributable to the low doses ofmycotoxins used, and to the detoxifying action of the rumen flora on OTA.28

INTRODUCTIONc) effects of AF and OTA on microscopic lesionsMicroscopic lesions following ingestion of AF and OTA contaminated feed were mostly evaluatedin the liver and kidneys, their respective target organs (Huff and Doerr, 1981). Unfortunately, thedifferent studies are not in agreement one to each other. For example, in chicken, OTA in the dietsprevented hepatic fatty infiltration that was caused by AF (Huff and Doerr, 1981). Pigs fed the cocontaminateddiet, presented the same hepatic lesions than those fed diet contaminated with AFalone (Tapia and Seawright, 1985). By contrast, Sakhare et al. (2007) recorded more severe hepaticlesions in chicken receiving the co-contaminated diet, with granular and vacuolar degenerativechanges, necrosis of liver parenchyma and areas of hemorrhages.The same discrepancy was noticed for the histology of the kidney. In pigs, Tapia and Seawright(1985) observed less severe renal lesions in animal fed the co-contaminated diet that in animalreceiving the OTA-contaminated diet. This is in agreement with the data obtained in pigs, by Harveyet al. (1989a) on the relative weight of kidney, and the data obtained by Tapia and Seawright (1985)on the creatinine and blood urea nitrogen concentrations (Table 2). Conversely, Sakhare et al. (2007)indicated that renal injuries appeared earlier and were more developed in chicken fed multicontaminateddiet that in animal receiving the mono-contaminated diets. This led to destruction oftubular epithelium, with detachment of tubular cells from basement membrane. The species usedmay explain these discrepancies.One study focused on the effect of AF and OTA on bruising in broiler chickens (Huff et al., 1983),as bloody thigh syndrome has been suspected to be associated with mycotoxicoses. This studyrevealed a severe coagulopathy, as measured by elevated prothrombin times, and, an additive andan antagonistic interaction between AF and OTA were observed for the incidence and severity ofbreast and right thigh bruises, respectively (Table 2).d) effects of AF and OTA on immunityAtrophy of lymphoid organs were observed in several studies (Kalorey et al., 2005; Raju andDevegowda, 2002; Sakhare et al., 2007; Verma et al., 2004) (Table 2). This seems to be due tonecrosis and cellular depletion, as suggested by microscopical observations with necrotic areas ingerminal centre, depopulation of lymphocytes in spleen, or depletion and necrosis of lymphoid cellsfrom follicle in bursa of Fabricius (Sakhare et al. 2007). In the bursa of Fabricius, thesehistopathological changes were more pronounced in animals fed the co-contaminated diet comparedto animal receiving the mono-contaminated diets (Sakhare et al., 2007).29

INTRODUCTIONThe depletion of lymphocytes suggested a suppression of cell mediated immunity. Indeed, thecontact sensitivity reaction, was reduced in chicken fed the co-contaminated diet with an additiveinteraction between AF and OTA (Kalorey et al., 2005; Sakhare et al., 2007; Verma et al., 2004).However, Verma et al. (2004) observed that, with increasing doses of toxins, the interaction wentfrom additive to less than additive and to antagonist (Table 2).In order to analyze the effect of the mycotoxins on the humoral response, the antibodies titersagainst Newcastle disease, Sheep red blood cells or Infectious Bursal disease were measured (Kaloreyet al., 2005; Raju and Devegowda, 2002; Sakhare et al., 2007; Verma et al., 2004). Mono- and multicontaminateddiets decreased the humoral response, and interaction between AF and OTA showedeither an additive (Sakhare et al., 2007; Verma et al., 2004) or antagonistic interaction (Kalorey et al.,2005; Raju and Devegowda, 2002; Verma et al., 2004) (Table 2).Also reported the effect on complement activity, but a less than additive interaction was obtainedbetween OTA and AF (Richard et al., 1975) (Table 2).e) effects of AF and OTA on teratologyBoth AF and OTA are able to cross the placental barrier and demonstrated some teratogenicproperties in some species (Wangikar et al., 2004b; Wangikar et al., 2005)In two separate experiments, Edrington et al. (1995) showed that embryonic mortality increasedfollowing injection to combination of AF and OTA, but led to different types of interaction accordingto doses and experiments. He supposed these differences were related to variations in embryosensitivities among batches of eggs. By contrast, exposure of chicken embryos to the combination oftoxins did mostly result in a synergistic interaction in the number of abnormalities.Wangikar in 2004b and 2005 performed two large studies in rats and rabbits, respectively, wheremycotoxins were administered orally on days 6-18 of gestation. Both mycotoxins interfered with thebody wall formation leading to gastroschisis as observed in both the high OTA and high AFcombination groups (Wangikar et al., 2004b). In the co-exposed group, more severe cardiac lesionswere noticed (Wangikar et al., 2004b; Wangikar et al., 2005) but skeletal, and visceral anomalieswere reduced or absent, when compared to the mono-exposed groups. Similarly, the presence of AFseems to prevent the exencephaly and incomplete closure of the skull caused by OTA; and thepresence of OTA prevent the head abnormality and open eye caused by AF. Another study of thesame group also indicates that brain, kidney, and liver lesions were less severe in rats fetuses coexposedto AF and OTA than in mono-exposed animals (Wangikar et al., 2004a). Two mechanismswere proposed to explain these observations. The first hypothesis is that AF may inhibit themetabolism of OTA in the liver and increase its excretion. The second hypothesis is based on the30

Table 3 : Interaction between Aflatoxins (AF) and Trichothecenes (TCT)MycotoxinsSpecies(exposure)AF-T2Chicken(35 d)AF-T2Pig(28 d)AF-T2Chicken(21 d)AF-T2Chicken(21 d)AF-T2Quail(35 d)AF-T2Quail(35 d)Doses2.0 – 1.02.5 – 102.5 – 4.03.5 – 8.03.0 – 4.03.0 – 4.0AF-T2Hamster(21 d) 1 1.0 – 1.0AF-T2Chicken(35 d)0.3 – 3.0SYNERGISTIC INTERACTIONANTAGONISTIC INTERACTIONLESS THANADDITIVEADDITIVEType 1 Type 2 Type 3 INTERACTIONType 1 Type 2INTERACTION- RW-K ↗ (P) - BWG ↘- RW-L ↗- RW-T,BF ↘- BWG ↘- RW-L,K,Pv,H ↗ (P)- RW-G ↗- TP, albumin, uric acid↘- cholesterol ↘ (P)- potassium ↘ (P)- CHL ↘ (P)- BWG ↘- RW-L ↗ (P)- RW-G ↗- hemoglobin ↘- ALP ↘- GGT ↗- calcium ↗- GGT ↘ - LDH ↘- bilirubin ↘- RW-L, K ↗ (P)- ALT ↘- hemoglobin ↘- CK ↗ - hemoglobin ↘- triglycerides ↘- calcium ↘- LDH, ALP ↘- feed intake ↘- RW-G,S ↗- Ab IB D ↘- BWG ↘ - BUN ↘- GGT ↗- CK ↗ - RW-Pv ↗- RW-BF ↘- uric acid,triglycerides, TP ↘- ALT, AST↘- coagulationtime ↘- TP, albumin,globulin ↘- oral lesions ↗- RW-S,Pc ↗- RBC ↘- magnesium ↘- RW-K,Pc ↗- cholesterol, albumin ↘- cholesterol ↘- AST ↗- potassium ↘REF- Ab ND ↘ Girish andDevegowda,2006- sodium, phosphorus,albumin ↘- cholesterol,triglycerides, TP ↘- ALP, CHL ↗, AST→- RBC, WBC ↗,hemoglobin →- RW-L ↗, RW-H,K→- glucose ↘ - sodium →- mortality ↗- LDH ↘- CHL ↗- glucose, BUN ↘- ALT ↘- oral lesions ↗- RW-H,S ↗- GGT ↗- ALP →- sodium →Harvey etal., 1990Huff et al.,1988Kubena etal., 1990Madheswaran et al.,2004- BW, feed intake ↘ - hemoglobin, RBC ↘ Madheswaran et al.,2005- BW ↘- GGT ↗- glucose ↘ - ALP ↗ - cholesterol ↗- feed intake ↘- mortality ↗- RW-A ↗- BUN ↘- RW-G →- TP, cholesterol ↘Rajmon etal., 2001Raju andDevegowda,2000

AF-T2Chicken 0.3 – 3.0(35 d)AF-T2Rat0.25 – 0.05(140 d) 2AF-DASLamb2.5 – 5.0(34 d)AF-DASPig2.5 – 2.0(28 d)AF-DASChicken(21 d)AF-DONPig(28 d)AF-DONChicken(21 d)3.5 – 5.03.0 – 3.02.5 – 16- RW-T ↘ - RW-BF ↘ - Ab IB D ↘ - Ab ND ↘ Raju andDevegowda,2002- TP ↗- GGT ↗ (P)- BWG ↘- mortality ↗- GGT ↗- cholesterol, glucose ↘(P)- WBC ↗ (P)- RW-Pv,G, L ↗- ALP ↘1) Rajmon et al. (2001) : intragastrically administration, twice a week for 3 weeks2) Tamimi et al. (1997) : intraperitoneally administration, twice a week for 20 weeks- BWG, feed intake ↘- BUN ↘- BWG ↘ - ALP, GGT ↗- AST ↘- RW-S ↗- RW-G ↗- LDH ↘- CK ↘- TP ↘- BWG ↘- hemoglobin ↘- AST ↘- RW-L, K →- C HL ↘ - cholesterol ↗- oral lesions ↗- RW-K,H,Pv,S,Pc ↗- glucose, TP ↘- BWG ↘- GGT, AST ↗- BUN ↘- calcium, magnesium ↘- RW-S,K ↗- TP, albumin, uric acid,cholesterol, triglyceride↘- calcium ↘- CHL ↗- creatinine ↗- triglycerides, albumin↘- AST ↘- WBC, hemoglobin ↗- RW-L ↗- RW-L ↗- cholesterol ↘- calcium →- CK →- albumin ↘ - ALP ↗- potassium,phosphorus ↘- RBC, hemoglobin,prothrombin time ↗- glucose ↘- LDH ↘- liver lipid ↗- RBC →- phosphorus →Tamimi etal., 1997Harvey etal., 1995aHarvey etal., 1991Kubena etal., 1993Harvey etal., 1989bHuff et al.,1986

INTRODUCTIONproperty of OTA and AF to inhibit the protein synthesis. Indeed, OTA is known to limit the proteinsynthesis through competitive inhibition of phenylalanine-t-RNA-synthesis with phenylalanine, whileAF prevented protein synthesis through an inhibition of RNA synthesis. Therefore, antagonisticeffects of the mycotoxins might be due to inhibition of transcription by AF, with a concomitantincrease in the cellular pool of phenylalanine available for translation, as inhibition of proteinsynthesis by OTA was inversely proportional to phenylalanine concentration (Wangikar et al., 2004b).3) Interaction between Aflatoxins (AF) and Trichothecenes (TCT)3.1) Interaction between Aflatoxins (AF) and T-2 toxin (T2)a) effects of AF and T-2 toxin on zootechnical parametersSeveral studies have investigated the combined effect of AF and T-2 toxin on body weight gain(Table 3), and all of them demonstrated at least an additive, or a synergistic interaction between thetwo toxins (Girish and Devegowda, 2006; Harvey et al., 1990; Huff et al., 1988; Kubena et al., 1990;Madheswaran et al., 2005; Raju and Devegowda, 2000). A similar interaction was also observedwhen looking at the relative weight of several organs of poultry (Girish and Devegowda, 2006; Huff etal., 1988; Kubena et al., 1990; Raju and Devegowda, 2000; 2002). By contrast, in pigs and in rats,Harvey et al. (1990) and Tamimi et al. (1997) observed an antagonistic interaction on the relativeweights of liver, heart and kidney (Table 3).b) effects of AF and T-2 toxin on biochemical parametersRegarding biochemical parameters, the combined treatment showed a significant decrease forseveral parameters, such as total proteins, albumin, cholesterol, triglycerides, as well as someenzyme concentrations (ALP, ALT, LDH) (Table 3). These reduced concentrations may be attributed tothe property of both AF and T-2 toxin to inhibit the protein synthesis, during transcription andtranslation respectively. Nonetheless, among these studies, different types of interaction wereobserved for these biochemical parameters, ranging from synergistic to antagonistic interactions(Table 3).c) effects of AF and T-2 toxin on microscopic lesionsBecause T-2 toxin is a potent irritant of the buccal cavity, the effect of AF and T-2 toxin wasinvestigated on the oral cavity. When chicken received mono-contaminated feed, oral lesions were31

INTRODUCTIONonly seen in birds given T-2 toxin, and the combination with AF resulted in an antagonistic interaction(Kubena et al., 1990) (Table 3). The authors suggested that the lower effect of T-2 toxin in thepresence of AF was likely due to a decreased intake of T-2 toxin in chicks that received the multicontaminateddiet. In swine, Harvey et al. (1990) reported necrotizing contact dermatitis on thesnout, buccal commissures, and prepuce of animals. However, in the publication, the authorsindicated that these dermal lesions were observed in T-2 and T-2+AF groups, but did not mention thetype of interaction.The interaction of AF and T-2 toxin led to contradictory results concerning the lesions observed inthe liver. Indeed, a synergistic effect was observed in rats with hepatic injuries characterized bycongestions with fatty acid changes, and significant bile duct proliferation (Tamimi et al., 1997). Anadditive effect was observed in quails characterized by fatty vacuoles, mitochondria degeneration,and indistinguishable rough endoplasmic reticulum (Madheswaran et al., 2006). By contrast, lessthan additive and antagonistic effects were reported in chicken and swine fed with multicontaminateddiets respectively, in comparison to the lesions observed in animals exposed to AFcontaminateddiet (Harvey et al., 1990; Kubena et al., 1990). These data correlate with theinteractions observed on biochemical parameters for the two toxins.A synergistic effect of the two toxins was also observed for renal lesions, with tubular epithelialdegeneration, congestion, swelling of the glomureli, as well as hypercellularity (Tamimi et al., 1997).d) effect on immunityOne research group has investigated the interaction between AF and T-2 toxin on the immunesystem (Girish and Devegowda, 2006; Raju and Devegowda, 2002). Conclusions between these twostudies were in agreement and showed different types of interaction of the two mycotoxinsdepending on the parameter : a synergistic or additive effect on the thymus weight, an additiveeffect on the bursa of Fabricius weight, a less than additive effect on the antibody titer against bursaldisease infection, and an antagonistic effect on the antibody titer against Newcastle disease (Table3).3.2) Interaction between Aflatoxins (AF) and Diacetoxyscirpenol (DAS)The interaction between AF and diacetoxyscirpenol (DAS) was reported in three studiesperformed in lamb, pig and chicken (Harvey et al., 1995a; Harvey et al., 1991; Kubena et al., 1993).The levels of mycotoxins used in these studies were quite similar and the association of toxins led toa synergistic or additive interaction on body weight gain (Table 3).32

Table 4 : Interaction between Aflatoxins (AF) and other mycotoxinsMycotoxinsSpecies(exposure)AF-CPAChicken(28 d)Doses1.0 – 200.1 bw/d –0.1 bw/d2.0 bw/d –0.1 bw/dAF-CPARat0.1 bw/d –(3 d) 1 4.0 bw/dAF-CPAGuinea pig(20 d)AF-CPAChicken(21 d)AF-CITChicken(28 d)AF-MONChicken(21 d)AF-STERGuinea pig(14 d)AF-RUBRat2.0 bw/d –4.0 bw/d0.045 – 2.23.5 – 500.5 – 1503.5 – 1000.01 mg/d –4.2 mg/d4.0 – 5.0SYNERGISTIC INTERACTIONANTAGONISTIC INTERACTIONLESS THANADDITIVEADDITIVEType 1 Type 2 Type 3 INTERACTIONType 1 Type 2INTERACTION- mortality ↗- incidence of lesions inkidney ↗- BWG ↘- mortality ↗- mortality ↗- AST ↗ (P)- inorganic phosphorus↘- albumin ↗- globulin ↘- RBC ↘ - albumin, glucose,hemoglobin ↘- BW ↘- RW-L,Pc ↗- incidence of lesions inliver ↗- BWG, feed intake ↘- incidence of lesions inliver ↗- incidence of lesions inliver ↗- BWG, feed intake ↘- mortality ↗- incidence of lesions inkidney ↗- glycocholic acid ↗- hemolytic complementtiters ↘- RW-K,Pv ↗- RW-BF ↘- phosphorus ↘- BUN ↗- incidence of lesions inliver ↗REF- TP, cholesterol ↘ Kumar andBalachandran, 2005- glycocholic acid ↗- DTH reaction →- albumin ↘ - RW-G →- TP, cholesterol,triglycerides →- uric acid ↗- LDH →- RW-K ↗ - TP ↘ - BWG, feed intake ↘- RW-L ↗- RBC ↘- albumin, cholesterol ↘- BWG ↘- RW-H ↗- RBC ↘- BWG ↘- complement titer ↘- albumin ↘- creatinine ↗- ALT ↗Morrissey etal., 1987Pier et al.,1989Smith et al.,1992- hemoglobin ↘- uric acid ↘ Ahamad etal., 2006- RW-G ↗ - RW-K →- RW-BF ↘- TP ↘- cholesterol, calcium→- ALP ↘Kubena etal., 1997cRichard etal., 1978- BW ↘ (P) Hayes et al.,1977

(30 d) 2AF-RUBGuinea pig 0.02 bw –(14 d) 3 8.0 bw- complement titer,bacteriostatic activity ↘(P)- AST ↗ (P)- ALP ↘Thurston etal., 1989AF-RUBChicken 2.5 – 500(21 d)1) Morrissey et al. (1987) : intragastrically administration2) Hayes et al. (1977) : initial exposure to RUB, followed by AF on day 153) Thurston et al. (1989) : 11 administrations- BW ↘- TP ↘- cholesterol ↘Wyatt et al.,1973

INTRODUCTIONLike T-2 toxin, DAS has been described as radiomimetic with regard to lymphoid tissues andgastrointestinal epithelium, as a contact necrotizing agent for lingual and buccosal mucosa. However,oral lesions were only observed in chicken, the most susceptible species (Kubena et al. 1993).The association of AF and DAS led to a less than additive or an antagonistic interaction on mostbiochemical parameters investigated, and also with either increased or decreased values for thesame compounds (cholesterol, TP, enzyme concentrations) depending on species (Table 3).In liver, lesions (vacuolar changes accompanied by early portal fibrosis) were reported similar innature and severity between animals exposed to AF and AF-DAS treatments (Harvey et al., 1995a;Harvey et al., 1991).3.3) Interaction between Aflatoxins (AF) and Deoxynivalenol (DON)Two experiments were conducted in pigs and chickens to study the interaction between AF andDON (Harvey et al., 1989b; Huff et al., 1986). In both experiments, the body weight gain wasdecreased with an additive or less than additive interaction between the two toxins (Table 3). Themain difference between the two experiments was the individual effect of DON. Harvey et al.(1989b) only observed minor effects of DON, and for most parameters he measured, the interactionwith AF was less than additive or antagonistic. Conversely, Huff et al. (1986), who used highconcentration of DON, reported an important effect of the toxin. He also observed less than additiveor antagonistic interaction between DON and AF for the different parameters he measured (Table 3).At the histological level, mono-contaminated feed with either DON or AF induced edema of thegastric mucosa. By contrast, animals fed the multi-contaminated diet did not show any gastric lesions(Harvey et al., 1989b). In liver, mild hepatic interlobular fibrosis, bile duct hyperplasia, and diffusehepatocellular lipidosis were only observed in the groups fed AF, either mono- or multicontaminatedgroups, and suggested less than additive interaction (Harvey et al., 1989b).4) Interaction between Aflatoxins (AF) and other mycotoxins4.1) Interaction between Aflatoxins (AF) and Cyclopiazonic Acid (CPA)The interaction between these two toxins was investigated as strains of Aspergillus flavus producethe toxins simultaneously. The experiments were performed on laboratory animals (rats or guineapigs) or on chickens. The body weight gain was always decreased in animals fed the co-contaminateddiet (Table 4). When rats were exposed for a short period to the toxins (3 days), a less than additiveinteraction was observed (Morrissey et al., 1987). But in longer period of exposure, the interaction33

INTRODUCTIONbetween AF and CPA on the body weight of animals was qualified as synergistic or additive (Pier etal., 1989; Smith et al., 1992).Biochemical and hematological alterations were also measured, and less than additive toantagonistic interaction between AF and CPA were reported (Kumar and Balachandran, 2005;Morrissey et al., 1987; Smith et al., 1992) (Table 4). Similar types of interaction were noted onimmune parameters, with especially an antagonism action of CPA on the depressant effect of AF oncell-mediated immunity (Pier et al., 1989) (Table 4). However, the author suggested that the reducednumbers of animals in the combination group, and that survivors had apparently greater resistanceto the toxins may have influenced the data obtained.Among these studies, a particular attention has been paid on microscopic lesions. When rats wereshortly exposed to the mycotoxins, less than additive or antagonistic interaction between the twotoxins were described for the incidence in liver lesions (Morrissey et al., 1987) (Table 4). Conversely,when animals were exposed for a longer period to these toxins, the interaction was suggested to beadditive in liver lesions, characterized by a marked cytoplasmic vacuolation (Kumar andBalachandran, 2009; Pier et al., 1989). Likewise, this greater effect was noticed in the severity andincidence of kidney lesions in broiler chickens (Kumar and Balachandran, 2009), and inconsistently inrats, depending on the ratio of mycotoxins used (Morrissey et al., 1987) (Table 4). Of note, lesionsrecorded in gizzard, crop and proventriculus suggested adverse effects on the digestive tract; theinteraction leading either to additive (Kumar and Balachandran, 2009) or less than additive effects(Smith et al., 1992).4.2) Interaction between Aflatoxins (AF) and - Moniliformin (MON), - Sterigmatocystin (STER), -Citrinin (CIT), - Rubratoxin (RUB)In order to a give complete picture of the published literature, we should mention that one studyhave investigated the combined effect of AF and moniliformin on chicken (Kubena et al., 1997c),another one the combined effect of AF and sterigmatocystin on guinea pig (Richard et al., 1978), anda third one the combined effect of aflatoxin and citrinin on chicken (Ahamad et al., 2006). Thecombined effect of AF and rubratoxin has also been investigated (Hayes et al., 1977; Thurston et al.,1989; Wyatt et al., 1973). The parameters measured and the type of interaction observed betweenAF and MON, STER, CIT and RUB are summarized in Table 4.34

Table 5 : Interaction between FusariotoxinsMycotoxinsSpecies(exposure)FB-MONTurkey(21 d)FB-MON:Pig(28 d)FB-MONLaying hen(420 d)Doses200 – 100100 – 100100 – 50100 – 100FB-MON100 – 200Chicken(21 d) 1 200 – 100200 – 200SYNERGISTIC INTERACTIONANTAGONISTIC INTERACTIONLESS THANADDITIVEADDITIVEType 1 Type 2 Type 3 INTERACTIONType 1 Type 2INTERACTION- feed intake ↘ (P) - RW-L ↗- AST ↗- feed intake ↘- creatinine ↗ (P)- RW-H ↗ (P)- albumin ↘- RW-L ↗- albumin, TP ↗- BWG ↘- RW-H ↗- RW-BF ↘- RBC ↘ - BWG ↘- mortality ↗- glucose ↘- LDH ↗- inorganic phosphorus↘- AST, ALP, LDH,GGT ↗- total iron →- AST ↗ - mortality ↘ - CK ↘ - RW-L, K ↘ - uric acid →- egg production ↘- egg weight →- RW-H ↗ - feed intake ↘- mortality ↗- AST ↗- RW-H,K, L ↗- mortality ↗- albumin, TP ↗ - AST ↗- RW-K ↗ - RW-H ↗ - feed intake, BWG ↘- mortality ↗- AST ↗- RW-K, L ↗ - RW-H ↗- albumin ↗- mortality ↗- AST ↗- BWG ↘- TP ↗REFBermudez etal., 1997Harvey et al.,2002Kubena et al.,1999Ledoux et al.,2003FB-MONTurkey(21-28 d)FB-MONQuail(35 d)FB-MONFish(70 d)FB-T2Turkey200 – 100200 – 10020 – 4040 – 40300 – 5.0- BWG ↘- serum pyruvate ↗- size of hepatocytenuclei ↘- RW-L ↗ (P)- RW-G ↗- Ab ND ↘ - feed intake, BWG ↘- RW-T,BF,S ↘- lymphocytesstimulation ↘- bacteria in tissue &blood, mortality p.i ↗- BW ↘- mortality ↗- size of hepatocytenuclei ↘- TP, cholesterol ↗- ALT, AST ↗- BWG, feed intake ↘- Sa/So liver ↗- feed intake ↘ - Sa/So liver ↗- BWG ↘- inorganic phosphorus- oral lesions ↗- RW-Pc ↗- albumin ↗- LDH, CK ↗Li et al., 2000- creatinine →- DTH reaction → Sharma et al.,2008Yildirim etal., 2000- uric acid ↗ Kubena et al.,1995a

(21 day) - RBC, hemoglobin ↗- AST ↗- LDH ↗ (P)FB-T2Chicken(19 d)FB-DASTurkey(21 d)FB-DONPig(28 d)FB-DONChicken(21 d)FB-DONPig(35 d)FB-FAChicken egg(21 d)T2-DASLaying hen(24 d)DON-T2Pig(35 d)DON-T2Chicken(21 d)DON-NIV300 – 5.0300 – 4.050 – 4.0300 – 156.0 – 3.01.0 – 1.0(µg/egg)5.0 – 5.0(µg/egg)25 – 25(µg/egg)50 – 50(µg/egg)2.0 – 2.0- RW-Pc ↗ (P) - BWG, feed intake ↘- mortality ↗- RW-G ↗- inorganic phosphorus↘- RBC ↗ (P)- BWG ↘- AST, CHL, ALP ↗ (P)- lymphocytesstimulation ↘ (P)- mortality ↗- RW-Pv ↗ (P)- cholesterol ↗ (P)- AST ↗ (P)- LDH ↗- embryos mortality ↗- embryos mortality ↗- embryos mortality ↗- embryos mortality ↗- egg production (at endof the trial) ↘- feed intake↗↘- RW-L ↘ - feed intake ↘- RW-Lg ↗- cholesterol ↘- RW-L, K ↗- cholesterol ↗- calcium ↗- BWG, feed intake ↘ - RW-L, G ↗- RW-S,H ↘- AST, LDH ↗- RW-G ↗- BUN ↗- GGT ↗- severity, extent ofliver & lung lesions ↗- specific Ab IgG ↘- cytokines expression↘- oral lesions ↗- feed intake ↗- triglycerides ↘- uric acid ↗- oral lesions ↗- RW-S →- TP, albumin ↗- AST, LDH, GGT ↗- oral lesions ↗- RW-Pc →- cholesterol ↘Kubena et al.,1997aKubena et al.,1997b- creatinine ↗ - albumin ↘ - cholesterol ↗- GGT ↗ Harvey et al.,1996- BWG ↘- RW-L, K ↗- TP ↗- neutrophils ↘- severity, extent ofkidney lesions ↗- specific lymphocytesstimulation ↘2.5 – 0.4 - BWG, feed intake ↘2.5 – 0.8 - BWG, feed intake ↘2.5 – 1.6 - BWG, feed intake ↘2.5 – 3.2 - BWG, feed intake ↘- BW ↘- RW-G,BF ↗16 – 4.0- cholesterol ↘ - TP, albumin ↘- LDH ↘- oral lesions ↗0.071 bw –0.071 bw- uric acid ↗ - TP ↗- PROD ↗- RW-H → - RW-BF →- LDH ↗ - egg production (atintermediate time) ↘- GDH ↘- Ig A ↗- CDNB ↗- total CO2 ↘- EROD ↗- creatinine, albumin→- specific Ab IgA →Kubena et al.,1997aGrenier et al.(2011)Bacon et al.,1995Diaz et al.,1994Friend et al.,1992Kubena et al.,1989bGouze et al.,

Mouse(28 d) 2 0.071 bw –0.355 bwDON-ZEAMouse(56 d)DON-ZEAMouse(14-21 d)DON-MONTurkey(21 d)0.355 bw –0.071 bw0.355 bw –0.355 bw5.0 – 1025 – 1020 – 100- uric acid ↗ (P) - TP ↗- Ig A ↗- CDNB ↗- phosphorus ↗- total CO2 ↘- uric acid ↗- TP ↗- Ig A ↗- DCNB ↗- feed intake↘- globulin ↘- TP ↗- phosphorus ↗- total CO2 ↘- PROD ↗- DCNB ↗- Ig A ↗- CDNB ↗- total CO2 ↘- uric acid ↗- CDNB ↗- EROD ↗ - feed intake ↘ 2005- EROD, PROD ↗- RBC ↘ - BW ↘- RW-L →- Ig A ↗1) Ledoux et al. (2003) : for the highest dose of MON in the combination, feed intake and BWG not took into account due to high mortality2) Gouze et al. (2005) : administration thrice a week for 4 weeks- EROD, PROD ↗ - feed intake ↘- bacteria in spleen ↗ - DTH reaction ↘Forsell et al.,1986Pestka et al.,1987- RW-K ↗ - BWG ↘ - RW-H ↗- calcium → Morris et al.,1999

INTRODUCTIONII. INTERACTIONS BETWEEN FUSARIOTOXINS1) Interaction between Fumonisins (FB) and other Fusariotoxins1.1) Interaction between FB and MONSeveral studies reported the effects of interaction between FB and MON, and mainly with highlycontaminated feeds and in poultry (Table 5).A less than additive and an antagonistic interactions in turkeys (Bermudez et al., 1997; Li et al.,2000) and in broiler chicks (Ledoux et al., 2003), respectively, were observed on the growth animalsdepression. In pigs, quails and catfish, the effect of the multi-contaminated diet on BWG ranged fromless than additive to additive and to synergistic (Harvey et al., 2002; Sharma et al., 2008; Yildirim etal., 2000) (Table 5). However, the interaction on feed consumption was not consistently associatedwith the interaction on BWG.Some authors reported animals mortality, especially due to MON presence in chickens (Ledoux etal., 2003). Surprisingly, no mortality in laying hens fed combination treatment has been observedcompared to individual treatments (Kubena et al., 1999). But as suggested by the authors, the highermortality in FB group alone may be attributed to the increased egg weights because 4 of the deathswere attributed to uterine prolapses in the later stages of production. In the same study, antagonisticinteraction between MON and FB was reported, with a sparing action of MON on the FB effects oneggs performance (Table 5).Interestingly, in comparison to other mycotoxin interactions both FB and MON induced anincrease of serum TP and albumin concentrations. Other mycotoxins mainly reported a decrease ofthese biochemical compounds, likely due to their property to inhibit the synthesis of proteins.Serum AST concentrations increased following exposure to the combined treatment, but differenttypes of interaction, from synergism type 2 to antagonism were recorded (Bermudez et al., 1997;Harvey et al., 2002; Kubena et al., 1999; Ledoux et al., 2003; Sharma et al., 2008) (Table 5). Of notethat Yildirim et al. (2000) analyzed the serum pyruvate concentrations in animals exposed tocontaminated diets, as MON is known to inhibit the pyruvate dehydrogenase. At the higher inclusionof FB in the co-contaminated diet, he reported a synergistic interaction on the level of serumpyruvate (Table 5), suggesting a stronger inhibition of this enzyme.This enzyme inhibition could be related in the increase of relative weight of hearts in animalsexposed to MON, as this enzyme is involved in the production of ATP and that the heart may have toincrease the blood flow to supply more oxygen to the body in order to increase ATP production(Ledoux et al., 2003). Cardiomegaly was thus observed, and MON combined with FB led either to35

INTRODUCTIONsynergistic and additive effects (Kubena et al., 1999; Ledoux et al., 2003) or less than additive effects(Bermudez et al., 1997; Ledoux et al., 2003) (Table 5). A lack of cardiomegaly was indicated byHarvey et al. (2002), showing that swine is less sensitive to MON than poultry.Cardiac lesions, mainly characterized by a diffuse loss of cardiomyocyte cross striations wererecorded in the MON and MON+FB groups, with the same degree of severity between these groups(Bermudez et al., 1997; Harvey et al., 2002; Ledoux et al., 2003). Similarly, liver lesions described bythese authors were attributed to FB and were not more severe in the FB+MON diet. In other hand,combined mycotoxins revealed smaller hepatocellular nuclei in fish, indicating an additive orsynergistic interaction according to the concentration of FB in the combination (Yildirim et al., 2000).Two studies focused on the interaction effect on the immune system (Li et al., 2000; Sharma etal., 2008) (Table 5). An additive effect was observed in the reduced antibodies against NewcastleDisease. Li et al. (2000) indicated that after an E. coli challenge, the ability of animals to eliminatebacteria from the blood system was diminished in animals exposed to mycotoxins. However, a lessthan additive interaction was reported in the multi-contaminated diet, as shown by the increasednumbers of bacterial colonies in blood and tissues. Likewise in this study, a less than additive effectwas reported in the reduced proliferation of lymphocytes upon mitogenic activation. By contrast,Sharma et al. (2008) observed that MON reversed the immunosuppressive effects of FB, in quailsundergoing skin hypersensitivity test, suggesting an antagonistic interaction on the cellular immuneresponse.1.2) Interaction between FB and TCT1.2.1) with TCT type AInteraction between FB and T-2 toxin was reported in turkeys and in chickens (Kubena et al.,1997a; Kubena et al., 1995a) (Table 5).Both experiments reported an additive effect on the BWG depression. Oral lesions, caused by T-2toxin, were also recorded, but effect on scores differs between studies. The antagonist effect waslikely due to the greater reduction in feed consumption, thereby diminishing the T-2 toxin ingestionin the combination group (Kubena et al., 1997a). In both studies, the increased RW-G was related tothe irritant property of T-2 toxin, typically described as focally reddened mucosa (Kubena et al.,1995a).The decreased or increased effect on cholesterol concentration reflects the individual effect of FBin each study. An important difference between these two studies is the activity of hepatic enzyme.In one hand, T-2 toxin seems to potentiate the FB effect (Kubena et al., 1995a) on AST and LDH,36

INTRODUCTIONwhereas in the experiment of Kubena et al. (1997a), the effect on these enzymes was lesser than FBalone. No histological analysis on liver was done to find out a possible link, but a partial explanationcould also be due to the greater reduction in feed intake in the combination group (Kubena et al.,1997a). Another reason may be the type of birds used and the sex, sensitivity being differentbetween male and female.However, the likelihood of encountering concentrations of 300 mg FB/kg in finished feed is verysmall.One study investigated the interaction of DAS with a very high content of FB in diets (Kubena etal., 1997b), and reported additive effect on performance and antagonistic effect for oral lesions(Table 5). Hematological and biochemical values showed different types of interaction, fromsynergism to antagonism.1.2.2) with TCT type BThree experiments, performed in pigs and in chickens, investigated the combined effect of DONand FB, the most frequently detected fusariotoxins (Grenier et al., 2011; Harvey et al., 1996; Kubenaet al., 1997a) (Table 5).The type of interaction on body weight gain differed depending on the experiment. A synergisticinteraction was reported by Harvey, no change in this parameter whatever the toxins included in thediet was observed by Grenier. In chicken, Kubena et al. (1997a) observed a less than additive effect,mainly due to FB in the combination group. Interestingly, in this latter study, the feed intake wasslightly increased in comparison to the other treatments (Table 5).An antagonisitic interaction has been observed for albumin. DON in the multi-contaminated dietseems to potentiate the FB effects, such as reflected by the hepatic enzymes concentrations (Harveyet al., 1996; Kubena et al., 1997a). It could be linked to the unexpected effect on the RW-L in theexperiment conducted by Harvey et al. (1996), where the combination led to an opposite effectcompared to individual effects (Table 5). The large and synergistic decrease in the RW-L may be dueto an important hepatic damage, as already mentioned in this review for AF-FB interaction. However,histologically in this experiment, hepatic lesions observed (necrosis) in animals fed the combinedmycotoxins, were not more severe than those observed in animals fed the FB diet. By contrast,Grenier et al. (2011) showed a greater effect in the severity and extent of lesions (mainlyhepatocytes vacuolization and megalocytosis) (Table 5), suggesting that DON, known to alter theintestinal permeability, had enhanced the intestinal absorption of FB, well known to be poorlyabsorbed. The differences in these two studies are most likely due to the different concentrations of37

INTRODUCTIONFB used. These two studies are in agreement regarding the kidney lesions, where pigs exposed to theDON+FB diet did not have an increased frequency of lesions compared to individual diets (Table 5).Immune system was also evaluated following the exposure to both DON and FB (Table 5). Harveyet al. (1996) recorded a potentialization of the effect of FB by DON in the reduced lymphocytesproliferation upon mitogenic stimulation. Grenier et al. (2011) reported a less than additive effect onthe decreased index when lymphocytes were stimulated by ovalbumin. In their experiment, pigswere immunized with ovalbumin and the specific immune response was thus impaired, as alsodisplayed by the additive effect on the content of specific IgG. On contrary, antagonistic interactionwas recorded in the levels of specific IgA, showing a sparing effect of FB on the DON-induced IgAelevation. Also in this study, an additive effect on the decreased expression of cytokines in spleenwas noted.1.3) Interaction between FB and Fusaric Acid (FA)A synergistic interaction was reported when FB and Fusaric Acid (FA) were combined in equalconcentrations and injected into chicken eggs (Table 5), resulting in increased toxicity, as shown bythe percentage of dead embryos (Bacon et al., 1995). In the same study, when a relatively non-toxicconcentration of FA was combined with graded doses of FB, a synergistic response was alsoobtained.2) Interaction between Trichothecenes (TCT)2.1) TCT type AOne study investigated the interaction between the two type A trichothecenes, T-2 toxin and DAS.Both toxins inhibit protein synthesis in eukaryotic cells and are highly toxic to poultry (Diaz et al.,1994). In laying hens, Diaz et al. (1994) observed an additive effect of T-2 toxin and DAS on feedintake and oral lesions (Table 5). In the experiment, the authors recorded egg production throughoutthe trial. The lower egg production observed in hens receiving either T-2 toxin or DAS recoveredgradually during the experiment. Interestingly, the decrease in egg production of hens fed the cocontaminateddiet became progressively worse (Table 5). The authors suggested that whereas hensexposed to a single mycotoxin at low concentrations may be able to sustain a satisfactory rate of eggproduction, they may not be able to tolerate a combination of mycotoxins at similarly lowconcentrations.38

INTRODUCTION2.2) TCT type A and BTwo reports focused on the interaction between type B and type A trichothecenes, DON and T-2toxin (Friend et al., 1992; Kubena et al., 1989b). In swine, except for the highest dose of T-2 toxinthat interact synergistically with DON in the decreased BWG and feed intake, a less than additiveeffect was observed on these zootechnic parameters following the combination exposure, reflectingpredominantly the DON effect (Friend et al., 1992) (Table 5). By contrast, in broiler chicks, most ofthe observed effects (oral lesions, total protein, albumin and LDH) were due to T-2 toxin in thecombined treatment (Kubena et al., 1989b). However, an additive effect was obtained on the bodyweight gain and on the serum cholesterol concentration (Table 5). Decreased cholesterol levels couldsuggest inhibition of biosynthesis, with liver involvement and perhaps a shift of concentration fromblood to the liver.2.3) TCT type BOne study investigated the interaction between the two type B trichothecenes, deoxynivalenoland nivalenol (NIV), using two different low doses of both toxins (Gouze et al., 2005) (Table 5).Although the lowest dose of NIV and the two doses of DON used did not affect the plasmatic levels ofuric acid, when used in combination the toxins act in a synergistic manner. Whatever the dose of NIVand DON used, when the toxins were used in combination an additive effect was observed for thetotal protein level.Ingestion of DON or NIV induces IgA nephropathy in mice. Depending on the doses of two toxins,when used in combination the interaction range from synergistic to additive or less than additive onIgA synthesis. The hepatic drug metabolism activity was also assessed. When mice were exposed toDON and NIV, the toxins showed an antagonistic interaction on the activities of two monooxygenase(EROD and PROD). By contrast, for the highest doses of DON and NIV, a synergistic effect wasobserved on the elevated activity of glutathione S-transferase (DCNB as substrates).3) Interaction between Deoxynivalenol (DON) and other Fusariotoxins3.1) Interaction between DON and Zearalenone (ZEA)Three publications investigated the interaction between DON and zearalenone (ZEA) (Boeira etal., 2000; Forsell et al., 1986; Pestka et al., 1987). Forsell et al. (1986) showed in mice that exposureto the combined treatment resulted mainly in antagonistic interaction (Table 5). The authorobserved an antagonistic effect of ZEA on the DON-induced elevation of serum IgA. Similarly,39

INTRODUCTIONantagonism was noted in the ability of DON to inhibit the delayed hypersensitivity response inpresence of ZEA (Pestka et al., 1987). However, the resistance to Listeria monocytogenes wasreduced in an additive manner by co-administration of DON and ZEA (Pestka et al., 1987) (Table 5).Interestingly, DON+ZEA combination was assessed on growth of brewing yeast, considering thatgrains contaminated with mycotoxins are used for beer production, and may be introduced atdifferent steps of the brewing process (Boeira et al., 2000). Despite the effect caused by combinationof DON and ZEA at high concentrations, shown to pass from antagonism to synergism depending onthe ratio of toxins in the mixture, when low concentrations were used, inhibition of yeast growth wasnot observed.3.2) Interaction between DON and MONDue to the relatively tolerance of DON in poultry, its interaction with MON reflected mostly theMON effects (Harvey et al., 1997; Morris et al., 1999). Poults fed diets containing MON alone and theDON-MON combination exhibited an increased incidence of variable sized cardiomyocyte nuclei, withnumerous large, giant nuclei and a generalized loss of cardiomyocyte cross striations (Morris et al.,1999). Isolated renal tubules in sections of kidney were noted to have diffuse mineralization inanimals fed MON and DON-MON (Harvey et al., 1997; Morris et al., 1999), and also extensive tubularepithelial degeneration (Harvey et al., 1997). But Harvey et al. (1997) indicated a moderation of theseverity of lesions in the tissues of chicks fed DON-MON, suggesting an antagonistic effect.40

Table 6 : Interaction between Ochratoxin A (OTA) and other mycotoxins; interaction between T-2 toxin (T2) and cyclopiazonic acid (CPA) isalso includedMycotoxinsSpecies(exposure)OTA-CITRabbit(60 d)OTA-CITChicken(21 d)Doses0.75 – 153.0 – 300OTA-CITRat(20 d) 1 1.0 – 30OTA-CITRat(21 d)0.026 – 0.1OTA-CITRat(1 d) 2 1.0 – 25OTA-T2Chicken(21 d)OTA-T2Pig(30 d)OTA-T2Chicken(21 d)OTA-T2Chicken0.567 –0.9272.5 – 8.02.0 – 4.02.0 – 3.0SYNERGISTIC INTERACTIONANTAGONISTIC INTERACTIONLESS THANADDITIVEADDITIVEType 1 Type 2 Type 3 INTERACTIONType 1 Type 2INTERACTION- maternal mortality ↗- resorptions ↗- live fetuses ↘- % malformed ↗- DNA adducts in kidney↗ (P)- renal Na + -K + -ATPaseactivity ↘- % repairing tissue inliver ↗- NADPHdehydrogenase, NADPHcytochrome creductase ↘- cytochromeP-450 content↘- RW-L ↗- potassium ↘- GGT ↘ - BWG, feed intake ↘- hemoglobin ↘- lymphocytesstimulation ↘- RW-G ↗- BUN ↘- triglycerides ↗- AST ↘ (P)- lymphocytesstimulation ↘- RW-K ↗- glucose ↗- calcium ↘- lesion score in kidney↗- renal Mg 2+ -ATPaseactivity ↘- BWG, feed intake ↘ - TP, albumin, globulin↘- % tumid, necrotic cellsin liver ↗- calcium → - BWG ↘- hemoglobin ↘- TP ↘- CK ↘- LDH ↘- RW-G ↗- hemoglobin ↘- cholesterol ↘- creatinine ↗- ALP ↘- phagocytosis activity↘- oral lesions ↗- RW-L,K,Pv ↗- creatinine ↗- cholesterol, albumin ↘- BW, feed intake ↘- mortality ↗- Ab SRBC ↘- feed intake, BW ↘- water consumption ↗- triglycerides ↗- phosphorus ↘- aniline oxidation ↘- TP, albumin,globulin ↘- cholesterol →- uric acid ↗REFKumar etal., 2008Manning etal., 1985Mayura etal., 1984- BWG → Pfohl-Leszkowiczet al., 2008- RW-L ↗ - uric acid →- RW-K ↗- % necrotic tubularcells in kidney ↗- inorganic phosphorus - TP →↘- phosphorus ↘ - RW-Pc ↗- magnesium ↘- uric acid ↗- GGT →- ALP ↘- RW-L ↗- TP, BUN →Siraj et al.,1981Garcia etal., 2003Harvey etal., 1994Kubena etal., 1989aRaju andDevegowda,

(35 d) - RW-K,A ↗- GGT ↗ ALT ↘OTA-T2- RW-T ↘ - RW-BF →Chicken 2.0 – 3.0- Ab ND,IBD ↘(35 d)OTA-DASChicken(19 d)OTA-DONChicken(21 d)2.0 – 6.02.0 – 165.10 -6 bw/d– 2.10 -4bw/dOTA-FB5.10Rat-6 bw/d(15 d) 3 – 0.05 bw/dOTA-FBTurkey(21 d)0.05 bw/d –0.05 bw/d3.0 – 300OTA-FBRabbit 2.0 – 10(45 d)OTA-PAChicken1.0 bw – 60(28d) 4 bwOTA-PAMouse10 – 40(21 d)OTA-CPAChicken(21 d)2.5 – 34- MDA, PC in liver ↗- CAT in kidney ↘- MDA, PC in liver ↗- MDA in kidney ↗- CAT in kidney ↘- PC in liver ↗- PC in kidney ↗- CAT in kidney ↘- BWG, feed intake ↘- RW-H ↘- cholesterol ↘ (P)- creatinine ↗- AST ↗ (P)- ALT ↗- mortality ↘ - RW-G ↗ - BW ↘- oral lesions ↗- RW-K ↗- TP, albumin ↘- CK ↘- RW-G ↗- RBC ↗- BWG ↘- RW-L,K,Pv ↗- glucose, TP, albumin↘- RW-Pv ↗- uric acid →- cholesterol ↘- PC in kidney ↗ - MDA in kidney ↗- PC in kidney ↗- MDA in liver ↗- MDA in kidney ↗- uric acid ↗ - BUN ↗ - RW-K ↗ - RW-L,Pc,G ↗- RW-S ↘- RBC ↗- TP ↘ - albumin, globulin ↘- ALP ↗- mortality ↗ - BW ↘- RW-S ↘ (P)- mortality ↗- RW-Pc ↗- RW-L,K ↗ (P)- TP, cholesterol ↘ (P)- albumin ↘- creatinine ↗- ALT, AST, LDH ↗- RW-K ↘- BW ↘- BW ↘ - uric acid, triglycerides↗- cholesterol ↘- coagulation time ↗- RW-L ↗- triglycerides ↗- BUN → - uric acid ↗- creatinine →- triglycerides ↗- cholesterol ↘- inorganic phosphorus→- hemoglobin ↗ - triglycerides →- LDH ↗- glucose ↘- RW-Pv ↗2000Raju andDevegowda,2002Kubena etal., 1994aKubena etal., 1988Domijan etal., 2007Kubena etal., 1997bSivakumaret al., 2009Kubena etal., 1984Shepherd etal., 1981Gentles etal., 1999

T2-CPAChicken(28 d)T2-CPAChicken(21 d)1.0 – 106.0 – 34- CK ↗ (P)- RW-K ↗- triglycerides ↗- RW-L ↗- cholesterol↗- % CD + 4 & CD + 8 inthymus ↘- % CD + 8 in spleen ↘- lymphocytesstimulation ↘- RW-Pc ↗- albumin ↘1) Mayura et al. (1984) : subcutaneously administration during the 20 d-period of pregnancy2) Siraj et al. (1981) : one administration intragastrically and measurements 12 d post administration3) Domijan et al. (2007) : 15 d-treatment with OTA, and treatment with FB for the last 5 days of OTA treatment4) Kubena et al. (1984) : 14 administrations- BW ↘- mortality, oral lesions↗- RW-Pv,G ↗- RW-BF ↘- TP ↘- GGT ↘- % CD 4 + in spleen ↘- Ab ND ↘ Kamalavenkatesh et al.,2005- glucose ↘Kubena etal., 1994b

INTRODUCTIONIII. INTERACTIONS BETWEEN OCHRATOXIN A AND OTHER MYCOTOXINS1) Interaction between OTA and Citrinin (CIT)Regarding zootechnical parameters, Manning et al. (1985) observed an antagonistic interactionfor body weight gain when chickens were fed with OTA and CIT (Table 6). This response was mostlikely the result of the improved feed intake in the OTA+CIT group. Birds receiving the cocontaminatedfeed also showed an antagonistic effect in the increase of water consumption that isnormally associated with CIT toxicosis.Changes in serum total protein, cholesterol, triglycerides, and uric acid concentration alsodemonstrated an antagonistic effect between the two toxins (Table 6), and suggested that thecombination of mycotoxins gave some advantage to the birds and allowed them to maintainbiochemical parameters close to the to normal ranges. A proposed mechanism would be that thecytoplasmic accumulation of smooth endoplasmic reticulum may be indicative of enzyme induced bythe exposure to CIT and would protect against the toxicity of OTA (Manning et al. 1985; Brown et al.1986).Less than additive and antagonistic interactions between the two toxins were also observed inrabbits for lymphocytes proliferation and specific antibody response (Kumar et al., 2008) (Table 6).Because both OTA and CIT target the kidney and have been implicated as causal agents for theBalkan endemic nephropathy, most of the experiments conducted with both OTA and CITinvestigated their effects on renal function and structure (Brown et al., 1986; Glahn et al., 1988;Kitchen et al., 1977a; b; Kumar et al., 2007; Manning et al., 1985; Siraj et al., 1981). A synergisticinteraction between the two toxins was observed in dog when looking at the severity of clinical signsand mortality (Kitchen et al., 1977a). Concentrations of urinary GOT and LDH were significantlyincreased in animals exposed to both toxins, signing renal damages. A synergistic interaction was alsoobserved for the inhibition of the renal Na + -K + -ATPase activity in neonatal rats (Siraj et al., 1981)(Table 6). An experiment also reported that the OTA pretreatment blunted the diuretic effect of CITon kidney of pullets (Glahn et al., 1988). But as indicated by the author, the sequential use of toxinsand mode of administration may create conditions that differ significantly from situations in whichOTA and CIT are present simultaneously.Ultrastructural assessment of kidney showed mainly degenerative and necrotic changes in theproximal and distal tubules. The nephrotoxicity was related with mitochondria damage (Kumar et al.2007; Brown et al. 1986). In the animal group receiving both toxins, these lesions were more severeand intense (Kitchen et al., 1977b; Kumar et al., 2007), similar to the ones observed in the OTA-41

INTRODUCTIONexposed group (Brown et al., 1986; Manning et al., 1985) or less severe than the ones observed inthe animal group receiving the highest dose of CIT alone (Kitchen et al., 1977b).An increased genotoxicity of OTA was also observed in the presence of CIT, with increased DNAadductsin kidney, suggesting synergism (Pfohl-Leszkowicz et al., 2008) (Table 6).Two studies investigated the interaction between OTA and CIT on the teratogenesis and led todifferent results. Mayura et al. (1984) observed synergism between the toxins when looking at fetalmalformations in rats (Table 6). By contrast, Vesela et al. (1983) observed the same teratogeniceffects in chicken embryos exposed to either OTA alone or OTA+CIT.2) Interaction between OTA and Fusariotoxins2.1) Interaction between OTA and TCT2.1.1) with TCT type AFive studies focused on the interaction between OTA and T-2 toxin (Garcia et al., 2003; Harvey etal., 1994; Kubena et al., 1989a; Raju and Devegowda, 2000; 2002) (Table 6).Except the study of Raju and Devegowda (2000), the experiments showed that the association ofOTA and T-2 toxin act in an additive manner to depress the body weight gain (Table 6). The serumconcentrations of albumin, creatinine, cholesterol, uric acid and phosphorus were affected in thesestudies, representing mostly the effect of OTA alone, but less than additive or antagonisticinteractions were reported. These studies noted an additive interaction on the decreasedhemoglobin concentration. Effect of both OTA and T-2 toxin ingestion on enzyme concentrationsinduced different type of interactions between these studies, but in most of the case a significantdecrease was reported (Table 6). It was suggested that the property to inhibit the protein synthesisby these mycotoxins was involved.It seems from these studies that T-2 toxin may interfere in the OTA-induced renal impairment, asshown by the antagonistic interaction on the RW-K, uric acid, phosphorus or also the percentage ofnecrotic tubular cells (Table 6). Nonetheless, microscopic lesions showing degenerative changes inrenal tubular epithelium and proximal convoluted tubules were similarly reported for OTA andOTA+T2 groups (Harvey et al., 1994; Kubena et al., 1989a). In liver, T-2 toxin appeared to slightlypotentiate the hepatotoxicity of OTA, visible on the percentage of repairing tissue (Garcia et al.,2003) (Table 6).Finally, their interaction on immune system was reported by Harvey et al. (1994) and Raju andDevegowda (2002). Although an additive effect was observed on the lymphocytes proliferation upon42

INTRODUCTIONmitogenic stimulation, the reduced activity of phagocytosis in the combined treatment was similar tothe individual OTA treatment (Harvey et al., 1994), and an antagonistic effect was reported for thelevel of specific antibodies (Raju and Devegowda, 2002) (Table 6).One experiment investigated the interaction between OTA and DAS in chicken (Table 6), andobserved in general less than additive or antagonistic effect (Kubena et al., 1994a). Surprisingly, nomortality was recorded in comparison to the individual treatments. A less than additive effect wasobserved for the oral lesions, induced by DAS. Like T-2 toxin, in association with OTA, DAS seems tospare the OTA-induced renal dysfunction, as shown with the antagonistic effect on uric acidconcentration.2.1.2) with TCT type BOne study focused on the interaction between OTA and DON in chicken (Kubena et al., 1988). Formany parameters, such as BWG, relative weight of organs and biochemistry, the interaction was lessthan additive or antagonistic in nature (Table 6). Similarly, lesions in liver and kidney were only dueto the OTA presence in the contaminated diet.2.2) Interaction between OTA and FBThree experiments were conducted in rats, turkey poults and rabbits to analyze the interactionbetween these two toxins (Table 6); however the authors used very different doses of mycotoxins(Domijan et al., 2007; Kubena et al., 1997b; Sivakumar et al., 2009). Using very high concentrations ofFB, Kubena et al. (1997b) showed synergistic effects in association with OTA compared to mycotoxinsalone (performance, some biochemical parameters and enzyme levels). Using lower doses of toxins,Sivakumar et al. (2009) observed less than additive or additive effects on biochemistry and enzymelevels. Interestingly, in both studies, the concentrations of serum enzymes were increased comparedto the interaction between OTA and T-2 toxin (Table 6).Using doses of toxins reflecting the European-type diet, Domijan et al. (2007) analyzed theinteraction between these two toxins on the oxidative stress, in liver and kidneys of rats (Table 6).They observed that although the lowest doses of OTA and FB given separately did not increase theconcentration of malondialdehyde (MDA) and protein carbonyls (PCs) in the liver, their combinationproduced a synergistic effect. Similarly, in the kidney, the combination further increased the PCsconcentration, ranging from synergistic to additive or to less than additive effect, depending on thedoses used. The catalase activity in rat kidney decreased in a synergistic manner after exposure toboth OTA and FB, while this parameter was not affected by separate OTA or FB treatments.43

INTRODUCTION2.3) Interaction between OTA and ZEAHalabi et al. (1998) studied the histopathological effects of OTA and ZEA, in liver and kidneys ofrats. Small amounts of mycotoxins, over a long period were given intraperitoneally to rats. Theauthors indicated that ZEA antagonizes the toxic effects of OTA for body weight gain and relativeweight of kidney, leading to no changes in the combination group. Likewise, less severe lesions inkidneys were observed for ZEA and ZEA+OTA in comparison to the severity induced by OTA.3) Interaction between OTA and Aspergillus/Penicillium toxins3.1) Interaction between OTA and Penicillic Acid (PA)Combination of OTA and PA showed less than additive effects on body weight in chickens (Kubenaet al., 1984) and in mice (Shepherd et al., 1981), but act in a synergistic manner in the increase of themortality in both species (Table 6). Shepherd et al. (1981) also indicated that the combinationproduced more extensive lesions in kidney within the proximal convoluted tubules, but with lessrenal damages at day 21 than at day 10, showing a recovery from the initial shock.3.2) Interaction between OTA and CPAIn chicken, synergistic interactions were often recorded after combination of both mycotoxins(Table 6), and some of the deleterious effects induced by OTA were exacerbated by the presence ofCPA (Gentles et al., 1999).Another interaction, that cannot be classified in these different sections, concern the associationbetween T-2 toxin and CPA. This interaction was investigated in chicken either on general parameters(Kubena et al., 1994b) or on immune parameters (Kamalavenkatesh et al., 2005) (Table 6). Kubena etal. (1994b) showed less than additive effects on animal performances and synergistic effects onrelative weight of liver and kidney, as well as lipid compounds. Kamalavenkatesh et al. (2005)reported some additive interactions in the reduced subpopulations of lymphocytes in lymphoidorgans, and in the lymphocytes stimulation. By contrast, he observed an antagonistic interaction inthe specific antibodies content.44

INTRODUCTIONCONCLUSIONIn the present review, we have analyzed the data published in more than 100 papers, tocharacterize for different parameters the interaction between mycotoxins. More than half of thepublished papers investigated the interactions between aflatoxins with other mycotoxins. Althoughthese experiments are relevant in terms of toxicity and natural co-occurrence, we were surprisedthat only few published papers investigated the interactions between other mycotoxins, especiallyinteraction between toxins from Fusarium spp. which are of major concern worldwide.Overall, the review highlights the complexity of the interaction between mycotoxins. Indeed, evenwithin the same experiment the type of interaction varies according to the parameter measured.Although, most of the studies have shown a synergistic or additive interaction on the adverse effectsof the animals performance, the results on other parameters, especially on biochemical compounds,led to different type of interaction, going from synergistic to antagonistic for a same association.Many contributing factors could explain these discrepancies: sensitivity of animal model tomycotoxins, age and sex, nutritional status, as well as the duration and the route of exposure to thetoxin. In addition, as reflected in the tables, the levels of mycotoxins used in the experiments maychange the type of interaction observed.Very few studies investigated the effect of mycotoxins, alone or in combination on immuneparameters, while in farms vaccination or response to pathogens are common situation.Histopathological analyses provide informations on the organs and cells injuries, but would need tobe related with physiologically consequences, especially on functions, not enough studied (oxidativestress, hepatic drug metabolizing enzymes, intestinal permeability…).Finally, the number of studies investigating effect of low doses of toxins, representative of fieldsituation is very low. Although the use of moderate to high concentrations of toxins providesinformation on the type of interaction, these doses are not expected in environmental conditionsand are largely over the limit set by regulation/recommendation in different countries. It has beennoted in this review that a combination of mycotoxins at low concentration may have negativeeffects, even though the concentrations of individual mycotoxins are below the concentrationsreported to cause negative effects (Domijan et al., 2007). Although not reported in this review, theanalysis of interaction with more than two mycotoxins would be also relevant and useful in terms ofrisk assessment.Nonetheless, it can be concluded that exposure to a co-contaminated food/feed result in agreater risk to human and animal health. The co-exposition to two toxins led finally to greater total45

INTRODUCTIONeffects in comparison to the total effect of each individual toxin, even in cases categorized less thanadditive or antagonistic.ACKNOWLEDGEMENTSB. Grenier was supported by a doctoral fellowship (CIFRE 065/2007), jointly financed by theBiomin company and the Innovation Network B.R.A.I.N, ANRT (Association Nationale de la RechercheTechnique) and INRA (Institut National de la Recherche Agronomique).46

INTRODUCTIONTABLES LEGEND- ↗, ↘, → : evaluation at the end of each experiment, of the effect of the co-contaminatedtreatment in comparison to the control treatment.- (P), potentialization- unless stated, doses are expressed in mg of toxins/kg of feed. In few experiments, doses areexpressed in mg of toxins/kg of body weight/day and in mg of toxins/day, and indicated as bw/d andmg/d, respectively.- zootechnical parameters : BW(G), body weight (gain)- RW, relative weight of organs : RW-L, liver; -K, kidney; -G, gizzard; -Pv, proventriculus; -Pc,pancreas; -H, heart; -Lg, lung; -A, adrenals; -S, spleen; -T, thymus; -BF, bursa of Fabricius.- biochemical and hematological parameters : TP, total proteins; BUN, blood urea nitrogen; RBC, redblood cells; WBC, white blood cells; Sa, sphinganine; So, sphingosine; MDA, malondialdehyde; PC,protein carbonyl; CDNB, 1-chloro-2,4-dinitrobenzene; DCNB, 1,2-dichloro-4-nitrobenzene.- enzymatic parameters : ALT, alanine aminotransferase; ALP, alkaline phosphatase; AST, aspartateaminotransferase; GGT, gamma-glutamyltransferase; LDH, lactate dehydrogenase; GDH, glutamatedehydrogenase; CK, creatine kinase; CHL, cholinesterase; CAT, catalase; SOD, superoxide dismutase;GSTP + , gluthatione-S-transferase placental form positive; EROD, ethoxyresorufin O-deealkylase;PROD, pentoxyresorufin O-depenthylase- immune parameters : SMC, spleen mononuclear cells; Ab, antibodies; ND, Newcastle disease; IDB,infectious bursal disease; SRBC, sheep red blood cells; DTH, delayed-type hypersensitivity.47

INTRODUCTION3. Procédés de décontamination des denrées contaminées et réduction desmycotoxines chez l’animalEn conséquence des préjudices causés par les mycotoxines (sanitaire, pertes économiques), denombreuses stratégies ont été développées pour empêcher la croissance des champignonstoxinogènes et pour décontaminer et/ou détoxifier les aliments contaminés en mycotoxines. Cesstratégies incluent : 1) la prévention de la contamination en mycotoxines, 2) la décontamination desmycotoxines présentes dans les aliments et 3) l’inhibition de l’absorption des mycotoxines dans letractus gastro-intestinal.Nous présentons ci-après les approches mises en œuvre pour satisfaire les points 2 et 3. Lestechniques décrites ont principalement été rapportées en termes d’efficacité pour éliminer ouréduire la teneur en mycotoxines, mais d’autres critères ont également été pris en compte. En effet,la FAO (Food and Agriculture Organization of the United Nations) a établi certaines lignes directricesdans l’évaluation de ces procédés de détoxification et décontamination. Le procédé doit (i) inactiver,détruire, ou éliminer la ou les mycotoxines ; (ii) ne pas aboutir à la production de substancestoxiques, métabolites ou sous-produits dans l’alimentation ; (iii) conserver les valeurs nutritives etl’acceptabilité des denrées alimentaires ; (iiii) ne pas altérer les propriétés technologiques desproduits ; et si possible (iiiii) détruire les spores fongiques. En plus de ces critères, les procédésdoivent être facilement accessibles, facile d’utilisation et peu coûteux.Cette étude bibliographique fait l’objet de deux chapitres dans l’ouvrage Mycotoxin Reduction inGrain Chains: A Practical Guide édité par Antonio F. Logrieco, et paraîtra courant 2011.3.1. Méthodes physiques et chimiques pour la décontamination du maïscontaminéLe premier chapitre de cet ouvrage traite des techniques permettant une décontamination desdenrées contaminées, et plus particulièrement celles évaluées sur le maïs. D’une part, nousreportons les traitements dits physiques, tels que le tri, le tamisage ou le lavage, et également desprocédés industriels, initialement non développés pour la lutte contre les mycotoxines, tels qu’enamidonnerie, les radiations ou l’extrusion. D’une autre part, nous présentons les traitements ditschimiques, tels que l’ammoniation, l’ozonation ou la nixtamalization. A noter que, l’ammoniation est48

INTRODUCTIONun procédé largement connu et utilisé dans certains pays d’Amérique, d’Europe ou d’Afrique, contrela contamination en aflatoxines.Nous avons également indiqué dans la mesure du possible, la toxicité résiduelle des alimentstraités par ces méthodes, en particulier par les méthodes chimiques. Des tests biologiques, ainsi quedes expériences animales ont été menés à cet effet.49

INTRODUCTIONPhysical and chemical methods for mycotoxins decontamination in maizeGRENIER, B. 1,3 , LOUREIRO-BRACARENSE 2 , A.P. & OSWALD, I.P 1 .1INRA, ToxAlim, Equipe Immuno-Mycotoxicologie, Toulouse, France.2Universidade Estadual de Londrina, Lab. Patologia Animal, Londrina, Brazil.3BIOMIN Research Center, Technopark 1, Tulln, Austria.Address correspondence toDr Isabelle P. OswaldINRA-Unité ToxAlim180 chemin de Tournefeuille BP 9317331027 Toulouse Cedex 3Phone : +33561285480E-Mail : isabelle.oswald@toulouse.inra.fr50

INTRODUCTIONABSTRACTMycotoxins are fungal secondary metabolites that have been associated with toxic effects inhuman and animal. The contamination of food and feed with mycotoxins is a worldwide problem.Because of the detrimental effects of these mycotoxins, a number of strategies have been developedto prevent the growth of mycotoxigenic fungi and to decontaminate and/or detoxify mycotoxincontaminated food and animal feed. These strategies include : 1) the prevention of mycotoxincontamination, 2) detoxification of mycotoxins present in food and feed and 3) inhibition ofmycotoxin absorption in the gastrointestinal tract. The present chapter mainly highlights processesinvolved in the second point, commonly classified as physical and chemical methods, and which havebeen evaluated in the maize grain chain. Efficiency and safety of these approaches are the maincriteria reported. Only a few of these methods are in practical use; this is probably due to thedifficulty to comply with the FAO requirements. In addition, no single method has the ability toremove simultaneously the wide variety of mycotoxins which may co-occur in maize.51

INTRODUCTIONINTRODUCTIONConsumption of mycotoxin-contaminated food or feed lead to many different toxic effects such ascancer, acute mortality, reproduction disorder or growth impairment in humans and/or animals(Oswald and Comera, 1998). The Food and Agricultural Organization (FAO) estimated that as much as25% of the world’s agricultural commodities are contaminated with mycotoxins. The economic lossesdue to mycotoxins contamination are estimated in billions dollars annually worldwide (Kabak et al,2006).Due to the multiple possible origins of fungal infection, any prevention strategy for fungal andmycotoxin contamination must be carried out at an integrative level all along the food productionchain. Three steps of intervention have been identified. The first step in prevention should occurbefore any fungal infestation; the second step is during the period of fungal invasion of plantmaterial and mycotoxin production; the third step is initiated when the agricultural products havebeen identified as heavily contaminated. Such hazard analysis has some similarity with the HACCPmanagement system of food safety (Jouany, 2007).The efforts must be concentrated on the two first steps since, once mycotoxins are present, it isdifficult to eliminate them in a practical way. However, the prevention of mycotoxin contaminationprior to harvest or during post-harvest and storage is not always possible necessitatingdecontamination before the use of such materials for food and feed purposes. Consequently,developing and implementing efficient detoxification methods became very important.Guidelines for evaluating mycotoxin detoxification and decontamination procedures have beenestablished by the FAO. The process should (1) inactivate, destroy, or remove the mycotoxin; (2) notresult in the deposition of toxic substances, metabolites, or byproducts in the food or feed; (3) retainnutrient value and feed acceptability of the product or commodity; (4) not result in significantalterations of the product’s technological properties; and if possible, (5) destroy fungal spores. Inaddition to these criteria, the process(es) should be readily available, easily utilized, and inexpensive(CAST, 2003).This chapter will present and discuss some of the physical and chemical methods, whichcombined most of the points mentioned above, especially the effectiveness of mycotoxinsdecontamination, the safety of these treatments, and the ability to be used on a commercial scale aswell. Nonetheless, in the following sections are only reviewed approaches assessed in the maizegrain chain and thereby, their efficiencies on mycotoxins that are of the greatest concern for thiscereal : aflatoxins, fumonisins, trichothecenes, zearalenone and ochratoxins.52

INTRODUCTIONI. PHYSICAL METHODS1) Removal from contaminated maizea) SortingSince the majority of mycotoxin contamination usually occurs in a relatively small number ofseeds or kernels, the segregation and sorting of damaged, discolored, crops containing visible mouldgrowth can lead to the removal of significant quantities of mycotoxins from these crops (Kabak et al.,2006). Sorting of the maize by fluorescence can achieve a partial removal of aflatoxins.Contamination can be observed by fluorescence following illumination with UV light (365 nmwavelenght). The Bright Greenish Yellow (BGY) fluorescence obtained, is not caused by aflatoxin itselfbut by a peroxydase transformation product of kojic acid, a metabolite of Aspergillus flavus and otherfungi (Scott, 1998). This test, known as the black light test, is considered as a presumptive indicatorof aflatoxin, but as mentioned by Bothast and Hesseltine (1975) both false positives (BGYfluorescence and no toxin detected) and false negatives (toxin detected and no BGY fluorescence)may occur during this evaluation, and results must be interpreted with caution.More recently, Pearson et al. (2004) proposed another sorting method, based on the fact thatmycotoxin-producing molds initially infect the oil-rich germ using grain lipids for their growth andmetabolism. Therefore, the free fatty acid content of grain has been proposed as a sensitive index ofincipient grain deterioration. Using optimal pair of filters (spectral absorbances at 750 and 1200 nm)the authors could distinguish kernels with aflatoxin contamination at >100 ppb from kernels with nodetectable aflatoxin, with >98% accuracy. When these two spectral bands were applied to sortingcorn at high speeds, reductions in aflatoxin averaged 82% for corn samples with an initial level ofaflatoxin >10 ppb and by removing ≈5% of the incoming grain (Pearson et al., 2004).In the same study, the method was applied for fumonisins-contaminated corn. The high-speeddual-wavelength sorter reduced fumonisin by an average of 88%, including those with low levels offumonisin.b) Sieving-cleaningCleaning grains removes kernels with extensive mold growth, broken kernels and fine materialssuch as dirt and debris, which helps to reduce mycotoxin concentrations. Murphy et al. (1993)reported the sieving of corn screenings (160 samples) into eight different particle sizes andconcluded that corn screenings or broken corn kernels usually contain fumonisins levels about 10fold higher than intact corn kernels. Similarly, fractionation of maize samples by sieving through a 353

INTRODUCTIONmm screen shows that fractions termed 'fines' (< 3 mm) had significantly higher total fumonisinsconcentrations and accounted for between 4.7 and 20.0% of the samples by mass (Sydenham et al.,1994). The data indicated that removal of the 'fines' resulted in overall reductions in total fumonisinlevels of between 26.2 and 69.4%.Regarding deoxynivalenol (DON) and zearalenone (ZEA), Trenholm et al. (1991) investigated theefficacy of sieving with a series of screens, on corn contaminated with 23 mg of DON/kg and with 1.2mg of ZEA/kg. Reduction of 73% and 79% of the total DON and ZEA respectively has been reported.c) Flotation and density segregationMould-damaged, mycotoxin-contaminated kernels may exhibit different physical properties thannon-damaged kernels. So they can be separated by density segregation in certain liquids, orfractionation according to specific gravity tables. Because fungi deriving nutrients from grainsthrough metabolism may alter the density of grains, it seems possible that fungal-infected grains canbe segregated from uninfected grains without specificity for any given fungus or grain. A 67%reduction in aflatoxin concentration of corn was obtained by removal of corn buoyant in water (Huff,1980; Huff and Hagler, 1982). The efficiency of density segregation was enhanced by increasing thespecific gravity of the suspending liquid. With the use of a 30% sucrose solution, the aflatoxinconcentration of corn was reduced by 87% (Huff, 1980). In another study, only 4 of 54 aflatoxincontaminatedsamples remained aflatoxin-positive after density segregation with 30% sucrose (Huffand Hagler, 1982). Similarly, density segregation using a saturated sodium chloride solution waseffective in reducing the aflatoxin concentration of contaminated corn (Huff and Hagler, 1985). Cornkernels which were buoyant in saturated sodium chloride represented 3% of the total sample, yetcontained 74% of the aflatoxin.Besides aflatoxin-contaminated grains removal, Huff and Hagler (1985) assessed their method ondeoxynivalenol-contaminated corn kernels and obtained for two naturally contaminated cornsamples, a reduction of 53% and 77% in DON concentration with water and 30% sucrose.The same principle was applied for fumonisins-contaminated maize (Shetty and Bhat, 1999). Withincreasing concentrations of NaCl, more maize entered the buoyant fraction and the removal offumonisins can reach 86% with saturated NaCl solution. The authors suggested washing grains withwater in order to remove the salty taste (Shetty and Bhat, 1999).54

INTRODUCTIONd) Washing with sodium carbonate solutionWashing methods take advantage of the fact that mycotoxins are primarly found on the outersurface of grains, but also involve contamination of commodities by soluble mycotoxins. In addition,sodium carbonate is inexpensive, widely available, and nontoxic to humans and animals. A simplemethod to reduce Fusarium mycotoxins such as deoxynivalenol and zearalenone, has beenproposed by Trenholm et al. (1992), where simple washing procedures, using distilled water orsodium carbonate solutions were applied in corn contaminated samples. Authors reported for corncontaminated with deoxynivalenol at 25 mg/kg, a removal from the grain (69-95% reduction) aftersoaking in a solution of sodium carbonate (0.1M) for 24-72h. Mycotoxin removal increased withincreased soaking time. A substantial reduction in zearalenone levels (87%) was also observed oncontaminated corn washed with 1M sodium carbonate solution for 30 minutes (Trenholm et al.,1992). ZEA is a weak phenolic acid whose solubility is greatly enhanced in alkaline conditions assodium carbonate solution.In a feeding trial, this procedure was shown to be effective in reducing the toxicity of Fusariumcontaminatedcorn when fed to growing pigs, as measured by the increase in daily feed consumptionand weight gains (Rotter et al., 1995). By contrast, Bennet et al. (1980) reported that sodiumbicarbonate had no effect on the ZEA concentration of naturally or artificially contaminated corn.e) DehullingDehulling is a milling technique that removes the outer layers of grain by abrasion. One of theearly practices of African food preparation was manual dehulling of grain before grinding. Thispractice is no longer in common use except in few countries, and has been replaced by use ofmechanical disc dehullers on a commercial and semi-commercial scale. The efficacy of this proceduredepends on the degree of fungal penetration into the kernel.The outer layers are most susceptible to fungal attack and aflatoxins accumulation. A reduction of92% in aflatoxin levels was reported after a physical dehulling procedure in contaminated maize(Siwela et al., 2005). The same reduction level of aflatoxin content in naturally contaminated maizewas obtained using “muthokoi”, a traditional dehulled maize dish prepared in Kenya (Mutungi et al.,2008).f) MillingMilling is traditionally used for grain processing. We can distinguish two milling processes : dryand wet milling, the latter being the major milling process used for maize. This method allows to55

INTRODUCTIONseparate the grain into different fractions. It is therefore, important to identify the fractions wherethe mycotoxin remains, so that they can be diverted to lower-risk uses or subjected todecontamination procedures.During the dry milling process, mycotoxins tend to be concentrated in germ and bran fractions. Bycontrast, during the wet milling process, mycotoxins may be dissolved into the steep water ordistributed among the byproducts of the process (Bullerman and Bianchini, 2007).After wet milling of aflatoxin-contaminated corn samples, the steepwater-solubles fractionscontained the highest concentration of aflatoxin (about 40% of the recovered aflatoxin). The fiberfraction accounted for about one-third (30 and 38%) of the total aflatoxin, the germ for 6 and 10%and 1% for the starch. Thus, after wet-milling of aflatoxin-contaminated corn, most of the aflatoxinwill ultimately end in the feed fractions. Thus, feed by-products from the milling industry need to bedecontaminated by others processes such as ammonia treatment (Bennett and Anderson, 1978).In the dry milling process, fumonsins tend to be concentrated in the bran and germ fractions ofcorn (used as animal feed or for oil extraction) and produced grits are relatively free ofcontamination (Katta et al., 1997).Bennett and Richard (1996) found that during wet milling, fumonisin (water-soluble) incontaminated corn was dissolved into the steep water or distributed to the gluten, fiber and germfractions, leaving no detectable amounts in the starch.Using a laboratory stimulated process of wet milling, Collins and Rosen (1981) found that 66% ofT-2 toxin in corn was removed by the steep and process water, 4% was found in the starch, andapproximatively 30% was in the germ.For ochratoxin A, concentration in cleanings, bran and other fractions derived from the seed coatwere the outcomes reported after maize milling (Scudamore, 2005).Dry milling was also an effective method for salvaging zearalenone contaminated corn (Bennett etal., 1976). The starch fraction, the largest and most important for food purposes, was essentially freeof ZEA. However, the germ and bran fractions, which are used as animal feed contained much higherZEA concentrations than the original corn.g) SteepingThe corn steeping process is the first step of the industrial wet-milling of corn, which involvessoaking the corn in water containing SO 2 to facilitate germ separation. Preliminary work by Canela etal. (1996) has shown that steeping in water may reduce the fumonisins content of naturallycontaminated corn, where FB1 migrates from kernels to the steeping water as a result of its high56

INTRODUCTIONpolarity. Pujol et al., (1999) concluded that steeping corn kernels in 0.2% SO 2 solution at 60°C for 6 his very effective to decrease the amount of FB1.Ochratoxin A is distributed differently in the corn steeping process, and the toxin remains in thesteeped corn, and do not reach into the steep water (Wood, 1982).2) Inactivation in contaminated maizea) Heat treatmentMost of mycotoxins are heat-stable within the range of conventional food/feed processingtemperatures (80-121°C), so little or no destruction occurs under normal cooking conditions such asboiling and frying, or even following pasteurization. The sensitivity of mycotoxins to heat treatment isaffected by many different factors including moisture, pH, and ionic strength of food. Degradation byheat treatment also depends on the type of mycotoxin, its concentration, the extent of bindingbetween the mycotoxin and the food constituents, the degree of heat penetration, and theprocessing time (Kabak et al., 2006). A heat treatment which has been widely studied for thedestruction of some naturally occurring food toxins, is extrusion cooking process. The reduction ofthe mycotoxin in the finished products depends on several factors including extruder temperature,screw speed, moisture content of the extrusion mixture, and transit time in the extruder. Of thesefactors, extrusion temperature and transit time seem to have the greatest effect. The highestreductions in mycotoxin concentrations in extruded products occur at temperatures of 160°C orgreater, and when time of transit is increased. During extrusion cooking, the raw material issubjected to high temperatures, high pressure, and severe shear forces. Two major types of extruderare used in the food industry : single-screw and twin-screw extruders.Hameed (1993) showed that extrusion alone was able to reduce aflatoxins content in naturallycontaminatedcorn by 50-80%, and with addition of ammonia, either as hydroxide (0.7 and 1.0%) oras bicarbonate (0.4%) the aflatoxin reduction achieved was greater than 95%. However, Cazzaniga etal., (2001) emphasizes the importance of the extruder design, which could be a significant factoraffecting the residence time and the degree of mixing.In addition to this heat process, roasting is also effective in the detoxification of aflatoxincontaminatedcorn. Reduction in aflatoxin content was higher when roasting was combined withammonia treatement (Conway et al., 1978).Extrusion appears to be an effective method for reducing fumonisins levels in corn (Castelo et al.,1998a; Katta et al., 1999). In the study of Katta (1999), the FB1 losses in corn grits at different57

Table 7 : toxicological evaluation of effective processes on maize decontaminationProcessing methodsPhysical methodsExtrusionMycotoxinsToxicity assessmentBioassayFeedingtrialToxicity outcomes after treatmentsReferencesFUM x in rats – decrease of the kidney lesions Bullerman et al. (2007)DONxloss of toxicity assessed by the MTT assay on CHO-K1 (Chinesehamster ovary) cell lineCetin and Bullerman(2006)ZEAxloss of biological activity assessed by the MTT assay on MCF-7(human breast cancer) cell lineCetin and Bullerman(2005)Chemical methodsAmmoniationAFxxmetabolites less toxic than AFB1 in the Salmonella microsomalassay – assessment of the mutagenic activityabsence of toxicity in some species (rats, ducks, swine, chickens,trouts) – performance, biochemistry & histopathology verifiedPark et al. (1993)Park et al. (1988)FUMxin rats - reduced weight gains, elevated serum enzyme levels andhistopathological lesionsNorred et al. 1991OzonationAFxabsence of toxicity of metabolites in the Hydra attenuata (HA)assay – toxicity endpoint determined by the “tulip” or“disintegration” stage of HAMcKenzie et al. (1997)

xin turkeys - improvement of body weight gain and relative organsweights, no liver discoloration and positive effect on biochemistryMcKenzie et al. (1998)FUMxtoxicity of metabolites in the Hydra attenuata assay and Sa/Soratio still elevatedMcKenzie et al. (1997)OTA x absence of toxicity of metabolites in the Hydra attenuata assay McKenzie et al. (1997)ZEA x uterine weights of mice not affected Lemke et al. (1999)NixtamalizationAF x mutagenic activity higher for extracts of acidified tortillasPrice and Jorgensen(1985)FUMxin rats - hepatotoxicity and nephrotoxicity of nixtamalized culturematerial (biochemistry and histopathology)Voss et al. (1996)Modified nixtamalizationFUMxtoxicity reduction in the brine shrimp assay and no positivemutagenic potential in the Salmonella microsomal assayPark et al. (1995)Sodium bisulfiteDONxin pigs - improvement of feed intake and body weight gain, andno emesisYoung et al. (1987)Reducing sugarsFUMxin pigs - improvement of body weight gain and feed intake,positive effect on biochemistry, no lesion in liver and kidney, anddecrease of Sa/So ratioFernandez-Surumay etal. (2005)Citric acidAFxxmutagenic activity of acidified samples greatly reduced in theSalmonella microsomal assayin ducks – improvement of body weight gain, positive effect onbiochemistry and decline in the severity of lesions in liverMendez-Albores et al.(2005)Mendez-Albores et al.(2007)AF = Aflatoxins; FUM = Fumonisins; DON = Deoxynivalenol; ZEA = Zearalenone; OTA = Ochratoxin A

INTRODUCTIONextrusion parameters ranged from 34 to 95%. Another approach proposed by Castelo et al. (2001),was to evaluate the effect of sugars on the stability of fumonisins during extrusion processing of corngrits. Both the screw speed and glucose concentration significantly affected the extent of FB 1reduction in extruded grits, with greater reductions of FB1 (up to 92.7%) observed at low screwspeeds and high glucose concentrations. However, extrusion may lead to modifications of themycotoxin structure, or interactions with the matrix, that may not be detectable or quantifiable byeither HPLC or ELISA methods. Identification and toxicological assessment of degradation products istherefore necessary. For example, Bullerman et al. (2007) showed for FB1 that extrusion with 10%glucose led to the formation of N-(1-Deoxy-D-Fructos-1-y1) FB1, the main FB1 derivative detected.When the rat kidney was used as a biosensor to detect residual toxicity, the authors observed that N-(1-Deoxy-D-Fructos-1-y1) FB1 was less toxic than unmodified FB1 (see Table 7).Other thermal processes have also been studied in the fumonisins inactivation. Castelo et al.(1998b) showed that roasting corn meal spiked with 5 ppm FB1 at 218°C for 15 min resulted inalmost complete loss of fumonisins. Corn flakes processing in the presence of glucose, gave 86-89%reduction of FB1 after cooking and toasting (Castelo, 1999). When canning was applied to wholekernel corn, a significant decrease in fumonisins content occurred as well (Castelo et al., 1998b).Finally, Scott and Lawrence (1994) reported that heating dry corn meal spiked with FB1 and FB2 at190°C (60 min) and 220°C (25 min) resulted in 60 and 100% loss of both toxins, respectively.Cazzaniga et al. (2001) studied the effect of extrusion cooking on the stability of deoxynivalenol inmaize in the presence or not of additives (sodium metabisulphite). In all conditions applied, whichincluded moisture contents of 15% and 30%, temperatures of 150 and 180°C, and sodiummetabisulphite concentrations of 0 and 1%, the detoxification achieved was greater than 95%.Toxicity assessment of degradation products following extrusion-processed corn grits contaminatedwith DON has been reported (Cetin and Bullerman, 2006; see also Table 7).Extrusion cooking of corn grits resulted in reductions of zearalenone ranging from 77 to 83%, 74to 83% and 66 to 77% at 120 0 C, 140 0 C and 160 0 C, respectively (Ryu et al., 1999). More recently, Cetinand Bullerman (2005) reported in an in vitro study that the extrusion process at temperatures of 150,175, and 200 0 C resulted in a reduction of ZEA ranging from 60 to 78%. In this study, the authorsdeveloped a bioassay in order to determine the toxicity obtained after extrusion processing (seeTable 7).b) Irradiation treatmentIonizing radiation is currently used to eliminate pathogenic microorganisms in foods. Herzallah etal. (2008) reported that the solar radiation was more effective in aflatoxins reduction when58

INTRODUCTIONcompared with γ-irradiation and microwave heating. More than 60% of the aflatoxins were found tobe degraded after 30h of exposure to sunlight. The degree of aflatoxin reduction was found to bedependent on the duration of exposure to sunlight. But solar radiation has potential as aninexpensive means of degrading aflatoxins in tropical countries. Regarding γ-irradiation treatment,40% of aflatoxin content was destroyed and these results were in agreement with Farag et al. (1995),who found that an 83% reduction of aflatoxin after a 20-kGy dose of γ-irradiation of yellow corn wasachieved. On the contrary, Aziz and Youssef (2002) found that the dose of 20 kGy was sufficient tocompletely destroy AFB1 in yellow corn.The effect of γ-irradiation on fumonisin B1 and fumonisin B2 occurring in maize materials hasbeen investigated together with the stability of fumonisins in γ-irradiated maize stored at differenttemperatures (-18 to +40°C) for different periods (2, 4, 13 and 26 weeks). Fifteen KGy γ-irradiationwas required to sterilize efficiently maize flour. This process caused a decrease in fumonisin contentof about 20%. The stability studies showed that fumonisins are stable in γ-irradiated maize for atleast 6 months at 25°C or at least 4 weeks at 40°C (Visconti et al., 1996).Paster et al. (1985) indicated that pure ochratoxin A (OTA) is stable even at 75 kGy but Refai et al.(1996) showed that a dose of 15 or 20 kGy is sufficient for complete destruction of OTA in yellowcorn. Similarly, Aziz and Youssef (2002) reported 75% of detoxification in yellow corn at a radiationdose of 20 kGy.For zearalenone, complete destruction was obtained when a dose of 20 kGY was applied in yellowcorn (Aziz and Youssef, 2002).Other ionizing radiations such as microwave and convection heating have been reported by Young(1986) to be partially successful in lowering deoxynivalenol level in naturally contaminated corn.In conclusion, physical methods outlined in this first part can be divided in two categories withregard to their initial applications. Some methods have been specifically developed for mycotoxinsdecontamination (such as sorting, cleaning or washing); whereas others are industrial processes(such as milling, irradiation, ethanol fermentation or extrusion) initially developed for other purposesthan mycotoxins reduction, that have shown an interest in the fighting strategy against mycotoxins.The first category can be considered as a first step in the removal of mycotoxins from maize. Besidesthe substantial efficiency of these approaches, most of the criteria listed by the FAO guidelines arefulfilled : cheap and simple, no production of toxic metabolites, and no change in the nutritionalvalue and properties of raw materials. Nonetheless, application of these techniques on a commercialscale raises some problems :59

INTRODUCTION- an important loss of grain material has been reported in some cases for sieving and cleaningprocedures. In their study, Trenholm et al. (1991) observed a 73% and 79% reduction for DON andZEA respectively, however in these conditions up to 69% of the total weights of the corn wereremoved as well.- after flotation and washing, the cost of drying grains is a significant problem; unless these methodsare used prior manufacturing processes that require the grain to be wetted or tempered such as wetmilling or alkaline processing of corn.- the duration of grains treatment is also a parameter to take into account.It is also important to stress that contaminated waste (broken kernels, residues, dust…) resultingfrom these decontamination procedures are highly contaminated, thus they must be destroyed andnot used in animal feed.Processes (milling, thermal treatments) from the second category are commonly used in theindustry for food production, intended to the human consumption. Many reports have also showntheir ability to reduce mycotoxins content. Of note, a few methods are already used in feed industry,such as extrusion in pet food manufacturing.The problems met with milling process, concern the toxicity of the different fractions obtainedafter grain separation. Indeed, mycotoxins tend to be concentrated in these fractions (bran andgerm) that are used in animal feed. Accordingly, decontamination procedures need to be applied onthese fractions.Regarding heat treatments, it is important to emphasize that this process may lead to theformation of unknown biologically active mycotoxin degradation products or, to the reversiblebinding of the toxin to sugar or proteins in the food/feed matrix (Humpf and Voss, 2004). Additionalinvestigations using an integrated approach combining chemical studies and appropriate bioassaymethods are needed to identify and characterize these “hidden” mycotoxins.Finally, another point to mention is the blend of contaminated crops with batches of good-qualitymaterial. Despite the effectiveness of this method in decreasing mycotoxins concentration, thepractice is prohibited within the European Union (Commission Regulation, 2001).60

INTRODUCTIONII. CHEMICAL METHODS1) AmmoniationTreatment with ammonia in the gaseous phase, or with substances capable of releasing it,achieved optimum results in detoxifying corn meals, especially for aflatoxins reduction. It has beenwidely demonstrated that the efficacy of mycotoxins detoxification with ammonia is positivelycorrelated with the quantity used, reaction time, temperature and pressure levels. Since few data onammonia treatment have been reported for other mycotoxins, the section on aflatoxindecontamination is presented in more details.First outcomes on the chemical inactivation of aflatoxin by ammoniation were reported forcottonseed and peanut meals (Dollear et al., 1969). Nowadays, it is an approved procedure and amethod of choice for the detoxification of aflatoxin-contaminated agricultural commodities andfeeds, in some of the North American states (Arizona, California, Georgia, Alabama), in France, inSenegal, in Sudan, in Brazil, in Mexico and in South Africa.While at least five types of ammoniation processes have been described or patented, the twoprocedures considered most practical are a high temperature/high pressure process used bycommercial treatment plants and an atmospheric pressure/ambient temperature process for in situfarm usage. Both procedures have the advantages of removing up to 90% of aflatoxin contaminationin the matrix.Aflatoxin B1 degradation by ammonia proceeds through hydrolysis of the lactone ring followed bydecarboxylation to aflatoxin D1 and loss of the cyclopentone ring to give a compound of a molecularweight of 206 daltons. It was also observed that the AFB1 molecular structure is irreversibly altered ifexposure to ammonia lasts long enough. In contrast, if exposure is not sufficiently protracted, themolecule can revert to its original state. To note that the addition of formaldehyde in the process,allows to reduce the use of ammonia, to break more easily the lactone ring and to avoid thereversibility of the reaction.The studies concerning the safety of ammoniation process began in 1983, where no signs oftoxicosis occurred in rats fed ammoniated aflatoxin-contaminated corn, in contrast to rats fed withuntreated aflatoxin-contaminated corn (Norred and Morrissey, 1983). A comprehensive review onfeeding trials carried out to evaluate the safety of ammoniated feed, was published by Park et al.(1988; see Table 7). In addition, toxicity assessment of metabolites obtained after aflatoxinprocessing was reported (Park, 1993; see also Table 7). More recently, an experiment on broilerchicks suggests that replacement of aflatoxin-containing maize with ammoniated grains significantlysuppress aflatoxicosis, leading to an improvement in production parameters in broilers (Allameh et61

INTRODUCTIONal., 2005). The fate of ammonia products in lactating cows was also investigated, and several studiesconcluded that the efficient reduction of aflatoxin levels by ammoniation, resulted in a strongreduction of aflatoxin-related residues in milk (Fremy et al., 1988; Hoogenboom et al., 2001).Fumonisin B1 is also degraded by ammoniation with a high pressure/ambient temperature oratmospheric pressure/high temperature. Ammonia treatment resulted in 79% of FB1 destruction oncontaminated maize (Park et al., 1992). However, Norred et al. (1991) found that ammoniationprocedure on fumonisins content of corn did not reduce the toxicity of the materials when fed to rats(see Table 7).Exposing corn to ammonia (100%) for at least 18h resulted also in substantial reductions (85%) indeoxynivalenol levels (Young, 1986). Toxicological studies are needed to evaluate the safety ofammoniated DON-contaminated corn.In barley, ochratoxin A was nearly completely degraded after ammonia treatment. However, instudies in which the ammoniated ochratoxin-contaminated barley was fed to pigs, some toxicity andlower nutritional value was observed (Madsen et al., 1983). Residues of OTA were also found in thekidneys, and it was thought to be reformed in the animals by recyclization.In 1981, Chelkowski et al. reported that addition of ammonia to corn in a final concentration of2% inactivates partially zearalenone at temperatures of 20-50 0 C.2) OzonationOzone is a powerful oxidizing agent capable of reaction with numerous chemical groups, though ithas an affinity for double bonds. Its industrial application as a disinfectant is already known andcurrently used. In addition, ozone decomposes to form oxygen and therefore, can be classified as anonpersistent chemical. Thus, several research studies have been undertaken to evaluate the effectof ozone in reducing mycotoxins in agricultural products. Ozone has a high potential for degradingaflatoxins. It reacts across the 8-9- double bond of the furan ring through electrophilic attack. In1997, McKenzie developed a novel and continuous source of O 3 gas through electrolysis. Totaldegradation of AFB1 in solution was obtained at 15 sec using 20 weight% ozone, and the same year,the author claimed using O 3 generated by electrolysis, the processing and the decontamination ofbulk quantities of corn that was contaminated with high levels of AFB1. Furthermore, the toxicityresulting from processing of aflatoxin by ozonation, was evaluated in a bioassay (see Table 7). In asimilar study, McKenzie et al. (1998) found that aflatoxins could be reduced by 95% in corn treatedwith 14 wt% ozone for 92h at a flow rate of 200 mg/min. Turkeys fed with ozone-treatedcontaminated corn did not show harmful effects as compared to animals fed with untreatedcontaminated corn (see Table 7). Similarly, phagocytosis by rat peritoneal macrophages, which was62

INTRODUCTIONfound to be suppressed in the presence of aflatoxin, remained unimpaired when the applied AFB1was pretreated with 1.2 mg/L ozone for 6 min at a flow rate of 40 mL/min (Chatterjee andMukhergee, 1993). However, the ozonation process may cause formation of fat-soluble reactionproducts with low mutagenic potential (Prudente and King, 2002).McKenzie et al. (1997) found that treating an aqueous solution of fumonisin B1 with 10 wt%ozone gas for 15 sec resulted in the conversion of the parent compound to the 3-keto FB1 derivative.In two separate toxicity tests, 3k-FB1 was found to retain most of the toxicity present in the parentcompound (see Table 7). Therefore, a further degradation of FB1 to a product less active than 3k-FB1is needed (i.e. removal of primary amine).Moisture appears to be essential in the reaction between deoxynivalenol and ozone. Moist ozone(1.1 mol %) in air resulted in a 90% reduction in DON-contaminated corn, at 1000 ppm, after 1h,while dry ozone resulted only in a 70% reduction (Young, 1986). For DON treatment with ozone,differences also occur between wheat and corn. A matrix effect was suggested (Young et al., 2006),the ground corn being more porous, whereas ozone may not be able to penetrate the whole wheatkernels as readily. Young et al. (2006) reported the degradation of several trichothecenes mycotoxinsby aqueous ozone and on the basis of UV and MS data, proposed that the degradation begins withattack of ozone at the C9-10 double bond with the net addition of two atoms of oxygen. However,toxicity of the breakdown products and efficiency of this technology in maize need to be evaluated.Ozone treatment induces a rapid degradation of zearalenone level in water solutions containing12 ppm of ZEA (Lemke et al., 1999), and prevents its estrogenic effects (see Table 7). In an in vitrostudy a reduction to undetectable levels was observed after ozone treatment for 15 s in ZEA solution.No new products were observed after the treatment (McKenzie et al., 1997).The same study showed treatment of aqueous solution of ochratoxin A with 10 wt% O 3 for 15seconds reduced the mycotoxin level to undetectable levels, and a bioassay revealed a decrease inthe toxicity (see Table 7).Using ozone to purify commodities, the commercial Oxygreen ® process claims the destruction ofmycotoxins found on grain by bringing them to levels tolerated by legislation. This procedure lead toa 94% reduction of OTA in grains. Toxicity of this process was evaluated by a four-week study in ratsby dietary administration of treated wheat, and the treated products were considered as safe for theconsumer (Gaou et al., 2005).3) NixtamalizationNixtamalization, a Mexican traditional process for making tortillas, consists of the cooking ofmaize grain in abundant water and lime (calcium hydroxide), at temperatures near boiling, followed63

INTRODUCTIONby a steeping period. Several studies reported a significant reduction in aflatoxins content (90-95%)in corn with this alkaline treatment. However, the apparent aflatoxin reduction in contaminatedmaize was not permanent, since it was reverted by an acidic treatment, which probably causedaflatoxin reformation by closing the open lactone ring (Mendez-Albores et al., 2004; Price andJorgensen, 1985). Acidification of the aflatoxin extracts, with a pH similar to the stomach as occursduring digestion, lead to a rebuilding of the aflatoxin molecule (see Table 7).Under alkaline conditions, fumonisins in contaminated corn are converted to the so-calledhydrolyzed fumonisins (HFB1). Dombrink-Kurtzman et al. (2000) showed that nixtamalizationreduced the FB1 concentration in tortillas by 81.5% and that the FB1 and HFB1 were mainly found inthe steeping and washing water. Cortez-Rocha et al. (2002) observed a 39% reduction of FB1concentration when raw corn was nixtamalized. Others found that the traditional nixtamalizationmethod used by the Mayan communities in Guatamala reduced total fumonisins (FB1 and HFB1) by50% and that the residual lime and washing water also contained 50% of the fumonisins initiallypresent in the corn (Palencia et al., 2003). However, when fed to rats, Voss et al. (1996) showed thatnixtamalized corn culture materials that contained HFB1, still induce toxicity (see Table 7). Thistoxicity was originally attributed to the hydrolyzed fumonisins, but the same author in 2009disclaimed the HFB1-induced toxicity, suggesting these effects were mediated by residual or“hidden” FB1 (matrix bound forms not detected by HPLC) remaining in the nixtamalizedpreparations, rather than HFB1.Park et al. (1995) observed that further decontamination and detoxification of FB1 contaminatedcorn was achieved by using a modified nixtamalization procedure. This procedure consisted in acombination of heat treatment with hydrogen peroxide/sodium bicarbonate with calcium hydroxideand gave up to 100% reduction of FB1 in contaminated maize (100 ppm). The best procedureinvolved treating corn with a combination of H 2 O 2 /NaHCO 3 alone for one hour. This is a simpletreatment that could be integrated into industrial processing procedures for corn. Furthermore,treatments of FB1-contaminated corn simulating modified nixtamalization have shown a reduction oftoxicity (see Table 7).Abbas et al. (1988) indicated that boiling deoxynivalenol-contaminated corn in lime-waterremoved 72-82% of DON and 100% of 15-acetyl-DON. In the same study, the authors observed areduction of 59-100% for zearalenone-contaminated corn. No data on the safety of treated feed,contaminated with either DON or ZEA, has been reported.64

INTRODUCTION4) Hydrogen peroxide/sodium bicarbonateAs mentioned above, the addition of oxidizing agents, such as hydrogen peroxide, is an effectivehelp in nixtamalization. Some studies have shown that hydrogen peroxide/sodium bicarbonate aloneis effective for simultaneous degradation/detoxification of aflatoxins and fumonisins, therebyreducing toxicity (Lopez-Garcia et al., 1999).Abd Alla (1997) investigated the efficiency of hydrogen peroxide for detoxification ofcontaminated corn with zearalenone. A degradation level of 84% was achieved with a treatmentwith 10% H 2 O 2 for 16 hours at 80 0 C.5) Sodium bisulfiteThe use of sodium bisulfite as a detoxification reagent is worth special mention as it is a commonfood additive. In maize, sodium bisulfite treatments used at 0.5% and 2% destroy 80% and 90% ofaflatoxin B1 respectively (Doyle et al., 1982). The main product of reaction is a sulfonate, AFB1-S,formed by addition of bisulfite to the furan ring present in AFB1 and AFG1, but not in AFB2 and AFG2.Piva et al. (1995) considered that although this process is less efficient than ammoniation, itovercomes some of the typical disadvantages of ammonia method (less hazardous to handle for thestaff, no undesirable brown color of treated feed or no impact on nutrient value) and is muchcheaper.Sodium bisulfite also destroys deoxynivalenol in maize. Indeed, Young et al. (1987) reported thatDON in contaminated maize was reduced by 85% through treatment with sodium bisulfite solutions.The greatest reductions (up to 95%) were achieved when the contaminated corn was autoclaved for1h at 121°C in the presence of 8.33% aqueous sodium bisulfite. In the same study, the authorsperformed two toxicological trials in order to assess the safety of this process; the first one consistedto a 7 days feeding trial in pigs with treated-contaminated corn (7.2 ppm DON), and the second onewith DON-sulfonate (the by-product obtained after treatment) which was administered orally toswine at the same level (molar equivalent) at which nonderivatized DON caused severe emesis (seeTable 7). However, some authors indicated that DON-sulfonate is stable in acid but hydrolyzed toDON under alkaline conditions.6) Reducing sugarsThe primary amine group in the fumonisin B1 molecule is responsible for its toxicity (Fernandez-Surumay et al., 2004). Transformation of FB1 into a FB1-reducing sugar adduct through anonenzymatic browning reaction (65 °C for 48 h) reduces significantly the toxicity of the native FB165

INTRODUCTIONon cell tissue cultures, on mice, on rats and on swine (Lu et al., 1997; Liu et al., 2001; Fernandez-Surumay et al., 2004). The products of FB1-reducing sugar reaction were characterized as N-(carboxymethyl) FB1, N-(1-deoxy-d-fructos-1-yl) FB1, N-methyl-FB1, N-(3-hydroxyacetonyl) FB1, andN-(2-hydroxy-2-carboxyethyl) FB1. In 2005, Fernandez-Surumay et al. examined the effects offumonisin B-glucose reaction products in swine fed with treated maize (see Table 7). The authorsuggested that dietary FB1-glucose might provide a detoxification approach in instances ofwidespread fumonisins grain contamination.7) Citric acidUp to 96% degradation of aflatoxin B1 occurred in maize contaminated with 93 ppb when treatedwith an aqueous citric acid (Mendez-Albores et al., 2005). In addition, they evaluated the toxicity ofacidified samples obtained in a bioassay (see Table 7). These results show the efficacy and safety ofthe acidification procedure in reducing aflatoxins levels in maize. The author suggested also that forwhole grain, it is likely that acid treatment may be less effective than for ground maize, since toxinsdeposited inside whole kernels are less likely to be exposed to acidic treatment than toxins in smallmaize particles. Toxicity of untreated aflatoxin-contaminated feed was reduced in ducks whenaqueous citric acid was applied to feed (Mendez-Albores et al., 2007; see Table 7).8) Other chemical treatmentsA complete destruction of deoxynivalenol was reported in corn contaminated with 1000 ppm ofDON, treated with 30% chlorine (v/v) for 30 min (Young, 1986). The author suggested that thetreatment might be too drastic for grains destined for human consumption but could be used inanimal feed production.Calcium hydroxide monomethylamine has been used successfully to decontaminate T-2 and HT-2toxin in maize (Bauer, 1994).Butylated hydroxytoluene (BHT) treatment significantly reduced the hepatocellular necrosis,biliary hyperplasia, and elevated serum enzymes commonly caused by aflatoxin B1 in turkeys. Thisdiminished pathology demonstrates a physiologic protective effect (Coulombe et al., 2005). Theseresults were supported by the fact that dietary BHT can cause significant reductions in AFB1bioavailability, AFB1-DNA adduct formation in the liver, and reduced AFB1 residues in tissues(Guarisco et al., 2008).66

INTRODUCTIONIn conclusion, although the chemical detoxification methods could greatly reduce mycotoxincontamination, they do not seem able to fulfill all the FAO requirements; thus limiting theirwidespread use. Critical points remains :- identification of mycotoxin degradation/transformation products with suitable methods ofdetection. For example, problems to detect the presence of “hidden” mycotoxins, such as fornixtamalization of FB1-contaminated feed (Voss et al., 2009).- bioassay and toxicological studies in target species to assess the safety of the treatment.Importance to evaluate the fate of processed contaminated food/feed after ingestion in animals. Forexample, mycotoxin reversion into its native toxic form may occur, as revealed after nixtamalizationof aflatoxin-contaminated feed (Price and Jorgensen, 1985).- impact on the nutrients content or on the fate of nutrients after processing. For example, reductionin the content of some amino acids (cystine, methionine and especially lysine) and increase in totalnitrogen and non protein nitrogen, are some of the effects reported for ammoniation (Piva et al.,1995).In addition, the handling of chemical products implies risks for the workers. It should be notedthat chemical treatment is not allowed within the European Community for commodities destinedfor human food.Finally, an approval procedure should also act efficiently on several types of mycotoxinssimultaneously, without eliciting residual toxicity. According to criteria listed above and to studiespresented in this chapter, ozonation and modified nixtamalization seem to be the most suitable.67

INTRODUCTIONCONCLUSIONDetoxification efforts for mycotoxins have focused primarily on the aflatoxins; many strategies todecontaminate aflatoxin-contaminated crops and products have been reported and reviewed. Bycontrast, informations on methods detoxifying other mycotoxins are limited. Multi-contaminationcommonly occurs in commodities and as the impact of these hazardous toxins is being recognized,successfully removing them from the food/feed supply represents an emerging area of interest andresearch. The effectiveness of a method in the detoxification of mycotoxins depends on the nature ofthe food/feed, the environmental conditions such as moisture content, temperature, as well as thetype of mycotoxin, its concentration and the extent of binding between mycotoxin and constituents.Although a number of effective processes have been developed, no single physical or chemicalmethod can remove simultaneously all mycotoxins from feed.Other approaches based on mycotoxins adsorption and biotransformation, which act directly inthe gastrointestinal tract of animals, have been reported. Consequently, when prevention ofmycotoxins contamination prior to harvest or during post-harvest and storage is unavoidable, anintegrated multipronged approach seems to be the most suitable strategy for controlling mycotoxinscontamination in feed : physical and/or chemical methods to lower the mycotoxins content in rawmaterials, and adsorption and/or biotransformation to reduce the bioavailability of mycotoxinsduring the digestive process of animals.To conclude, mycotoxin contamination of food/feed will remain a global problem in the future,especially with the global warming. The decontaminated or detoxified crops are mostly considered tobe a lower quality, and thus fetch a lower price than a normal uncontaminated crop. This is a seriousconstraint on the practical use of decontamination and detoxification methods, and the processedcrops are mainly used for feed production and animal feeding. Therefore, a partial solution tomycotoxins would be to use contaminated crops in more cases for purposes other than direct foodand feed, such as biofuel industry.68

INTRODUCTION3. Procédés de décontamination des denrées contaminées et réduction desmycotoxines chez l’animal3.2. Adsorption et détoxification biologiqueLe deuxième chapitre de cet ouvrage présente des méthodes alternatives mais complémentairesà celles décrites dans le chapitre précédent, et à la différence, consistent en une action directe dansl’organisme de l’animal. Ces approches plus récentes sont en plein essor, et nous pouvons distinguerdans ces catégories d’additifs, les adsorbants et les micro-organismes/enzymes.Les adsorbants sont des molécules permettant de séquestrer les mycotoxines et ainsi d’éviterl’absorption intestinale des toxines. Le complexe adsorbant-toxine est finalement excrété via lesfèces de l’animal. On distingue deux classes d’adsorbants, inorganiques et organiques. La premièreclasse concerne essentiellement les argiles, tels que les aluminosilicates de sodium ou les bentonites,qui sont largement utilisés contre les aflatoxines. Les ligands organiques sont plus nombreux, telsque le charbon activé, les polymères ou les parois de levures et de bactéries.La détoxification biologique par des micro-organismes entiers ou des enzymes purifiées reposesur une dégradation par voie enzymatique des mycotoxines dans le système digestif de l’animal, etaboutit à des métabolites moins ou non toxiques. De nombreux micro-organismes (incluant leslevures, bactéries et champignons) ont été isolés à partir du sol, de compost, de rumen, ou encored’insectes, et ont été screenés pour leur capacité à modifier ou inactiver les mycotoxines.69

INTRODUCTIONMycotoxin Reduction in Animals – Adsorption and biological detoxificationKatia Pedrosa 1 , Dian Schatzmayr 1 , Didier Jans 2 , Gérard Bertin 3 & Bertrand Grenier 1,41BIOMIN Research Center, Technopark 1, Tulln, Austria.2FEFANA (EU Feed Additives and Premixtures Association), Belgium.3Alltech-France, European Regulatory Department, Levallois-Perret, France.4INRA, ToxAlim, Equipe Immuno-Mycotoxicologie, Toulouse, France.70

INTRODUCTIONABSTRACTMycotoxins are toxic chemical products formed by fungal species that colonize crops in the fieldor after harvest and thus pose a potential threat to human and animal health. Even thoughrecommended agricultural practices have been implemented to decrease production of mycotoxinsduring crop growth, harvesting and storage, the potential for significant contamination still exists.The significance of these unavoidable, naturally occurring toxicants to human and animal health, theincrease in mycotoxin regulations and global trans-shipment of agricultural commodities highlightthe need to provide successful counteracting strategies. The approaches to this are varied and maybe categorized as physical methods, chemical decontamination, treatments with adsorbents (organicor inorganic) and biological detoxification to decrease the bioavailability of mycotoxins to the hostanimal.Certain treatments have been found to reduce levels of specific mycotoxins, however, no singlemethod has been developed that is equally effective against the wide variety of mycotoxins whichmay co-occur in different commodities.71

INTRODUCTIONINTRODUCTIONMycotoxin-producing fungi grow on staple raw materials used for animal feeds and human foods.Subsequently, mycotoxins must be considered persistent and common contaminants of feed andfoods that world-wide affect both, animal and human health (Adams, 2004; Galvano et al., 2005).Extreme vigilance and careful management is vital to ensure that levels of mycotoxins are keptacceptably low (Adams, 2001).The complex processes involved in mycotoxin production by moulds make it difficult to predictwhich toxin will be produced, when and in what concentration.Unfortunately it is not possible to entirely prevent the production of mycotoxins before harvest ofagricultural crops, in storage, or during processing operations. Thus, numerous strategies areevolving for the management and field-practical control of mycotoxins. Some are clearly morepractical and effective than others.A variety of physical, chemical and biological approaches to counteract the mycotoxin problemhave been reported, but large-scale, practical, and cost-effective methods for a completedetoxification of mycotoxin-containing feedstuffs are currently not available. Substances used tosuppress or reduce the absorption of mycotoxins by the animal should either promote the excretionof mycotoxins in feces (in case of binders) or modify their mode of action (in case of biologicaldetoxifiers).One of the most promising approaches to the problem has been the use of sorbents in the dietthat sequester mycotoxins and reduce the absorption of these toxins from the gastrointestinal tract,avoiding the toxic effects for livestock and the carryover of these fungal metabolites into animalproducts. For an adsorbent to successfully prevent the absorption of mycotoxins from thegastrointestinal tract, it should have a high binding capacity and affinity for the mycotoxin, avoiddissociation and thus reduce the bioavailability. Adsorbents that may appear effective in vitro do notnecessarily retain their efficacy and safety when tested in vivo.Besides, biological detoxification has become stronger and more important during the last decadeand it is nowadays an emerging area of interest and research in animal feed production. Isolation andcharacterization of microorganisms or enzymes that are able to biotransform mycotoxins canpossibly be the breakthrough for the practical application of biotechnology in specificdecontamination processes taking place directly in the intestinal tract of animals. This biologicaldecontamination may become a technology of choice, as enzymatic reactions offer a specific,efficient and environmentally friendly way of detoxification.72

INTRODUCTIONIndependent of the kind of product one of the main targets is to evaluate efficacy in multicontaminationconditions which is closer to reality.Thus, this chapter comments on the main binders used in mycotoxin adsorption, their in vitroefficacy and in vivo applications in a range of animals. Moreover, the biological detoxification isdescribed and presented as a promising strategy in counteracting mycotoxins that cannot beinhibited with inorganic and organic adsorbents.73

INTRODUCTIONI. ADSORPTION OF MYCOTOXINSThe ever-increasing number of reports on the presence of mycotoxins in foods and feed dictatesthe necessity for practical and economical deactivation procedures.The high costs and limitations of physical and chemical treatments prompted a search for othersolutions to the mycotoxin hazard. Up to now, the most widely investigated method in this field isthe addition of nutritionally inert sorbents with the capacity to tightly bind and immobilizemycotoxins in the gastrointestinal tract of animals, resulting in a major reduction in toxin bioavailability.1) Use of inorganic adsorbentsIn several scientific studies clay and zeolitic minerals, hydrated sodium calcium aluminosilicates(HSCAS) and other similar montmorillonite and bentonite clays have proven to be the most promisingadsorbents for aflatoxins (AF) (Grant and Phillips, 1998; Ramos and Hernandez, 1996; Phillips et al.,1988, 1994; Vekiru et al., 2007). Mixed into feed they markedly diminished AF uptake by the bloodand distribution to target organs, thus avoiding aflatoxin-related diseases and the carryover ofaflatoxins into animal products such as milk (Phillips et al., 1990 and 1991). These binders have theproperty of adsorbing organic substances either on their external surfaces or within theirinterlaminar spaces, by the interaction with or substitution of the exchange cations present in thesespaces. However, clay and zeolitic minerals, which comprise a broad family of diversealuminosilicates, are not produced equally thus do not possess the same physical properties.While good and scientifically explained results were obtained in counteracting aflatoxins,adsorption of other mycotoxins (e.g. zearalenone, ochratoxin A, fumonisins) was limited or evenfailed under field conditions (e.g. trichothecenes such as deoxynivalenol, T-2 toxin) (Friend et al.,1984; Kubena et al., 1990; Huff et al., 1992; Kubena et al., 1993; Ramos et al., 1996).a) Hydrated sodium calcium aluminosilicate (HSCAS)HSCAS is a phyllosilicate clay. Its discovery as an effective enterosorbent of aflatoxins, its chemicalcomposition and its mechanism of aflatoxin sorption at interlayer surfaces is described in numerousscientific publications by the T.D. Phillips group. The stability of the aflatoxin-HSCAS complexes mayexplain the in vivo effectiveness of the adsorbent in preventing the toxic effects of aflatoxins inseveral animal species (Pettersson, 2004).Since early studies, HSCAS has been reported to diminish the effects of aflatoxins in a variety ofyoung animals including rodents, chicks, broilers, turkey poults, ducklings, lambs, pigs, minks and74

Table 8 : Outcome of HSCAS effects in AFB 1 contaminated diets in different speciesAnimalspeciesChickensTurkeypoultsAFB 1concentrationHSCASconcentration1000 ppb 0.5%4000 ppb 1%7500 ppb 2.5 g/kg7500 ppb 0.5%2500-5000 ppb 0.5%5000 ppb 0.375%500-1000 ppb 0.5%Pigs 840 ppb 0.5%Outcome of HSCAS effectsPerformance: improvement of feed intake and body weightOrgans: reduction of toxic effects on liver weight, no cell damage and decline inthe severity of lesions in liverSystemic response: positive effects on serum chemical parametersPerformance: improvement of feed intake and body weightOrgans: protection against changes in organ weights (liver, kidney, proventriculusand pancreas) and decline in the severity of lesionsSystemic response: positive effects on serum chemical parametersPerformance: improvement of feed intake and body weightOrgans: reduction of toxic effects on liver weight and decline in the severity oflesions in liverSystemic response: no dataPerformance: decrease the growth inhibitory effectOrgans: protective effects on gross hepatic changesSystemic response: no dataPerformance: decrease the growth inhibitory effectOrgans: reduction of toxic effects on liver, kidney and proventriculus weightsSystemic response: protection against changes in serum biochemical parametersPerformance: improvement in body weight gainOrgans: reduction of toxic effects on liver, kidney, pancreas and proventriculusweightsSystemic response: protection against changes in serum biochemical parametersPerformance: 68% decrease in mortality, improvement in body weight gainOrgans: protection against changes in organ weightsSystemic response: improvement in hematological and biochemical parametersPerformance: decrease the growth inhibitory effectOrgans: improvement in liver functionsReferencesGowda etal. 2008Ledoux etal. 1998Dixon et al.2008Phillips etal. 1988Kubena etal. 1993Kubena etal. 1998Kubena etal. 1991Lindemannet al. 1993

Rats3000 ppb 0.5 %2500 ppb 0.5 %2500 ppb 0.5%Lambs 2600 ppb 2.0%Minks34-102 ppb(AF B 1 + B 2 )0.5%Systemic response: restored cell-mediated responsivenessPerformance: improvement in body weightOrgans: no dataSystemic response: positive effect on lymphocytes proliferation, on macrophageactivity and functionPerformance: no dataOrgans: any microscopic lesions observed in liver and in kidneySystemic response: improvement in hematological and biochemical parametersPerformance: improvement of feed intake and body weight, ability of reproductionwarrantedOrgans: no dataSystemic response: positive effects on serum chemical parametersPerformance: improvement in body weightOrgans: no dataSystemic response: positive effects on serum biochemical parametersPerformance: prevention of mortality, improvement of feed intake and body weightOrgans: decline in the severity of lesions in liverSystemic response: no dataHarvey etal. 1994Abdel-Wahhab etal. 2002Abdel-Wahhab etal.1999Harvey etal. 1991Bonna et al.1991Chickens included broiler chicks, laying hens.Rats included male and female.Pigs included piglets, sow, barrow.

INTRODUCTIONtrouts (Table 8). Levels of aflatoxin M 1 in milk from lactating dairy cattle and goats decreased withHSCAS clay presence (CAST, 2003) and, also, when present at 0.5% in the diet it significantly reducedaflatoxins (M 1 + B 1 + B 2 ) in liver, kidney, muscle and fat (Beaver et al., 1990).Several authors found limited effects of HSCAS against zearalenone and ochratoxin A and provedeven their ineffectiveness against trichothecenes (Huwig et al., 2001; Satin et al., 2002a; Santin et al.,2002b; CAST, 2003; Pettersson, 2004). This can be explained by the HSCAS-mycotoxin-bindingmechanism: a β-keto-lactone or bilactone system, which no other mycotoxins but aflatoxins have, isthought to be essential for chemisorption to HSCAS. Results have shown that ochratoxin A (OTA) inthe diet impaired the productivity indexes (heavier kidneys, enlarged livers) and that HSCAS did notimprove these parameters (Santin et al., 2002b). Additionally, according to Bursian et al. (1992)HSCAS did not significantly alter the hyperestrogenic effects of zearalenone (ZEA). The increase ingestation length, decrease in litter size and increase in kit mortality of mink were some of the effectsreported.Adding HSCAS at either 0.5 or 1.0% did not positively influence the average daily gain of pigsexposed to deoxynivalenol (DON) (CAST, 2003).b) Bentonite (montmorillonite)Bentonites are clay minerals which result from the decomposition of volcanic ash consistingmainly of phyllosilicate minerals belonging to the smectite group; in 70% of the cases this smectite ismontmorillonite.Promising in vitro results were obtained with bentonites concerning effective adsorption ofaflatoxins, fumonisins and ergot alkaloids (ergin, ergotamine, ergovalin). Vekiru et al. (2007) provedtheir high aflatoxins adsorption capacity and affinity. Bentonites were also proven to be able toadsorb fumonisins at an extent of 90% in a complex media such as gastrointestinal fluid (CAST, 2003).According to Lindemann et al. (1993) and Schell et al. (1993b), an inclusion rate of 0.5% sodium orcalcium bentonite seemed to be beneficial in restoring growth performance and serum profiles ofgrowing pigs fed diets containing 800 ppb of aflatoxin B 1 . Further feeding trials suggested thatbentonites and thus montmorillonite can effectively reduce the toxicity of aflatoxins in broiler chicks,poultry and pigs, improve growth performance, average daily weight gain, average daily feed intake,feed conversion ratio and reduce the negative immunosuppressive effect (Ibrahim et al., 2000; Rosaet al. 2001; Gupta and Gardner, 2005; Shi et al., 2006; Yu et al., 2008). The addition ofmontmorillonite in aflatoxin-contaminated diets diminished the adverse effects on relative organweights, hematological, serum and liver biochemical values as well as enzymatic activities associatedwith aflatoxicosis (Shi et al., 2006).75

INTRODUCTIONIn studies of the potential of bentonites to reduce the carry-over of aflatoxins into milk of dairycows variable reductions have been observed (Veldman, 1992). Petterson (2004) thus concluded thatit will be hazardous to trust the efficiency of these adsorbents at the low EU tolerance limits foraflatoxins in feed and milk. However, significant reduction of milk aflatoxin M 1 content was verifiedby Pietri et al. (2009) using a blend of bentonites. Regarding trichothecenes, in 1983 Carson andSmith observed general improvements in performance parameters and an increase in the toxinexcretion in rats fed a diet with T-2 toxin and bentonite incorporation. However, these results couldonly be achieved when the incorporation level in the diet was about 10-20 times higher (10%) thanthe efficient level for aflatoxins.Contradictory, bentonites have shown low or no effect against zearalenone and although theymay be useful in situations where moulds cause non-specific unpalatability, bentonites appearineffective where feed refusal is due to direct effects on appetite as appears to be the case withtrichothecenes such as nivalenol (Williams et al. 1994; Guerre, 2000).c) ZeoliteZeolites, other aluminosilicate clays, are similar to molecular sieves as well as to ion exchangeresins and are suitable for the distinction of different molecules by size, shape and charge.Depending on dosage, physical structure, type of zeolite and respective concentration of aflatoxins inthe contaminated feed, the ability of zeolites in reducing their toxicity can differ widely.Zeolites have been studied as adsorbents for aflatoxins. They normally have less capacity thanHSCAS or bentonites to adsorb aflatoxins in vitro, but showed some in vivo efficacy under practicalconditions (Pettersson, 2004). Natural zeolite (1% in the diet) given to broilers consuming 2.5 ppmAFB 1 has alleviated the growth depression and reduced the increase in liver lipid concentrationcaused by aflatoxins (Scheideler, 1993). Indeed, several studies with natural zeolite, the clinoptilolite,which is frequently added to animal feed, confirmed improvement on performance, decline of liverand kidney lesions, heterophilia and lymphopenia reduction and improvements on humoralimmunity (Sova et al., 1991, Parlat et al., 1999; Oguz and Kurtoglu, 2000; Oguz et al., 2000; Oguz etal., 2003; Ortatatli et al., 2001; Ortatatli et al., 2005).However, other studies reported contradictory results being ineffective and causing organ lesions(Mayrua et al., 1998; Harvey et al., 1993). Incorporation of a synthetic anion-exchange zeolite into ratdiets at 5% completely eliminated the deleterious effect of zearalenone on body weight, feedconsumption and feed efficiency. In opposite, the use of a synthetic cation-exchange zeolite was notable to protect rats against ZEA toxicosis (Smith, 1980), presenting higher uterus weight in piglets fed76

INTRODUCTIONdiets containing ZEA and zeolite than in piglets fed diets with ZEA and without the zeolite (Guerre,2000; Coenen and Boyens, 2001).This observation supported the hypothesis that an anion-exchange zeolite, which together withzearalenone is anionic at intestinal pH, treated zearalenone toxicosis by reducing absorption of thetoxin and increasing excretion of zearalenone in feces (Ramos, 1996a).d) Other claysTo some extent other clays such as diatomaceous earth have also been studied for their ability tobind mycotoxins in vitro and in vivo.Some commercial products have been reported to have positive effects against aflatoxins inbroilers (Denli et al., 2009) and ochratoxin A in hens leading to an increase in egg albumen height,redness color of egg yolk and serum calcium concentration (Denli et al., 2008).e) Nutrient interactions in animalsThe adsorbents’ activity has raised many questions regarding their influence on the utilization ofnutrients such as carbohydrates, proteins, vitamins and minerals and controversy results have beenreported.No nutrient interactions have been reported with HSCAS at the 0.5% level in the diet (Phillips,1999); it did not impair phytate or inorganic phosphorous utilization. In other studies, the addition ofHSCAS to basal diets at concentrations of 0.5% or 1% did not impair the utilization of riboflavin,vitamin A or manganese; however there was a slight but significant reduction in zinc utilization in thepresence of 1% clay (Chung et al., 1990). According to Schell et al. (1993a) significant interactions formany minerals indicated that the effects on mineral metabolism were more pronounced whenaflatoxin-contaminated corn was fed. Feeding bentonite with aflatoxin-contaminated corn resultedin partial restoration of performance and liver function without greatly influencing mineralmetabolism (Schell et al., 1993a). Nonetheless, zeolites seemed to have a significantly negative effecton vitamin and mineral uptake and on their distribution in the body of sows (Papaioannou et al.,2002). Thus, one of the most important criteria to take into account when evaluating a potentialbinder is the absence of affinity for vitamins, minerals or other nutrients (Huwig et al., 2001).2) Use of organic adsorbents77

INTRODUCTIONSubstances investigated as potential organic mycotoxin-binding agents include among othersactivated charcoal, alfalfa, canola oil bleaching clays, organic polymers such as cholestyramin,polyvinylpolypyrrolidone (PVPP), yeast cell walls and components thereof as well as bacterial cells.a) Activated charcoalActivated charcoal is known as one of the most effective and non-toxic, but also mostunspecifically acting group of sorbents with a high surface to mass ratio (500-3500 m 2 /g). They areformed by pyrolysis of several organic compounds. They can efficiently adsorb most of themycotoxins in aqueous solution, whereas different activated charcoals have less or even no effectsagainst mycotoxicosis (Kolossova et al., 2009).Activated charcoal has been investigated for its ability to adsorb mycotoxins both, in vitro and invivo, being verified to adsorb zearalenone, deoxynivalenol and nivalenol in vitro, using agastrointestinal model, and aflatoxins, ochratoxin A and fumonisins rather efficiently in vivo (CAST,2003, Pettersson, 2004; Kolossova et al., 2009).In vitro aflatoxin adsorption tests with different activated charcoals showed good results atdifferent pH values (Huwig et al., 2001). In vivo, certain granulated activated carbons (GACs)decreased conversion of aflatoxin B1 to aflatoxin M1 in Friesian cows by 40.6 to 73.6% whenincluded in the diet at concentrations of 2.0% (CAST, 2003). Improvements in body weight gains andfeed intake of chickens were verified when activated charcoal was added to AF-contaminated diets(Edrington et al., 1997; Dalvi and Ademoyero 1983; Dalvi and McGowan 1984 and Jindal et al., 1993).However conflicting results have been verified, possibly due different types of activated charcoalsused.Successful in vitro OTA adsorption tests with activated charcoal showed that 1% product additionleads to complete adsorption of ochratoxin A from aqueous solutions not being influenced by pHvaluesranging from 3-8 (Plank et al., 1990).In addition, some trials have been performed with superactivated charcoal which differs fromactivated charcoal by its reduced particle size, higher surface area and a chemical modification duringthe manufacturing process. Superactivated charcoal was evaluated for its effectiveness in preventingdeath in rats given an oral lethal dose of 8 mg/kg body weight of T-2 toxin. The median effectivedose of oral superactive charcoal in preventing deaths in rats was found to be 0.175 g/kg bodyweight. At the end of the trial it was concluded that one gram per kilogram body weight oralsuperactive charcoal enhanced survival times and survival rates in rats given T-2 toxin (Galey et al.,1987). In vivo, orally administered activated charcoal was assessed for treatment of acute oral orparenteral exposure to T-2 toxin in mice. Charcoal treatment (7 g/kg) either immediately or 1 hr after78

INTRODUCTIONtoxin exposure resulted in a significant improvement of survival rate with values of 100% and 75%,respectively (Fricke et al., 1990). This is also in agreement with Bratich et al. (1989) who verified nolesions in rats fed diets contaminated with about 25 mg/kg T-2 toxin.No adsorbent materials, with the exception of activated carbon, showed relevant ability in bindingdeoxynivalenol and NIV (Binder, 2007). At 2% inclusion level the absorption with respect to theintake was lowered from 51% to 28% for DON and from 21% to 12% for NIV (Avantaggiato et al.,2004). However, the binding activity of activated carbon for these trichothecenes was lower thanthat observed for zearalenone (Avantaggiato et al., 2004).In vitro efficacy of activated carbon toward fumonisins was not confirmed in vivo using respectivebiomarkers (Binder, 2007).In a study of Vekiru et al. (2007) activated charcoal was investigated with regard to aflatoxinbinding and adsorption of vitamin B5 (Panthothenic acid), B12 and vitamin H (biotin). These watersolublevitamins play an important role in animal growth and productivity. Since they are not storedin the tissues in appreciable amounts they have to be regularly added to animal diets. Whencompared to other binders, charcoal presented a very high unspecific binding, with a high adsorptionof nutrients. Charcoal absorbed 100% AFB1 and extremely high amounts of vitamin B12 (99%) andbiotin (78%).b) CholestyramineCholestyramine is an anion exchange resin pharmaceutically used to decrease cholesterol. It hasbeen shown to efficiently bind zearalenone, ochratoxin A and fumonisins in vitro and in vivo (CAST,2003; Pettersson, 2004; Binder, 2007). Only a small number of adsorbent materials possessed theability to bind more than one mycotoxin.In in vitro tests, cholestyramine has shown to be the best adsorbent for zearalenone followed bycrospovidone, montmorillonite, bentonite, sepiolite and magnesium trisilicate (Ramos et al., 1996a).ZEA was tested in experiments using a dynamic gastrointestinal model. It is known that formation ofcholestyramine-ZEA complexes occurs rapidly (within 1 min) being stable for 24 h and is notinfluenced by pH or temperature (Ramos et al., 1994).Studies carried out in mice and rats fed diets containing different amounts of cholestyramineresulted in a reduction of estrogenic effects, toxin urinary excretion as well as renal and hepaticresidues and also in a reduction of 19 to 52% of intestinal absorption of ZEA (Underhill et al., 1995;Guerre, 2000; Avantaggiato et al., 2003; CAST, 2003).79

Table 9 : Outcome of yeast cell walls effects in mycotoxin contaminated diets in different speciesAnimalspeciesmycotoxinconcentrationAdditivesconcentrationChickens AF (5 ppm) 0.05-0.1%Chickens AF (2 ppm) 0.05-0.1%Chickens AF (2 ppm) 0.075 %Chickens T-2 (8.1 ppm) 0.1%PigsPigsT-2 (up to 2.1ppm)DON (4.44ppm)0.2%0.2 %Outcome of feed additive effectsPerformance: improvement in body weightOrgans: restoration of relative weights of liver, proventriculus and heartSystemic response: positive effects on serum biochemical parametersPerformance : no dataOrgans : decline in the number and severity of pathological lesions in liver, bursaof Fabricius, thymus and spleenSystemic response : no dataPerformance: no dataOrgans: Subacute dietary intake of AF altered nonenzymatic components (CP, alb,vit A, â-carotene, Vit E) of antioxidant defense systems.Systemic response: yeast glucomannan was not sufficient to ameliorate theoxidative damagePerformance: no dataOrgans: partial protection of the feed additive alone and further protection incombinaison with organic selenium against antioxidant depletion and lipidperoxydation in liverSystemic response: no dataPerformance: no difference on performance whatever the dietsOrgans: no lesions were observed regardless of T-2 toxin dose or glucomannansupplementationSystemic response: protective effect against AFB 1 immunotoxicity during thevaccinal responsePerformance: Feeding of the mycotoxin-contaminated diets resulted in a decreasein feed intake and live weight gain by 28% and 14% when compared to the controlgroup.Organs: no dataSystemic response: The addition of glucomannan mycotoxin adsorbent to pigletReferencesStanley etal. (1993)Karaman etal. (2005)Çinar et al.(2008)Dvorska etal. (2007)Meissonnieret al. (2009)Dänicke etal. (2007)

Cows AF (100 ppb) 1%, 5%Cows AFB 1 (112ppb)0.56%diets which are mainly contaminated with DON could not be recommendedPerformance: evaluation of milk aflatoxin concentrations in lactating Holsteincows with the inclusion of non-digestible yeast oligosaccharidesOrgans: no dataSystemic response: No significant changes (P>0.25) in AFM 1 concentrations inresponse to non-digestible yeast oligosaccharides were verified.Performance: reduction of AFM1 concentrations in dairy cowsOrgans: no dataSystemic response: the commercial product was not effective in reducing milkAFM1 concentrations (4%), AFM1 excretion (5%), or AF transfer from feed to milk(2.52%)Hopkins etal. (2008)Kutz et al.(2009)Chickens included broiler chicks, laying hens.Rats included male and female.Pigs included piglets, sow, barrow.

INTRODUCTIONFurthermore, cholestyramine has shown to increase ochratoxin A excretion in rats (Madhyasta etal., 1992), to inhibit the enterohepatic circulation of OTA in mice (Roth et al., 1988) and to reduceOTA absorption in the gastrointestinal tract of rats (Kerkadi et al., 1998).However, cholestyramine demonstrated just a little effect on the reduction of ochratoxin Aconcentration in blood, bile and tissues (Huwig et al., 2001).Several adsorbent materials were tested for their in vitro capacity to adsorb fumonisin B 1 andcholestyramine showed to be an effective binder with best adsorption capacity (Solfrizzo et al., 2001;Avantaggiato et al., 2005).c) Polyvinylpolypyrrolidone (PVPP)Adsorption of mycotoxins by PVPP is based on the hydration hull formed by the adsorbent.Particles must be polar to be attracted towards this hull and consequently to be bound to theadsorbent. PVPP reduces the toxicity of aflatoxins by decreasing its absorption in the gastrointestinaltract thus preventing negative effects on aflatoxicosis (Pasteiner et al., 1994 and Kiran et al., 1998).In fact, results of broiler chickens fed a contaminated diet with 2.5 ppm AF and PVPP (3 g/kg diet)suggested a prevention of depressive effects on peritoneal macrophage functions (Celik et al., 1996)and also some immune organ functions (Celik et al., 2000).In vitro 0.3 mg/g zearalenone were adsorbed by PVPP. Regarding respective in vivo efficacy,however, this polymer has rarely been tested. Pettersson (2004), Friend et al. (1984) and Ramos(1996) did not find any effects of PVPP in pigs fed deoxynivalenol-contaminated feed.d) Yeast cell walls and components thereof (Table 9)Yeasts, and particularly the cell wall of Saccharomyces cerevisiae, are an environmentally friendlyalternative to inorganic adsorbents, which are not extensively biodegradable and are associated withthe risk of contaminants (Yiannikouris et al., 2007).The structure of the cell wall of S. cerevisiae and the nature of its polysaccharide (glucan, mannan)components was subject of intensive scientific research, however, their role in the adsorption ofmycotoxins is a concept that has been little studied and has shown conflicting results.The walls of yeast are used to form complexes with dietary toxins which limit their absorption inthe digestive tract and consequently their negative impact on the animal and the quality of animalproducts used in food. Their ability to bind is due to their large surface of exchange. Recently,Yiannikouris et al. (2004; 2006) demonstrated that the β-D-glucan fraction of yeast cell wall is directly80

INTRODUCTIONinvolved in the binding process of mycotoxins and that the structural organisation of β-D-glucansmodulates the binding strength (Jouany, 2007).Beneficial effects were verified in counteracting aflatoxins in several animal species with live yeastculture preparations based on a Saccharomyces cerevisiae strain. Yeast glucomannan showed bindingability for various mycotoxins in vitro and/or in vivo (Stanley, 2004; Meissonnier et al., 2009,Pettersson, 2004). A binding capacity trial for T-2 toxin concluded that supplementation of modifiedglucomannan at 1 kg/t of feed is beneficial in preventing the absorption of T-2 toxin. Up to 35% T-2binding capacity was observed in the gastro-intestinal tract of broilers (Reddy et al., 2003). Contrary,yeast glucomannan investigated by Cinar et al. (2008) was not sufficient to prevent or ameliorate theoxidative damage caused by aflatoxins in broilers.An experiment, conducted to determine the effect of non-digestible yeast oligosaccharides (NYO)on milk aflatoxin concentrations in lactating Holstein cows consuming aflatoxins, concluded thatthere were no significant (P>0.25) changes in AFM 1 concentrations in response to NYO (Hopkins etal., 2008). These results are in accordance with a study of Stroud et al. (2006).Several studies have been carried out in order to assess the efficacy of yeast glucomannansagainst multi-mycotoxin contaminations. In vitro mycotoxin binding capacity by a commercialesterified glucomannan on aflatoxins, ochratoxin A and T-2 toxin was performed by Raju andDevegowda (2002). Results showed that the binding values of the mycotoxins decreased significantlywhen they were present in combination. The cumulative binding of the mycotoxins by the productwas dependent on the mycotoxin present.Esterified glucomannan (0.05-0.1%) in chickens reduced or eliminated many of the combinedtoxic effects (on performance, serum biochemistry and haematology) of AF, OTA and T-2 toxin (Rajuand Devegowda, 2000; Aravind et al., 2003). Using 0.5% of the same product also alleviated growthdepression in broilers associated with naturally contaminated diets containing aflatoxins, ochratoxinA, zearalenone and T-2 toxin (Bursian et al., 2006). According to a pig trial carried out byMeissonnier et al. (2009) glucomannan dietary supplementation demonstrated protective effectsagainst AFB 1 and T-2 toxin immunotoxicity. According to Bursian et al. (2004), however, a glucanpolymer product (0.2%) did not alleviate the toxic effects on mink consuming diets contaminatedwith fumonisins B 1 , ochratoxin A, moniliformin and zearalenone.e) Adsorption by bacteriaUntil now, most of the research regarding the use of bacteria as adsorbents has been focused onin vitro tests, and only a few studies have been performed in vivo. Trials with some dairy strains oflactic acid bacteria and bifidobacteria showed that they are able to effectively bind aflatoxins. Cell81

INTRODUCTIONwall peptidoglycans and polysaccharides have been suggested to be the two most importantelements responsible for the binding by lactic acid bacteria. According to Haskard et al. (2001) andPeltonen et al. (2001), Lactobacillus amylovorus strains and Lb. rhamnosus strains removed morethan 50% AFB 1 .Furthermore, probiotics such as Lactobacillus rhamnosus strain GG and a mixture of Lactobacillusrhamnosus LC705 and Propionibacterium freudenreichii ssp. shermanii JS (LC705+JS) wereinvestigated in order to know whether these probiotic bacteria bind AFB 1 under physiologicalconditions in the gut. Results concluded that both probiotic preparations were able to bind AFB 1 invitro (Gratz, 2007). Studies carried out by Pierides et al. (2000) suggested that six dairy strains oflactic acid bacteria can reduce AFM 1 content in liquid media. Their ability to bind AFM 1 gave apotential approach for either decontamination of foods and feeds or reduction of the bioavailabilityof aflatoxins in the diet, presuming that the bacteria are able to remove the toxin from thegastrointestinal tract.Several successful studies were conducted with bacteria regarding binding of zearalenone. Theinteraction between two Fusarium mycotoxins, zearalenone and its derivative zearalenol with twostrains of Lactobacillus was investigated by El-Nezami et al. (2002b). After mycotoxin incubation (2mg/ml) with two strains of Lb. rhamnosus, approximately 55% of the toxins were bound instantlyafter mixing with the bacteria. The mechanism linking these toxins to the surface of bacteria wasshown not to be due to metabolic activity of living cells. In addition, removal of ZEA from medium byfive Propionibacterium, three Lactobacillus and two Bifidobacterium strains has shown to be possible.All examined strains caused a decrease of toxin concentration in medium after incubation with both,viable and nonviable cells. Usually higher efficiency was observed with Propionibacteria (El-Nezami etal., 2002a; Gwiazdowska et al., 2006). Lactobacillus and Propionibacterium also removedtrichothecenes (DON, 3-AcDON, NIV, FX, DAS, T-2) from liquid media (El-Nezami et al., 2002a).A trial conducted by El-Nezami et al. (2000) indicated that probiotic lactic acid bacteria arecapable of binding AFB 1 under in vivo conditions. They observed that in the presence of Lb.rhamnosus GG 74% of the reduction in the uptake of AFB 1 by the intestinal tissue takes place within60 min. Also, probiotic Lb. rhamnosus GG administration in rats demonstrated that fecal AFB 1excretion in GG-treated rats was increased via bacterial AFB 1 binding and that AFB 1 -associatedgrowth faltering and liver injury were alleviated with GG treatment (Gratz et al., 2006).f) Other organic adsorbentsOther organic adsorbents such as humic acid, divinylbenzene-styrene, chlorophyllin, alfalfa andfibers were studied. Van Rensburg (2006) investigated oxihumate (pure, high quality humic acid)82

INTRODUCTIONwhich slightly differs chemically from humic acids obtained from other sources. Efficient results wereobtained in broiler chickens fed aflatoxin-contaminated diets and it was proven that oxihumate ishighly effective in the amelioration of aflatoxicosis in broilers. Furthermore, divinylbenzene-styreneshowed to reduce the absorption and the toxic effects of ZEA (Guerre, 2000) and T-2 toxin in rats(Carson and Smith, 1983) when using anion-exchange and not cation-exchange resins. This anionexchange resin has been shown to prevent zearalenone toxicosis by binding the toxin in the digestivetract to prevent absorption (Smith, 1980). Little information is available concerning chlorophyllin(CHL) binding capacity. Breinholt et al. (1995) supported that CHL is a potent dose-responsiveinhibitor of aflatoxin I, DNA adduction and hepatocarcinogenesis in the rainbow trout.Furthermore, Jouany (2007) indicated that alfalfa and oat fibers are capable of reducing theestrogenic effect of ZEA on rats and bleaching clays that had been used to process canola oil werefound to lessen the effects of T-2 (CAST, 2003).A trial conducted with rats and swine to determine the potential of dietary alfalfa as a treatmentfor zearalenone toxicosis showed that alfalfa inclusion in diet reduced the inhibitory effects of ZEA ongrowth and feed consumption, minimized ZEA-induced liver enlargement, increased hepatic 3 -HSDactivity, reduced concentrations of residual ZEA liver and decreased uterine enlargement. Thesestudies proved that dietary alfalfa promotes ZEA metabolism in rats and that this feedstuff may alsobe useful for treating ZEA toxicosis in livestock (James and Smith, 1982).Besides micronized wheat fibers (MWF) were evaluated to decrease the levels of ochratoxin A inplasma, kidney and liver of piglets fed a naturally contaminated diet. Results demonstrated asignificantly protective effect against usual increased kidneys and liver weights caused by OTAcontaminateddiets. MWF significantly protected against high OTA concentration in plasma (45.6%decrease), kidney (40.8% decrease) and liver weights (26.5% decrease). Thus, these results suggestthat the addition of MWF is effective in decreasing the bioavailability of OTA from contaminateddiets in piglets (Aoudia et al., 2009).83

INTRODUCTIONII. BIOLOGICAL DETOXIFICATIONEnzymatic or microbial degradation of mycotoxins (“biotransformation”) leading to less- or evennon-toxic metabolites has already been a subject of research for more than 40 years as this strategyis a quite attractive method for the decontamination of crops. Microorganisms including yeast,moulds and bacteria have been screened for their ability to modify or inactivate differentmycotoxins. Only a few microorganisms with respective activity were isolated, the first wasFlavobacterium aurantiacum with the ability to detoxify aflatoxins (Ciegler et al., 1966). Thisorganism has since then been studied extensively for possible degradation products (Bata andLasztity, 1999). Apart from F. aurantiacum, a number of bacterial and especially fungal species havebeen found to detoxify aflatoxins (Karlovsky, 1999). Rhizopus sp. has been claimed to be particularlysuitable for large-scale detoxification of aflatoxin-contaminated feeds by solid-state fermentation(Knol et al., 1990, as cited by Karlovsky, 1999).Ochratoxin A is rapidly degraded by micro-organisms in the rumen to ochratoxin α (OTα) andphenylalanine (Hult et al., 1976; Kiessling et al., 1984). Wegst and Lingens (1983) proved degradationof OTA by the aerobic bacterium Phenylobacterium immobile. Cheng-An and Draughon (1994) havescreened bacteria, yeast and moulds for their ability to detoxify ochratoxin A and foundAcinetobacter calcoaceticus to be able to degrade OTA in an ethanol-containing medium. Differentstrains of Lactobacillus, Bacillus and Saccharomyces have also been shown to degrade ochratoxin A invitro to varying degrees up to 94 % (Böhm et al., 2000). The same strains were also tested fordegradation of trichothecenes, but with less success. Styriak et al. (2001) showed partial degradationof ochratoxin A, nivalenol, deoxynivalenol, zearalenone and fumonisins by yeast strains.A yeast strain isolated from the hindgut of a termite and identified as a member of the genusTrichosporon showed a potential deactivation of both, OTA and ZEA, in animal feeds (Molnar et al.,2004; Schatzmayr et al., 2006). Due to the yeast’s main property to degrade mycotoxins this strainwas named T. mycotoxinivorans (lat. vorare = degrade). The yeast detoxifies OTA by cleavage of thephenylalanine moiety from the isocumarin derivate ochratoxin α (OTα). This metabolite has beendescribed to be nontoxic or at least 500 times less toxic than the parent compound (Bruinink et al.,1999; Schatzmayr et al., 2003). Feeding trials with the aim to test the efficacy of T. mycotoxinivoransto suppress ochratoxicosis proved that the dietary inclusion of this yeast blocks ochtratoxin-inducedimmune suppression in broiler chicks (Politis et al., 2005; Binder, 2007). More recently, a feeding trialwith broiler chicks was performed in order to evaluate the toxic effects of OTA and attenuatingeffects of a commercial toxin deactivator containing the yeast Trichosporon mycotoxinivorans onperformance parameters, serum enzymes and clinic-pathomorphology of internal organs.84

INTRODUCTIONImprovement in FCR and less pronounced histological changes in kidneys, liver, bursa and spleenwere verified in animals fed the toxin deactivator. Thus in the presence of this feed additive, theharmful effects in the pathomorphological and histological changes in internal organs wereattenuated (Hanif et al., 2008).Zearalenone has no acute toxicity, but it mimics the reproduction hormone estrogen, andtherefore causes substantial fertility problems. The metabolisation of this toxin by T.mycotoxinivorans leads to a compound that is no longer estrogenic. This has been proven in an invitro assay with breast cancer cells (Schatzmayr et al., 2003).Reduction of zearalenone to α- and β-zearalenols has been shown in ruminal fluid and for manymixed and pure cultures of bacteria, yeast and fungi. However, this transformation cannot beconsidered as a detoxification as zearalenols still show significant estrogenic activity. Non-estrogenicZEA-metabolites were obtained from degradation by e.g. Thamnidium elegans, Mucor baineri,Rhizopus sp., Streptomyces rimosus, Cunninghamella baineri and Gliocladium roseum (Kamimura,1986; EI-Sharkawy and Abul-Hajj, 1987; 1988). The latter strain detoxified zearalenone by ringopening with subsequent decarboxylation in yields ranging between 80 and 90% (El-Sharkawy andAbul-Hajj, 1988).A great deal of literature is available concerning the biotransformation of trichothecenes, whichare among the world’s most important agricultural toxins. Their toxicity can mainly be attributed totheir 12,13-epoxide ring. Thus, reductive de-epoxidation by ruminal and intestinal flora of pigs, hensand rats (King et al., 1984; Kollarczik et al., 1994; Swanson et al., 1987, 1988; Yoshizawa et al., 1983)or by a new strain of Eubacterium isolated from bovine rumen contents (Binder et al., 2000) results ina significant loss of toxicity. The latter bacterium referred to as BBSH 797 is actually the first bacterialstrain cultured, produced and stabilized in order to be applied as feed additive to counteract thenegative effects of trichothecenes by biotransformation. Fuchs et al. (2000; 2002) identified nontoxicmetabolites after the microbial degradation of type A and type B trichothecenes by BBSH 797and showed that de-acetylation occurs simultaneously to de-epoxidation when the strain wasapplied. The in vivo efficacy of the additive has been tested in trials with pigs and chickens.Significant feed gain and improved feed conversion ratios were determined in piglets fed 2.5 mg/lDON and the chickens fed 10.5 mg/l DON showed reduced mortality and a positive influence onweight development (Binder et al., 2001; Plank et al., 2009). Moreover, a BBSH 797-containing feedadditive was capable of counteracting the adverse effects on performance of growing broilerchickens caused by the dietary administration of 2 ppm T-2 toxin (Diaz et al., 2005).Other investigations concerning microbial detoxification of trichothecenes concentrated on theuse of aerobic soil bacteria. He et al. (1992) incubated soil with DON and measured a significant85

INTRODUCTIONdecrease in toxin-concentration during incubation. Attempts to isolate a pure culture failed. In 1983,Ueno et al. isolated the Curtobacterium sp. (strain 114-2), a soil bacterium capable of deacetylatingT-2 toxin to the less toxic metabolites HT-2 toxin and T-2 triol. Finally, in 1997, Shima et al. weresuccessful in isolating a soil bacterium, belonging to the Agrobacterium-Rhizobium group, capable ofbiotransforming DON to the less-toxic metabolite 3-keto-DON. Protective effect of a feed additiveagainst the toxic effects of 4,15-diacetoxisciepenol (DAS) in broiler chickens was also investigated.Results suggested that feed additive supplementation in the diet protected against the adverse effectof DAS on feed intake and body weight at different levels of inclusion.Enzymes capable of degrading fumonisins have been isolated from a filamentous saprophyticfungus growing on maize (Blackwell et al., 1999; Duvik, 2001). Recently scientists isolated andcharacterized new fumonisin-metabolizing bacterial strains. Some of these isolates were found to beactive in the gastrointestinal tract of animals. One of the strains with the highest technologicalpotential belongs to the family of Sphingomonadaceae. It degrades fumonisins by first cleaving offtricarballylic acid side chains and subsequently catabolising the rest of the molecule into non-toxicproducts (Moll et al., 2009).Moreover, biological detoxification was also tested in multi-mycotoxin contaminated feeds.Cheng et al. (2006) investigated the combined effects of deoxynivalenol and zearalenone on growthperformance, blood biochemistry and immune response of weaning piglets and the alleviating effectsof a mycotoxin degrading enzyme (MDE) on the toxicity of these Fusarium mycotoxins. Based on theresults, it was suggested that the combination of DON and ZEA confers a chemical multi-organtoxicity in pigs and MDE provides a partial or complete toxic sparing effect of mycotoxins.Biotechnology in animal feed production has become stronger and more important during the lastdecades. The uses of microbiological (enzymes and microbes) in feed have dramatically increasedand are prime examples for the necessity of new approaches in animal production.Isolation and characterization of microorganisms or enzymes that are able to biotransformmycotoxins could possibly be the breakthrough for the practical application of biotechnology inspecific decontamination processes taking place directly in the intestinal tract of animals. Thisbiological decontamination may become a technology of choice, as enzymatic reactions offer aspecific, efficient and environmentally friendly way of detoxification.Ensiling of mycotoxin-contaminated crops for detoxification has been proposed as an interestingand possible method for elimination or reduction of mycotoxins. Normal ensiling has, however, onlyrarely been studied for its mycotoxin degrading potential. A study by Rotter et al. (1990) showed thatensiling of ochratoxin-contaminated barley could reduce the toxin by approximately 68%. However,86

INTRODUCTIONin feeding studies with chicken, no improvement in performance or mortality could, be foundcompared with the non-ensiled diets. Yeasts in grass silage have been found to degrade patulin insilage inoculated with Paecitomyces sp. to induce patulin production (Dutton et al., 1984). Both,bacteria and yeasts from maize silage have been shown to be able to degrade fumonisins (Camilo etal., 2000). Stimulation of mycotoxin degradation by naturally occurring micro-organisms in silage orthe addition of yeasts or bacteria with known mycotoxin degradation ability to silage may in thefuture become practical means to detoxify mycotoxins in certain crops.87

INTRODUCTIONCONCLUSIONPrevention and reduction of mycotoxin contamination during crop and feed production hasbecome more important, good agricultural practices and HACCP in the production process hasbecome more a requirement and the need of feed additives to prevent absorption and toxic effectsof mycotoxins in farm animals has increased significantly.Actually, there is already an excellent potential for organic and inorganic binders to help managethe mycotoxin problem in a safe, economic and easy way. Adsorbents can be used as a complement,acting in the animal´s organism, to physical and chemical methods used mainly in thedecontamination of raw materials.Since feedstuffs are commonly contaminated with more than one mycotoxin, it is important totake into account that a good mycotoxin inactivator has to be as effective as possible against severalmycotoxin contaminations, incorporated in a small amount in the complete diet, have a high bindingcapacity and be free of impurities and odors.It is now known that adsorbents belonging to the same category have different physical andchemical properties, thus their efficacy can vary markedly. To assure efficiency and safety, alladditives have to be carefully tested before coming into the market, being the in vitro tests mostlyimportant for their scientific development and improvement. Moreover, sensitive parameters haveto be measured such as biochemistry, gross pathology, histopathology and immune parameters.As adsorption is not a viable option regarding trichothecenes, zearalenone and ochratoxins,mycotoxin inactivation by biotransformation is a very promising strategy to detoxify thesemycotoxins. Indeed, biotransformation is currently the most recent and challenging process ofmycotoxin deactivation being under development and therefore, requiring further investigations.Another important target for the future is the development of products that are effective againstmulti-mycotoxin contaminations.There are many factors that negatively influence and complicate the setting of safe levels for thedifferent mycotoxins in different countries, however regulations exist worldwide and maximumlevels for certain contaminants in foodstuffs were already set by EU legislation.Currently, a large part of the feed industry has already applied for production standards, whichare often stricter than required by legislation. The HACCP analysis of the production practices that isrequired to comply with the EU Feed Hygiene legislation, and the quality systems it triggered such asFami-QS, is an obvious qualifier to operate in the European feed industry.Mycotoxin contamination, however, still occurs despite the most strenuous efforts of prevention,therefore there is an obvious need for mycotoxin inactivating products to complement it in the88

INTRODUCTIONperspective of HACCP analysis conclusions. It is believed that mycotoxin inactivating products mayreduce or even close the gaps left open by the introduction of maximum limits, preventive actionsand analytical control.89

TRAVAILEXPERIMENTAL90

TRAVAIL EXPERIMENTALOBJECTIFS DE LA THESELes mycotoxines présentent autant de propriétés toxiques qu’elles divergent structurellement.Néanmoins, ces substances ont quasiment toutes la capacité de moduler la réponse immunitaire del’organisme, et ce, même lors de l’exposition à de faibles doses. Les mycotoxines peuvent agir surl’expression et la fonction de différents médiateurs et cellules du système immunitaire, impliquésdans le système de défense de l’hôte, à savoir l’inflammation, la réponse à médiation cellulaire et laréponse à médiation humorale. En conséquence, il a été montré que ces effets immunotoxiquespeuvent sensibiliser les animaux aux infections, diminuer l’efficacité vaccinale et thérapeutique, ouencore réactiver des infections.Dans ce contexte, l’équipe d’immuno-mycotoxicologie au sein du pôle ToxAlim de l’INRA deToulouse, s’intéresse à ces effets immunomodulateurs lors d’exposition à faible doses, et plusparticulièrement après un challenge antigénique. La thématique de l’équipe s’est également orientéedepuis peu sur l’effet des mycotoxines sur le tractus gastro-intestinal. En effet, la voie principaled’exposition aux mycotoxines étant l’ingestion d’aliments contaminés, l’intestin est un organe cibledes mycotoxines, et peut être exposé à de fortes concentrations en toxines. Considérant lesdifférentes fonctions du système digestif, telles que l’immunité, la perméabilité intestinale oul’absorption des nutriments, nos recherches ont porté une attention particulière à certains de cescritères.L’activité de la société BIOMIN est basée sur le développement de stratégies de lutte contre lesmycotoxines majeures présentes dans les aliments. Contrairement à de nombreuses sociétés, quiélaborent des produits à base d’argile ou de levures permettant l’adsorption des mycotoxines,BIOMIN axe ses travaux sur l’élimination des toxines par voie enzymatique. Un long travail en amontd’isolement et de screening de micro-organismes, capable de détoxifier les mycotoxines, leur permetde proposer des solutions innovantes, et ont l’avantage de cibler spécifiquement chaque mycotoxine,considérant leur grande diversité structurelle. Leur intérêt dans ce partenariat résidait dans leursouhait d’avoir une expertise plus approfondie de l’efficacité de leurs produits, et ce, sur desparamètres plus fins que des paramètres zootechniques. De plus, une de leurs dernières approches,encore non-commercialisée, cible spécifiquement les fumonisines, mycotoxines difficilementéliminées par les méthodes physiques, chimiques et d’adsorption.Ainsi, la thèse a comporté différents objectifs. Le premier a été de déterminer la toxicitéindividuelle et combinée de faibles doses de déoxynivalénol et de fumonisines (fumonisines B1 + B2)chez le porcelet. Pour cela, différents critères ont été évalués. L’impact des mycotoxines sur la91

TRAVAIL EXPERIMENTALtoxicité générale, telle que la performance, l’hématologie et la biochimie, et les organes cibles (foie,rein, poumon), a été analysé. L’impact sur la mise en place d’une réponse immunitaire spécifique ouacquise, suite à un challenge antigénique a été étudié. Nous avons pour cela intégré dans notreétude un protocole de vaccination avec un antigène de laboratoire, l’ovalbumine ; l’utilisation de cetantigène ayant déjà fait ses preuves dans des expériences antérieures. Et finalement, l’impact sur lecompartiment intestinal, avec des analyses histopathologiques et immunologiques, a également étédéterminé.Le second objectif a été de comparer la toxicité de la fumonisine B1 (FB1), la mycotoxineprédominante des fumonisines, et de son dérivé hydrolysé (HFB1). En effet, BIOMIN a purifié uneenzyme, une carboxylestérase issue d’une bactérie isolée de compost, douée de propriétésd’hydrolyse et permettant ainsi l’élimination des deux chaînes latérales d’acide tricarballyle de laFB1. Après hydrolyse totale de la FB1 par cette approche, le métabolite obtenu HFB1 a été confrontéà la molécule mère FB1 (à une dose aigüe en comparaison de celle utilisée dans la premièreexpérience) lors d’une expérience in vivo menée chez le porcelet. Les deux substances ont étéadministrées oralement aux animaux, et la toxicité a été déterminée au niveau hépatique etintestinal.Enfin, l’efficacité d’un produit commercial développé par BIOMIN et combiné avec lacarboxylestérase récemment purifié, a été évaluée lors de leurs incorporations dans des alimentscontaminés en déoxynivalénol et fumonisines, et ingérés par des porcelets. Les mêmes paramètresque dans l’expérience de co-contamination décrite ci-dessus ont été étudiés.Le porc a été notre modèle d’étude expérimental du fait de sa sensibilité, de son expositionnaturelle aux mycotoxines via son régime alimentaire, et aussi de son degré de similitude avec lessystèmes biologiques chez l’homme, notamment immunitaire et digestif.92

CHAPITRE 1Toxicité in vivo du Déoxynivalénol etde la Fumonisine, seuls ou encombinaison chez le porcelet93

TRAVAIL EXPERIMENTALRESUME DES ETUDESL’objectif de cette première partie du travail expérimental a été de déterminer le risque et le typed’interaction suite à l’exposition simultanée à deux mycotoxines majeures, le déoxynivalénol (DON)et la fumonisine (FB). Des porcelets ont ainsi ingéré durant cinq semaines des aliments contaminés àdes doses qui n’entraînent pas de manifestations cliniques. Cette exposition mime donc une situationde terrain, avec des concentrations en toxines proches des valeurs individuelles fixées par lacommission européenne pour un aliment complet chez le porc.Afin d’évaluer au mieux ce risque, nous avons analysé les effets des régimes contaminés sur denombreux paramètres, avec une attention particulière sur le développement de la réponseimmunitaire suite à l’immunisation des animaux, et aussi sur le tractus gastro-intestinal. Ces deuxchamps d’investigations sont présentés séparément ci-après.1. Toxicité in vivo du déoxynivalénol et de la fumonisine, seuls ou encombinaison sur la réponse immunitaireVingt quatre porcelets ont été répartis aléatoirement dans quatre groupes et ont reçu pendant 35jours des régimes différents. Le premier a servi de régime contrôle, le second était contaminé avec3,0 mg DON/kg d’aliment, le troisième avec 6,0 mg FB/kg d’aliment, et le dernier avec la présencedes deux toxines aux concentrations respectives des régimes individuels. Pour stimuler la réponseimmunitaire, les animaux ont reçu à 12 jours d’intervalle deux injections sous-cutanéesd’ovalbumine.Les régimes mono et co-contaminés n’ont pas modifié la croissance des animaux et n’ont euqu’un effet mineur sur les paramètres hématologiques et biochimiques. En revanche, l’ingestion desmycotoxines a provoqué des lésions hépatiques, rénales et pulmonaires, avec une interaction detype additive sur les lésions du foie lorsque les toxines étaient en association.En ce qui concerne les effets sur le système immunitaire, aucune altération de la réponseimmunitaire totale et non-spécifique n’a été observée. A l’inverse, nos résultats sur la proliférationdes lymphocytes après stimulation in vitro avec de l’ovalbumine et les titres en IgG anti-ovalbuminenous laissent penser que les toxines sont capable d’affecter la réponse immunitaire spécifique,cellulaire et humorale. De plus, le dosage par RT-PCT temps réel des ARNs de la rate, codant pour descytokines, a révélé une baisse de l’expression de certains médiateurs impliqués dans le recrutement94

TRAVAIL EXPERIMENTALdes cellules présentatrices d’antigène et dans l’activation et la prolifération des lymphocytes. Cesobservations couplées aux données de la littérature impliquent que l’établissement de la réponseimmunitaire suite à la sensibilisation par notre antigène a été significativement affecté par l’ingestiondes mycotoxines, et en particulier avec le régime co-contaminé.95

These are not the final page numbersMol. Nutr. Food Res. 2011, 55, 1–11 DOI 10.1002/mnfr.2010004021RESEARCH ARTICLEIndividual and combined effects of subclinical dosesof deoxynivalenol and fumonisins in pigletsBertrand Grenier 1,2 , Ana-Paula Loureiro-Bracarense 3 , Joelma Lucioli 3 ,Graziela Drociunas Pacheco 1,3 , Anne-Marie Cossalter 1 , Wulf-Dieter Moll 2 , Gerd Schatzmayr 2and Isabelle P. Oswald 11 INRA, Unité de Pharmacologie-Toxicologie, Toulouse, France2 BIOMIN Research Center, Tulln, Austria3 Universidade Estadual de Londrina, Lab Patologia Animal, Londrina, BrazilScope: Deoxynivalenol (DON) and fumonisins (FB) are the most frequently encounteredmycotoxins produced by Fusarium species and most commonly co-occur in animal diets.These mycotoxins were studied for their toxicity in piglets on several parameters includingplasma biochemistry, organ histopathology and immune response.Methods and results: Twenty-four 5-wk-old animals were randomly assigned to four differentgroups, receiving separate diets for 5 wk, a control diet, a diet contaminated with either DON(3 mg/kg) or FB (6 mg/kg) or both toxins. At days 4 and 16 of the trial, the animals weresubcutaneously immunized with ovalbumin to assess their specific immune response. Thedifferent diets did not affect animal performance and had minimal effect on hematologicaland biochemical blood parameters. By contrast, DON and FB induced histopathologicallesions in the liver, the lungs and the kidneys of exposed animals. The liver was significantlymore affected when the two mycotoxins were present simultaneously. The contaminateddiets also altered the specific immune response upon vaccination as measured by reducedanti-ovalbumin IgG level in the plasma and reduced lymphocyte proliferation upon antigenicstimulation. Because cytokines play a key role in immunity, the expression levels of IL-8, IL-1b, IL-6 and macrophage inflammatory protein-1b were measured by RT-PCR at the end ofthe experiment. The expression of these four cytokines was significantly decreased in thespleen of piglets exposed to multi-contaminated diet.Conclusion: Taken together, our data indicate that ingestion of multi-contaminated dietinduces greater histopathological lesions and higher immune suppression than ingestion ofmono-contaminated diets.Received: August 25, 2010Revised: November 6, 2010Accepted: November 25, 2010Keywords:Co-contamination / Deoxynivalenol / Fumonisins / Immunity / Subclinical doses1 IntroductionMycotoxins are secondary metabolites of fungi that maycontaminate animal feeds and human foods. They arefrequently detected in grains, but also in fruits, vegetables,nuts and silages. The Food and Agricultural Organizationestimated that as much as 25% of the world’s agriculturalcommodities are contaminated with mycotoxins and theeconomic losses due to mycotoxin contamination are estimatedin billions of dollars annually worldwide [1]. Clinicalsigns caused by mycotoxins range from mortality to slowgrowth and reduced reproductive efficiency. ConsumptionCorrespondence: Dr. Isabelle P. Oswald, INRA-Laboratoire dePharmacologie-Toxicologie, 180 chemin de Tournefeuille BP93173, 31027 Toulouse Cedex 3, FranceE-mail: isabelle.oswald@toulouse.inra.frFax: 133-5-61-28-53-10Abbreviations: APC, antigen-presenting cells; BALT, bronchioleassociatedlymphoid tissue; DON, deoxynivalenol; FB,fumonisins; HE, hematoxylin–eosin; MAPK, mitogen-activatedprotein kinase; MIP-1b, macrophage inflammatory protein-1b;OVA, ovalbumin& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.mnf-journal.com

These are not the final page numbers2 B. Grenier et al. Mol. Nutr. Food Res. 2011, 55, 1–11& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimof mycotoxins may also result in impaired immunity anddecreased resistance to infectious diseases [2].Worldwide surveys on the occurrence and contaminationlevels of mycotoxins in raw materials indicate that toxinsproduced by Fusarium mold species are of concern [3–5].Among the fusariotoxins, deoxynivalenol (DON) and fumonisins(FB) are frequently detected with concentrations up to927 mg DON/kg and 300 mg FB/kg [4]. Among cerealsamples collected from European countries, 54% were cocontaminatedwith DON and FB [6]. Similarly, in France, 65%of the maize kernels harvested during 2004–2006 were cocontaminatedwith DON and FB (Arvalis-Institut du végétal,unpublished data). These two mycotoxins are of majorconcern not only in terms of their ubiquitous distribution butalso because of their effects on human and animal health.At high concentrations, FB cause equine leukoencephalomalaciaand porcine pulmonary edema, and it is nephroandhepatotoxic and carcinogenic in rats and mice. FB1 hasbeen classified as a potential human carcinogen (class 2B)by the International Agency for Research on Cancer. Inhumans, consumption of FB-contaminated food has beenlinked with human esophageal cancer and neural tubedefects [7]. Disruption of sphingolipid biosynthesis appearsto be one mechanism involved in FB toxicity, with inhibitionof ceramide synthase [7] leading to accumulation of sphingoidbases (sphinganine and sphingosine). The effects ofingestioning low doses of FB are less documented butrevealed pathological alterations of the lungs and anincrease in intestinal colonization by opportunistic pathogenicbacteria in piglets [8–10].Acute exposure to high doses of DON induces diarrhea,vomiting, leukocytosis and gastrointestinal hemorrhage.Anorexia, growth retardation and immunotoxicity occur inrodents and pigs following chronic DON ingestion [11]. At thecellular level, DON interferes with the active site of peptidyltransferase on ribosomes, and inhibits protein synthesis [11].Further, binding of DON to the ribosome in eukaryotic cellstriggers a ‘‘ribotoxic stress response’’, which involves phosphorylationof the mitogen-activated protein kinases (MAPKs)[12]. MAPK activation modulates the expression of genesassociated with the immune response, chemotaxis, inflammationand apoptosis. The cellular and molecular mechanismsof the immunomodulating action of DON have beendescribed in numerous studies using mice and murine celllines [13]. Depending on the dose and frequency of exposure,DON can be either immunosuppressive or immunostimulatory[11, 14]. Prolonged ingestion of DON produces elevationof immunoglobulin A in plasma [13–15] while increasing thesusceptibility to infectious diseases [11].The toxicity of combinations of mycotoxins cannot alwaysbe predicted based upon their individual toxicities [1]. Interactionsbetween concomitantly occurring mycotoxins can beantagonistic, additive or synergistic. The data on combinedtoxic effects of mycotoxins are limited and, therefore, theactual combined health risk from exposure to mycotoxins isunknown. Assessment of the interaction of Fusarium mycotoxinshas been investigated in vitro on immune cells andintestinal epithelial cells [16, 17]. In vivo experiments have alsobeen done on mice, pigs and poultry using high doses oftoxins in which the authors mainly looked for differences inanimal performance. Among them, few studies wereconcerned with the interaction between DON and FB [18, 19].The purpose of this study was to compare the effects of lowdoses of DON and FB in pigs when fed individually and incombination with particular emphasis on their effects on theimmune response. The experimental design was a factorialassay including control feed and feed contaminated with 3and 6 mg/kg DON and FB individually and in combination,respectively. These contamination levels correspond to levelsthat frequently occur naturally in cereals [1]. Results havebeen reported in terms of both general toxicological parametersincluding weight gain, hematology, plasma biochemistryand organ histology as well as specific parametersdescribing immune system responses (total and specificantibody, lymphocyte proliferation, cytokine expression).2 Materials and methods2.1 AnimalsAll animal experimentation procedures were carried out inaccordance with the European Guidelines for the Care andUse of Animals for Research Purposes (Directive 86/609/EEC). Twenty-four 4-wk-old weaned castrated male pigs(Pietrain/Duroc/Large-white) were obtained locally. Malepigs were used in this protocol as it was previouslydemonstrated that a greater effect of DON and FB occurs inmale pigs compared to female pigs [20]. Animals wereacclimatized for 1 wk in the animal facility of the INRALaboratory of Pharmacology and Toxicology (Toulouse,France), prior to being used in experimental protocols. Sixpigs were allocated to each treatment on the basis of bodyweight. During the 35-day experimental period each treatmentgroup was given free access to water and the assigneddiet. The pigs were observed daily and weighed weekly.2.2 Experimental dietsDiets were manufactured at INRA facilities in Rennes(France), and formulated according to the energy and aminoacid requirements for piglets. Feed composition is detailedin Table 1. Four different batches were prepared, onecontrol batch and three batches artificially contaminatedwith the mycotoxins. Two strains of Fusarium, F. graminearumDSM-4528 and F. verticillioides M-3125 were used toproduce DON and FB, respectively. These strains weregrown separately on rice. FB were produced as previouslydescribed [21]. DON was extracted with ethyl acetate, andthe extract dried on silica gel 60 (Merck, Darmstadt,Germany). The homogenized extracts contained 24 andwww.mnf-journal.com

These are not the final page numbersMol. Nutr. Food Res. 2011, 55, 1–11 3Table 1. Composition of the experimental dietIngredient (%)Wheat 47.50Soybean meal 24.30Barley 22.90Calcium phosphate 1.12Calcium carbonate 1.00Vitamin and mineral premix a) 0.50Vegetable oil 1.40Sodium chloride 0.40Phytase 0.01Lysine 0.465Methionine 0.165Threonine 0.195Tryptophan 0.045Composition b)Starch (g) 476.8Crude protein (g) 218.3Crude fiber (g) 37.5Ca (g) 10.5P (g) 6.5K (g) 8.7Net energy (MJ) 15.6a) Vitamin A, 2 000 000 IU/kg; vitamin D3, 400 000 IU/kg; vitaminE, 4000 mg/kg; vitamin C, 8000 mg/kg; vitamin B1, 400 mg/kg;vitamin K3, 400 mg/kg; iron, 20 000 mg/kg; copper, 4000 mg/kg; zinc, 20 000 mg/kg; manganese, 8000 mg/kg.b) Corresponding to 1000 g dry matter/kg.21 g/kg DON and FB, respectively. The extracts containingthe mycotoxins were mixed into the vitamin mineralsupplements and then incorporated into the cereal mixturebefore granulation.The feed was analysed for mycotoxin content by QuantasAnalytik (Tulln, Austria) and by using a multi-mycotoxinmethod [22]. DON, zearalenone and enniatin were found tobe naturally present in the cereals used, resulting inconcentrations of 500, 50 and 100 mg/kg of feed, respectively.All other mycotoxins, including aflatoxins, T-2 toxin, HT-2toxin and ochratoxin A were below the limit of detection.The mono-contaminated diets contained 2.8 mg of DON/kgof feed and 5.9 mg of FB/kg of feed (4.1 mg FB1/kg11.8 mgFB2/kg of feed) while the co-contaminated diet contained3.1 mg of DON and 6.5 mg of FB/kg of feed (4.5 mg FB1/kg12.0 mg FB2/kg of feed).2.3 Experimental design and sample collectionOn the 4 th and 16 th day of the experiment, all piglets wereimmunized by subcutaneous inoculation with 1 and 2 mg ofovalbumin (OVA), respectively (Sigma, St-Quentin Fallavier,France), dissolved in sterile PBS and mixed with incompleteFreund’s adjuvant (Sigma). At weekly time intervals, bloodsamples were aseptically collected from the left jugular vein.Blood was collected in tubes containing sodium heparin or& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, WeinheimEDTA (Vacutainer s , Becton-Dickinson, USA) for blood cultureor blood formula, respectively. Plasma samples were obtainedafter centrifugation of heparinized blood and stored at 201Cfor later analysis. After 35 days of dietary exposure to mycotoxins,immediately after electrical stunning, pigs were killedby exsanguination. Samples of lungs, liver and kidneys werecollected from all groups and fixed in 10% buffered formalinfor histopathological analysis. In addition, a portion of thespleen was collected from euthanatized animals, flash-frozenin liquid nitrogen and stored at 801C until processed formeasurements of cytokine mRNA.2.4 Hematology and biochemistryHematological analysis was carried out using the impedancecoulter LH500 (Beckman Coulter, Villepinte, France). Subpopulationsof white blood cells (lymphocytes, monocytes,neutrophils, eosinophils and basophils) were also studiedand made manually on 100 leukocytes on May-Gr.unwaldGiemsa stained smears.Plasma concentrations of total proteins, albumin, urea,creatinin, cholesterol, triglycerides and activity of g-glutamyltransferase were determined by a Vitros 250 analyzer (OrthoClinical Diagnostics, Issy les Moulineaux, France) at theVeterinary School of Toulouse (France).2.5 HistologyThe tissue pieces were dehydrated through graded alcoholsand embedded in paraffin wax. Sections of 3 mm werestained with hematoxylin–eosin (HE) for histopathologicalevaluation. For each organ, three slides per animal wereprepared for analysis, and an area of 2000–2500 mm 2 perslide was observed. As displayed in Table 2, microscopicobservations led to the identification of different lesions inthe different organs, and allowed for establishing a lesionscore per animal. Based on a recent method published [23],we calculated the lesion score by taking into account thedegree of severity (severity factor) and the extent of eachlesion (according to intensity or observed frequency, scoredfrom 0 to 3). For each lesion, the score of the extent wasmultiplied by the severity factor. For each tissue, the minimalscores were 0 and the maximal scores were 21, 33 and15 for liver, lung and kidney, respectively (Table 2).2.6 Measurement of hepatocyte proliferationThe cellular proliferation activity was assessed by countingKi-67-positive nuclei on formalin-fixed embedded liversections as already described [24]. Briefly, the sections wereincubated with the primary antibody (Zymed (South SanFrancisco, CA, USA) Ki-67 Clone 7B11 – diluted 1:50) at 41Covernight in a humidity chamber; then the secondarywww.mnf-journal.com

These are not the final page numbers4 B. Grenier et al. Mol. Nutr. Food Res. 2011, 55, 1–11antibody (Kit Super Picture TM Zymed) was applied andfollowed by the addition of a chromogen (3,3 0 -diaminobenzidine).Finally, the tissue sections were counterstainedwith hematoxylin and mounted under coverslips using apermanent mounting medium.The number of Ki-67-positive nuclei among the total of100 nuclei was counted on the sections under light microscopyat 40 magnification. The proliferative index wascalculated by Ki-67-positive cells/total cells 100.Table 2. Establishment of a lesion score – endpoints used toassess histological lesions a)Tissue Type of lesions (severity factor) Maximal scoreLiver Disorganization of hepatic cords (1)Hepatic cell vacuolization (1)Apoptosis (2) 21Megalocytosis (2)Nuclear vacuolation (1)Lung Alveolar edema (2)Interstitial pneumonia (2)BALT depletion (2) 33Hypertrophy muscle cell (2)Hemorrhage (2)Vascular congestion (1)Kidney Nuclear change (1)Mitosis (1)Cytoplasmic vacuolization (1) 15Tubular casts (1)Congestion (1)a) The score for each lesion was obtained by multiplying theseverity factor with the extent of the lesion. The organ scorewas then obtained by the sum of each lesion score. Severityfactor (or degree of severity), 1 5 mild lesions, 2 5 moderatelesions; the extent of each lesion (intensity or observedfrequency) was evaluated and scored as 0 5 no lesion, 1 5 lowextent, 2 5 intermediate extent, 3 5 large extent.2.7 Measurement of total and specificimmunoglobulin subsetsThe total concentration of the immunoglobulin subsets wasmeasured by ELISA as already described [25]. Briefly, thedifferent isotypes were detected with the appropriateperoxidase anti-pig IgA or IgG (Bethyl, Interchim, Montlucon,France) and were quantified by reference to standardcurves constructed with known amounts of pig immunoglobulinclasses. Titers of specific antibody anti-OVA werealso measured by ELISA [14]. Briefly, the anti-OVA antibodieswere detected with peroxidase-labeled anti-pig IgG orIgA (Bethyl). Absorbance was read at 450 nm using anELISA plate reader (Spectra thermo, Tecan, NC, USA) andthe Biolise 2.0 data management software.2.8 Determination of lymphocyte proliferative indexLymphocyte proliferation was measured on blood samplescollected at different times of the experimental period. Thequantification was performed in 96-well plates as alreadydescribed [15, 26]. The results were expressed as stimulatingindex of lymphocyte proliferation calculated as counts perminute in stimulated culture/cpm in control non-stimulatedculture.2.9 Determination of the expression of mRNAencoding for cytokines by real-time PCRTissue RNA was processed in lysing matrix D tubes (MPBiomedicals, Illkirch, France) containing guanidine-thiocyanateacid phenol (Extract-All s , Eurobio, les Ulis, France) for use withthe FastPrep-24 (MP Biomedicals). Concentrations, integrityand quality of RNA were determined spectrophotometrically(OD 260 ) using Nanodrop ND1000 (Labtech International, Paris,Table 3. Nucleotide sequences of primers for real-time PCR a)Gene Primer sequence Genbank no. ReferencesRPL32 F (300 nM) TGCTCTCAGACCCCTTGTGAAG NM_001001636 [46]R (300 nM) TTTCCGCCAGTTCCGCTTAb2-microglobulin F (900 nM) TTCTACCTTCTGGTCCACACTGA NM_213978 [27]R (300 nM) TCATCCAACCCAGATGCAIL-12p40 F (300 nM) GGTTTCAGACCCGACGAACTCT NM_214013 [27]R (900 nM) CATATGGCCACAATGGGAGATGIL-8 F (300 nM) GCTCTCTGTGAGGCTGCAGTTC NM_213867 Present studyR (900 nM) AAGGTGTGGAATGCGTATTTATGCIL-1b F (300 nM) GAGCTGAAGGCTCTCCACCTC NM_001005149 [27]R (300 nM) ATCGCTGTCATCTCCTTGCACMIP-1b F (300 nM) AGCGCTCTCAGCACCAATG AJ311717 Present studyR (300 nM) AGCTTCCGCACGGTGTATGIL-6 F (300 nM) GGCAAAAGGGAAAGAATCCAG NM_214399 Present studyR (300 nM) CGTTCTGTGACTGCAGCTTATCCa) RPL32, ribosomal protein L32.& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.mnf-journal.com

These are not the final page numbersMol. Nutr. Food Res. 2011, 55, 1–11 5France). Besides this inspection, a 200 ng sample of RNA wasanalyzed by electrophoresis. The reverse transcription and realtimePCR steps were performed as already described [26]. RNAnon-reverse transcripted was used as the non-template controlfor verification of a no genomic DNA amplification signal.Specificity of PCR products was checked out at the end of thereaction by analyzing the curve of dissociation. In addition, thesize of amplicons was verified by electrophoresis. The sequencesof the primers used are detailed in Table 3. Primers for macrophageinflammatory protein-1b (MIP-1b), IL-8 and IL-6 detectionwere designed using Primer Express s software (AppliedBiosystems, Courtaboeuf, France). Primers were purchasedfrom Invitrogen (Cergy Pontoise, France). Amplification efficiencyand initial fluorescence were determined by DataAnalysis for Real Time-PCR method; then values obtained werenormalized by both house-keeping genes, b2-microglobulin andribosomal protein L32, and finally, gene expression wasexpressed relative to the control group as already described [27].2.10 StatisticsFollowing the Fisher test on equality of variances, one-wayANOVA was used to analyze the differences between thedifferent groups of animals at each time point. p-Values of0.05 were considered significant.Table 4. Individual or combined effects of DON and FB on weightgain a)Bodyweightgain/day(kg)Days1–14Days14–35Animal dietsControl DON FB DON1FB0.3670.05 a 0.3570.03 a 0.4370.05 a 0.3270.07 a0.7670.05 a 0.6570.03 a 0.7470.06 a 0.6870.03 aa) Results are expressed as mean7SEM for five animals.3 Results3.1 Individual or combined effects of DON and FB onweight gain, hematological and biochemicalparametersDuring the experiment, piglets were weighed weekly and asreported in Table 4, ingestion of individual or combinedDON- and FB-contaminated diets did not significantlyimpair animal growth.At the end of the experiment, blood samples were takenfrom all piglets to investigate the effects of mycotoxins onhematological and biochemical variables (Tables 5 and 6).Piglets fed either FB- or FB1DON-contaminated dietsdisplayed a significant decrease in neutrophil number(Table 5). An increase in creatinin concentration (p 5 0.047)and a decrease in albumin concentration (p 5 0.015) wasalso observed in the animal groups fed with FB- or DONcontaminateddiets, respectively. These alterations were notobserved in animals fed the diet contaminated with bothtoxins (Table 6).3.2 Individual or combined effects of DON and FB onorgan histopathologyLiver, lungs and kidneys were collected at the end of thetrial for histopathological analysis. The lesions observedin these three organs were mild to moderate for animal fedany of the three contaminated diets (DON, FB, DON1FB)(Fig. 1).The main histological lesions observed in the livers werea disorganization of hepatic cords, cytoplasmatic andnuclear vacuolization of hepatocytes and megalocytosis(Figs. 1A and B). Piglets fed either DON- or FB-contaminateddiets displayed significant liver lesions whencompared to animals fed the control diet. The lesion scorewas further increased for animals fed diet contaminatedwith both toxins. The proliferation of hepatocytes wasassessed by counting Ki-67-positive cells in liver sections.Table 5. Individual or combined effects of DON and FB on hematological parameters a)Hematological parameters Animal diets (wk 6)Control DON FB DON7FBWhite blood cells (thousands/mL) 21.271.9 a 19.672.3 a 20.372.8 a 18.271.6 aLymphocytes (thousands/mL) 12.471.9 a 11.471.4 a 14.772.1 a 12.671.0 aNeutrophils (thousands/mL) 7.371.1 a 7.071.1 a,b 4.570.9 b 4.670.6 bRed blood cells (thousands/mL) 6.270.3 a 5.770.2 a 6.170.4 a 5.970.4 aMean corpuscular volume (fL) 47.670.8 a 47.170.7 a 46.270.5 a 50.471.9 aHematocrit (%) 29.871.6 a 27.070.5 a 28.071.9 a 29.571.6 aHemoglobin (g/dL) 9.670.5 a 9.070.2 a 9.470.5 a 9.770.5 aMean corpuscular hemoglobin (pg) 15.470.2 a 15.670.2 a 15.670.2 a 16.570.8 aMean corpuscular hemoglobin concentration (%) 32.470.4 a 33.270.3 a 33.870.6 a 32.870.4 aa) Results are expressed as mean7SEM for six animals. Values in rows with different letters are significantly different.& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.mnf-journal.com

These are not the final page numbers6 B. Grenier et al. Mol. Nutr. Food Res. 2011, 55, 1–11Table 6. Individual or combined effects of DON and FB on biochemical parameters a)Biochemical parameters Animal diets (wk 6)Control DON FB DON1FBUrea (mmol/L) 3.870.4 a 3.370.4 a 4.270.3 a 4.070.4 aCreatinin (mmol/L) 102.575.3 a 98.074.1 a 120.575.6 b 101.675.5 aCholesterol (mmol/L) 2.670.2 a 2.470.2 a 2.370.1 a 2.370.1 aTriglycerides (mmol/L) 0.5170.07 a 0.3470.04 a 0.3970.06 a 0.4170.06 aTotal proteins (g/L) 59.871.0 a 57.172.1 a 59.972.5 a 57.672.5 aAlbumin (g/L) 34.370.7 a 29.271.5 b 35.172.1 a 32.872.1 a,bGGT (IU/L) 65.478.6 a 88.6714.4 a 79.4715.0 a 77.0711.5 aa) GGT, g-glutamyl transferase. Results are expressed as mean7SEM for five animals. Values in rows with different letters aresignificantly different.The mean proliferation indexes were 16.471.5 in thecontrol group, 18.873.3 in the DON-treated group,22.871.7 in the FB-treated group and 39.4712.8 in theDON1FB-treated group (po0.001, po0.01 and po0.05, forcomparison between DON1FB and control, DON or FBgroups, respectively).In the lungs, depletion of bronchiole-associated lymphoidtissue (BALT) and vascular disorders (peribronchiolar, alveolarhemorrhage and congestion) were the most frequentobserved lesions (Fig. 1C). Of note, BALT structures werechecked and were present in all individual pigs, evaluated in acomparable size and area between experimental groups.Alveolar edema showed a focal presentation (Fig. 1D). Asdemonstrated by the lesion scores, lung lesions were onlyobserved in animals receiving FB- or FB1DON-contaminateddiets. In this latter group, a medial hypertrophy of pulmonaryarterioles was observed in half of the animals.Lesions in the kidneys were mild as indicated by lowlesion scores. The main observed lesions were degenerativechanges in tubular epithelial cells (vacuolization of thecytoplasm and nucleus, Figs. 1E and F) and interstitialinfiltration of lymphocytes with a focal or multifocal pattern.These lesions were observed in animals receiving dietscontaminated with DON, FB and both toxins.3.3 Individual or combined effects of DON and FB onthe immune responseThe main objective of this study was to assess the individualand combined effects of DON and FB on the immuneresponse in piglets. Ingestion of diets contaminated withindividual or combined mycotoxins neither altered the totalplasmatic concentration of IgG and IgA nor modulated thelymphocyte proliferation upon concanavalin A stimulation(data not shown).The immunization protocol with OVA allowed us toinvestigate the effects of mycotoxins on antigen-specificimmunity [14, 26]. The ingestion of diets contaminated withDON or FB individually or in combination significantlyaltered the production of immunoglobulins after OVAvaccination (Fig. 2). Animals fed mycotoxin-contaminateddiets displayed a reduced anti-OVA IgG concentration intheir plasma. However, because of high individual variability,the decrease was only significant for animals receivingFB-contaminated feed. This decrease was furtherpronounced for animals fed the diet containing both toxins.Concerning the effect of mycotoxins on the specific IgAconcentration, as expected, we observed a significantincrease of this immunoglobulin isotype in piglets fed theDON-contaminated diet. However, when DON was fed incombination with FB, the increase of plasmatic-specific IgAconcentration was not observed (Fig. 2).As already observed [14, 26], piglets receiving the controldiet displayed a significant increase in the lymphocyteproliferation upon OVA stimulation after the secondimmunization (1.4-fold increase, p 5 0.191; 3.3-foldincrease, p 5 0.012 and 2.8-fold increase, p 5 0.020 at days21, 28 and 35 of the experiment, respectively). By contrast,the lymphocyte proliferation upon OVA stimulation in theanimals receiving any of the three contaminated diets(DON, FB and DON1FB) remained as low as in controlunstimulated lymphocytes (Fig. 3).3.4 Individual or combined effects of DON and FB onthe expression of cytokinesCytokines play a key role in regulating both humoraland cell-mediated immunity. The mRNA expression of fivecytokines (IL-12p40, IL-8, IL-1b, IL-6 and MIP-1b) wasmeasured by real-time RT-PCR in spleen samples collectedat the end of the experiment (Fig. 4). Animals fed thediet containing both DON and FB demonstrated a significantdecrease in mRNA for all tested cytokines whencompared to the control pigs (p 5 0.009 for IL-8; p 5 0.035for IL-1b; p 5 0.004 for IL-6; p 5 0.031 for IL-12p40;p 5 0.006 for MIP-1b). Animals fed the diet contaminatedwith DON demonstrated a significant decrease in mRNAencoding for IL-8, whereas animals fed the diet contaminatedwith FB demonstrated a significant decrease inmRNA encoding for IL-1b and IL-6.& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.mnf-journal.com

These are not the final page numbersMol. Nutr. Food Res. 2011, 55, 1–11 7LIVER76cABLesional score54321abb,c0LUNGKIDNEYCDLesional scoreLesional scoreE F a9b,c c87654 a,b3a2103 bb2b10Control DON FB DON+FBAnimal treatmentsFigure 1. Individual and combinedeffects of DON and FB onliver, lungs and kidneys. Pigsreceived a control diet (&), or aDON-contaminated diet ( ), oran FB-contaminated diet ( )or a diet contaminated withboth toxins (&). (A) Hepatocytecytoplasmatic vacuolizationand (B) hepatocytemegalocytosis (arrow). HE40 . (C) BALT depletion andperibronchiolar hemorrhage.HE 10 and (D) Alveolaredema. HE 40 . (E) Cytoplasmaticvacuolization of tubularcells and mitosis (arrow) and(F) nuclear change (arrow)in tubular cells. HE 40 .Lesion scores were establishedafter histological examinationaccording to the severity andthe extent of the lesions. Valuesare mean7SEM for fiveanimals.4 DiscussionIn the present 5-wk study, piglets were exposed to low dosesof two major Fusarium mycotoxins, DON and FB, at levelscommonly found in crops. Most of the current dataconcerning the effects of DON or FB on animals, includingrodents, have been obtained using highly mono-contaminatedfeeds [9, 12, 13, 28]. It was thus of interest to determinethe effect of ingestion of feeds contaminated with lowlevel of these toxins, present alone or in combination, onzootechnical, hematological, biochemical, histopathologicaland immune parameters of piglets.We did not observe any effect of mycotoxin-contaminateddiets (DON, FB, DON1FB) on the body weight gain ofthe animals. Considering the low contamination levelswe used, these results are not surprising. Indeed, noeffect on body weight gain has been reported in pigsand in poultry fed with up to 70 mg FB/kg feed [18, 29]. Theeffects of DON on body weight gain are more controversial,especially in pigs. Some studies indicate that dietaryconcentrations of DON above 1–2 mg/kg have an effect onweight gain, whereas in other studies no effect is observedfor up to 4.5 mg DON/kg feed [30]. A weight gain reductionhas also been described when DON and FB were giventogether to growing barrows [18]. However, in this study, thedose of FB was ten-fold higher than the one used in thepresent experiment.Exposure of piglets to low doses of either DON or FB didnot have a major impact on the hematological andbiochemical parameters investigated. For blood hematology,only a reduction in neutrophil numbers was noticed in FBexposedpiglets. This observation is in relation with the& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.mnf-journal.com

These are not the final page numbers8 B. Grenier et al. Mol. Nutr. Food Res. 2011, 55, 1–11Arbitrary UnitsArbitrary Units200015001000500050403020100Specific IgGSpecific IgAaa,breduced viability measured in human neutrophils exposedin vitro to this toxin [31]. For blood biochemistry, there was adecrease in albumin concentration in DON-exposed animalsand an increase in creatinin concentration in FB-exposedpiglets in accordance with previously published studies[18, 20, 32, 33]. Ingestion of diets co-contaminated withDON and FB had less effect on hematology and biochemistryparameters than did mono-contaminated diets. Somestudies have already reported a weaker effect on plasmabiochemical parameters for piglets fed multi-contaminateddiets than for piglets receiving mono-contaminated feeds[18, 19], which suggests an opposite effect of the twomycotoxins.Despite the absence of effects on zootechnical, hematologicaland biochemical parameters, ingestion of feedscontaminated with low concentrations of either DON or FBinduced histopathological lesions in liver, lungs andkidneys. Toxic effects of FB on liver have been reported inseveral papers using highly contaminated materials [9, 28].The effects included a disorganization of hepatic cords,hepatocellular vacuolation, megalocytosis, apoptosis,a1 7 14 21 28 35Length of exposure (days)Figure 2. Individual and combined effects of DON and FB onplasma concentrations of specific immunoglobulin (IgA and IgG)anti-OVA. Pigs received a control diet (J), or a DON-contaminateddiet ( ), or an FB-contaminated diet ( ) or a contaminateddiet with both toxins (K). At days 4 and 16 of the trial,animals receiving either control or contaminated feeds weresubcutaneously immunized with OVA. Plasma samples werecollected weekly and the levels of IgA and IgG specific for OVAwere determined by ELISA and normalized against a standardizedreference plasma. Values are mean7SEM for five animals.Statistics are mentioned when significant changes wereobserved.& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimbaa,bbbba,ba,baaa,bbbbaaaaa,bb,ccbaaaLymphocytes proliferative index654310Ovalbumin stimulation2 b1 7 14 21 28 35Length of exposure (days)Figure 3. Individual and combined effects of DON and FB onlymphocyte-specific (OVA) proliferation. Pigs received a controldiet (J), or a DON-contaminated diet ( ), or an FB-contaminateddiet ( ) or a contaminated diet with both toxins (K). At days 4and 16 of the trial, animals were subcutaneously immunizedwith OVA. Blood samples were taken weekly to measure thelymphocyte proliferation. Results are expressed as stimulatingindex of lymphocyte proliferation calculated as counts perminute in stimulated culture/cpm in control non-stimulatedculture. Values are mean7SEM for five animals. Statistics arementioned when significant changes were observed.necrosis and cell proliferation. In the present study, it wasobserved that even when present at a lower dose, FBinduced similar liver histopathological lesions. Liverlesions, such as hepatic cell vacuolation, were alsoobserved in piglets fed the DON-contaminated diet [34].These lesions were not associated with major biochemicalalterations. The biological meaning of the hepatic lesionsremains to be determined. Histopathological analysis oflung confirmed that this is a target organ for FB. At highdoses (Z92 mg/kg of feed for 4–7 days), FB induce lethalpulmonary edema in swine [9]. In the present study, the lowdose of FB also induced pulmonary damages, mainly BALTdepletion and vascular disorders. By contrast, when presentat a low dose in the diet, DON did not induce any lesion inthe lung.For the three organs investigated, the damages elicitedfrom the ingestion of the diet co-contaminated with DONand FB were equal to or higher than the ones elicited by theingestion of a single mycotoxin. Very few publicationsanalyzed the effects of mixed mycotoxins on histopathologicalparameters, especially at low doses [35, 36]. Thehistopathological lesions observed in the lungs ofco-exposed piglets were slightly more pronounced than theones observed in the lungs of FB-exposed animals. In theliver, ingestion of the co-contaminated diet induced significantlyhigher lesions than ingestion of either of the monocontaminatedfeeds as demonstrated by the lesion score andthe hepatocyte proliferation. One explanation for the highliver toxicity of DON and FB when present simultaneouslycould be the higher absorption of FB in the presence ofDON. Indeed, DON has recently been shown to decrease thebarrier function of the intestine [37]. Thus, ingestion ofabbwww.mnf-journal.comab b b

These are not the final page numbersMol. Nutr. Food Res. 2011, 55, 1–11 9Cytokine expression levels (A.U)IL 12 p40IL 81.41.2 a1.2a,b aa1.01.00.8b a,b0.8b0.6 b0.60.40.40.20.20.00.0IL 1βIL 61.21.21.2aa1.0a,b1.0a,b1.0a0.8b b 0.8b,c0.80.60.6c0.60.40.40.40.20.20.20.0 0.00.0Control DON FB DON+FB Control DON FB DON+FB ControlAnimal treatmentsMIP 1βaaFBbDON+FBFigure 4. Individual andcombined effects of DON andFB on splenic mRNA expressionof cytokines. Pigs receiveda control diet (&), or a DONcontaminateddiet ( ), or anFB-contaminated diet ( )oradiet contaminated with bothtoxins (&). Quantification ofthe relative cytokine mRNAlevel for each sample isexpressed in arbitrary units(A.U). Values are mean7SEMfor five animals.DON may increase the absorption of FB, mycotoxins alreadyknown to be poorly absorbed [7, 9].The main objective of this study was to investigate theeffect of low doses of DON or FB ingested separately or incombination on the immune response of piglets. As inprevious experiments, it was observed that at low doses,mycotoxins have little or no effect on the total non-specificimmune responses as measured by lymphocyte proliferationupon mitogenic stimulation and the plasmatic concentrationof immunoglobulin classes. Immunization protocols,as already described, were needed to observe an effect of lowdoses of mycotoxins, fed either alone or in combination onthe immune responses [14, 26, 38].A very low proliferative index, close to the one observedin unstimulated cells, was obtained in cells isolated fromanimals fed either DON-, FB- or DON1FB-contaminateddiets. This alteration of lymphocyte proliferation might bedue to an effect of these toxins on antigen-presenting cells(APC) as suggested by recent in vitro studies on monocytederivedAPC treated with DON [39, 40] or in vivo studieswith piglets acutely exposed to FB [27].Interestingly, the diet co-contaminated with DON and FBappeared to be able to counteract the increased level ofspecific IgA observed in the animal receiving only the DONcontaminateddiet. Indeed, consumption of theDON-contaminated diet increased the level of specific IgA inthe plasma [11, 14] whereas ingestion of diet contaminatedwith both DON and FB did not alter the plasma level of thisimmunoglobulin isotype. We can hypothesize that FBinterfere with the DON-induced IgA elevation at theintestinal level through its action on sphingolipids. Indeed,FB are known to disrupt the sphingolipid metabolismleading to depletion of ceramide and all ceramide-derivedcomplex sphingolipids, such as sphingomyelin [41, 42]. Thislatter compound has been recently reported to control theamount of IgA in the large intestine [43].Depending on the mycotoxin, DON or FB significantlyimpaired the specific IgG concentration and the level ofcytokine expression. Nonetheless, the diet co-contaminatedwith DON and FB led to a strong decrease of specific IgGconcentration, greater than the one observed in animalsreceiving only one toxin. Similar effects were observed forthe five cytokines investigated, where the impact of the cocontaminateddiet was higher than either of the monocontaminateddiets. Several studies investigated cytokineexpression during chronic exposure to mycotoxins [14, 15,25, 27], but none of them concern the co-contamination.Cytokines are important mediators in the immuneresponse. Expressions of IL-8 and MIP-1b, which areinvolved in cell chemotaxis, were significantly inhibited inanimals fed the co-contaminated diet, and it can be anticipatedthat in these animals, recruitment and migration ofAPC to peripheral lymphoid tissue were reduced. Similarly,the decreased mRNA levels of IL-1b and IL-6 mRNA inpiglets receiving the co-contaminated diet may lead to adefective antigen presentation and an impaired activation oflymphocytes and may explain the decreased IgG responseobserved in this study.Find a mechanism that explains the observed effects afterthe combination of both toxins is not easy, but at the cellularlevel, it might be hypothesized that MAPK activation couldbe involved. Indeed, both DON and FB have been shown toactivate MAPKs [12, 44], and these kinases are well known tomodulate numerous physiological processes, such as cellgrowth, apoptosis or immune response [45].In conclusion, chronic exposure to low doses of DON orFB, either alone or in combination did not elicit importantclinical signs (body weight gain, hematology, biochemistry),but induced microscopic lesions and altered the immuneresponse, especially when the mycotoxins were fed incombination. The modulation of the immune response wasonly observed when the immune system was activated.Considering (i) that vaccination or infection by pathogens isa common situation encountered in animal husbandry and(ii) the natural occurrence of these mycotoxins in feedstuffs,the present experiment suggests a significant disruption in& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.mnf-journal.com

These are not the final page numbers10 B. Grenier et al. Mol. Nutr. Food Res. 2011, 55, 1–11the establishment of an appropriate specific response inanimals receiving mycotoxin-contaminated diets. This studyalso highlights the complexity of mycotoxin interactions;some effects are not enhanced by the combination of toxins(biochemistry, lung and kidney lesions, specific IgAcontent), while others are (specific IgG content, cytokinesexpression, liver lesions). These results may have someimpact on the current regulation/recommendation that onlytakes into account individual mycotoxins and not multimycotoxincontamination.B.G. was supported by a doctoral fellowship (CIFRE 065/2007), jointly financed by the Biomin company, ANRT (AssociationNationale de la Recherche Technique) and INRA(Institut National de la Recherche Agronomique). This studywas supported in part by a CAPES/COFECUB Grant (No.593/08) and a CNDT Grant (No. 472048/2008-2). Theauthors thank M. Kainz and E. Pichler from Quantas AnalytikGmbH and M. Sulyok from IFA-Tulln for mycotoxin analysis,G. H.aubl and G. Jaunecker from Biopure (Romer Labs) formycotoxin production, G. Guillemois from INRA Rennes for hisassistance with feed manufacture, P. Pinton, J. Laffitte,R. Solinhac and M. Gallois for technical assistance during theanimal experiments and Dr. Mike Watkins for his help with theEnglish text.The authors have declared no conflict of interest.5 References[1] CAST, Council for Agricultural Science and Technology,Mycotoxins, Risks in Plant, Animal, and Human System,Task Force Report 139, Ames Iowa 2003.[2] Oswald, I. P., Comera, C., Immunotoxicity of mycotoxins.Rev. Med. Vet. 1998, 149, 585–590.[3] Binder, E. M., Tan, L. M., Chin, L. J., Handl, J., Richard, J.,Worldwide occurrence of mycotoxins in commodities,feeds and feed ingredients. Anim. Feed Sci. Technol. 2007,137, 265–282.[4] Placinta, C. M., D’Mello, J. P. F., Macdonald, A. M. C.,A review of worldwide contamination of cereal grains andanimal feed with Fusarium mycotoxins. Anim. Feed Sci.Technol. 1999, 78, 21–37.[5] Schothorst, R. C., van Egmond, H. P., Report from SCOOPtask 3.2.10 ‘‘collection of occurrence data of Fusariumtoxins in food and assessment of dietary intake by thepopulation of EU member states’’ – Subtask: trichothecenes.Toxicol. Lett. 2004, 153, 133–143.[6] Monbaliu, S., Van Poucke, C., Detavernier, C., Dumoulin, F.et al., Occurrence of mycotoxins in feed as analyzed by amulti-mycotoxin LC-MS/MS method. J. Agric. Food Chem.2010, 58, 66–71.[7] Voss, K. A., Smith, G. W., Haschek, W. M., Fumonisins:toxicokinetics, mechanism of action and toxicity. Anim.Feed Sci. Technol. 2007, 137, 299–325.[8] Halloy, D. J., Gustin, P. G., Bouhet, S., Oswald, I. P., Oralexposure to culture material extract containing fumonisinspredisposes swine to the development of pneumonitiscaused by Pasteurella multocida. Toxicology 2005, 213,34–44.[9] Haschek, W. M., Gumprecht, L. A., Smith, G., Tumbleson,M. E., Constable, P. D., Fumonisin toxicosis in swine: anoverview of porcine pulmonary edema and currentperspectives. Environ. Health Perspect. 2001, 109, 251–257.[10] Oswald, I. P., Desautels, C., Laffitte, J., Fournout, S. et al.,Mycotoxin fumonisin B-1 increases intestinal colonizationby pathogenic Escherichia coli in pigs. Appl. Environ.Microbiol. 2003, 69, 5870–5874.[11] Pestka, J. J., Smolinski, A. T., Deoxynivalenol: toxicologyand potential effects on humans. J. Toxicol. Env. HealthCrit. Rev. 2005, 8, 39–69.[12] Zhou, H. R., Islam, Z., Pestka, J. J., Rapid, sequential activationof mitogen-activated protein kinases and transcriptionfactors precedes proinflammatory cytokine mRNAexpression in spleens of mice exposed to the trichothecenevomitoxin. Toxicol. Sci. 2003, 72, 130–142.[13] Pestka, J. J., Zhou, H. R., Moon, Y., Chung, Y. J., Cellularand molecular mechanisms for immune modulation bydeoxynivalenol and other trichothecenes: unraveling aparadox. Toxicol. Lett. 2004, 153, 61–73.[14] Pinton, P., Accensi, F., Beauchamp, E., Cossalter, A. M.et al., Ingestion of deoxynivalenol (DON) contaminated feedalters the pig vaccinal immune responses. Toxicol. Lett.2008, 177, 215–222.[15] Accensi, F., Pinton, P., Callu, P., Abella-Bourges, N. et al.,Ingestion of low doses of deoxynivalenol does not affecthematological, biochemical, or immune responses ofpiglets. J. Anim. Sci. 2006, 84, 1935–1942.[16] Kouadio, J. H., Dano, S. D., Moukha, S., Mobio, T. A.,Creppy, E. E., Effects of combinations of Fusarium mycotoxinson the inhibition of macromolecular synthesis,malondialdehyde levels, DNA methylation and fragmentation,and viability in Caco-2 cells. Toxicon 2007, 49,306–317.[17] Marzocco, S., Russo, R., Bianco, G., Autore, G., Severino, L.,Pro-apoptotic effects of nivalenol and deoxynivalenoltrichothecenes in J774A.1 murine macrophages. Toxicol.Lett. 2009, 189, 21–26.[18] Harvey, R. B., Edrington, T. S., Kubena, L. F., Elissalde, M. H.et al., Effects of dietary fumonisin B-1-containing culturematerial, deoxynivalenol-contaminated wheat, or theircombination on growing barrows. Am. J. Vet. Res. 1996, 57,1790–1794.[19] Kubena, L. F., Edrington, T. S., Harvey, R. B., Buckley, S. A.et al., Individual and combined effects of fumonisin B1present in Fusarium moniliforme culture material and T-2toxin or deoxynivalenol in broiler chicks. Poult. Sci. 1997,76, 1239–1247.[20] Marin, D. E., Taranu, I., Pascale, F., Lionide, A. et al., Sexrelateddifferences in the immune response of weanlingpiglets exposed to low doses of fumonisin extract. Br.J. Nutr. 2006, 95, 1185–1192.& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.mnf-journal.com

These are not the final page numbersMol. Nutr. Food Res. 2011, 55, 1–11 11[21] Leslie, J. F., Plattner, R. D., Desjardins, A. E., Klittich, C. J. R.,Fumonisin B1 production by strains from different matingpopulations of Gibberella-Fujikuroi (Fusarium SectionLiseola). Phytopathology 1992, 82, 341–345.[22] Sulyok, M., Krska, R., Schuhmacher, R., A liquid chromatography/tandemmass spectrometric multi-mycotoxinmethod for the quantification of 87 analytes and itsapplication to semi-quantitative screening of moldyfood samples. Anal. Bioanal. Chem. 2007, 389,1505–1523.[23] Kolf-Clauw, M., Castellote, J., Joly, B., Bourges-Abella, N.et al., Development of a pig jejunal explant culture forstudying the gastrointestinal toxicity of the mycotoxindeoxynivalenol: histopathological analysis. Toxicol. Vitro2009, 23, 1580–1584.[24] Makino, H., Togo, S., Kubota, T., Morioka, D. et al., A goodmodel of hepatic failure after excessive hepatectomy inmice. J. Surg. Res. 2005, 127, 171–176.[25] Taranu, I., Marin, D. E., Bouhet, S., Pascale, F. et al.,Mycotoxin fumonisin B-1 alters the cytokine profile anddecreases the vaccinal antibody titer in pigs. Toxicol. Sci.2005, 84, 301–307.[26] Meissonnier, G. M., Pinton, P., Laffitte, J., Cossalter, A. M.et al., Immunotoxicity of aflatoxin B1: Impairment of the cellmediatedresponse to vaccine antigen and modulation ofcytokine expression. Toxicol. Appl. Pharmacol. 2008, 231,142–149.[27] Devriendt, B., Gallois, M., Verdonck, F., Wache, Y. et al.,The food contaminant fumonisin B-1 reduces thematuration of porcine CD11R1(1) intestinal antigenpresenting cells and antigen-specific immune responses,leading to a prolonged intestinal ETEC infection. Vet. Res.2009, 40, 40.[28] Voss, K. A., Riley, R. T., Norred, W. P., Bacon, C. W. et al., Anoverview of rodent toxicities: liver and kidney effects offumonisins and Fusarium moniliforme. Environ. HealthPerspect. 2001, 109, 259–266.[29] Broomhead, J. N., Ledoux, D. R., Bermudez, A. J., Rottinghaus,G. E., Chronic effects of fumonisin B-1 in broilersand turkeys fed dietary treatments to market age. Poult. Sci.2002, 81, 56–61.[30] Etienne, M., Wache, Y., in: Oswald, I. P., Taranu, I. (Eds.),Mycotoxins in Farm Animals, Research Signpost, Kerala2008, pgs. 113–130.[31] Odhav, B., Adam, J. K., Bhoola, K. D., Modulating effects offumonisin B1 and ochratoxin A on leukocytes andmessenger cytokines of the human immune system. Int.Immunopharmacol. 2008, 8, 799–809.[32] Bergsjo, B., Langseth, W., Nafstad, I., Jansen, J. H., Larsen,H. J. S., The effects of naturally deoxynivalenol-contaminatedoats on the clinical condition, blood parameters,performance and carcass composition of growing pigs. Vet.Res. Commun. 1993, 17, 283–294.[33] Rotter, B. A., Thompson, B. K., Lessard, M., Effects ofdeoxynivalenol-contaminated diet on performance andblood parameters in growing swine. Can. J. Anim. Sci.1995, 75, 297–302.[34] Zielonka, L., Wisniewska, M., Gajecka, M., Obremski, K.,Gajecki, M., Influence of low doses of deoxynivalenol onhistopathology of selected organs of pigs. Pol. J. Vet. Sci.2009, 12, 89–95.[35] Chen, F., Ma, Y. L., Xue, C. Y., Ma, J. Y. et al., The combinationof deoxynivalenol and zearalenone at permitted feedconcentrations causes serious physiological effects inyoung pigs. J. Vet. Sci. 2008, 9, 39–44.[36] Tiemann, U., Brussow, K. P., Kuchenmeister, U., Jonas, L.et al., Influence of diets with cereal grains contaminated bygraded levels of two Fusarium toxins on selected enzymaticand histological parameters of liver in gilts. Food Chem.Toxicol. 2006, 44, 1228–1235.[37] Pinton, P., Nougayrede, J. P., Del Rio, J. C., Moreno, C.et al., The food contaminant deoxynivalenol, decreasesintestinal barrier permeability and reduces claudin expression.Toxicol. Appl. Pharmacol. 2009, 237, 41–48.[38] Marin, D. E., Gouze, M. E., Taranu, I., Oswald, I. P., FumonisinB1 alters cell cycle progression and interleukin-2synthesis in swine peripheral blood mononuclear cells.Mol. Nutr. Food Res. 2007, 51, 1406–1412.[39] Bimczok, D., Doll, S., Rau, H., Goyarts, T. et al., The Fusariumtoxin deoxynivalenol disrupts phenotype and functionof monocyte-derived dendritic cells in vivo and in vitro.Immunobiology 2007, 212, 655–666.[40] Wache, Y. J., Hbabi-Haddioui, L., Guzylack-Piriou, L.,Belkhelfa, H. et al., The mycotoxin Deoxynivalenol inhibitsthe cell surface expression of activation markers in humanmacrophages. Toxicology 2009, 262, 239–244.[41] Loiseau, N., Debrauwer, L., Sambou, T., Bouhet, S. et al.,Fumonisin B-1 exposure and its selective effect on porcinejejunal segment: sphingolipids, glycolipids and transepithelialpassage disturbance. Biochem. Pharmacol. 2007,74, 144–152.[42] Soriano, J. M., Gonzalez, L., Catala, A. I., Mechanism ofaction of sphingolipids and their metabolites in the toxicityof fumonisin B1. Prog. Lipid Res. 2005, 44, 345–356.[43] Furuya, H., Ohkawara, S., Nagashima, K., Asanuma, N.,Hino, T., Dietary sphingomyelin alleviates experimentalinflammatory bowel disease in mice. Int. J. Vitam. Nutr.Res. 2008, 78, 41–49.[44] Pinelli, E., Poux, N., Garren, L., Pipy, B. et al., Activation ofmitogen-activated protein kinase by fumonisin B-1 stimulatescPLA(2) phosphorylation, the arachidonic acidcascade and cAMP production. Carcinogenesis 1999, 20,1683–1688.[45] Dong, C., Davis, R. J., Flavell, R. A., MAP kinases in theimmune response. Annu. Rev. Immunol. 2002, 20, 55–72.[46] Pinton, P., Braicu, C., Nougayrede, J. P., Laffitte, J. et al.,Deoxynivalenol impairs porcine intestinal barrier functionand decreases the protein expression of claudin-4 through amitogen activated protein kinase dependent mechanism.J. Nutr. 2010, 140, 1956–1962.& 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheimwww.mnf-journal.com

TRAVAIL EXPERIMENTAL2. Toxicité in vivo du déoxynivalénol et des fumonisines, seuls ou encombinaison sur la morphologie et la réponse locale de l’intestinL’expérimentation animale présentée dans la première étude a également fait l’objet derecherches sur l’effet des quatre régimes sur l’intestin. Des échantillons d’intestin grêle ont étéprélevés lors de l’autopsie des animaux, et des analyses histologiques et immunologiques ont étéréalisées.Les résultats histologiques ont montré une augmentation du nombre de lésions après l’ingestiondes mycotoxines, au niveau du jéjunum et de l’iléon. L’effet observé sur les cellules en mitosepourrait expliquer l’aplatissement des villosités dans la partie jéjunale. Néanmoins, ces conclusionssur les analyses histologiques ont été principalement confirmées pour le régime mono-contaminéavec le DON. Et il semblerait que la présence de FB ait un effet antagoniste sur l’effet du DON dans lerégime co-contaminé. Ce même type d’interaction a été observé sur la diminution du nombre decellules caliciformes dans le jéjunum.L’analyse de l’expression des cytokines par RT-PCR temps réel a montré de manière intéressante,une augmentation du niveau des ARN dans le jéjunum et l’iléon. L’augmentation de l’IL-6 causée parle DON pourrait être corrélée à l’élévation du titre en IgA spécifiques observée dans la premièreétude. L’augmentation générale des ARNs codant pour diverses cytokines tout au long de l’intestingrêle, suggère un état inflammatoire chronique suite à l’exposition continue aux régimes contaminéspendant les cinq semaines. Dans ce sens, il a été reporté dans la littérature que lors d’inflammationsdigestives chroniques, l’expression exacerbée de certaines cytokines (IL-1β, TNF-α et IFN-γ) induisaitune plus grande perméabilité intestinale. Basé sur ces observations et sur les récents travaux denotre équipe sur l’altération de l’expression de protéines de jonctions par le DON, nous avons ainsimontré qu’in vivo, les régimes contaminés diminuaient l’expression protéique de l’E-cadhérine et del’occludine au niveau de l’iléon; un effet additif étant obtenu pour le régime co-contaminé.107

TRAVAIL EXPERIMENTALChronic ingestion of deoxynivalenol and fumonisin B1 induce morphologicaland immunological changes in the intestine of pigletsJoelma LUCIOLI 1,2 , Bertrand GRENIER 2,3 , Anne-Marie COSSALTER 2 , Wulf-Dieter MOLL 3 ,SCHATZMAYR 3 , Isabelle P. OSWALD 2 & Ana-Paula F. R. LOUREIRO-BRACARENSE 1Gerd1Universidade Estadual de Londrina, Lab. Patologia Animal, Londrina, Brazil.2INRA, ToxAlim, Equipe Immuno-Mycotoxicologie, Toulouse, France.3BIOMIN Research Center, Technopark 1, Tulln, Austria.Address correspondence toPr. Ana-Paula F. R. L. BracarenseLab. Patologia Animal, Universidade Estadual de LondrinaCP 6001CEP 86051-990, Londrina, PR, BrazilPhone : 55 43 3371-4062E-Mail : anauel02@yahoo.com.brAbbreviations: AJ, adherens junctions ; DON, deoxynivalenol ; FB, fumonisins ; HE, hematoxylin-eosin; TJ, tight junctions ; WB, western blottingKeywords: co-contamination, deoxynivalenol, fumonisins, intestinal tract, morphology, immunology108

TRAVAIL EXPERIMENTALABSTRACTAlthough low levels of multiple mycotoxins are frequently encountered in commodities, there is alack of data on the potential risk of mycotoxin multi-contamination, especially at low doses.Deoxynivalenol (DON) and Fumonisins (FB) are both produced by Fusarium species and naturally cooccurin animal diets. Individually, these mycotoxins exert numerous toxicological effects accordingto doses ingested, and are therefore of concern in terms of animal health. Because the gastrointestinaltract represents the first barrier met by exogenous food/feed compounds, the purpose ofthis study was to investigate the effects of DON and FB, alone and in combination on some intestinalparameters, including morphology, expression of tight junction or cytokines expression. Twenty-four5-wk-old piglets were randomly assigned to four different groups, receiving separate diets for 5weeks, a control diet, a diet contaminated with either DON (3mg/kg) or FB (6mg/kg) or both toxins.Chronic ingestion of these contaminated diets induced histological changes, as shown by thedecreased villi height and cell proliferation in jejunum, and by the reduced number of goblet cells inboth jejunum and ileum. The low doses used potentiated the cytokines secretion all along theintestine, especially IL-1β and TNF-α. Upregulation of these two cytokines could be the cause of thedefective expression observed in the ileal segment for two junction proteins, occludin and E-cadherin. This alteration in cell adhesion suggests that consumption of DON and FB could predisposeanimals to enteric infection through impairment of the intestinal barrier function. To conclude, theintestinal response observed for the co-contaminated diet was similar to the response to monocontaminateddiets for the majority of parameters evaluated, indicating no synergistic effect.109

TRAVAIL EXPERIMENTALINTRODUCTIONMycotoxins are secondary metabolites of various fungi commonly found in feed and foodstuffs.Based on their known and suspected effects on human and animal health, aflatoxin, fumonisin,deoxynivalenol, ochratoxin A and zearalenone are recognized as the five most important agriculturalmycotoxins (Shephard, 2008). The toxic effects of Fusarium mycotoxins in animals include reducedgrowth, feed refusal, immunosupression, gastrointestinal lesions, and neurological and reproductivedisorders (Rocha et al., 2005).Recent surveys demonstrated regular occurrence of low levels of multiple mycotoxins in cereals(Roscoe et al., 2008; Tabuc et al., 2009; Scudamore and Patel, 2009). Also, mycotoxins may bepresent in grains in conjugated chemical forms that escape detection through conventional analyticalmethods (Zhou et al., 2007), resulting in an underestimation of the total amount of mycotoxins andfeed toxicity. Fusarium mycotoxins in combination can exert more pronounced effects in animalsthan individual mycotoxins (Smith et al., 1997; Grenier et al., 2011).The intestinal tract is the first barrier against ingested antigens, including mycotoxins andpathogenic bacteria. Following ingestion of mycotoxin-contaminated food, enterocytes may beexposed to high concentrations of toxins (Bouhet and Oswald, 2005). Studies focusing on theinfluence of food-derived antigens on intestinal morphology as an indicator of animal health arecommon; meanwhile, there are few publications on the effects of chronic exposure to a mycotoxinco-contaminated diet.Fumonisins (FB) are toxic and carcinogenic mycotoxins produced by Fusarium verticillioides, acommon contaminant of maize. Fumonisin B1 (FB1) causes porcine pulmonary edema and equineleukoencephalamalacia. An association between human esophageal cancer and FB1 exposure indeveloping countries has been reported (Zhang et al., 1997). Fumonisins are structurally related tosphingoid bases and cause inhibition of ceramide synthase, leading to accumulation of correspondingfree sphingoid bases, sphingoid base metabolites, and depletion of more complex sphingolipids(Wang et al., 1991). In vitro, FB1 induces apoptosis, necrosis, and inhibition of proliferation in pigrenal epithelial (LLC-PK1) cells and human colonic cells (HT29) (Gopee et al., 2003; Schmelz et al.,1998). FB1 can impair the intestinal absorption of nutrients. This can be explained by the villousfusion and atrophy observed in the intestines of pigs treated with 30 mg FB1/kg of feed and/or bysphingolipid disturbances (Taranu et al., 2005).Deoxynivalenol (DON) causes toxic and immunotoxic effects in a variety of cell systems andanimal species. DON is produced by F. graminearum and F. culmorum mainly in wheat, barley andmaize (Pestka and Smolinski, 2005). Swine is more sensitive to DON than other species, in part110

TRAVAIL EXPERIMENTALbecause of differences in the metabolism of DON. Chronic low dietary concentrations induceanorexia, decreased weight gain, and immune alterations, while acute higher doses induce vomiting,hemorrhagic diarrhea and circulatory shock (Rotter et al., 1996a). At the cellular level, the maineffect is inhibition of protein synthesis via binding to the ribosomes. Low exposure to DON wasshown to upregulate expression of cytokines and inflammatory genes with concurrent immunestimulation, whereas high exposure promoted leukocyte apoptosis associated with immunesuppression (Pestka et al., 2004).The purpose of this study was to compare the effects of low doses of DON and FB in pigs whenfed individually and in combination with particular emphasis on their effects on the intestine. Theexperimental design was a factorial assay including control feed and feed contaminated with 3 and 6mg/kg DON and FB individually and in combination, respectively. These contamination levelscorrespond to levels that frequently occur naturally in cereals. We investigated the effect of DON andFB1 on intestine morphology, on the expression of tight junction as well as on the intestinalexpression of cytokines.111

TRAVAIL EXPERIMENTALMATERIAL & METHODS1) Animals and dietsA total of 24 crossbred weaned piglets were used in this study. Animals were kept in batch pensfor 35 days. Pigs were acclimatized for 1 week in the animal facility of the INRA Laboratory ofPharmacology and Toxicology (Toulouse, France) prior to being used in experimental protocols. Feedand water were provided ad libitum throughout the experimental period. The animals weresubmitted to one of four dietary treatments for 35 days: control non-contaminated diet, dietcontaining 2.8 mg of deoxynivalenol/kg (DON) of feed, diet containing 5.9 mg of fumonisins (4.1 mgFB1 + 1.8 mg FB2)/kg of feed and diet containing 3.1 mg of DON plus 6.5 mg FB (4.5 mgFB1 + 2.0 mgFB2)/kg of feed. The diets were artificially contaminated with the mycotoxins. The diet formulationsand nutrient contents were described elsewhere (Grenier et al., 2011).The experimental design used in this study was entirely randomized with six repetitions (eachanimal represented one repetition). At the end of the experiment, pigs were submitted to electricalstunning and euthanized by exsanguination. The institutional Ethics Committee for AnimalExperimentation approved the study.2) Histological assessmentAt postmortem examination, samples from the jejunum and ileum were collected from all groupsand fixed in 10% buffered formalin. The tissue pieces were then dehydrated through graded alcoholsand embedded in paraffin wax. Sections of 3 µm were stained with hematoxylin-eosin (HE) forhistopathological evaluation. To evaluate mucus production, sections of intestine were stained withalcian blue. Positively stained goblet cells were counted randomly in five fields per sample at 40xmagnification, and the means were submitted to statistical analysis.Villi height and crypt depth were measured randomly on thirty villi using a MOTIC Image Plus 2.0ML ® image analysis system. The numbers of lymphocytes, plasma cells, and eosinophils were countedrandomly on three fields per sample at 40x magnification. The number of mitotic figures was countedin 20 fields per slide using 40x magnification. The means of intestinal morphometry, number ofgoblet cells, inflammatory infiltrate and mitosis were utilized for statistical analysis.3) Immunohistochemical assessmentE-cadherin expression was analyzed on formalin-fixed embedded intestinal sections to evaluateintestinal cell adherens junctions. Tissue sections were deparaffinized with xylene and dehydrated112

Table 10 : Origins and dilutions of primary antibodies used for detecting tight junction proteins andβ-actin by Western blottingProtein Origin DilutionE-Cadherin (24E10) Rabbit mAbRabbit anti-Occludin (672381A)β-actin mAB MOUSE (8H10D10)Cell Signaling Technology® - Massachusetts, USA(ref. 3195)Invitrogen Corporation® - California, USA(ref. 71-1500)Cell Signaling Technology® - Massachusetts, USA(ref. 3700)1:5001:5001:1000Table 11 : Nucleotide sequences of primers for real-time PCRGENE PRIMER SEQUENCE GENBANK NO. REFERENCESRPL32F (300 nM) TGCTCTCAGACCCCTTGTGAAGR (300 nM) TTTCCGCCAGTTCCGCTTANM_001001636 Pinton et al.(2010)β2-μglobulin F (900 nM) TTCTACCTTCTGGTCCACACTGAR (300 nM) TCATCCAACCCAGATGCANM_213978Devriendt etal. (2009)IL-12p40F (300 nM) GGTTTCAGACCCGACGAACTCTR (900 nM) CATATGGCCACAATGGGAGATGNM_214013Devriendt etal. (2009)IL-8F (300 nM) GCTCTCTGTGAGGCTGCAGTTCR (900 nM) AAGGTGTGGAATGCGTATTTATGCNM_213867Grenier et al.(2011)IL-1βF (300 nM) GAGCTGAAGGCTCTCCACCTCR (300 nM) ATCGCTGTCATCTCCTTGCACNM_001005149 Devriendt etal. (2009)MIP-1βF (300 nM) AGCGCTCTCAGCACCAATGR (300 nM) AGCTTCCGCACGGTGTATGAJ311717Grenier et al.(2011)IL-6IFN-γTNF-αIL-2IL-10F (300 nM) GGCAAAAGGGAAAGAATCCAGR (300 nM) CGTTCTGTGACTGCAGCTTATCCF (300 nM) TGGTAGCTCTGGGAAACTGAATGR (300 nM) GGCTTTGCGCTGGATCTGF (300 nM) ACTGCACTTCGAGGTTATCGGR (300 nM) GGCGACGGGCTTATCTGAF (300 nM) GCCATTGCTGCTGGATTTACR (300 nM) CCCTCCAGAGCTTTGAGTTCF (300 nM) GGCCCAGTGAAGAGTTTCTTTCR (300 nM) CAACAAGTCGCCCATCTGGTNM_214399NM_213948NM_214022AY294018NM_214041Grenier et al.(2011)Royaee et al.(2004)Meissonnier etal. (2008)Meurens et al.(2008)Present study

TRAVAIL EXPERIMENTALthrough a graded ethanol series. The sections were placed in a microwave-resistant container andimmersed in EDTA buffer (pH 9.0). Endogenous peroxidase activity was blocked by incubation inmethanol/H 2 O 2 solution. After adding the primary antibody, the secondary antibody (Kit SuperPicture TM Zymed) was applied. The sections were incubated with the primary antibody (Zymed anti-EcadherinClone 4A2C7 - diluted 1:50) at 4ºC overnight, followed by the addition of a chromogen (3,3′-diaminobenzidine). Finally, the tissue sections were counterstained with hematoxylin and mountedon coverslips using a permanent mounting medium. Tissue sections were examined, and theproportion of samples expressing E-cadherin was estimated. Each sample was assessed as showingeither normal or reduced staining. Normal staining was considered when a homogeneous and strongbasolateral membrane staining of enterocytes was detected. Heterogeneous and weak staining wasconsidered to indicate reduced expression.4) Western BlottingProteins were extracted from ileum and assayed as described previously (Schaffner andWeissmann, 1973). Briefly, the extraction was carried out on ice in extraction buffer. The proteaseinhibitors (antipain, pepstatin, benzamidine, AEBSF, aprotinine and leupeptin) were added to theextraction buffer just before use. Extracts tissue proteins were then separated by SDS-PAGEelectrophoresis. Equal amounts of proteins were loaded on a 12,5% acrylamide gel. Migration wasconducted in a 250 mM Tris buffer (pH 7.6) containing 1% SDS and 1,92 M Glycine. After separation,proteins were transferred onto Optitran BA-S 83 membrane (Whatman®, Germany). The primaryantibodies against E-Cadherin, Occludin and β-actin used in this study are presented in Table 10.Expression of β-actin was used for checking the equal protein load across gel tracks. Band densitieswere obtained by scanning the membranes using Odyssey® Infrared Imaging System (LI-CORBiosciences, USA). Density data were standardized within membranes by expressing the density ofeach band of interest relative to that of β-actin in same lane.5) Determination of the expression of mRNA encoding for cytokines by real-time PCRTissue RNA was processed in lysing matrix D tubes (MP Biomedicals, Illkirch, France) containingguanidine-thiocyanate acid phenol (Extract-All®, Eurobio, les Ulis, France) for use with the FastPrep-24 (MP Biomedicals, Illkirch, France). Concentration, integrity and quality of RNA were determinedspectrophotometrically (O.D. 260 ) using Nanodrop ND1000 (Labtech International, Paris, France). Inaddition to this inspection, 200 ng of RNA was analyzed by electrophoresis. The reverse transcriptionof 2 µg of total RNA was performed using M-MLV reverse-transcriptase, Rnasin® plus (Promega,113

TRAVAIL EXPERIMENTALCharbonnière, France) and random primers (Invitrogen, Cergy Pontoise, France) (5 min at 37°C, 1hour at 42°C, 15 min at 70°C) as already described (Oswald et al., 2001). Real-time PCR assays wereperformed on 8 ng of cDNA (RNA equivalent) in a 25-µl volume reaction per well using Power SYBR®Green PCR Master Mix as the reporter dye and the automated photometric detector ABI Prism 7000Sequence Detection System for data acquisition (Applied Biosystems, Courtaboeuf, France). Theamplification conditions were as follows: 95°C for 10 min followed by 40 cycles of 95°C for 15 sec and60°C for 1 min. RNA non-reverse transcript was used as a non-template control (NTC) for verificationthat no genomic DNA amplification signal existed. Specificity of PCR products was checked at the endof the reaction by analyzing the curve of dissociation. In addition, the size of amplicons was verifiedby electrophoresis. The sequences of the primers used are detailed in Table 11. Primers for MIP-1β,IL-8 and IL-6 detection were designed using Primer Express® software (Applied Biosystems). Primerswere purchased from Invitrogen (Cergy Pontoise, France). Amplification efficiency and initialfluorescence were determined by the DART-PCR method (Peirson et al., 2003), and the valuesobtained were then normalized by two housekeeping genes, beta2-μglobulin and ribosomal proteinL32 (RPL32); and finally, gene expression was calculated relative to the control group as alreadydescribed (Le Gall et al., 2009).6) Statistical analysisData were analyzed with Statview software, version 5.0 (SAS Institute Inc, Cary, NC), usingANOVA, Tukey and PLSD Fisher test. P values < 0.05 were considered significant.114

Figure 5ABCDVilli height (µm)5432abaaaaaa10JEJUNUMILEUMEFLesion score65432bbbbbb1aa0JEJUNUMILEUM

Figure 60246810121416JEJUNUMILEUMGOBLET CELLS051015202530JEJUNUMILEUMaLYMPHOCYTES0510152025303540JEJUNUMILEUMPLASMA CELLS0246810121416JEJUNUMILEUMEOSINOPHILSControl DON FB DON+FBbaa,ba,baabbba,baa,baabbaa,baba,baaba,baaba,baa,b

TRAVAIL EXPERIMENTALRESULTS1) Histomorphometrical analysisSamples of jejunum and ileum were collected for histomorphometrical analysis. Piglets fed dietscontaminated with mycotoxins showed mild to moderate intestinal lesions. The main histologicalchanges observed were lymphatic vessel dilation, mild eosinophil granulation within the cytoplasm ofenterocytes, and prominent lymphoid follicles. The lesional score increased for animals fedcontaminated diets (Figure 5).Changes in villous height and crypt depth are indicative of enterocyte loss and impairedabsorption of nutrients. As shown in Figure 5, villi height decreased significantly in the jejunum of theanimals that received DON and DON+FB when compared with controls. Focal apical necrosis of villi(Figures 5C and 5D) was also observed in the intestine of piglets fed mycotoxins diets. No change incrypt depth was observed in any intestinal region. Goblet cells synthesize and secrete mucin, which isinvolved in gut physiology. The number of goblet cells decreased (Figures 5F and 6) significantly inthe DON- and FB+DON-treated animals in the jejunum and ileum.Inflammatory infiltrate of lymphocytes, plasma cells and eosinophils was observed in all regions ofthe intestine. In the groups treated with mycotoxins, a reduction in lymphocytic infiltrate wasobserved in all regions of the intestine. However, only the jejunum was significantly affected (Figure6). The number of eosinophils in the lamina propria decreased significantly in the ileum of the groupthat received DON+FB. The mean number of inflammatory cells per field in each region of theintestine is summarized in Figure 6.2) Cell proliferationCell proliferation was estimated by counting the number of mitosis figures in enterocytes on HEslides. The mean number of mitosis figures in the jejunum were 2.36 ± 1.64 in the control group, 1.73± 1.35 in the DON treated group, 1.66 ± 1.11 in the FB treated group and 1.91 ± 1.19 in the DON+FBtreated group. In the ileum the mean number were 1.75 ± 1.26, 1.78 ± 1.46, 1.62 ± 1.17 and 1.89 ±1.11 for the control group, DON treated group, FB treated group and DON+FB treated group,respectively. A significant decrease (p

Figure 7JEJUNUMC y t o k i n e e x p r e s s i o n l e v e l s ( A . U.)2.01.61.20.80.40.01.61.20.80.40.01.41.21.00.80.60.40.2IL-10a, baMIP-1βbaIL-4a aba, baba, ba1.61.20.80.40.02.52.01.51.00.50.02.52.01.51.00.5IFN-γaIL-2aIL-8aa, bbaba, baa, ba, ba2.01.61.20.80.40.02.52.01.51.00.50.02.01.61.20.80.4IL-1βbaIL-12baTNF-αaaaa, baa, ba, ba3.02.52.01.51.00.50.0IL-6baaaCont DON FB DON+FB0.0Cont DON FB DON+FB0.0Cont DON FB DON+FB0.0Cont DON FB DON+FBA n i m a l t r e a t m e n t s

Figure 8ILEUM2.01.6TNF-αbbb2.52.0IL-1βbbb2.52.0IL-6b1.2a1.5a1.5aaaC y t o k i n e e x p r e s s i o n l e v e l s ( A . U.)0.80.40.01.61.41.21.00.80.60.40.20.01.61.41.21.00.80.60.40.20.01.00.50.0IL-12a2.0 IL-10aaa a1.6aa1.2a0.80.40.0IL-2MIP-1β2.0aaaaaa1.6a1.2 a0.80.41.00.50.02.0IL-8a a1.6aa1.20.80.40.02.5IFN-γa2.0a1.5aa1.00.50.00.0Cont DON FB DON+FB Cont DON FB DON+FB Cont DON FB DON+FBA n i m a l t r e a t m e n t s

Figure 9IMMUNOHISTOCHEMICAL ANALYSISWESTERN BLOT ANALYSISE-CadherinOccludinABβ-actinDON+FB1***E-cadherin**Band intensity (A.U.)1.21.00.80.60.40.20.01.41.21.00.80.60.40.20.0aa,bE-cadherinbbab b OccludinbControl DON FB DON+FBFB1DONControlDON+FB1FB1DON**0% 20% 40% 60% 80% 100%++++*********Strong andhomogeneous stainingWeak andheterogeneous stainingE-cadherinStrong andhomogeneous stainingWeak andheterogeneous stainingAnimal treatmentControl0% 20% 40% 60% 80% 100%

TRAVAIL EXPERIMENTAL3) Intestinal immune responseTo evaluate the mechanisms of porcine intestinal defense against multi-mycotoxin exposure, wequantified the expression of genes coding for pro-inflammatory cytokines after chronic ingestion ofmycotoxin co-contaminated feed. Statistically significant changes in gene expression in differentregions relative to control and treated animals are shown in Figures 7 and 8.In general, the jejunum showed greater induction of cytokines as more cytokines were expressedin this region after mycotoxin exposition. DON treated group demonstrated significant increase inmRNA encoding for IL-1β, IL-6, MIP-1β, IL-2, and IL-12p40 in the jejunum, whereas FB induced asignificant increase in IL-10 and IFN-γ. In the ileum, all treatments induced significant increases inTNF-α and IL-1β.4) Expression of adherens junction proteinsThe expression of E-cadherin and occludin, two junction proteins involved in the adherence andpermeability of enterocytes, was analyzed in the ileum of animals by western blotting (WB). Afternormalization by the housekeeping gene β-actin, the data revealed a significant decrease ofexpression of both proteins compared to the control animals. In animals exposed to the cocontaminateddiet, the defective expression of E-cadherin and occludin was more pronounced(Figure 9).As WB indicated a significant decrease in the total amount of E-cadherin expression in the ileum,we decided to evaluate the expression of this protein in enterocytes, using an immunohistochemicalassay. The expression of E-cadherin in the jejunum and ileum was significantly reduced in the groupsthat received mono-contaminated or co-contaminated diets (Figure 9). The reduction was mostpronounced in the ileum of the DON+FB group, in which all animals showed weak and irregularstaining.116

TRAVAIL EXPERIMENTALDISCUSSIONCo-contamination of grains and feed is frequently reported by analytical laboratories around theworld. In fact, the occurrence of single-mycotoxin contamination seems to be rare (Kubena et al.,1997). However, most studies in pigs have employed mono-contaminated diets. Fumonisins (FB) anddeoxynivalenol (DON) exert effects on different mechanisms in the intestinal tract. FB blockssphingolipid synthesis, which is essential for the formation of cell membranes, while DON inhibitsprotein synthesis via ribosome binding (Scott, 1993). Enterocytes have a continuous replicating cycle,making the intestine a main target for the toxins’ action (Bouhet et al., 2004). Considering the cocontaminationof FB and DON in feed and the intestine as a main target for toxin action, it isimportant to know whether these mycotoxins have additive, synergistic or antagonistic effects in theintestines of pigs.The main histological findings observed were villi flattening and a reduction in the number ofgoblet cells. Decreased villi height was significant only in the jejunum with the diet monocontaminatedwith DON. Similar changes were observed in in vivo and ex vivo studies with DONexposition (Awad et al., 2006; Obremski et al., 2008; Kolf-Clauw et al., 2009). The mode of toxicaction of DON is inhibition of protein synthesis, thus primarily affecting rapidly dividing cells such asthose of the gastrointestinal tract and immune system (Leeson et al., 1995). The villi flattening in thejejunum is likely due to impairment of cell proliferation, as could be observed by the decrease in thenumber of mitotic figures in the same region.Swine are considered as possessing an intermediate level of sensitivity to fumonisins, and levelsunder 10 mg/kg in the total ration are considered safe (USFDA, 2001). There were no differences invilli height or crypt depth between the FB group and the control group. The absence of effects in theintestine of the FB group is probably related to the low dose (6 mg/kg) used, which is associated withpoor transport of this mycotoxin across the epithelium of the intestine (Bouhet and Oswald, 2007).Villous fusion and atrophy were observed in the intestines of pigs treated with diets containing highlevels of FB1 (Dilkin et al., 2004; Piva et al., 2005).Mucin production can be estimated by the number of goblet cells in the intestinal wall.Considering that goblet cells originate from intestinal stem cells, the decrease observed here couldbe explained by the inhibitory effect of DON on protein synthesis in cells under proliferation. Indomestic animals, hyperplasia of intestinal goblet cells has been observed in piglets receiving 10 to30 mg FB1/kg of feed for 4 weeks and in broiler chicks receiving 300 mg FB1/kg of feed for 2 weeks(Dilkin et al., 2004; Brown et al., 1992). In the present study, a decrease in the number of goblet cellswas shown in the DON-contaminated and co-contaminated diets in the jejunum, while in the FB-117

TRAVAIL EXPERIMENTALtreated group there was no difference. Although not significant, it seems that an antagonisticinteraction between FB and DON affects the number of goblet cells as FB interfered with DON’seffect of decreasing the number of goblet cells.Controversial results have been reported with respect to intestinal proliferation in in vivo and invitro studies of mycotoxicosis. In this study, we have evaluated cell proliferation by counting mitoticfigures in intestinal crypts, which showed a significant decrease in the jejunum of the FB and DONtreatedgroups. Theumer et al. (2002) observed that subchronic experimental FB1 mycotoxicosisincreased the number of mitotic figures in rat intestinal crypts, while Obremski et al. (2008) observednumerous dividing cells within the glandular epithelium in pigs fed a diet containing DON,zearalenone and T-2 toxin. Using in vitro models, Bouhet et al. (2004) verified a decrease in theproliferation of undifferentiated porcine epithelial cells treated with FB1 due to a blockade in theG0/G1 cell cycle phase. Because DON acts as a protein synthesis inhibitor, it is expected that cellswith high turnover such as crypt enterocytes will be a target for the toxin’s action. Recently, it wasshown in Caco-2 cells that DON causes a concentration-dependent decrease in total protein contentassociated with a reduction in the incorporation of [3H]-leucine, demonstrating its inhibitory effecton protein synthesis (Van de Walle et al., 2010). The modes of action of FB and DON to inhibit cellproliferation are quite different. FB lead to accumulation of free sphingoid bases that are proapoptotic,cytotoxic and growth cell inhibitors (Lalles et al., 2010), while DON acts as a proteinsynthesis inhibitor.Mononuclear and eosinophil inflammatory infiltrate has been reported in FB mycotoxicosis inseveral species. In addition, proliferation of lymphoid nodules in the ileum and cecum was alsoobserved (Theumer et al., 2002; Piva et al., 2005). We have shown that the number of lymphocytesin the lamina propria decreased significantly in the jejunum of the treated groups, mainly in the DONtreatment. Studies of macrophages and lymphocytes have shown that the trichothecene-mediatedimmunosuppressive effect was associated with induction of apoptosis (Rocha et al., 2005). DONinitiates its toxic effects by inducing ribotoxic stress, which activates c-jun terminal kinase and p38mitogen-activated protein kinase, stimulating apoptosis. DON was particularly potent in this respect,elevating caspase levels more than twelvefold (Shifrin and Anderson, 1999). Because lymphoid cellsare constantly renewing, lymphocytes could be particularly sensitive to DON. It is interesting that ourresults show that even low doses of both mycotoxins induced a decrease in lymphocyte infiltration.This effect also occurred in the jejunum along with other histological alterations, supporting thefinding that this region is a main target for toxin action.In the present study, the chronic treatment of piglets with FB, DON or both by the oral routeinduced activation of the proinflammatory cytokine network in the intestine. Increases in the mRNA118

TRAVAIL EXPERIMENTALlevels of the nine cytokines evaluated (TNF-α, MIP-1β, IFN-γ, IL-1β, IL-2, IL-6, IL-8, IL-10, IL-12p40)were observed in the jejunum and ileum. A systemic intestinal cytokine mRNA profile indicative ofmacrophage and TH1 activation has been reported after a single oral dose of DON in mice(Azconaolivera et al., 1995). It has been established that the intestine has its own immune network,which can cause localized induction of various cytokines and chemokines (Stadnyk, 2002). Low-doseexposure to trichothecenes generally results in stimulatory effects, causing increased resistance topathogens, elevated serum IgA levels (mediated by IL-6) and up-regulated expression of genesencoding for cytokines and chemokines; on the other hand, high-dose exposure causesimmunosuppression, characterized by decreased resistance to pathogens, reduced IgM and IgGlevels and impaired and delayed hypersensitivity responses (Pestka et al., 2004; Shifrin andAnderson, 1999; Rocha et al., 2005). This diverse spectrum of effects is likely to result fromdifferences in the intensity and duration of kinase signaling and the resultant gene expression(Pestka, 2010).Although all of the cytokines showed elevated expression, only TNF-α and IL-1β showed increasesin all treatments. A dose-dependent increase in TNF-α mRNA levels was observed in fumonisin B1-treated murine peritoneal macrophages and in the liver (Dugyala et al., 1998; Bhandari et al., 2002).TNF-α and IL-1β are known to stimulate the production of IL-8 and other cytokines (Fiers, 1991;Azconaolivera et al., 1995), but also to induce apoptosis via the receptor-ligand-mediated mechanism(Van Cruchten and Van den Broeck, 2002). We hypothesize that, besides the known apoptoticmechanisms, FB and DON could induce TNF-mediated lymphocyte apoptosis in the intestine, whichcould explain the decrease in the number of these cells observed in exposed pigs. A relationshipbetween clinically relevant concentrations of TNF-α (1–10 ng/ml) and IL-1β and an increase inintestinal tight junctions (TJ) permeability has been demonstrated in Caco-2 cells, mediated by anincrease in myosin light chain kinase (MLCK) protein expression (Ye et al., 2006; Al-Sadi et al., 2008).With regard to this association, we can consider that the increased levels of TNF-α and IL-1β verifiedafter the ingestion of DON and FB1 could also contribute to TJ intestinal barrier defects.In a previous study, we observed alterations in enterocyte TJ in in vitro and in vivo DON exposuredue to reduced expression of claudins (Pinton et al., 2009). Considering that paracellularpermeability of the intestinal epithelium is regulated by intercellular TJ and adherens junctions (AJ),we wish to know if other transmembrane proteins, as occludin and E-cadherin were also affected bymycotoxins exposition. The AJ, which is immediately subjacent to the TJ, comprises thetransmembrane protein E-cadherin and the associated cytoplasmic proteins, the catenins, and playsan important role in the formation of TJs.119

TRAVAIL EXPERIMENTALThe potencial effects of DON, FB and both mycotoxins on AJ function were analyzed usingwestern blotting and immunohistochemical assays. DON and DON+FB treatments altered totalcellular amount of E-cadherin, whereas all treatments induced changes in the amount of occludin. Areduction of E-cadherin and occludin, therefore, suggests a loss of enterocytes’ adhesive propertiesthat permits increased intestinal permeation by toxic luminal antigens, promoting intestinalinflammation. However, to the best of the author’s knowledge, no data have been reportedconcerning reduced expression of E-cadherin in the intestinal tract after ingestion of a mycotoxinmono- or co-contaminated diet. It has been suggested that derangement of the apical junctioncomplex of the intestinal epithelium may be involved in the generation of an aberrant immuneresponse due, for example, to a loss of epithelial cell polarity or an abnormal delivery of antigens viaa paracellular pathway (Hershberg and Mayer, 2000; Soderholm et al., 2002).Considering that the intestine is the first tissue involved in the absorption of mycotoxins, a proinflammatoryeffect, as we observed in this study, is expected as a consequence of TJ barrierdysfunction. Increased intestinal permeability could allow luminal antigens to enter into the mucosa,inducing proinflammatory cytokine expression. Considering that TNF-α, IFN-γ and IL-1β causeddecreases in ZO-1 and occludin protein expression in Caco-2 cells (Han et al., 2003; Cui et al., 2010),we can hypothesize that increased levels of these cytokines also affect E-cadherin expression.Multi-contamination with low doses of mycotoxins is more likely to occur in naturalcontamination, but there are few data indicating the effects of co-contaminated mycotoxin diets inpigs. Taken together, the present data provide strong evidence that chronic ingestion of low doses ofmycotoxins induces histological changes in an in vivo model, potentiates cytokines secretion, andaffects the expression of proteins involved in cell adhesion, suggesting that ingestion of DON, FB andboth toxins could predispose animals to infections by enteric pathogens through alteration ofintestinal barrier function. Moreover, the response observed for the co-contaminated diet wassimilar to the response to mono-contaminated diets for the majority of the parameters evaluated,indicating no synergistic effect in the intestine.120

TRAVAIL EXPERIMENTALFIGURES LEGENDFigure 5 : Effect of individual and combined DON and FB exposure on jejunum and ileum histology.Pigs received a control diet ( ), or a DON-contaminated diet ( ), or a FB-contaminated diet ( ), ora contaminated diet with both toxins ( ).(A) Jejunum of a control piglet and (B) DON treated piglet. Villi flattening (arrow). HE. 10x. (C) Villiapical necrosis (arrow). HE. 10x and (D) Bacterial adhesion in the area with necrosis (arrow). HE. 40x.(E) Globet cells in a control piglet and (F) Decrease in the number of globet cells (arrow). Alcian-Blue.20x.Morphometry of villi height in the jejunum and ileum. Data are mean height (µm) ± SEM for 6 pigs.Lesional score after histological examination according to the occurrence and the severity of lesions.Values are mean scores for 6 pigs. Statistics are mentioned when significant changes were observed.Figure 6 : Effect of individual and combined DON and FB exposure on the number of inflammatorycells and globet cells in jejunum and ileum.Pigs received a control diet ( ), or a DON-contaminated diet ( ), or a FB-contaminated diet ( ), ora contaminated diet with both toxins ( ).Values are mean number of inflammatory and globet cells ± SEM for 6 pigs. Statistics are mentionedwhen significant changes were observed.Figure 7 : Effect of individual and combined DON and FB exposure on jejunum mRNA expression ofcytokines.Pigs received a control diet ( ), or a DON-contaminated diet ( ), or a FB-contaminated diet ( ), ora contaminated diet with both toxins ( ).Quantification of the relative cytokine mRNA level for each sample is expressed in arbitrary units(A.U). Values are mean ± SEM for 5 pigs. Statistics are mentioned when significant changes wereobserved.Figure 8 : Effect of individual and combined DON and FB exposure on ileum mRNA expression ofcytokines.Pigs received a control diet ( ), or a DON-contaminated diet ( ), or a FB-contaminated diet ( ), ora contaminated diet with both toxins ( ).121

TRAVAIL EXPERIMENTALQuantification of the relative cytokine mRNA level for each sample is expressed in arbitrary units(A.U). Values are mean ± SEM for 5 pigs. Statistics are mentioned when significant changes wereobserved.Figure 9 : Effect of individual and combined DON and FB exposure on intestinal expression of E-cadherin and occludin.Pigs received a control diet ( ), or a DON-contaminated diet ( ), or a FB-contaminated diet ( ), ora contaminated diet with both toxins ( ).Quantification of the relative protein level for each sample is expressed in arbitrary units (A.U).Values are mean ± SEM for 5 pigs.(A) Jejunum of a control piglet. Strong and homogeneous immunoreactivity to E-cadherin.Immunoperoxidase, 20x. (B) Jejunum of a piglet fed DON-contaminated diet. Weak immunoreactivityto E-cadherin. Immunoperoxidase, 10x.Quantification of the expression of immunoreactivity to E-cadherin (%) in intestinal slides. Statisticsare mentioned when significant changes were observed.122

CHAPITRE 2Évaluation de la toxicité du produitd’hydrolyse de la Fumonisine B1 chezle porcelet123

Figure 10 :(a) Réaction de deestérification par la carboxylestérase sur la Fumonisine B1 (FB1), résultant enFumonisine B1 totalement hydrolysée (HFB1)(b) Cinétique in vitro d’hydrolyse de la Fumonisine B1 par la carboxylestérase(a)FB1HFB1(b)600µM500400300200100FB1 µMHFB1 µM00 5 10 15 20Temps d'incubation (en heures)

TRAVAIL EXPERIMENTALRESUME DE L’ETUDEComme déjà souligné dans le présent manuscrit, les fumonisines et plus particulièrement lafumonisine B1 (FB1) sont des mycotoxines relativement stables et difficiles à éliminer par destraitements physiques et chimiques, et des phénomènes d’adsorption.Ainsi, notre partenaire industriel BIOMIN a développé au cours des dernières années uneapproche de détoxification de la fumonisine B1 par voie enzymatique. Cette réaction debiotransformation a été rendue possible après le screening et l’identification d’une souchebactérienne capable de dégrader complètement la FB1. Cette souche a initialement été isolée àpartir de compost et s’est révélé appartenir au genre Sphingopyxis. L’analyse du cluster de gènes decette souche a permis d’isoler un gène encodant pour une carboxylestérase, et responsable ducatabolisme de la FB1. Ainsi, après expression de ce gène en système hétérologue, une dégradationcomplète et rapide de la FB1 a été obtenue in vitro, résultant en la production de fumonisine B1totalement hydrolysée (Figure 10). De plus, ce processus de détoxification enzymatique semble agirindépendamment de la présence d’oxygène, tels que dans des conditions limitées en oxygène, àsavoir les fourrages ou le tractus intestinal des animaux.Afin d’évaluer la toxicité du produit d’hydrolyse de la FB1 (HFB1) et étant donné qu’aucune étuden’avait reporté l’effet de la HFB1 sur des animaux domestiques, nous avons traité oralement desporcelets durant deux semaines avec une solution tampon contrôle, une solution de HFB1 et unesolution avec la molécule mère FB1. La solution HFB1 a été obtenue après hydrolyse totale de la FB1,avec la carboxylestérase produite par BIOMIN. Afin de comparer la toxicité entre FB1 et HFB1, nousavons traité les animaux avec des doses élevées. En effet, la FB1 a été administrée à 2 mg/kg pv/jour,ce qui correspond dans nos conditions à une alimentation contaminée approximativement avec 40mg FB1/kg d’aliment. La HFB1 a quant à elle été administrée de façon équimolaire à 1,1 mg/kgpv/jour.La toxicité a principalement été évaluée sur le foie et sur le tractus gastro-intestinal. Peu dedonnées ont été rapportées sur l’effet de la FB1, et encore moins sur l’effet de la HFB1, sur l’intestin.Ainsi, cette étude apporte également de nouveaux éléments sur la toxicité intestinale aigüe de cessubstances. De manière générale, nos données confirment chez le porc que l’hydrolyse de la FB1réduit fortement sa toxicité. Tout d’abord, le ratio des bases sphingoides Sa/So (Sa, sphinganine etSo, sphingosine), considéré comme un bon marqueur de toxicité, était similaire à celui des animauxdu groupe contrôle, contrairement à ceux des animaux traités avec la FB1, et ce, quelque soit lamatrice biologique (plasma, foie, reins et poumons). Au niveau hépatique, les analyses n’ont pas124

TRAVAIL EXPERIMENTALrévélé de lésions microscopiques, et très peu d’effets sur des marqueurs de fonction etd’inflammation lorsque les porcelets avaient ingéré la solution de HFB1, à l’inverse de ceux ayantingéré la solution de FB1. De plus, l’intestin ne présentait pas ou très peu de lésions et d’altérationsdes villosités, contrairement aux intestins des animaux du groupe FB1. De la même manière,l’immunosuppression reflétée par une diminution de l’expression des cytokines, n’a pas été observéele long de l’intestin grêle après administration de la HFB1.125

TRAVAIL EXPERIMENTALHydrolysis of fumonisin B1 strongly reduced the toxicity for piglets at bothintestinal and hepatic levelsBertrand GRENIER 1,3 , Ana-Paula LOUREIRO-BRACARENSE 2 , Heidi Elisabeth SCHWARTZ 4 , Anne-MarieCOSSALTER 1 , Martine KOLF-CLAUW 5 , Gerd SCHATZMAYR 3 , Isabelle P. OSWALD 1 & Wulf-Dieter MOLL 31INRA, ToxAlim, Equipe Immuno-Mycotoxicologie, Toulouse, France.2Universidade Estadual de Londrina, Lab. Patologia Animal, Londrina, Brazil.3BIOMIN Research Center, Technopark 1, Tulln, Austria.4University of Natural Resources and Life Sciences, Center for Analytical Chemistry, Department forAgrobiotechnology IFA, Tulln, Austria.5Université de Toulouse, Ecole Nationale Vétérinaire, Toulouse, France.Address correspondence toDr Isabelle P. OswaldINRA-ToxAlim180 chemin de Tournefeuille BP 9317331027 Toulouse Cedex 3Phone : +33561285480E-Mail : isabelle.oswald@toulouse.inra.frAbbreviations: FB1, fumonisin B1 ; HFB1, hydrolyzed fumonisin B1 ; Sa, sphinganine ; So, sphingosineKeywords: fumonisin, hydrolyzed fumonisin, sphingoid bases, liver, digestive tract126

TRAVAIL EXPERIMENTALABSTRACTFumonisins, produced by Fusarium verticillioides, are of major worldwide concern in terms ofubiquitous and frequency distribution, and toxicity. The need to restrict in feeds the levels offumonisins, and thereby to ensure animal health and productivity, is a continuous challenge.Biotechnological approaches through microorganisms or enzymes can provide an alternativesolution. In this study, the comparative toxicity of fumonisin B1 (FB1), the predominant fumonisins inthis family, and its hydrolyzed form (HFB1) was investigated in swine over a short-term trial. Eighteen5-week-old animals were orally and daily treated for two weeks with tree different solutions, acontrol buffer solution, a FB1 solution (2 mg/kg body weight/day) and a HFB1 solution (equimolar to2 mg FB1/kg body weight/day). The HFB1 solution was initially obtained after a deesterificationreaction on FB1 solution, through the use of a carboxylesterase. Results were reported in terms ofliver and intestinal toxicity. Besides, exposure was assessed by determining the sphingoid basescontent in plasma and liver samples. On contrary to FB1, HFB1 did not trigger hepatotoxicty asrevealed by biochemical compounds, inflammatory markers and microscopical lesions. Similarly,HFB1 did not impair the intestinal integrity and intestinal immune system, as evaluated in differentsegments of gastro-intestinal tract by villi and lesions characterization and cytokines expression.These results are supported by the ratio of sphinganine and sphingosine in biological samples, whichwere not affected at intermediate and end points of the in vivo experiment. It can be concluded thathydrolysis of FB1 strongly reduced toxicity in piglets.127

TRAVAIL EXPERIMENTALINTRODUCTIONFumonisins are a family of mycotoxins produced by Fusarium verticillioides (formerly F.moniliforme), which are common fungal contaminants of corn and some other grains (Marasas,2001). Fumonisins exert complex biological effects. The toxic effects of fumonisins range fromhepatotoxicity and renal toxicity to species-specific effects such as pulmonary edema in pigs andleukoencephalomalacia in horses (Voss et al., 2007). In humans, exposure to fumonisins has beenlinked to esophageal cancer and neural tube defects (Voss et al., 2007; Waes et al., 2005). The effectof low doses has been lesser investigated but revealed reduced performance, pathologicalalterations of the lungs and an increase in intestinal colonization by opportunistic pathogenicbacteria in piglets (Rotter et al., 1996b; Halloy et al., 2005; Oswald et al., 2003; Haschek et al., 2001).Fumonisins are structurally similar to sphingoid bases, and were determined to be potentinhibitors of sphinganine N-acyl transferase (ceramide synthase) (Wang et al., 1991). Research sincethis discovery has provided substantial evidence that the toxicity and carcinogenicity of fumonisins isrelated to the disruption of sphingolipid metabolism that occurs as a result of inhibition of ceramidesynthase (Riley et al., 1993, 1994a; Voss et al., 2007).As much as 59% of the corn and corn-based products worldwide have been estimated to becontaminated with variable amounts of fumonisin B1 (FB1), the most prevalent of the fumonisinsubspecies (Desai et al., 2002; Haschek et al., 2001), and therefore, it is not surprising that livestockfarming and feed processing industries are interested in efficient and applicable methods forreducing fumonisin concentrations in animal feed. Most physical and chemical methods, such asthermal food processing, ammoniation or alkali treatment, reduce but do not completely eliminatefumonisins present in feed samples (Norred et al., 1991; Sydenham et al., 1995; Visconti et al., 1996).Besides, most of the methods are difficult to implement into the production process. New approachconsists to find organism or enzymes able to metabolize fumonisins and that could lead to theformation of hydrolyzed fumonisins (HFB) (Heinl et al., 2010).Except in one case in which FB1 metabolism led to a modest amount of HFB1 in vervet monkeys(Shephard et al., 1994), hydrolysis of fumonisins occurs after the well known alkali treatment namednixtamalization. This process is widely used in regions such as Central and South America, to producetortillas or other corn products made from corn cooked and steeped in a lime solution (Dombrink-Kurtzman et al., 2000; Palencia et al., 2003; Cortez-Rocha et al., 2002).As never investigated in domestic animals, we proposed in the present study to compare thetoxicity of fumonisin B1 (FB1) and hydrolyzed fumonisin B1 (HFB1) in piglets orally exposed tomoderate doses. In terms of toxicity assessment, we also reported effects of these treatments at the128

TRAVAIL EXPERIMENTALintestinal level, the first compartment of the organism to be exposed to natural contaminants.Furthermore, the number of publications on the effects of FB1 and especially its hydrolyzed form onthe gastrointestinal tract is very poor.129

TRAVAIL EXPERIMENTALMATERIAL & METHODS1) AnimalsAll animal experimentation procedures were carried out in accordance with the EuropeanGuidelines for the Care and Use of Animals for Research Purposes (Directive 86/609/EEC). Eighteen,4-week-old weaned female pigs (Pietrain/Duroc/Large-white) were obtained locally. Animals wereacclimatized for 1 week in the animal facility of the INRA Laboratory of Pharmacology and Toxicology(Toulouse, France), prior to being used in experimental protocols. Six pigs were allocated to eachtreatment on the basis of body weight. During the 14-day experimental period, animals were givenfree access to water and were fed with a basal diet ad libitum, as previously described (Grenier et al.,2011).2) Experimental treatmentsThree solutions were prepared and orally administered to piglets. These solutions consisted toone control solution, one FB1 solution and one HFB1 solution. Lyophilized culture material ofFusarium verticillioides, containing 13.7 g FB1/kg, was used to prepare the FB1 solution (Biopure –Romer Labs Diagnostic GmbH, Tulln, Austria). Briefly, culture material was resuspended in sodiumphosphate buffer, centrifuged and supernatant collected. This supernatant extract was separatedinto two equal volumes. From one of this extract, the HFB1 solution was obtained by hydrolysis ofthe FB1 with a carboxylesterase, FumD. This fumonisin carboxylesterase was first prepared byfermentation of recombinant Pichia pastoris as previously described (Heinl et al., 2010), and thelyophilized fermentation supernatant containing the FumD enzyme was dissolved in water at 100mg/mL, and added to one of the FB1 extract. Both extracts (FB1 and HFB1) were incubated overnightand heat-inactivated by boiling in a microwave. A portion of the FumD solution was heat-inactivated,and the same amount of enzyme that was used for preparation of HFB1 extract was added to theintact FB1 extract. Samples of both extracts were analyzed by LC-MS to confirm that the FB1 washydrolyzed and intact, respectively, in the two extracts. The extracts were found to contain 530 mgFB1/L and HFB1 equimolar to this concentration. The control solution contained sodium phosphatebuffer and the same concentration of heat-inactivated FumD as added to the two extracts.The three solutions were administered to animals by gavage at 3.77 mL/kg body weight/day,concentration equivalent to 2 mg FB1/kg b.w/day (equals HFB1 equimolar to 2 mg FB1/kg b.w/day).130

TRAVAIL EXPERIMENTAL3) Experimental design and sample collectionAnimals were treated daily with the different solutions and body weight of piglets was measuredevery two days in order to adjust the volume of treatment to administer. At weekly time intervals,blood samples were aseptically collected from the left jugular vein. Blood was collected in tubescontaining sodium heparin for biochemistry, except fibrinogen collected in citrate tubes(Vacutainer®, Becton-Dickinson, USA). Plasma samples were obtained after centrifugation ofheparinized blood and stored at -20°C for later analysis. Upon termination of the experiment,corresponding to 14 days of dietary exposure to treatments, immediately after electrical stunning,pigs were killed by exsanguination. Samples of liver, small intestine (proximal and distal jejunum, andileum) and mesenteric nodes were collected from all groups and fixed in 10% buffered formalin forhistopathological analysis. Besides, these tissues were flash-frozen in liquid nitrogen and stored at -80°C until processed for measurements of cytokine mRNA, and also for determination ofsphingolipids concentration in liver (between 5 and 15 g of hepatic tissues for each animal).4) BiochemistryPlasma concentrations of creatinin, albumin, total proteins, cholesterol, triglycerides, fibrinogenand activity of gamma-glutamyl transferase were determined by a Vitros 250 analyzer (Ortho ClinicalDiagnostics, Issy les Moulineaux, France) at the Veterinary School of Toulouse (France).5) Determination of the sphinganine (Sa)/sphingosine (So) ratioAssessment of sphingolipids metabolim, by measurement of the sphingoid bases content inplasma and liver samples, were carried out by HPLC-FLD (University of Natural Resources and LifeSciences, Center for Analytical Chemistry, Department for Agrobiotechnology IFA, Tulln, Austria).Liver samples were homogenized on ice in phosphate buffer and aliquots of the homogenates wereworked up and analyzed as previously described for plasma samples (Schwartz et al., 2009) withminor modifications. Concentration of Sa and So were determined and thereby, the ratio Sa/So wascalculated for each animal in the biological samples.6) HistologyThe tissue pieces were dehydrated through graded alcohols and embedded in paraffin wax.Sections of 3µm were stained with hematoxylin-eosin (HE) for histopathological evaluation.Microscopic observations in liver led to the identification of some different lesions. Establishment ofa lesion score per animal was determined, as previously described (Grenier et al., 2011), according to131

Table 12 : Nucleotide sequences of primers for real-time PCRGENE PRIMER SEQUENCE GENBANK NO. REFERENCESRPL32F (300 nM) TGCTCTCAGACCCCTTGTGAAGR (300 nM) TTTCCGCCAGTTCCGCTTANM_001001636 Pinton et al.(2010)β2-μglobulin F (900 nM) TTCTACCTTCTGGTCCACACTGAR (300 nM) TCATCCAACCCAGATGCANM_213978Devriendt etal. (2009)IL-12p40F (300 nM) GGTTTCAGACCCGACGAACTCTR (900 nM) CATATGGCCACAATGGGAGATGNM_214013Devriendt etal. (2009)IL-8F (300 nM) GCTCTCTGTGAGGCTGCAGTTCR (900 nM) AAGGTGTGGAATGCGTATTTATGCNM_213867Grenier et al.(2011)IL-1βF (300 nM) GAGCTGAAGGCTCTCCACCTCR (300 nM) ATCGCTGTCATCTCCTTGCACNM_001005149 Devriendt etal. (2009)MIP-1βF (300 nM) AGCGCTCTCAGCACCAATGR (300 nM) AGCTTCCGCACGGTGTATGAJ311717Grenier et al.(2011)IL-6IFN-γTNF-αIL-2IL-10F (300 nM) GGCAAAAGGGAAAGAATCCAGR (300 nM) CGTTCTGTGACTGCAGCTTATCCF (300 nM) TGGTAGCTCTGGGAAACTGAATGR (300 nM) GGCTTTGCGCTGGATCTGF (300 nM) ACTGCACTTCGAGGTTATCGGR (300 nM) GGCGACGGGCTTATCTGAF (300 nM) GCCATTGCTGCTGGATTTACR (300 nM) CCCTCCAGAGCTTTGAGTTCF (300 nM) GGCCCAGTGAAGAGTTTCTTTCR (300 nM) CAACAAGTCGCCCATCTGGTNM_214399NM_213948NM_214022AY294018NM_214041Grenier et al.(2011)Royaee et al.(2004)Meissonnier etal. (2008)Meurens et al.(2008)Present study

TRAVAIL EXPERIMENTALthe importance degree of the lesion (severity factor) and its extent (intensity or observed frequency ;scored from 0 to 3). For hepatic tissue, the minimal score was 0 and the maximal score was 21.The same method was applied for intestinal sections. The severity factor was equal for everylesion identified, and the lower score was 0 and 12 for the higher. As we accumulated scores fromthe two segments of jejunum and ileum, the maximal accumulated score was 36. Morphometry wasevaluated in the different segments of intestine, by measuring the villi height randomly on thirty villiusing a MOTIC Image Plus 2.0 ML® image analysis system, as already mentioned (Lucioli et al.manuscript in preparation, cf chapitre 1).7) Determination of the expression of mRNA encoding for cytokines by real-time PCRTissue RNA was processed in lysing matrix D tubes (MP Biomedicals, Illkirch, France) containingguanidine-thiocyanate acid phenol (Extract-All®, Eurobio, les Ulis, France) for use with the FastPrep-24 (MP Biomedicals, Illkirch, France). Concentrations, integrity and quality of RNA were determinedspectrophotometrically (O.D. 260 ) using Nanodrop ND1000 (Labtech International, Paris, France).Besides this inspection, 200 ng of RNA was analyzed by electrophoresis. The reverse transcription of2 µg of total RNA was performed using M-MLV reverse-transcriptase, Rnasin® plus (Promega,Charbonnière, France) and random primers (Invitrogen, Cergy Pontoise, France) (5 min 37°C, 1 hour42°C, 15 min 70°C) as already described (Oswald et al., 2001). Real-time PCR assays were performedon 8 ng of cDNA (RNA equivalent) in 25 µl volume reaction per well using Power SYBR® Green PCRMaster Mix as reporter dye, and the automated photometric detector ABI Prism 7000 SequenceDetection System for data acquisition (Applied Biosystems, Courtaboeuf, France). The amplificationconditions were as follows: 95°C for 10 min, and 40 cycles of 95°C for 15 sec and 60°C for 1 min. RNAnon-reverse transcript was used as non-template control for verification of no genomic DNAamplification signal. Specificity of PCR products was checked out at the end of the reaction byanalyzing the curve of dissociation. In addition, the size of amplicons was verified by electrophoresis.The sequences of the primers used are detailed in Table 12. Primers were purchased from Invitrogen(Cergy Pontoise, France). Amplification efficiency and initial fluorescence were determined by DART-PCR method (Peirson et al., 2003), then values obtained were normalized by both house-keepinggenes beta2-μglobulin and ribosomal protein L32 (RPL32) and finally, genes expression wasexpressed relative to the control group as already described (Le Gall et al., 2009).132

TRAVAIL EXPERIMENTAL8) StatisticsFollowing the Fisher test on equality of variances, one way ANOVA was used to analyze thedifferences between the different groups of animals at each time point. P values of 0.05 wereconsidered significant.133

Figure 11 : Effects of FB1 and HFB1 treatments on liverlesionsLesion scores were established after histologicalexamination according to the severity and the extent of thelesions. Values are mean ± SEM for five animals.Lesion score86420baacontrol FB1 HFB1Table 13 : Effects of FB1 and HFB1 treatments on biochemical parametersValues are mean ± SEM for five animals. GGT, γ‐glutamyl transferase.Control FB1 HFB1Creatinin (µmol/L)Albumin (g/L)Total proteins (g/L)Triglycerides (mmol/L)Cholesterol (mmol/L)Fibrinogen (g/L)GGT (U/L)Day 7Day 14Day 7Day 14Day 7Day 14Day 7Day 14Day 7Day 14Day 7Day 14Day 7Day 1485.0 ±4.2 a 106.0 ±6.9 b 87.0 ±3.3 a95.8 ±4.0 a 117.0 ±7.8 b 89.5 ±1.8 a26.2 ±0.8 a 31.3 ±0.7 b 26.1 ±0.9 a26.5 ±1.1 a 31.6 ±1.0 b 25.1 ±0.7 a49.7 ±0.8 a 58.7 ±1.4 b 49.9 ±1.2 a50.4 ±0.7 a 61.4 ±1.2 b 50.7 ±0.6 a0.89 ±0.12 a 1.19 ±0.16 a 0.76 ±0.11 a0.82 ±0.17 a 1.34 ±0.14 b 0.82 ±0.09 a2.70 ±0.22 a 3.78 ±0.21 b 2.16 ±0.34 a2.63 ±0.26 a 4.44 ±0.30 b 2.12 ±0.25 a2.16 ±0.13 a 2.91 ±0.21 b 1.89 ±0.27 a1.68 ±0.05 a 2.96 ±0.12 b 1.72 ±0.05 a46.0 ±2.4 a 71.2 ±6.2 b 53.2 ±5.3 a,b51.2 ±1.4 a 146.2 ± 32.3 b 71.2 ± 12.8 a,bControl FB1 HFB1Table 14 : Effects of FB1 and HFB1treatments on cytokines expression in liverQuantification of the relative cytokinemRNA level for each sample is expressed inarbitrary units (A.U). Values are mean ±SEM for five animals.IL‐1β 1.00 ±0.10 a 1.55 ±0.20 b 1.27 ±0.21 a,bIL‐8 1.00 ±0.09 a 1.82 ±0.55 a 1.26 ±0.10 aIL‐6 1.00 ±0.08 a 0.72 ±0.08 b 0.71 ± 0.07 bIL‐10 1.00 ±0.15 a 0.49 ±0.06 b 0.98 ±0.08 aIFN‐γ 1.00 ±0.11 a 0.69 ±0.04 b 1.00 ±0.25 a,bIL‐18 1.00 ±0.01 a 0.59 ± 0.06 b 0.86 ±0.05 a

TRAVAIL EXPERIMENTALRESULTS1) Comparative effects of FB1 and HFB1 on the hepatic parametersa) Histopathology results in liverLesions observed in the liver were considered as mild to moderate, and allowed to establish alesion score. This score for HFB1 group did not show any important damages, and was similar to thescore of the control group (Figure 11). By contrast, FB1 elicited significant injuries in liver, asdisplayed by its lesion score (p

Table 15 : Effects of FB1 and HFB1 treatments on the sphinganine/sphingosine ratio in biological samplesValues are mean ± SEM for six animals.PLASMA Day 0 Day 7 Day 14Control FB1 HFB10.26 ±0.02 a 0.24 ±0.02 a 0.26 ±0.02 a0.24 ±0.02 a 2.49 ±0.20 b 0.30 ±0.02 a0.27 ±0.02 a 2.14 ±0.07 b 0.33 ±0.03 aLIVER 0.11 ±0.01 a 3.08 ±0.45 b 0.25 ±0.06 c12Figure 12 : EffectsofFB1andHFB1treatmentsonthesmall intestineAccumulated lesion score1086420control FB1 HFB1ileumdistalLesion scores were established after histologicalexamination according to the extent of the lesions.Values are mean for five animals.proximalTable 16 : Effects of FB1 and HFB1 treatments on the villi height in the small intestineValues are mean ± SEM for five animals.Control FB1 HFB1Proximal Jejunum 300 ±16 a 259 ±17 a 255 ±19 aDistal Jejunum 321 ±13 a 259 ±21 b 297 ±10 a,bIleum 265 ±13 a 182 ±13 b 241 ±7 a

TRAVAIL EXPERIMENTAL2) Comparative effects of FB1 and HFB1 on the sphingolipids metabolismConcentrations of sphinganine and sphingosine were evaluated in plasma and liver samples.Following the analysis, the Sa/So ratio was established, and results in the Table 15 show similar ratiosfor the control and HFB1 groups in the biological samples. By contrast, and as expected, treatmentwith FB1 solution greatly affected the content in sphingoid bases, leading to an accumulation of Saand So, both in plasma and liver (p

Table 17 : Effects of FB1 and HFB1 treatments on cytokines expression in the small intestine and themesenteric lymph nodesValues are mean ± SEM for five animals.Control FB1 HFB1IL‐1βIL‐8TNF‐αIL‐6IFN‐γIL‐2IL‐12p40IL‐10Proximal jejunumDistal jejunumIleumMesenteric nodesProximal jejunumDistal jejunumIleumMesenteric nodesProximal jejunumDistal jejunumIleumMesenteric nodesProximal jejunumDistal jejunumIleumMesenteric nodesProximal jejunumDistal jejunumIleumMesenteric nodesProximal jejunumDistal jejunumIleumMesenteric nodesProximal jejunumDistal jejunumIleumMesenteric nodesProximal jejunumDistal jejunumIleumMesenteric nodes1.00 ±0.26 a1.00 ±0.18 a,b1.00 ±0.26 a1.00 ±0.13 a 0.41 ±0.06 b0.62 ±0.09 a0.44 ±0.13 b0.49 ±0.07 b 1.03 ±0.28 a1.02 ±0.07 b0.66 ±0.07 a,b0.94 ±0.16 a1.00 ±0.12 a,b1.00 ±0.13 a1.00 ±0.13 a1.00 ±0.17 a 0.84 ±0.10 a0.76 ±0.12 a0.58 ±0.08 b0.41 ±0.06 b 1.23 ±0.04 b1.07 ±0.15 a0.87 ±0.12 a,b0.88 ±0.16 a1.00 ±0.12 a1.00 ±0.04 a1.00 ±0.15 a1.00 ±0.04 a 0.93 ±0.12 a0.90 ±0.06 a1.42 ±0.16 a0.86 ±0.04 b 1.01 ±0.09 a0.96 ±0.05 a1.01 ±0.13 a0.74 ±0.02 c1.00 ±0.18 a1.00 ±0.10 a1.00 ±0.15 a1.00 ±0.07 a 0.88 ±0.16 a0.67 ±0.07 b1.26 ±0.17 a0.73 ±0.04 b 1.26 ±0.18 a0.83 ±0.13 a,b1.24 ±0.16 a0.82 ±0.05 b1.00 ±0.18 a1.00 ±0.18 a1.00 ±0.12 a1.00 ±0.09 a 0.54 ±0.05 b0.54 ±0.06 b1.55 ±0.10 b0.71 ±0.06 b 1.00 ±0.21 a1.00 ±0.12 a1.61 ±0.17 b1.01 ±0.08 a1.00 ±0.04 a1.00 ±0.13 a1.00 ±0.18 a1.00 ±0.09 a 0.60 ±0.06 b0.62 ±0.06 b0.64 ±0.10 a0.76 ±0.02 b 0.86 ±0.11 a0.89 ±0.24 a,b0.76 ±0.07 a1.16 ±0.16 a1.00 ±0.13 a,b1.00 ±0.08 a1.00 ±0.07 a1.00 ±0.21 a 0.70 ±0.09 a0.94 ±0.07 a1.34 ±0.16 a,b0.55 ±0.06 b 1.08 ±0.15 b1.91 ±0.29 b1.54 ±0.08 b0.77 ±0.08 a1.00 ±0.05 a1.00 ±0.06 a1.00 ±0.13 a1.00 ±0.12 a 0.57 ±0.05 b1.03 ±0.15 a,b1.18 ±0.17 a0.73 ±0.08 a 1.00 ±0.10 a1.36 ±0.13 b1.10 ±0.13 a0.92 ±0.05 a

TRAVAIL EXPERIMENTALexpression were noticed according to the different tissues investigated. Expression of thesemediators was quite similar to those of control group, or in some cases not so much reduced incomparison to FB1. Of note for the ileum, treatment with HFB1 seems to slightly promote theexpression of two Th1 cytokines, IL-12p40 and IFN-γ.136

TRAVAIL EXPERIMENTALDISCUSSIONThe aim of the current study was to compare the in vivo toxicity of FB1 and its fully hydrolyzedform, HFB1 (also named aminopentol or AP1). Despite controversial results on this compound, weshowed that the hydrolysis of FB1 strongly reduce toxicity in swine, either in liver or ingastrointestinal tract. To our knowledge, this is the first report of the HFB1 toxicity on domesticanimals.HFB1 is produced during a traditional corn treatment with calcium hydroxide and heat, alsonamed nixtamalization, the alkaline hydrolysis process removing the tricarballylic acid side chainsfrom FB1 contained in corn. Nixtamalization is the major way of processing corn in both Mexico andCentral America, as well as in the United States, to make masa and tortillas. In addition to theoccurrence of HFB1 in nixtamalized material, detoxification of FB1-contaminated foods/feeds bymicroorganisms may lead to the formation of HFB1 (Heinl et al., 2010), and the effect of its ingestionneeds to be evaluated. Considering swine shows a great sensitivity to fumonisins and because of thehigh percentage of corn in pig diets, these alternative methods of decontamination might be appliedin husbandry. Besides, considering that many biological systems (intestinal, metabolic or immune) ofpigs are very similar to that of humans, swine can be regarded as a good model that can be appliedto humans (Almond, 1996; Verma et al., 2011).In this experiment, the dose of 2 mg FB1/kg bw/day allowed us to elicit toxicity and thus tocompare at equimolar concentration the HFB1 effects. Based on averaged feed consumption for pigsof this age, this dose corresponds to feed contaminated with a concentration of 37-44 mg of FB1/kg.Over the 14-day exposure, no effect was noticed on the animal growth or the feed intake (datanot shown). No effect on the body weight gain has been already reported in pigs and poultry fed upto 50 mg FB1/kg feed (Harvey et al., 1996; Broomhead et al., 2002). The hepatotoxicity elicited byFB1 administration was not observed in HFB1-treated piglets. Indeed, biochemical changesattributed to the disturbance of some hepatic functions, such as protein and lipid metabolism, wasnot noticed in HFB1 group. Similarly, liver was not damaged after HFB1 ingestion as shown by theabsence of microscopic lesions. Inflammation has been assessed in this short-term exposure throughbiochemical markers and liver cytokines as well. The plasmatic concentrations of GGT and fibrinogenwere unaffected following HFB1 exposure. The level of cytokines at the end of experiment wasslightly modulated in livers from animals treated with HFB1, but less than in those from FB1 group. Inthis latter group, levels of IL-1β and IL-8, well known as pro-inflammatory mediators, were stillelevated as already described after FB1 exposure (Bhandari et al., 2002). Levels of IL-6 and IL-10,known among others as mediators with anti-inflammatory properties (de Vries, 1995; Xing et al.,137

TRAVAIL EXPERIMENTAL1998), were greatly reduced in liver and thereby, could confirm the inflammatory state in pigletsexposed to FB1. By contrast, HFB1 administration did not modulate the cytokines expression as muchas FB1 did, and to our knowledge this is the first report on the effect of HFB1 on cytokines in liver.The lack of toxicity in liver after HFB1 exposure is in agreement with some authors (Collins et al.,2006; Howard et al., 2002; Voss et al., 2009), but also in discordance with others (Hendrich et al.,1993; Voss et al., 1996; Voss et al., 1998). Nonetheless, in these controversial studies, it is importantto clarify that the toxicity was only demonstrated in animals exposed to nixtamalized culture materialand not to pure HFB1. That’s why, it was suggested (Kim et al., 2003; Park et al., 2004; Seiferlein etal., 2007; Burns et al., 2008; Voss et al., 2009) that these effects were mediated by residual or“hidden” FB1 (matrix bound forms not detected by HPLC) remaining in the nixtamalizedpreparations, rather than HFB1.We also determined the potential toxicity of these compounds on the intestine. Indeed, as thegastrointestinal tract is a primary site of mycotoxins exposure and that FB1 has been also reported toexert its toxicity, through the ceramide synthase inhibition, in the intestine (Enongene et al., 2000;Loiseau et al., 2007), evaluation of HFB1 effects on intestine was necessary. Following ingestion ofcontaminated food or feed, intestinal epithelial cells could be exposed to a high concentration oftoxicants, potentially affecting intestinal functions. Some studies focused on the bowel, but most ofthem concern effects of HFB1 on intestinal cell lines by assessing the number of viable cells, HFB1acylation or sphingoid bases content (Schmelz et al., 1998; Humpf et al., 1998; Seiferlein et al., 2007).In the current study, HFB1 ingestion did not alter the intestinal integrity in the different segmentsexamined, as shown by the villi morphometry and the lesion scores. Among lesions, villi atrophy andfusion were evaluated. Occurrence of these lesions was considerably high in the intestines exposedto FB1, especially in the second part of the small intestine. These findings are in agreement with thevillous fusion and atrophy observed in the intestine of pigs treated with 30 ppm FB1 (Piva et al.,2005) and in the intestine of chicks fed with 61-546 ppm FB1 (Javed et al., 2005). In addition, a reporton human consumption of moldy maize (containing fumonisins) resulted in abdominal pain,borborygmi, and diarrhea (Bhat et al., 1997). Therefore, it can be suggested that on contrary to FB1,HFB1 consumption would not lead to impairment of the intestinal absorption of nutrients.The intestine is also an immune privileged site where immunoregulatory mechanismssimultaneously defend against pathogens but also preserve tissues homeostasis to avoid immunemediatedpathology in response to environmental challenges (Ramiro-Puig et al., 2008). To assessthe intestinal immunity following exposure to FB1 and HFB1, we focused on different sections of thejejunum, and on ileum and mesenteric lymph nodes (MLN) as well. In contrast to the jejunum, whichprovides a local response, the ileum containing Peyer’s patches and MLN take part from the gut-138

TRAVAIL EXPERIMENTALassociated lymphoid tissue (GALT) and act as inductive sites for intestinal immune response. Thesespecific tissues can be considered as specialized meeting places, where antigen-presenting cells andlymphocytes interact in presence of intestinal antigen. We investigated cytokines expression, whichare important mediators in the regulation of the immune and inflammatory responses. They areproduced by immune system cells (lymphocytes, macrophages, …) but also by cells not traditionallyconsidered as part of the immune system such as intestinal epithelial cells.Few data are available with regard to the effects of FB1 on the intestinal immunity, and evenlesser on cytokines production. FB1 decreases the expression and the synthesis of IL-8 in the porcineepithelial intestinal cell line, IPEC-1 (Bouhet et al., 2006). The same study confirmed this result invivo; ingestion of low doses of FB1 by piglets decreased the expression of IL-8 mRNA in the ileum.The FB1-induced IL-8 decrease may lead to a reduced recruitment of inflammatory cells in theintestine during infection and may participate in the observed increased susceptibility ofcontaminated piglets to intestinal infections (Oswald et al., 2003). Recently, another study showed inpiglets that following FB1 exposure, animals revealed a reduced intestinal expression of IL-12p40, animpaired function of intestinal antigen presenting cells (APC), with decreased upregulation of majorhistocompatibility complex class II molecule (MHC-II) and reduced T cell stimulatory capacity uponstimulation (Devriendt et al., 2009). As outlined by the authors, these results indicate an FB1-mediated reduction of in vivo APC maturation. In the current study, for most of the cytokinesinvestigated, we observed after FB1 ingestion a significant decrease of their expression in all thecompartments of the intestinal tract. By contrast, the profile of cytokines after exposition to HFB1revealed very few changes, close to the one observed in non-treated animals. These observationsimply that animals consuming HFB1 would be able to defend efficiently against potential invaders. Toour knowledge, this is the first study on the effects of HFB1 on some key components of theintestinal immune system.By contrast to FB1, the lack of effects, either at the hepatic or intestinal levels, observed followingHFB1 treatment might be explained by its ability to unaffect the sphingoid bases content. Indeed,FB1 acts as an inhibitor of the ceramide synthase, leading to an accumulation of sphinganine and to alesser extent of sphingosine. The establishment of the Sa/So ratio has been widely described to be arelevant and sensible biomarker of FB1 toxicity (Soriano et al., 2005, Tran et al., 2006), and illustratesthe degree of sphingolipid metabolism disruption. This elevation in sphinganine, a highly bioactivecompound, initiates a cascade of cellular alterations that are thought to be largely responsible for thetoxicity and carcinogenicity of this mycotoxin. To inhibit the ceramide synthase, aminopentolbackbone of FB1 competes for binding of the sphingoid base substrate, whereas the tricarballylicacids side chains (TCA) interfere with binding of the fatty acyl-CoA (Humpf et al., 1998; Desai et al.,139

TRAVAIL EXPERIMENTAL2002). Removal of the tricarballylic acids, which leads to HFB1, diminishes the potency of ceramidesynthase inhibition, and therefore the accumulation of sphingoid bases. In the current experiment,we may suppose that ceramide synthase inhibition did not occur, as displayed by the Sa/So ratio,either in plasma samples or in liver samples. This indicated that the TCA moieties are required formaximal inhibition of the enzyme. Our results are in agreement with studies which showed that HFB1was at least tenfold less potent compared to FB1 (Norred et al., 1992; Flynn et al., 1997; van derWesthuizen et al., 1998; Howard et al., 2002; Collins et al., 2006; Voss et al., 2009). As mentioned byVoss et al. (1998), increased liver Sa/So ratio is directly correlated with the increased of serumchemical indicators, leading to hepatotoxicity. This observation confirms our results on FB1 andHFB1.However, some findings showed that HFB1 becomes not only an inhibitor but also a substrate foracylation by ceramide synthase (Humpf et al., 1998; Abou-Karam et al., 2004; Seiferlein et al., 2007).This observation was firstly reported in HT29 cells (Humpf et al., 1998), where FB1 was notdetectably acylated on contrary to HFB1. The author suggested at this moment that the TCA of FB1occupy the acyl-CoA binding site and thereby block access of this co-substrate. Therefore, theremoval of TCA allowed to HFB1 to be acylated, and the first acylation observed was in presence ofpalmitoyl-CoA which led to the formation of N-palmitoyl-AP1 (PAP1 ; hydrolyzed fumonisin B1 alsonamed AP1 for aminopentol). As outlined by Humpf et al. (1998), a somewhat unexpected finding ofthis study was that PAP1 was highly cytotoxic for HT29 cells, and a more potent inhibitor of ceramidesynthase, causing sphinganine accumulation. In vivo acylation was also evaluated in rats, and whileformation of N-acyl-AP1 occurs and produces metabolites with fatty acids of various chain lengths,no toxicity was observed (Seiferlein et al., 2007). Considering observations of Humpf and Seiferlein,we cannot exclude in our experiment that acylation did not occur, but neither increase insphinganine concentrations nor toxicity was detected. Nonetheless, this aspect warrants furtherinvestigations, considering that there are multiple of isoforms of ceramide synthase (CerS) that aredifferently expressed according to tissues and differ in fatty acyl-CoA selectivity. Besides, a differentoutcome depending on animal species does not have to be excluded.To conclude, we showed for the first time in domestic animals that the hydrolyzed form of FB1 ismuch less toxic than the parent compound, and therefore confirmed results observed in rodentsexposed to pure HFB1. Furthermore, the HFB1 in this study was originally obtained after anenzymatic treatment on FB1 with a carboxylesterase. It is thus of concern to consider the potential ofthese biotransformation approaches in order to counteract adverse effects of FB1-contaminatedfeed in farms. Recently, it has also been suggested that the carboxylesterase could be combined withan enzyme, expressed from the same cluster gene of the carboxylesterase FumD (Heinl et al., 2010;140

TRAVAIL EXPERIMENTALHartinger et al., 2010). This enzyme possess the ability to deaminate the HFB1, following thedeesterification step, leading to the removal of the C2-amino group. This chemical group has beendescribed as responsible of the toxicity of FB1 (Voss et al., 2007), and also to be acylated on the HFB1by fatty acid (Humpf et al., 1998).141

CHAPITRE 3Évaluation des effets d’agentsdétoxifiants lors d’une exposition auDéoxynivalénol et à la Fumonisinechez le porcelet142

Figure 13 :(a) Transformation par voie enzymatique de l’ochratoxine A (OTA) et de la zéaralénone (ZEA), via l’utilisationdu microorganisme Trichosporon mycotoxinivorans(b) Transformation par voie enzymatique des trichothécènes (TCT) de type A et B, via l’utilisation dumicroorganisme Eubacterium BBSH 797(a)Trichosporonmycotoxinivorans(levure isolé du tractusdigestif des termites)OTA PhenylalanineOTA phenylalanineOTα OTαZEAForme detoxifiée(b)EubacteriumBBSH797(bactérie isolée durumen de bovins)TCTTCTforme detoxifiéeForme detoxifiée

TRAVAIL EXPERIMENTALRESUME DE L’ETUDECette étude s’inscrit dans la continuité des précédentes, avec pour objectif d’évaluer l’ajout deproduits désactivateurs de mycotoxines dans des aliments contaminés. Comme mentionnéprécédemment, BIOMIN développe et commercialise des produits composé de micro-organismescapable de dégrader les mycotoxines. Outre l’enzyme présentée dans l’étude précédente maisencore non commercialisée, BIOMIN propose des approches essentiellement basées par voieenzymatique et qui cible spécifiquement chaque mycotoxine ou famille de mycotoxine. En effet, cesmétabolites secondaires présentent une incroyable diversité structurelle, et ainsi chaque approchebiologique nécessite l’utilisation d’un micro-organisme/enzyme adapté à chacun. A noter que leprincipe de la méthode utilisée par notre partenaire pour éliminer les aflatoxines est différente desautres toxines, et repose sur des phénomènes d’adsorption par des argiles. En effet, comme résumédans l’introduction de ce mémoire de thèse, les adsorbants inorganiques et organiques ont étélargement reconnus comme efficaces vis-à-vis de leurs capacités à lier les aflatoxines et à les excréterhors de l’organisme animal. En revanche, ces produits ont essentiellement fait leurs preuves sur lesaflatoxines, mais rarement sur les autres toxines majeures. Ainsi, BIOMIN a développé des approchespar voie enzymatique contre l’ochratoxine A, la zéaralénone et les trichothécènes de type A et B,transformant ces métabolites toxiques en des substances non-toxiques (Figure 13).Dans la première étude sur l’exposition du déoxynivalénol et des fumonisines, seuls ou encombinaison, nous avons montré que ces toxines ont des effets toxiques sur le système immunitaireet sur différents organes cibles (foie, reins, poumons et intestin). L’objectif était donc de reproduirecette expérimentation (menée simultanément avec celle de la première étude) et d’évaluerl’efficacité d’agents détoxifiants (DA), par le biais d’une supplémentation de ces produits dans lesaliments mono- et co-contaminés. Le premier agent désactivateur utilisé était spécifique du DON etdéjà commercialisé. Il a été inclus dans l’aliment à 0,25%. Le deuxième agent désactivateur utiliséétait l’enzyme expérimentale isolée récemment par BIOMIN, spécifique des FB, et présentée dans lechapitre précédent. Il a été inclus via sa pulvérisation directe sur l’aliment (250 mL/25 kg d’aliment).143

Figure 14 :Illustrations de la procédure d’incorporation dans l’aliment de l’agent désactivateur ciblant spécifiquementles fumonisinesPulvérisation du produit DAdirectement sur les lots d’aliments250 mL de DA sur 25 kg de granulésétalés et formant une monocouche

TRAVAIL EXPERIMENTALMATERIEL ET METHODES1) Fabrication des alimentsSur la base d’une formule d’aliment porcelet 2 ème âge, 8 lots d’aliments contaminés ou non avecles mycotoxines et supplémentés ou non avec les DA ont été produits sur le site de l’INRA de Rennes(UMR SENAH). Comme détaillé dans la première étude, les extraits contenant le DON et les FB ontété mélangés au pré-mix de vitamines et minéraux, puis incorporés dans le mélange de céréalesavant l’étape de granulation. Dans le cas des aliments supplémentés, le premier DA ciblant le DON(Mycofix, Biomin GmbH, Tulln, Autriche) a aussi été inclus à cette étape de la préparation à la dosede 2,5 g/kg d’aliment fini (0,25%); alors que le DA ciblant les FB, conditionné sous forme lyophilisée,a été repris dans de l’eau et pulvérisé sur l’aliment fini à raison de 250 mL/25 kg aliment à l’INRA deToulouse (Figure 14).Les matières premières et les aliments finis ont été contrôlés pour leurs contaminations naturelleset/ou artificielles en mycotoxines au laboratoire Quantas Analytik (Tulln, Autriche) et aussi par uneméthode expérimentale d’analyse en multi-mycotoxines (Sulyok et al., 2007) comme décrit dans lapremière étude. A noter que nous avons été contraints de formuler notre aliment sans maïs, due àune contamination naturelle importante en DON (700 µg/kg de maïs) et en FB (900 µg/kg de maïs) ;et que le DON, la zéaralénone et l’enniatin ont été trouvés naturellement présents, à desconcentrations de 500, 50 et 100 µg/kg dans l’aliment fini, respectivement.Ci-dessous, le Tableau 18 résumant les différents régimes fabriqués avec leurs teneurs finalesrespectives en mycotoxines :en µg de toxines/kg d’alimentSansagentsdésactivateurs(DA-)Avecagentsdésactivateurs(DA+)DON FB1 FB2A1: Contrôle 539

Figure 15 :Plan expérimental de la phase animale avec le protocole de vaccinationDA‐DA+contrôlecontrôleDONFBDONFB8 régimes formulés etdonnés aux 48 porceletsRécupération dusang (10-15 mL)DON+FBDON+FBJours 0 4 7 14 16 21 28 35ovalbumine(OVA)EuthanasieimmunizationboostCollecte desorganes

TRAVAIL EXPERIMENTAL2) Plan expérimental de la phase animale et prélèvement des échantillons (Figure 15)Toutes les procédures d’expérimentations animales ont été menées en accord avec lesrecommandations européennes au regard de l’utilisation des animaux à des fins de recherches(Directive 86/609/EEC). Des porcelets mâles castrés (Pietrain X Duroc X Large-white) et sevrés ont étéobtenus dans un élevage local, et familiarisés à leur nouvel environnement les dix jours précédents ledébut de la phase expérimentale. Des lots homogènes basés sur le poids des animaux ont ensuite étéformés. Pendant les 35 jours de l’expérience, les porcelets ont eu libre accès à l’eau et au régimealimentaire désigné. Une surveillance quotidienne, ainsi qu’une pesée hebdomadaire, ont étéréalisées au cours de cette période. Les phases d’expérimentations animales avec les aliments nonsupplémentés et les aliments supplémentés en DA ont été menées simultanément.Afin de stimuler et d’étudier la réponse immunitaire, deux injections contenant une solutiond’ovalbumine mélangée à l’adjuvant incomplet de Freund, ont été administrées à 12 joursd’intervalle (J4 et J16 de la période expérimentale) à tous les animaux.Des prélèvements sanguins à la veine jugulaire ont été effectués chaque semaine pour la mise enculture du sang total et pour la récupération du plasma (stocké à -20°C).Après les 35 jours d’expositions aux différents régimes, tous les animaux ont été sacrifiés parexsanguination après électronarcose. Des échantillons de foie, poumons, rate et des sectionsd’intestin ont été prélevés et fixés dans un tampon de formaldéhyde à 10% pour les analyseshistologiques. De plus, des échantillons de rate ont été directement congelés dans l’azote liquide,puis stockés à -70°C avant leurs traitements pour les analyses transcriptomiques.3) Analyses hématologiques et biochimiquesL’analyse hématologique a été effectuée dans un laboratoire d’analyses médicales (Ferrandery,Tournefeuille, France) via l’utilisation d’un automate à impédance, Coulter LH500 (Beckman Coulter,Villepinte, France). La sous-population des globules blancs (lymphocytes, neutrophiles, monocytes,basophiles et éosinophiles) a été étudiée et comptée manuellement sur 100 leucocytes par lacoloration de May-Grünwald Giemsa.Les concentrations plasmatiques de l’albumine, des protéines totales, de l’urée, de la créatinine,du cholestérol, des triglycérides et de la gamma-glutamyl transférase, ont été déterminées à l’EcoleNationale Vétérinaire de Toulouse via l’utilisation d’un automate, Vitros 250 (Ortho ClinicalDiagnostics, Issy les Moulineaux, France).145

TRAVAIL EXPERIMENTAL4) Analyses du contenu plasmatique en bases sphingoïdesComme introduit dans le chapitre 2, tout au long de notre expérience les plasmas collectéshebdomadairement ont été également utilisés pour l’analyse de la concentration des basessphinganine (Sa) et sphingosine (So), et le ratio Sa/So a ainsi pu être déterminé. Ces analyses ont étéréalisées dans le laboratoire du centre de chimie analytique, du département d’agrobiotechnologieIFA à Tulln (Autriche). Brièvement, les échantillons de plasma ont été traités par hydrolyse avec unebase forte (Yoo et al., 1996), dérivés avec l’o-phthalaldéhyde (OPA, se lie aux groupements amines)(Riley et al., 1994b) et analysés par HPLC couplée à un appareil de détection à fluorescence (Schwartzet al., 2009). Les concentrations en Sa et So ont été déterminées par correction avec lespourcentages de récupération des standards internes C17-sphingosine et C20-sphinganine.5) Analyses histopathologiquesDans le cadre d’une collaboration avec l’Ecole Vétérinaire de Londrina au Brésil, les analyseshistologiques et immunohistochimiques des tissus ont été réalisées au laboratoire de PathologieAnimale. Les tissus fixés dans le formaldéhyde ont été inclus dans des blocs de paraffine aprèsdéshydratation par plusieurs gradients d’alcool. Des sections de 3µm ont été colorées àl’hématoxyline-éosine pour l’évaluation histologique. Comme présenté dans le chapitre 1, afind’évaluer l’effet des toxines sur les organes, nous avons établi un score lésionnel pour chaque tissurecensant l’ensemble des lésions observées. Ce score a été calculé en fonction du degré de sévéritéet de l’étendu de chaque lésion.La prolifération cellulaire des hépatocytes a été évaluée en comptant le nombre de noyaux Ki-67positifs sur les sections de foie. La méthode a été présentée dans le chapitre 1.L’évaluation de la hauteur des villosités, la profondeur des cryptes, le dénombrement des cellulesimmunitaires et des entérocytes en prolifération a été réalisée de la même manière que dans lechapitre 1.6) Mesure de la concentration des immunoglobulines plasmatiques totales et spécifiquesLes teneurs en immunoglobulines (Ig) totales A et G dans le plasma ont été déterminées par latechnique ELISA comme décrit au chapitre 1.Pour la mesure de réponse humorale spécifique suite à la vaccination, la technique ELISA aégalement été utilisée avec l’ovalbumine comme antigène de capture des anticorps plasmatiques, etdes anticorps anti-IgA et IgG porcins couplés à la peroxydase HRP pour la détection (cf. chapitre 1).146

TRAVAIL EXPERIMENTAL7) Détermination de l’index de prolifération des lymphocytesComme expliqué dans le chapitre 1, la prolifération des lymphocytes a pu être déterminée aprèsla mise en culture du sang total, et stimulé avec un mitogène et/ou un antigène (ovalbumine). Lastimulation mitogénique du sang total (dilué au 1/15 ème ) avec la concanavaline A (10 µg/mL) nous apermis d’évaluer la réponse cellulaire non-spécifique. A l’inverse, la réponse cellulaire spécifique aété étudiée après la stimulation du sang total (dilué au 1/15 ème ) en présence d’ovalbumine (5 et 10µg/mL). Des cultures cellulaires contrôles, non stimulées ont aussi été inclus. Après 48h destimulation, nous avons rajouté de la thymidine radio-marquée (tritium 3 H, 0,5 µCi/puit) au milieu deculture, et laissé à incuber 24 heures supplémentaires. Les cultures cellulaires ont ensuite étéstockées à -20°C. Après filtration des cultures, les niveaux de thymidine tritiée incorporée ont étémesurés à l’aide d’un compteur à scintillation (Kontron Instrument, Mila, Italie). L’index deprolifération des lymphocytes a été exprimé en cpm mesurés dans les cultures stimulées/cpmmesurés dans les cultures contrôles non stimulées.8) Détermination de l’expression des ARNm codant des cytokinesAfin d’extraire les ARNs de la rate, les échantillons collectés à l’autopsie et congelés à -70°C, ontété broyés au FastPrep-24 (MP Biomedicals, Ilkirch, France) via l’utilisation de tubes de lyse (MPBiomedicals) contenant du thiocyanate de guanidine (Extract-All®, Eurobio, les Ulis, France), et suivid’une extraction au phénol chloroforme. Comme mentionné dans le chapitre 1, les concentrations,l’intégrité et la qualité des ARNs ont été déterminées au spectrophotomètre Nanodrop ND1000(Labtech International, Paris, France). Les étapes de transcription inverse et de PCR temps réel ontété réalisées comme précédemment. Des ARNs non-inversement transcrits ont été utilisés commecontrôle pour vérifier que nous n’amplifions pas d’ADN génomique. En plus de cela, nos amorcesétaient situées sur deux exons différents, limitant ainsi l’amplification d’ADN génomique. Laspécificité des produits PCR a été contrôlée à la fin de la réaction par l’analyse de la courbe dedissociation. Basé sur la méthode de Peirson et al. (2003), également nommé DART-PCR (DataAnalysis for Real Time-PCR), nous avons pu déterminer l’efficacité d’amplification de chaqueéchantillon, et ainsi la fluorescence initiale de chaque matrice d’ADNc. Cette approche rend possiblel’analyse de chaque échantillon expérimental, basé sur sa propre réaction de cinétique plutôt que surcelle de standards artificiels. Enfin, les valeurs de fluorescence ont été normalisées par celle de deuxgènes de ménage, la β2-µglobuline et la protéine ribosomale L32, et l’expression des gènes a étéexprimée par rapport au groupe contrôle.147

TRAVAIL EXPERIMENTAL9) Analyses statistiquesLes données entre les groupes d’animaux à chaque temps de prélèvements ont été analysées paranalyse de variance (ANOVA) et test t de student, après vérification de l’égalité des variances par letest de Fisher. Une valeur de p

Tableau 20 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non en agents désactivateurs sur certains paramètreshématologiques et biochimiques à J35HEMATOLOGIESans agents désactivateursAvec agents désactivateursContrôle DON FB DON+FB Contrôle DON FB DON+FBGlobules rouges (milliers/µL) 6,2 ± 0.3 a 5,7 ± 0,2 a 6,1 ± 0,4 a 5,9 ± 0,4 a6,3 ± 0,2 a 5,8 ± 0,4 a 6,3 ± 0,4 a 6,1 ± 0,2 aGlobules blancs (milliers/µL) 21,2 ± 1,9 a 19,6 ± 2,3 a 20,3 ± 2,8 a 18,2 ± 1,6 aLymphocytes (milliers/µL) 12,4 ± 1,9 a 11,4 ± 1,4 a 14,7 ± 2,1 a 12,6 ± 1,0 aNeutrophiles (milliers/µL) 7,3 ± 1,1 a 7,0 ± 1,1 a,b 4,5 ± 0,9 b 4,6 ± 0,6 b19,9 ± 1,8 a 18,7 ± 1,5 a 24,9 ± 3,6 a 21,1 ± 2,5 a11,5 ± 1,4 a 11,0 ± 0,7 a 13,3 ± 1,3 a 11,4 ± 1,2 a6,7 ± 0,9 a,b 6,3 ± 0,9 a,b 9,8 ± 2,6 a 8,2 ± 1,6 aBIOCHIMIEUrée (mmol/L) 3,8 ± 0,4 a 3,3 ± 0,4 a 4,2 ± 0,3 a 4,0 ± 0,4 a 3,7 ± 0,6 a 3,5 ± 0,3 a 3,6 ± 0,4 a 3,3 ± 0,4 aCréatinine (µmol/L) 102,5 ± 5,3 a 98,0 ± 4,1 a 120,5 ± 5,6 b 101,6 ± 5,5 aCholestérol (mmol/L) 2,6 ± 0,2 a 2,4 ± 0,2 a 2,3 ± 0,1 a 2,3 ± 0,1 a101,7 ± 3,7 a 96,5 ± 6,9 a 100,0 ± 3,1 a 101,6 ± 5,1 a2,6 ± 0,2 a 2,1 ± 0,2 a 2,3 ± 0,2 a 2,3 ± 0,2 aTriglycérides (mmol/L) 0,51 ± 0,07 a,b 0,34 ± 0,04 a 0,39 ± 0,06 a 0,41 ± 0,06 a,b 0,42 ± 0,07 a,b 0,45 ± 0,06 a,b 0,43 ± 0,04 a,b 0,58 ± 0,07 bProtéines totales (g/L) 59,8 ± 1,0 a 57,1 ± 2,1 a 59,9 ± 2,5 a 57,6 ± 2,5 aAlbumine (g/L) 34,3 ± 0,7 a 29,2 ± 1,5 b 35,1 ± 2,1 a 32,8 ± 2,1 a,bGGT (IU/L) 65,4 ± 8,6 a 88,6 ± 14,4 a 79,4 ± 15,0 a 77,0 ± 11,5 a60,8 ± 2,2 a 58,8 ± 2,4 a 63,4 ± 4,5 a 58,8 ± 2,2 a34,6 ± 1,1 a 33,4 ± 2,1 a,b 35,3 ± 2,8 a,b 30,2 ± 1,0 b70,6 ± 1,8 a 86,4 ± 7,2 a 82,8 ± 23,7 a 67,8 ± 7,4 aNotes : les résultats sont présentés en moyenne ± erreur type pour 6 animaux (Hématologie) et 5 animaux (Biochimie). a,b En ligne, les valeurs ne présentant pas le mêmeexposant sont significativement différentes (p

TRAVAIL EXPERIMENTALRESULTATS1) Effet des agents détoxifiants sur le gain de poids, l’hématologie et la biochimieLe gain de poids cumulé sur la période de 35 jours est présenté dans le Tableau 19 ci-dessous.Tableau 19 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non en agentsdésactivateurs sur le gain de poids totalSans agents désactivateursAvec agents désactivateursContrôle DON FB DON+FB Contrôle DON FB DON+FB21,0 ± 1,7 a,b 18,5 ± 1,0 a 21,6 ± 1,5 a,b,c 18,8 ± 1,5 a 25,3 ± 0,9 c 22,3 ± 0,6 b 20,5 ± 1,9 a,b 18,8 ± 0,7 aNotes : les résultats sont présentés en moyenne ± erreur type pour 5 animaux. Le gain de poids est exprimé en kilogramme.a,b,c Les valeurs ne présentant pas le même exposant sont significativement différentes (p

Figure 16 : effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ounon en agents désactivateurs sur les bases sphingoïdes, sphinganine (Sa) et sphingosine (So)(a) Exemple de chromatogrammes obtenues en HPLC, C‐17 So et C‐20 Sa étant les standards (solutionsétalons), et les pics entourés correspondant à la détection des bases Sa et So dans les échantillons.(b) Ratio Sa/So dans les plasmas des porcelets exposés aux différents régimes et pendant toute lapériode expérimentale.(a)CONTRÔLE DA‐(b)1,21,0Contrôle DA‐DON DA‐FB DA‐DON+FB DA‐******FB DA‐0,8Contrôle DA+DON DA+***0,6FB DA+DON+FB DA+**0,40,20,01 7 14 21 28 35Temps d’exposition (en jours)Notes : DA, Agents Désactivateurs. Les résultats sont exprimés en moyenne ± erreur type pour 6 animaux.*** Différence significative entre les régimes FB et DON+FB sans DA, avec les autres régimes (p

TRAVAIL EXPERIMENTAL2) Effet des agents détoxifiants sur la concentration en bases sphingoïdesLa concentration des bases sphingoïdes, sphinganine (Sa) et sphingosine (So), a été mesurée dansles plasmas des porcelets tout au long de la période expérimentale. Afin d’évaluer l’effet sur lemétabolisme des sphingolipides, les ratios Sa/So ont été calculés et sont présentés dans la Figure 16.Une nette augmentation de ce ratio, caractérisé principalement par une augmentation dessphinganines, a été observée dès le 14 ème jour d’exposition pour les animaux exposés aux FB et auxFB associées au DON (Figure 16). L’élévation du ratio est dépendante du temps d’exposition et estcinq fois supérieur au contrôle pour ces deux régimes à J35. Comme attendu, le régime monocontaminéavec le DON n’a pas induit de modification de la concentration en bases sphingoïdes. Dela même manière, aucun des régimes supplémentés en DA n’a montré de différences dans les ratios(Figure 16). Les résultats révèlent une différence largement significative entre les groupes FB etFB+DON et leurs homologues respectifs contenant les DA.3) Effet des agents détoxifiants sur l’histopathologie des organes‣ Sur le FOIELa prolifération des hépatocytes a été mesurée par immunohistochimie, le noyau des cellules enprolifération ayant été marqué par un anticorps spécifique anti-Ki-67. Les résultats sont présentés cidessousdans le Tableau 21.Tableau 21 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non en agentsdésactivateurs sur l’index de prolifération des hépatocytesSans agents désactivateursAvec agents désactivateursContrôle DON FB DON+FB Contrôle DON FB DON+FB16,4 ± 1,5 a 18,8 ± 3,3 a,b 22,8 ± 1,7 b 39,4 ± 12,8 c 16,7 ± 2,4 a 18,6 ± 2,3 a,b 21,4 ± 1,7 a,b 19,0 ± 0,9 a,bNotes : les résultats sont présentés en moyenne ± erreur type pour 5 animaux. L’index de prolifération a été calculé selon lenombre de cellules positives Ki-67/nombre total de cellules (100). a,b,c Les valeurs ne présentant pas le même exposant sontsignificativement différentes (p

TRAVAIL EXPERIMENTALEn plus de l’analyse immunohistochimique, l’observation des lames colorées à l’hématoxylineéosinea permis d’identifier différents types de lésions. Comme mentionné dans la partie Matériel etMéthodes, ces lésions ont pu être évaluées numériquement afin d’établir un score lésionnel paranimal. Ainsi, nous présentons ci-dessous dans la Figure 17, le score lésionnel en fin d’expériencepour chaque groupe :adb,db,c,db,cb,dcb,c Figure 17 :Effet de l’exposition aux mycotoxines avec unrégime alimentaire supplémenté ou non enagents désactivateurs sur le score lésionnel dufoieNotes : DA, Agents Désactivateurs. Les résultatssont présentés en moyenne ± erreur type pour 5animaux. a,b,c,d Les valeurs ne présentant pas lemême exposant sont significativementdifférentes (p

TRAVAIL EXPERIMENTALb,cca,ba,b,da,da,baFigure 18 :Effet de l’exposition aux mycotoxines avec unrégime alimentaire supplémenté ou non enagents désactivateurs sur le score lésionnel despoumonsdNotes : DA, Agents Désactivateurs. Les résultatssont présentés en moyenne ± erreur type pour 5animaux. a,b,c,d Les valeurs ne présentant pas lemême exposant sont significativementdifférentes (p

Tableau 22 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non en agents désactivateurs sur la population cellulaire etla prolifération cellulaire du jéjunum et de l’iléon.Sans agents désactivateursAvec agents désactivateursContrôle DON FB DON+FB Contrôle DON FB DON+FBJEJUNUMLymphocytes 26,3 ± 3,0 a 17,2 ± 1,1 b,c 19,3 ± 1,7 a,b 18,9 ± 1,0 b,d 21,8 ± 3,0 a,b 14,2 ± 0,8 c 18,9 ± 1,6 a,b 21,8 ± 1,3 a,dPlasmocytes 24,9 ± 1,3 a 25,3 ± 3,4 a,b 32,4 ± 2,8 b 24,8 ± 0,8 a 26,5 ± 2,4 a,b 22,4 ± 1,9 a 24,7 ± 0,9 a 24,8 ± 2,5 a,bEosinophiles 8,3 ± 0,4 a 9,2 ± 1,1 a,b 10,7 ± 0,9 b 7,4 ± 0,8 a 9,0 ± 1,2 a,b 8,9 ± 0,8 a,b 8,7 ± 0,9 a,b 9,0 ± 1,2 a,bEntérocytes en mitose 2,6 ± 0,3 a 1,5 ± 0,2 b 1,8 ± 0,2 b,c 1,8 ± 0,2 b,c 1,5 ± 0,2 b 2,3 ± 0,3 a,c 2,5 ± 0,3 a,c 2,8 ± 0,2 aILEONLymphocytes 20,3 ± 2,7 a 16,3 ± 1,1 a,b 18,2 ± 1,4 a 12,9 ± 1,9 b 18,4 ± 1,9 a 17,1 ± 2,0 a,b 18,6 ± 1,9 a 19,1 ± 2,7 a,bPlasmocytes 22,0 ± 2,0 a 22,7 ± 1,9 a 24,8 ± 3,0 a,b 26,8 ± 0,8 b 20,0 ± 1,3 a 22,7 ±1,6 a 22,2 ± 1,7 a 22,4 ± 1,8 aEosinophiles 12,0 ± 0,9 a 12,5 ± 1,8 a,b 11,4 ± 1,6 a,b 8,9 ± 0,6 b 15,5 ± 2,5 a 11,1 ± 1,7 a,b 11,7 ± 2,4 a,b 9,9 ± 1,3 a,bEntérocytes en mitose 1,9 ± 0,3 a,b 1,9 ± 0,4 a,b 1,7 ± 0,2 a 2,0 ± 0,1 a 1,9 ± 0,3 a,b 2,1 ± 0,3 a,b 1,9 ± 0,3 a,b 2,5 ± 0,2 bNotes : les résultats sont présentés en moyenne ± erreur type pour 5 animaux. a,b,c,d En ligne, les valeurs ne présentant pas le même exposant sont significativementdifférentes (p

TRAVAIL EXPERIMENTALL’exposition individuelle au DON a réduit significativement la hauteur des villosités dans lejéjunum des porcelets (Figure 19). La présence des DA dans le régime respectif a partiellementneutralisé l’altération induite par le DON seul. Toutefois, une différence non négligeable entre lesdeux groupes contrôles a été mesurée.Le nombre de lymphocytes, de cellules plasmatiques et d’éosinophiles a été évalué dans la laminapropria de l’intestin. Le dénombrement de ces cellules est reporté dans le Tableau 22, ainsi que laprolifération des cellules, caractérisée par le nombre d’entérocytes en mitoses.La population lymphocytaire était significativement réduite après consommation du régimemono-contaminé en DON dans le jéjunum, et du régime co-contaminé en DON et FB dans le jéjunumet l’iléon (Tableau 22). Un nombre plus élevé de plasmocytes et d’éosinophiles après exposition auxFB seules a été noté dans le jéjunum des animaux. Dans l’iléon, le groupe co-contaminé a augmentéle nombre de plasmocytes et diminué le nombre d’éosinophiles. De manière intéressante, le jéjunumdes porcelets était plus affecté par les régimes ne contenant qu’une mycotoxine, alors qu’au niveaude l’iléon les deux toxines en association ont montré un effet plus important. Ces effets sur lapopulation cellulaire de la lamina propria du jéjunum et de l’iléon n’ont pas été observés en présencedes DA, excepté sur les lymphocytes de la section jéjunale pour le régime mono-contaminé en DON(Tableau 22).Le dénombrement des entérocytes en mitoses a montré une baisse significative de la proliférationcellulaire pour les régimes mono- et co-contaminés dans le jéjunum (Tableau 22). Cet effet n’a pasété observé dans la section iléale de l’intestin. L’addition des DA a contrecarré la diminution de laprolifération jéjunale observée dans les régimes homologues non supplémentés. Toutefois, le régimecontrôle avec les DA présentait un nombre de cellules en mitoses aussi faible que dans les régimesmono-contaminés non-supplémentés. A noter que dans l’iléon, la prolifération des entérocytes étaitplus élevée pour le groupe DON+FB+DA (Tableau 22).4) Effet des agents détoxifiants sur le système immunitaireAucun effet des mycotoxines n’a été observé sur la réponse immunitaire totale et non-spécifique.L’ingestion des régimes contaminés n’a ni altéré la synthèse des immunoglobulines (Ig) G et A totauxdans le plasma, et ni modifié la prolifération des lymphocytes après stimulation par le mitogèneConcanavaline A.Néanmoins, le protocole d’immunisation avec l’ovalbumine nous a permis d’examiner ledéveloppement de la réponse cellulaire et humorale, spécifique de cet antigène. Les résultats sontreportés dans les sections ci-après.153

Figure 20 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non en agents désactivateurs sur la prolifération deslymphocytes après stimulation antigénique, à J28 et J35aJ28J35aa,cb,cb,c,db,c,db,cba,dab,ca,bba,bba,cNotes : DA, Agents Désactivateurs. Les résultats sont exprimés en index de prolifération des lymphocytes (cpm dans les cultures stimulées / cpm dans les cultures contrôlesnon-stimulées). Les valeurs sont présentées en moyenne ± erreur type pour 5 animaux. a,b,c,d Les valeurs ne présentant pas le même exposant sont significativementdifférentes (p

Figure 21 : Effet de l’exposition aux mycotoxines avec un régime alimentaire supplémenté ou non en agents désactivateurs sur la concentration plasmatiquedes immunoglobulines A et G spécifiques de l’ovalbumine, à J28 et J35bIgA : J28bIgA : J35aaa,baaaa,baaa,baaaaaa,d a,b a,c,db,cb,dIgG : J28bb,daaa,bab,ca,cIgG : J35cb,cNotes : DA, Agents Désactivateurs. Les résultats sont exprimés en unités arbitraires, normalisés avec un plasma standard de référence. Les valeurs sont présentées enmoyenne ± erreur type pour 5 animaux. a,b,c,d Les valeurs ne présentant pas le même exposant sont significativement différentes (p

TRAVAIL EXPERIMENTAL‣ Sur la réponse spécifique CELLULAIRELa composante cellulaire de la réponse vaccinale a été évaluée par la mesure de la proliférationdes lymphocytes, suite à la mise en culture et la stimulation des cellules du sang total avec l’antigènevaccinal, l’ovalbumine. Comme attendu, les cellules du sang n’ont pas proliféré en présenced’ovalbumine avant la vaccination des animaux (jour 1, cf. chapitre 1). Suite à l’injectiond’ovalbumine et notamment après la deuxième injection (rappel), une augmentation significative del’index de prolifération (cpm dans les cultures stimulées / cpm dans les cultures contrôles nonstimulées)a été observée pour les animaux contrôles. Dans la Figure 20 est représentée l’index deprolifération des lymphocytes, 12 jours après la seconde injection d’ovalbumine.Les lymphocytes des porcelets exposés aux régimes mono- et co-contaminés montrent tous unediminution significative, par rapport au contrôle, de la capacité à proliférer après stimulation avecl’ovalbumine (Figure 20). La prolifération reste aussi faible que celle des lymphocytes contrôles nonstimulés.Pour les régimes supplémentés avec les DA, le contrôle est non significativement différentde son homologue non supplémenté, mais l’index prolifératif est néanmoins inférieur. Une netteamélioration de la prolifération a été obtenue lorsque les groupes contaminés DON et DON+FBcontenaient les DA, en comparaison des régimes respectifs sans DA (DON+DA à J35, 1,9 foissupérieur ; DON+FB+DA à J28 et à J35, 2,8 et 2,2 fois supérieur, respectivement).‣ Sur la réponse spécifique HUMORALELa composante humorale de la réponse vaccinale a été évaluée par la mesure du contenu en IgAet IgG plasmatiques reconnaissant spécifiquement l’antigène vaccinal, l’ovalbumine. Comme pour laprolifération des lymphocytes, la concentration en Ig spécifiques de l’ovalbumine était insignifianteavant la vaccination des animaux (Jour 1, cf. chapitre 1). Leurs concentrations ont fortementaugmenté suite à la deuxième administration d’ovalbumine, en particulier pour les IgG. Dans laFigure 21, nous présentons de la même manière que pour les résultats de la prolifération, lesconcentrations des IgA et IgG dirigées contre l’ovalbumine, 12 jours après la seconde injection del’antigène.Concernant les IgA, une élévation significative de leurs concentrations a été observée dans lesplasmas des animaux ayant ingéré le DON seul (Figure 21, +73% et +61% à J28 et J35respectivement). En revanche, le DON combiné aux FB n’a pas entraîné de différences significativesavec le contrôle. L’ajout de DA dans le régime mono-contaminé en DON a permis de réduire cettealtération des IgA.154

1,41,21,00,80,60,40,20,0IL‐8a,cc ccca,bbbDA‐ DA+ DA‐ DA+ DA‐ DA+ DA‐ DA+1,61,41,21,00,80,60,40,20,0MIP‐1βcc,da,d a,da,d a,dabDA‐ DA+ DA‐ DA+ DA‐ DA+ DA‐ DA+Figure 22 :Effet de l’exposition aux mycotoxines avec unrégime alimentaire supplémenté ou non enagents désactivateurs sur l’expression desARNs spléniques codant pour des cytokinesNotes : DA, Agents Désactivateurs. Les valeurssont présentées en moyenne ± erreur type pour 5animaux. a,b,c,d Les valeurs ne présentant pas lemême exposant sont significativement différentes(p

TRAVAIL EXPERIMENTALConcernant les IgG, une diminution significative de leurs concentrations a été observée dans lesplasmas des animaux ayant ingéré les FB seules, et surtout les FB associées au DON (Figure 21, -43%et -30% à J28 et J35 pour FB, respectivement, et -58% et -49% à J28 et J35 pour DON+FB,respectivement). L’ajout de DA n’a que très faiblement amélioré le changement induit par les FBseules, mais a partiellement restauré la dépression observée par les FB associées au DON (1,6 fois et1,3 fois mieux à J28 et J35 respectivement).‣ Sur l’expression des ARNm codant des CYTOKINESAfin de compléter le tableau sur le développement de la réponse vaccinale, nous avons évaluél’expression dans la rate, organe lymphoïde secondaire, des ARNm codant pour les cytokines IL-1β,IL-6, IL-12p40, IL-8 et MIP-1β. Ces cytokines sont connues pour être impliquées, en tant quemessagers entre les cellules immunitaires, dans le développement de la réponse immunitairespécifique.Tous les traitements en mycotoxines ont affecté à J35 l’expression de ces messagers dans la rate,en particulier le traitement co-contaminé (Figure 22, -39%, -46%, -33%, -47% et -32% d’expressiondes transcrits codant respectivement pour IL-1β, IL-6, IL-12p40, IL-8 et MIP-1β dans la rate desanimaux traités avec DON+FB en comparaison des animaux contrôles). L’ajout des DA dans lesrégimes contaminés en mycotoxines a en revanche nettement amélioré l’expression de ces ARNm,avec un profil de cytokines similaire à celui des deux groupes contrôles (Figure 22). Toutefois, uneexception a été remarquée, avec le traitement FB+DA provoquant une augmentation significative del’expression de MIP-1β.155

TRAVAIL EXPERIMENTALDISCUSSIONLes objectifs de nos travaux conduits avec différents groupes expérimentaux étaient (i) d’évaluerla toxicité induite par une combinaison de mycotoxines, le déoxynivalénol (DON) et la fumonisine(FB) deux fusariotoxines d’intérêt majeur, en comparaison de leur toxicité individuelle ; (ii) d’évaluerces expositions chez le porc et à de faibles doses, le modèle animal choisi étant pertinent par rapportà son exposition naturelle via la composition de sa ration alimentaire, sa sensibilité et ses similitudesavec l’homme, et le choix des doses étant approprié par rapport aux situations rencontrées dans lesélevages ;(iii) d’évaluer l’efficacité d’agents désactivateurs (DA) dans la lutte menée contre les effetsnocifs de ces mycotoxines ;(iiii) et d’analyser la réponse des animaux sur des paramètres plus fins etplus ciblés que l’étude de paramètres zootechniques.Toutefois, nous ne discuterons ici que de l’effet des agents désactivateurs, conçus par la sociétéBIOMIN, considérant que les effets des mycotoxines, seules ou en combinaison, sur la réponsesystémique et intestinale ont déjà été discutés dans le chapitre 1. Pour rappel, un des agentsdésactivateurs cible spécifiquement la mycotoxine DON, et l’autre spécifiquement les FB.Les régimes contaminés en mycotoxines n’ont pas eu d’effets significatifs sur la croissance desporcelets tout au long de la période expérimentale. Pourtant l’ajout des DA dans les régimes a eu uneffet non spécifique mais bénéfique sur la prise de poids des animaux, en particulier pour ceux desgroupes contrôle et DON. Cet effet sur le gain de poids a déjà été reporté lors de l’utilisation deproduits similaires développés par BIOMIN (Hanif et al., 2008). Dans notre étude, nous pouvonsattribuer le peu d’effets observés sur les paramètres hématologiques et biochimiques à l’utilisationde faibles doses de mycotoxines. Néanmoins, ces effets mineurs ont été totalement neutralisés par laprésence des DA dans les aliments contaminés. La baisse du nombre de neutrophiles dans lesrégimes FB et FB+DON n’a pas été observée dans les régimes respectifs supplémentés. De même, lesmodifications biochimiques provoquées par les FB seules sur la concentration en créatinine, et par leDON seul sur la concentration en albumine, n’ont pas été observées dans les régimes respectifssupplémentés.La perturbation du métabolisme des sphingolipides par les fumonisines a largement été décritedepuis de nombreuses années (Wang et al., 1991), et reconnue comme responsable de la toxicité deces toxines. Comme précédemment reporté dans le manuscrit, les fumonisines inhibent la céramidesynthase et provoquent une accumulation des bases sphingoïdes, sphinganine (Sa) et sphingosine(So). Ainsi, la détermination du ratio Sa/So dans les fluides et tissus biologiques représente un bonmarqueur d’exposition et de toxicité (Tran et al., 2006). Comme attendu, les animaux exposés auxrégimes contenant des FB présentaient un ratio largement supérieur aux animaux non exposés aux156

TRAVAIL EXPERIMENTALFB. A noter que la présence de DON avec les FB n’a pas entraîné de potentialisation ou encore d’effetantagoniste ; le ratio étant égal à celui du régime mono-contaminé pendant les 35 jours d’exposition.A l’inverse, la supplémentation en DA dans ces deux régimes a semblé totalement réprimer l’actiondes FB sur le métabolisme des sphingolipides, comme en témoigne la mesure des ratios Sa/So. Ainsi,il est fortement probable que la carboxylestérase, isolée par BIOMIN et pulvérisée sur les alimentscontaminés, ait hydrolysée totalement les fumonisines dans les conditions du tractus gastrointestinal.Le foie et les poumons étant des organes cibles de ces mycotoxines, en particulier des FB, etl’intestin étant un organe naturellement exposé aux toxines suite à l’ingestion d’alimentscontaminés, l’examen microscopique de ces tissus était ainsi nécessaire dans l’évaluation du risquemycotoxique. De plus, peu d’études ont reporté l’effet de faibles doses sur l’histopathologie desorganes. Les résultats de nos analyses ont révélé que les mycotoxines seules ou en associationinduisaient des lésions hépatiques et pulmonaires, et altéraient l’intégrité intestinale (cf. chapitre 1).Les effets sur le foie et les poumons supposent une biodisponibilité et/ou une activation desmycotoxines suite à leur absorption intestinale. Ainsi, une efficacité optimale des agents détoxifiantssuggère une biotransformation totale des mycotoxines dans les conditions du tube digestif, réduisantainsi leur biodisponibilité. Le procédé ne doit pas non plus conduire à la production de métabolitestoxiques. L’analyse des sections de foie des animaux ayant ingéré les régimes supplémentés en DA arévélé la présence de lésions hépatiques. De manière surprenante et inattendue, les animauxcontrôles présentaient un score lésionnel élevé, significativement différent du score du groupecontrôle non supplémenté. Cette différence de scores entre les deux groupes contrôles étaitessentiellement attribuée à la présence de mégalocytoses, comme en témoigne l’examen des lameshistologiques des contrôles DA+. Par conséquent, la présence de mégalocytoses chez les porceletsnourris avec les régimes contaminés et supplémentés, est vraisemblablement liée à un effet propredes agents détoxifiants qu’à un effet des mycotoxines. Dans ce sens, nous pouvons supposer que labiodisponibilité des FB a été fortement réduite, si l’on soustrait le score du groupe contrôle DA+ àcelui du régime FB DA+. De la même manière, les scores obtenus pour les groupes DON et DON+FBsupplémentés en DA pourraient être attribués d’une part à l’effet des DA, et d’autre part à un effetpartiel du DON et/ou de son métabolite de biotransformation.A l’inverse, la présence des DA dans les différents régimes n’a pas induit une augmentation de laprolifération des hépatocytes, et a grandement réduit l’index de prolifération observé dans le groupeco-contaminé en DON et en FB non supplémenté. L’addition des DA dans l’alimentation a égalementneutralisé l’effet des FB dans les poumons des animaux. Comme déjà mentionné dans ce mémoire etlargement décrit dans la littérature, les FB ingérées à de fortes concentrations peuvent induire des157

TRAVAIL EXPERIMENTALœdèmes pulmonaires chez le porc (Haschek et al., 2001). A faibles doses, les effets sont beaucoupmoins connus sur le poumon, mais dans notre étude que ce soit seules ou en combinaisons, les FBont aussi provoqué des lésions pulmonaires. Ces lésions étaient absentes des groupes traités avec lesproduits détoxifiants. Ici également, nous pouvons supposer que la biodisponibilité des FB a étéréduite, via son hydrolyse totale dans l’intestin par la carboxylestérase.Bien que la principale voie d’exposition aux mycotoxines soit la voie entérale, l’impact de cesmétabolites sur le tractus gastro-intestinal est une thématique encore peu étudiée. Nos résultats (cfchapitre 1) ont pourtant montré que même à faibles doses, les mycotoxines pouvaient affectercertaines fonctions et caractéristiques des cellules de l’épithélium. L’effet sur la diminution de lahauteur des villosités dans le jéjunum, après ingestion du régime mono-contaminé en DON, a étébloqué lors de l’ajout des DA dans ce régime. En revanche, une baisse de la hauteur des villosités aété notée pour le groupe contrôle supplémenté en DA, avec une valeur intermédiaire entre cellesdes deux groupes DON supplémenté ou non. Cette observation pourrait être liée à un plus faiblenombre de cellules en mitoses dans la même région, puisque les entérocytes qui composent lesvillosités et les cryptes sont des cellules à division rapide. De plus, l’effet du DON avec ou sans DA surles cellules en mitoses corrobore les données sur les villosités. Aucun effet sur ces paramètres n’a étéconstaté dans la partie iléale.La supplémentation en DA dans le régime mono-contaminé en DON n’a pas permis de réduirel’effet sur la diminution de l’infiltration lymphocytaire dans le jéjunum. Elle l’a néanmoins réduitedans le groupe co-contaminé que ce soit dans le segment jéjunal ou iléal. A l’inverse de l’effet sur leslymphocytes, une plus forte infiltration de cellules plasmatiques et d’éosinophiles dans le jéjunum aété détectée dans le régime mono-contaminé en FB. Aucune infiltration significative n’a été reportéepour les régimes supplémentés en DA. De la même manière, les effets sur ces populations cellulairesdans l’iléon ont été partiellement ou totalement résorbés en présence de DA.Un des principaux objectifs de cette étude était d’étudier la réponse des animaux traités suite àun protocole d’immunisation. En effet, la présence de pathogènes et la vaccination sont dessituations courantes dans les élevages, et considérant l’effet immunomodulateur des mycotoxines, ledéveloppement de la réponse spécifique suite à un challenge antigénique a été examiné. Commeexpliqué dans le chapitre 1, nous avons montré que suite à l’immunisation des porcelets avecl’ovalbumine, les faibles doses de DON et de FB utilisées affectaient significativement la mise enplace d’une réponse adaptée, et notamment lorsque ces toxines étaient en combinaison. Cette miseen place de la réponse vaccinale chez les animaux contrôles était optimale après la seconde injectiond’ovalbumine. Les données après ce rappel vaccinal ont montré que la prolifération des lymphocytesen culture après stimulation avec l’antigène vaccinal était altérée par les mycotoxines. Un effet158

TRAVAIL EXPERIMENTALbénéfique des DA n’a été observé qu’en fin d’expérience, et en particulier pour le régime cocontaminé.En plus de l’analyse de la réponse cellulaire, nous avons analysé la composante humorale par lebiais du titrage des immunoglobulines A et G reconnaissant spécifiquement l’ovalbumine. Commelargement décrit dans la littérature (Pestka et al., 2003, 2004, 2005; Pinton et al., 2008) et confirmédans notre étude, l’ingestion continue en DON induit la synthèse des IgA. Cette altération de laconcentration plasmatique des IgA anti-ovalbumine n’a pas été constatée après traitement durégime mono-contaminé en DON avec les DA. L’analyse des IgG spécifiques, qui est la classed’anticorps principalement produite en réponse à un antigène, a montré une nette diminution deleurs synthèses après exposition aux régimes FB et DON+FB. Ces régimes supplémentés en DA ontprésenté un meilleur profil d’IgG en fin d’expérience, bien que l’amélioration n’ait été que partielle.Une restauration partielle du titre d’anticorps anti-pseudo-rage a déjà été reportée chez desporcelets alimentés avec un régime co-contaminé en DON et zéaralénone (ZEA) et supplémenté avecl’agent désactivateur ciblant spécifiquement le DON, mais aussi avec un spécifique de la ZEA (Chenget al., 2006).Considérant que les cytokines sont des messagers importants dans le développement desréponses immunitaires, nous avons évalué à la fin de l’expérience l’expression de transcrits codantpour des cytokines dans la rate des animaux. Au vu des résultats des régimes contaminés et nonsupplémentés sur les cytokines IL-8, MIP-1β, IL-6, IL-1β et IL-12p40, nous nous sommes restreints àl’analyse de leur expression pour les régimes supplémentés. Les effets observés après l’ingestion desmycotoxines ont été considérablement atténués en présence des DA pour ces cinq médiateurs del’immunité.En conclusion, dans notre étude le bénéfice apporté par ces méthodes de détoxification parbiotransformation pouvait être de partiel à total en fonction des paramètres étudiés. Cette étude apermis de confirmer l’action efficace de la carboxylestérase sur la FB dans les conditons du tractusgastro-intestinal, et ainsi, de réduire la biodisponibilié de cette mycotoxine. Néanmoins, lasupplémentation des deux DA dans l’aliment semblerait induire quelques effets non spécifiques etjustifierait ainsi quelques études complémentaires.159

DISCUSSIONGENERALE160

Figure 23 :(a) Cheptel porcin des principaux pays producteurs en 2005 (source : SNCP, SyndicatNational du Commerce du Porc)(b) Évolution des échanges mondiaux de viande de porc(a)(b)

DISCUSSION GENERALEDans ce travail de thèse, nous avons étudié l’effet des mycotoxines chez le porcelet, notre modèleanimal lors des phases d’expérimentations, ainsi que des stratégies de lutte contre les mycotoxines.Nos résultats ont principalement été présentés sous la forme de publications scientifiques, etdiscutés par type d’étude. Ainsi dans cette discussion générale, nous nous attacherons à discuter ducontexte et des critères expérimentaux, puis nous considérerons les résultats par systèmes dedéfense de l’organisme, d’abord le système immunitaire et ensuite le système digestif. Enfin, nousnous interrogerons sur de possibles apports des effets protecteurs des micro-organismes/enzymesde biotransformation.1. Critères expérimentauxLE MODELE ANIMALL’élevage de porcs et la production de viande porcine représentent un intérêt majeur dansl’économie mondiale, et ce secteur de l’agro-alimentaire ne cesse de croître depuis une décenniesuite à l’intensification des échanges mondiaux (Figure 23). A l’échelle mondiale, la filière porcine estle chef de file des filières animales, avec 37% de la production mondiale de viande, devant lesviandes de volailles et bovines (année 2008, source FAO). Ainsi, cette économie de marché suscite denombreuses attentions quant à la qualité et au rendement de la production porcine, mais égalementen amont, à la qualité nutritionnelle de son alimentation. Du fait de sa composition riche en céréales(blé, orge, maïs, seigle), le porc est naturellement exposé à la contamination mycotoxique.Considérant qu’une grande partie de la production céréalière mondiale est destinée aux animauxd’élevage, dont le porc, la problématique des mycotoxines est au cœur du secteur de l’agroalimentaireet entraîne des pertes économiques considérables (e.g. pertes totales associées auxmycotoxines estimées entre 630 millions et 2,5 milliards de dollars annuellement aux Etats-Unis (Wu,2004)).De plus, même si l’on peut affirmer que toutes les espèces peuvent être affectées par lesmycotoxines, le porc est l’espèce la plus sensible à la plupart des mycotoxines. Cette observation esttraduite par l’établissement de seuils en mycotoxines, fixés par la commission européenne, plus basque pour les autres espèces cibles (Tableau 23). Cette sensibilité du porc aux effets des mycotoxines,est attribuée à l’absence de rumen qui joue un rôle important dans la détoxification des toxines. Lesruminants sont ainsi moins sensibles aux effets des mycotoxines, via l’action des micro-organismesdu rumen. Cependant, bien que les ruminants soient tolérants, le stress lié à la haute productivité161

Tableau 23 : Teneurs maximales recommandées en mg/kg (ppm) pour un aliment pour animauxayant un taux d’humidité de 12% (basé sur la recommandation de la Commission Européenne du 17août 2006 pour le déoxynivalénol, la zéaralénone, l’ochratoxine A et les fumonisines,recommandation 2006/576/CE ; basé sur la directive de la Commission Européenne du 31 octobre2003 pour l’aflatoxine B1, directive 2003/100/CE)ALIMENTS COMPLEMENTAIRES ETCOMPLETS POUR :Porcs Porcelets, Jeunes truies Truies, Porcs d’engraissementMycotoxines évaluées par l’UEDON ZEA OTA FB AFB10,90,90,10,250,050,0555 0,02Volaille 5 0,1 20 0,02Ruminants Adultes (bovins, ovins, caprins) Jeunes (veaux, agneaux, chevreaux)520,50,550200,020,01Bétail laitier 5 0,5 0,005Equidés, Lapins 5 5Poissons 5 10Notes : DON, Déoxynivalénol ; ZEA, Zéaralénone ; OTA, Ochratoxine A ; FB, Fumonisines ; AFB1, Aflatoxine B1.Les champs vides représentent des mycotoxines non évaluées par l’UE chez certaines espèces, par manque detoxicité ou de données toxicologiques.

DISCUSSION GENERALEdes vaches laitières et à la croissance rapide des bouvillons ou des agneaux lourds diminue lepotentiel de détoxification du rumen.Cette sensibilité du porc aux mycotoxines varie également selon le stade de production et laprésence de facteurs de stress dans le troupeau. Les porcelets sont généralement plus sensibles auxeffets d’une contamination par les mycotoxines dans leur alimentation. Leur système immunitaireétant plus faible, ils sont moins aptes à combattre les effets néfastes causés par ces mycotoxines.S’ajoute également les animaux en stade de reproduction et ce, peu importe l’espèce. Il sembleraitaussi que les mâles soient plus sensibles que les femelles à une intoxication par les mycotoxines(Cote et al., 1985; Marin et al., 2006).Outre ces facteurs agronomiques qui justifient le choix de notre modèle animal, le porc présentedes caractéristiques très proches de celles de l’homme pour différents systèmes biologiques. Eneffet, de nombreuses études attirent l’attention sur le choix du porc comme modèle expérimental derecherche, plutôt que le modèle rongeur par défaut. Il est acquis que l’utilisation des rongeurs enlaboratoire repose plus sur une faisabilité économique et technique que sur un choix dicté par lascience. Ainsi, du fait des similitudes en taille et en caractéristiques physiologiques des organes,l’espèce porcine Sus scrofa offre de nombreux avantages en tant que modèle animal dans l’étude desxénotransplantations, des maladies pulmonaires et cardiovasculaires, du système reproductif, desfonctions métaboliques, des désordres digestifs, ou encore des pathologies immunologiques (Millerand Ullrey, 1987; Rothkotter et al., 2002; Guilloteau et al., 2010; Verma et al., 2011). Par ailleurs,cette espèce partage de fortes homologies de séquences avec celles de l’homme, permettant ainside développer et d’utiliser des outils de transcriptomique et de protéomique dérivés des banques dedonnées humaines, dont les séquences sont très largement accessibles contrairement à celles duporc. Et enfin, notre équipe Immuno-Mycotoxicologie au sein du pôle ToxAlim de l’INRA a développédes approches in vitro (lignées de cellules épithéliales intestinales porcines, IPEC-1) et ex vivo(explants de jéjunum porcin (Kolf-Clauw et al., 2009)) à partir de tissus porcins, et permettent ainside comparer et valider ces modèles avec le modèle in vivo.162

DISCUSSION GENERALEL’EXPOSITION AUX MYCOTOXINES ET A LEURS DERIVES1) Modes d’exposition choisiesDans les phases d’expérimentations animales, le mode d’administration des mycotoxines a étédifférent d’une étude à l’autre. D’une part, nous avons formulé différents régimes alimentaires,artificiellement contaminés en mycotoxines, et les animaux étaient nourris avec ces régimes sansrestriction dans la prise alimentaire. D’autre part, nous avons administré oralement, par gavage desanimaux, des solutions de fumonisine B1 et de son dérivé hydrolysé, et ce en fonction du poids vifdes porcelets. Les volumes à administrer étaient ainsi ajustés tous les deux jours pendant la périodeexpérimentale.Cette dichotomie dans le mode d’administration des mycotoxines a comme principaleconséquence une différence dans l’exposition réelle aux toxines. En effet, dans le cas d’ingestiond’aliments contaminés ad libitum, nous pouvons aisément supposer que chaque animal dans unmême régime n’a pas ingéré la même quantité de mycotoxines tout au long de la phased’intoxication. Ce type d’exposition, dépendant de la quantité de nourriture consommée, pourraitêtre un des facteurs expliquant la variabilité de nos résultats. Toutefois, ces conditions reflètentexactement celles des élevages, et le peu d’effets négatifs observés dans certains cas sur les animauxpeut être expliqué par un refus alimentaire, et donc une diminution de la consommation. Lorsd’études d’association de mycotoxines, cette observation a déjà été reportée, et mise en cause dansle manque d’effets du régime combiné (effets moins qu’additif ou antagoniste). A l’opposé, le moded’administration orale a permis d’exposer chaque animal d’un même groupe à la mêmeconcentration de substances tout au long de l’expérience. Ce protocole expérimental permet deréaliser des études courtes et de comparer précisément la toxicité de deux substances à doseségales.2) Doses d’exposition choisiesDe la même manière, dans nos approches expérimentales nous avons utilisé des concentrationsde mycotoxines différentes. A noter que nous avons utilisé des aliments artificiellement contaminés(dans études chapitre 1 et 3) plutôt que des aliments formulés à base de lots de céréalesnaturellement contaminés. Considérant que le développement naturel de souches fongiques sur unecéréale modifie ses qualités nutritionnelles, et qu’il est difficile d’éviter les multi-contaminations, cechoix nous a permis d’assurer une homogénéité dans la qualité des lots et d’attribuer leschangements observés à la seule exposition alimentaire au DON et à la FB.163

Tableau 24 : Analyse de denrées agricoles de janvier à décembre 2009 et détermination des niveaux decontaminations mondiales en mycotoxines (enquête réalisée par la société BIOMIN). Au total, 2727échantillons ont été traités et évalués pour la présence de mycotoxines majeures. Les produits analysés étaientde nature diverses, comprenant des céréales (blé, maïs, orge, riz) et leurs co-produits (tourteaux de soja, glutende maïs, DDGS), mais aussi des fourrages (paille, ensilage) et des produits finis.ASIE AFB1 ZEA DON FB OTANombre d’échantillons testés 988 936 829 1006 856Positifs (%) 34 48 47 49 26Moyenne des positifs (µg/kg) 109 255 826 1694 13Maximum (µg/kg) 6105 7422 11836 32510 1582EUROPE AFB1 ZEA DON FB OTANombre d’échantillons testés 125 758 933 44 102Positifs (%) 10 17 56 45 41Moyenne des positifs (µg/kg) 3 94 725 2930 4Maximum (µg/kg) 6 973 6000 11050 21AMERIQUE DU NORD AFB1 ZEA DON FB OTANombre d’échantillons testés 102 148 163 85 37Positifs (%) 20 34 79 65 35Moyenne des positifs (µg/kg) 48 538 1286 1744 4Maximum (µg/kg) 831 8952 29300 11381 13AMERIQUE DU SUD AFB1 ZEA DON FB OTANombre d’échantillons testés 217 218 213 217 45Positifs (%) 47 51 9 87 9Moyenne des positifs (µg/kg) 9 185 372 4965 247Maximum (µg/kg) 342 1740 1027 23100 355MOYEN-ORIENT AFB1 ZEA DON FB OTANombre d’échantillons testés 122 121 119 126 6Positifs (%) 16 14 37 39 67Moyenne des positifs (µg/kg) 66 94 402 716 22Maximum (µg/kg) 388 392 1620 2948 31AFRIQUE AFB1 ZEA DON FB OTANombre d’échantillons testés 95 88 88 88 7Positifs (%) 85 44 52 80 86Moyenne des positifs (µg/kg) 91 60 569 1482 10Maximum (µg/kg) 556 310 3859 10485 12AFRIQUE DU SUD AFB1 ZEA DON FB OTANombre d’échantillons testés 81 71 82 82 2Positifs (%) 4 24 88 62 0Moyenne des positifs (µg/kg) 3 93 1403 831 0Maximum (µg/kg) 7 293 11022 4398 0

DISCUSSION GENERALEDans l’étude des effets du DON et de la FB, seuls ou en combinaison, les doses ingérées par lesanimaux représentent des niveaux de contamination naturellement retrouvés dans les matièresbrutes utilisées dans les exploitations agricoles et d’élevages (Tableau 24), et sont proches desrecommandations individuelles fixées par la commission européenne pour un aliment complet chezle porc (Tableau 23). Par ailleurs, notre objectif n’était pas d’induire des manifestations cliniques,comme très souvent reporté pour de fortes concentrations en mycotoxines, mais plutôt dedéterminer l’effet d’une faible multi-contamination sur des paramètres plus ciblés ; paramètres dontles professionnels de la santé animale et de l’agro-alimentaire sont encore peu sensibilisés.Parallèlement, si l’on se place dans le cadre d’expositions alimentaires chez l’homme tellesqu’elles peuvent être rencontrées dans les pays en voie de développement ou dans des régionsassurant leur autonomie via une production alimentaire locale, les doses que nous avons choisiesprésentent aussi un intérêt en termes de santé publique (Creppy, 2002). Il n’est pas rare dans cespopulations que des épisodes d’intoxications aigües en mycotoxines surviennent ou que desexpositions chroniques à des doses faibles ou modérées engendrent des troubles digestifs (Bhat etal., 1997) ou encore des effets cancérigènes (Li et al., 2001).Comme précédemment mentionné, l’étude sur la comparaison de la toxicité entre la FB1 et sondérivé hydrolysé nécessitait une analyse à concentrations égales, et aussi à concentrationsprovoquant une toxicité aigüe après ingestion de FB1. Pour cela nous avons décidé d’utiliser unedose de 2 mg FB1/kg de poids corporel, équivalente à une alimentation contaminée àapproximativement 40 mg/kg d’aliment (en tenant compte de la consommation alimentaire d’unporcelet âgé de 4 à 6 semaines), et qui a été déjà reportée comme toxique chez le porc sur unecourte période (Riley et al., 1993).3) Périodes d’exposition choisiesL’étude des effets du DON et de la FB, seuls ou en combinaison et supplémentés avec ou sans DA(Agents Désactivateurs), s’est réalisée sur une période couvrant 35 jours d’exposition aux différentsrégimes. Cette durée nous a permis d’examiner les effets des régimes suite à une expositionchronique à des doses subcliniques, et surtout de réaliser un protocole d’immunisation avec unantigène de laboratoire, l’ovalbumine. Les animaux au cours de ces 5 semaines d’expérimentationont pu développer leur propre réponse vaccinale, telle que la réponse mise en place suite à unchallenge antigénique. Dans notre équipe, plusieurs études ont en effet montré que les mycotoxinesà des doses faibles ou modérées induisaient peu d’effets sur le système immunitaire non-activé, maisqu’une fois stimulé les mycotoxines ciblaient spécifiquement les processus intervenant dans la miseen place d’une réponse immunitaire spécifique (Taranu et al., 2005; Pinton et al., 2008; Meissonnier164

DISCUSSION GENERALEet al., 2008b). Ce constat a été confirmé lors de ce travail de thèse, avec un effet amplifié lors del’association des deux mycotoxines. Nous avons ainsi pu suivre l’évolution et l’altération de cetteréponse vaccinale tout au long des 5 semaines, par le biais de prélèvements sanguinshebdomadaires. Cette durée d’exposition permet également de rendre compte de l’équilibrehoméostatique sur une période assez longue, en déterminant si les animaux deviennent tolérants àl’ingestion chronique de ces contaminants alimentaires ou au contraire s’ils ne parviennent pas àréguler ou contrôler certaines variables et fonctions biologiques. En effet, certaines études, enparticulier chez les rongeurs et pour le DON, ont montré des altérations importantes dans lespremières heures d’une exposition aigüe (Amuzie et al., 2008; Kinser et al., 2004; Zhou et al., 1997,2003), mais nos résultats confirment également que certains critères restent affectés lors d’uneexposition chronique (valeurs hématologiques et biochimiques, histopathologie, immunité, intégritéintestinale).L’étude de FB1 et HFB1 a quant à elle était réalisée sur une période d’exposition plus courte, 14jours. Ici, nous nous sommes placés dans des conditions d’exposition où la dose de FB1 utilisée et sonadministration quotidienne permettaient d’induire rapidement une toxicité aigüe et de maintenircette toxicité tout au long des deux semaines d’expérience. Contrairement à l’étude précédente,nous n’avons pas immunisé les animaux, la durée de l’expérience ne s’y prêtait pas et s’orientait plusvers une étude de toxicologie générale.165

DISCUSSION GENERALELES SUPPLEMENTS ALIMENTAIRES : MICRO-ORGANISME/ENZYME A PROPRIETESDETOXIFIANTESDeux types d’agents désactivateurs des mycotoxines ont été utilisés dans nos études. Le premiercorrespondait à une enzyme, plus particulièrement une carboxylestérase, qui a initialement étéisoléé à partir d’une bactérie de compost, du genre Sphingopyxis. Après séquençage du génomeentier de cette bactérie et identification dans le cluster du gène codant pour cette enzyme etresponsable de la dégradation de la FB1, le gène a été cloné puis exprimé en système hétérologue.Afin d’améliorer la solubilité de la protéine recombinante et d’éviter la formation de corpsd’inclusion chez E. coli, l’enzyme a été exprimée chez Pichia pastoris, et récupérée dans lessurnageants de culture (Heinl et et al., 2010). La solution de fumonisine B1 hydrolysée (HFB1) a étéobtenue après hydrolyse totale de la FB1 avec cette enzyme, et a été utilisée dans notre étudecomparative de toxicité entre FB1 et HFB1. Dans l’étude avec la supplémentation en agentsdésactivateurs des régimes DON et FB, seuls ou en combinaison, l’agent ciblant les FB correspondaitaux surnageants de culture lyophilisés, repris dans l’eau puis pulvérisés directement sur les lotsd’aliments.Le second agent désactivateur utilisé dans cette étude correspondait à une bactérie, du genreEubacterium, isolée initialement du rumen de bovins (Fuchs et al., 2002; Schatzmayr et al., 2006).Cette bactérie synthétise une époxydase capable de biotransformer le groupe époxyde destrichothécènes en un diène. Ainsi, le DON peut être enzymatiquement réduit par l’époxydased’Eubacterium en un métabolite non toxique, le de-époxy-déoxynivalénol (DOM-1). Incorporée dansun produit commercial développé par BIOMIN, cette souche bactérienne a été la première à êtreappliquer comme un additif alimentaire contre les effets négatifs des trichothécènes. Tout commel’enzyme expérimentale décrite ci-dessus, les cultures bactériennes obtenues après fermentation ontété stabilisées par lyophilisation, et inclus telles quelles dans nos régimes alimentaires.Le développement de ces produits biotechnologiques nécessite néanmoins un long travail enamont dans la recherche et l’identification de souches capables de dégrader des mycotoxines, et doitensuite satisfaire plusieurs critères en aval. Les principales exigences pour une applicationcommerciale sont :• une réaction de dégradation rapide (généralement évaluée au préalable dans des essais invitro et/ou ex vivo), spécifique de la mycotoxine et irréversible• une dégradation des mycotoxines en métabolites moins ou non toxiques• une absence de pathogénicité des micro-organismes utilisés166

DISCUSSION GENERALE• une réaction de dégradation optimale dans les conditions du tractus gastro-intestinal,quelque soit le potentiel d’oxydo-réduction, la teneur en oxygène et la valeur du pH• un métabolisme des micro-organismes qui s’active rapidement dans le système digestif poursynthétiser l’(s) enzyme(s) détoxifiante(s)• ne pas altérer les qualités organoleptiques et la valeur nutritionnelle des alimentsEn plus de ces critères d’efficacité, le micro-organisme sélectionné doit être stabilisé avantapplication commerciale par différents processus tels que le séchage (e.g. à lit fluidisé), lalyophilisation, et également être incorporé dans des substances protectives (essentiellement despolymères organiques). Cela permet de garantir une stabilité suffisante du micro-organisme contreles conditions environnementales lors du stockage des aliments, et d’assurer son passage à traversl’acidité du tractus gastrique pour rejoindre indemne son site d’action, l’intestin. Le micro-organismene doit pas non plus être inhibé par la microflore intestinale, et dans le cas d’emploi de bactéries, lasouche doit être résistante à certains antibiotiques qui pourraient être utilisés dans les élevages àdes fins thérapeutiques. C’est pourquoi l’utilisation d’enzymes peut présenter un certain nombred’avantages en termes d’exigences, par rapport à un micro-organisme entier.Toutefois, ces approches de biotransformation affichent des atouts majeurs par rapport auxautres méthodes d’élimination des mycotoxines présentées dans l’étude bibliographique. Ellespeuvent être facilement implémentées dans les procédés de fabrication d’aliments à un coûtrelativement bon marché, et sont de plus en plus reconnues et acceptées par les instanceseuropéennes (e.g. rapport de l’European Food Safety Authority (EFSA) sur la sécurité du produitBIOMIN Eubacterium BBSH797 en tant qu’additif alimentaire – N°EFSA-Q-2003-052 ; création récented’un nouveau groupe d’additif sous la dénomination officielle de produits désactivateurs demycotoxines). Les méthodes chimiques sont interdites dans l’UE, principalement du fait de la toxicitépotentielle des métabolites produits et du manque de données sur l’innocuité des denrées traitées.Les adsorbants inorganiques ont soulevé certaines interrogations quant à leur potentiel d’adsorptionnon-spécifique, notamment vis-à-vis des nutriments, et également les possibles contaminations pardes substances toxiques lors des extractions et purifications de ces adsorbants (e.g. cas de dioxine).La place occupée dans la ration alimentaire oblige également de réduire l’apport de certainsnutriments aux dépens de ces additifs. Les adsorbants organiques, et plus particulièrement les paroisde levures, présentent une solution alternative intéressante, mais comme les adsorbantsinorganiques, l’efficacité d’adsorption a principalement été montrée vis-à-vis des aflatoxines, et laquestion de l’impact sur l’environnement lors de l’excrétion fécale du complexemycotoxines/adsorbants se pose.167

Figure 24 :Effets toxicologiques provoqués chez le porc par l’ingestion des mycotoxines majeures (source WorldNutrition Forum, Salzburg 2010)AFB1, Aflatoxine B1OTA, Ochratoxine ADON, DéoxynivalénolFB1, Fumonisine B1ZEA, Zéaralénone

DISCUSSION GENERALE2. Les systèmes de défense de l’organismeLes mycotoxines présentent un large spectre d’effets toxiques (Figure 24, exemple chez le porc).Les syndromes toxicologiques cliniques associés à l’ingestion de quantité moyenne à élevée sont biencaractérisés, ils s’étendent de mortalité aigüe, aux retards de croissance et aux troubles dereproduction. En revanche, l’ingestion chronique de faibles doses est beaucoup moins documentée,car n’induisant peu ou pas de manifestations cliniques, l’évaluation des effets nécessite l’examen demarqueurs plus spécifiques, notamment au niveau cellulaire. Pour la plupart des mycotoxines, lemécanisme de toxicité repose sur une inhibition de la synthèse protéique, que ce soit au niveau de latranscription ou de la traduction des séquences nucléiques. Ainsi, les cibles cellulaires de ces toxinespeuvent être nombreuses et sont principalement des tissus et des cellules en prolifération etdifférenciation, présentant un taux de renouvellement protéique élevé, tels que l’on peut retrouverdans l’intestin grêle, le foie et le système immunitaire. En conséquence, certains processuscellulaires, tels que la réponse au stress oxydatif, le renouvellement et la perméabilité del’épithélium digestif ou encore la réponse immunitaire face à des pathogènes peuvent êtreconsidérablement affectés lors d’une exposition répétée à des aliments faiblement contaminés.L’IMMUNITÉ SYSTÉMIQUE ACQUISE ET SPÉCIFIQUEAu niveau du système immunitaire, les cibles cellulaires des mycotoxines sont les monocytes, lesmacrophages et les lymphocytes. En conséquence, l’immunosuppression induite par les mycotoxinespeut avoir diverses conséquences en termes de santé animale :‣ sensibilité aux maladies infectieuses : apparition spontanée d’infections à Salmonella ouCampilobacter chez des porcelets traités avec l’ochratoxine A (Stoev et al., 2000) ;augmentation de la sévérité des lésions pulmonaires lors d’une infection à Pasteurellamultocida chez des porcelets traités avec les fumonisines (Halloy et al., 2005)‣ réactivation des infections chroniques : réactivation de l’infection toxoplasmique dans lecerveau (augmentation du nombre de kystes) de rongeurs traités avec la toxine T-2 oul’aflatoxine B1 (Venturini et al., 1996)‣ réduction de l’efficacité thérapeutique : baisse de l’efficacité d’un anti-coccidien chez despoulets traités avec la toxine T-2 et challengés avec Eimeria tenella (Varga and Vanyi, 1992)‣ réduction de l’efficacité vaccinale : baisse du titre en anticorps spécifiques – de la maladie deNewcastle chez des poussins traités avec les aflatoxines (Gabal and Azzam, 1998) – de168

Figure 25 :Hypothèses proposées dans l’altération de la mise en place de la réponse spécifique, d’après les résultats obtenus dans le chapitre 1 sur l’effet duDON et des FB, seuls ou en combinaisoncontrol DON FB DON+FB1,5aa aa1,0ba,bb0,5b0,0IL-8 MIP-1βArbitrary units2000150010005000a aa,b a,baa,bb,cbbbcbD21 D28 D35Cytokines impliquées dans lerecrutement et la migration descellules présentatrices d’antigèneANTIGENEexemple ovalbumineEn conséquence, la production d’anticorps spécifiquesanti‐ovalbumine a été grandement diminuéeIncapacité des lymphocytes àproliférer après une stimulationavec l’antigène ovalbumine1,51,00,50,0aa,bb,cIL-6ca a,bbIL-1βbCytokines impliquées dans l’activation et ladifférenciation des lymphocytes B et TProliferative index6420aaaa a bba bbbD21 D28 D35b

DISCUSSION GENERALEMycoplasma agalactiae chez des porcelets traités avec les fumonisines (Taranu et al., 2005)– de l’ovalbumine chez des porcelets traités avec la toxine T-2 (Meissonnier et al., 2008a)Cependant, certains aspects de l’immunosuppression provoquée par les mycotoxines restent trèslargement inexplorés. Par exemple, les mycotoxines sont très souvent produites simultanément,mais seulement très peu d’études reportent les effets de l’interaction de ces toxines sur le systèmeimmunitaire.Une partie du travail de thèse a donc été de déterminer l’efficacité de la réponse vaccinale aprèsexposition à une combinaison de deux toxines majeures, le DON et la FB. Nos résultats présentésdans le chapitre 1 suggèrent que les mycotoxines et spécialement la combinaison ontconsidérablement affecté le développement de la réponse vaccinale (Figure 25). L’absence d’effetssur la réponse totale non-spécifique, c'est-à-dire sur la population totale des lymphocytes du sang etsur les anticorps totaux issus de la population totale des plasmocytes, pourrait être attribuée auxfaibles doses utilisées, et avant tout au fait qu’à ces doses, les mycotoxines agissent initialement surles cellules en division, comme par exemple celles qui sont impliquées lors de l’activation du systèmeimmunitaire suite à un challenge antigénique.L’initiation de la réponse acquise fait intervenir les cellules dendritiques, mais aussi lesmacrophages ou les lymphocytes B, qui exercent des fonctions de sentinelles permanentes dans lestissus de l’organisme. Elles vont capturer l’antigène dans les tissus grâce aux récepteurs TLR (Toll-LikeReceptor), puis devenir matures, on parlera alors de cellules présentatrices d’antigène (CPA). Cettematuration s’accompagne de modifications morphologiques et phénotypiques (expression dedifférents récepteurs et de molécules de co-stimulation). Les CPA vont ensuite migrer vers lesorganes lymphoïdes secondaires, tels que la rate, où elles vont rencontrer les lymphocytes T. Suite àla présentation de l’antigène aux lymphocytes T via les molécules du complexe majeurd’histocompatibilité (CMH) de classe I et II, et à l’activation par des molécules de co-stimulation etdes médiateurs tels que les cytokines, les lymphocytes T spécifiques de l’antigène vont proliférer.Dans notre étude, les différents résultats que nous avons obtenus suggèrent que la présentationde l’antigène ovalbumine et l’interaction avec les cellules T a été considérablement affecté parl’ingestion chronique des mycotoxines, en particulier quand le DON et la FB étaient associés, et à enconséquence entraîner une réponse vaccinale défectueuse. Nos données sur l’expression descytokines spléniques montrent une diminution des transcrits codant pour IL-8 et MIP-1β. Ceschémokines sont synthétisées lors de l’activation des macrophages avec l’antigène, et assurent lerecrutement d’autres cellules immunitaires sur le site de reconnaissance de l’antigène grâce à despropriétés chimiotactiques. Il est ainsi possible de supposer que la mobilisation des CPA n’était pas169

Figure 26 :Mécanisme d’inhibition de la céramide synthase par les fumonisines et implication dans le métabolismedes sphingolipidesMEMBRANE CELLULAIREEUCARYOTIQUEVoie enzymatique (céramide synthase) bloquée par les fumonisines

DISCUSSION GENERALEoptimale. Les autres données sur l’expression des cytokines dans la rate ont également montré unediminution des transcrits codant pour IL-1β, IL-6 et IL-12p40. Ces trois interleukines sont égalementproduites par des cellules présentatrices d’antigène et permettent dans les organes lymphoïdesd’induire l’immunité à médiation cellulaire grâce à leurs propriétés d’activation et de stimulation dela différenciation des lymphocytes T naïfs (Trinchieri, 2003). De même ici, il est possible d’admettreque l’absence de prolifération des lymphocytes, que nous avons mis en culture et stimulés in vitroavec l’ovabumine, était liée à une reconnaissance de l’antigène et une activation des lymphocytes Tdéfectueuses. Ces observations sont supportées par les données individuelles du DON et des FB dansla littérature. En effet, le DON a été montré in vitro comme capable d’inhiber l’expression desmolécules du CMH de classe II et des molécules de co-stimulation, sur des macrophages (Wache etal., 2009) et sur des cellules dendritiques (Hymery et al., 2006) dérivés de monocytes humains. De lamême manière, il a été démontré que le DON interférait avec la maturation phénotypique et lacapture des antigènes de cellules dendritiques issues de monocytes porcins (Bimczok et al., 2007), etaffectait le potentiel de phagocytose de macrophages de souris et de dinde (Ayral et al., 1992; Kiddet al., 1995). En ce qui concerne les FB, l’exposition à la FB1 altérait la maturation de CPA intestinalesin vivo chez le porc, ainsi que la capacité des cellules T à être stimulées (Devriendt et al., 2009) ; etégalement diminuait l’activité de phagocytose de macrophages de poulet (Qureshi and Hagler, 1992)et la viabilité de macrophages murins (Dresden-Osborne and Noblet, 2002) in vitro.L’immunosuppression liée à l’exposition de la FB1 peut être expliquée par son mode d’action sur lemétabolisme des sphingolipides, via l’inhibition de la céramide synthase. Comme plusieurs foismentionné dans ce manuscrit, cette inhibition provoque l’accumulation des bases sphingoides,sphinganine (Sa) et sphingosine (So), mais aussi de leurs métabolites Sa-1-phosphate et So-1-phosphate (Riley et al., 2001; Merrill et al., 1997, 2001), et ces composés ont été déjà décrits commeimmunomodulateurs. Par exemple, une augmentation de la sphinganine pourrait inhiber la protéinekinase C, impliquée dans l’activation des cellules T (Martinova and Merrill, 1995). Plusieurs études sesont focalisées sur le rôle de la sphingosine-1-phosphate (S1P) dans la réponse immunitaire, et ontmontré que ce médiateur lipidique pouvait interférer dans la différenciation des monocytes en CPAmatures et compétentes (Martino et al., 2007), inhiber la sécrétion de cytokines dans des cellulesdendritiques matures, et ainsi bloquer la différenciation des lymphocytes Th1 (Idzko et al., 2002), ouencore inhiber la prolifération des cellules T (Jin et al., 2003). En plus de ces considérations, lessphingolipides sont des constituants majeurs des membranes cellulaires, les fumonisines par leuraction sur la céramide synthase bloquent la voie de synthèse des sphingolipides complexes tels queles glycosphingolipides ou les sphingomyélines (Soriano et al., 2005) (Figure 26), et par conséquent170

DISCUSSION GENERALEpourraient moduler l’expression des marqueurs et/ou des récepteurs à la surface des membranesdes cellules immunitaires.Nos résultats expérimentaux sur les immunoglobulines spécifiques nous laissent penser que laforte diminution observée dans la concentration plasmatique en IgG anti-ovalbumine serait ainsi uneconséquence de ces effets adverses des mycotoxines sur l’initiation et la mise en place de la réponseimmunitaire spécifique. En effet, le développement de la réponse humorale se produit lorsque leslymphocytes T activés par les CPA interagissent ensuite avec les lymphocytes B dans les organeslymphoïdes. L’activation des cellules B par les T permet ainsi la formation de plasmocytes qui sontcapable de produire des anticorps dirigés contre l’antigène. Par ailleurs, l’IL-6 stimule les cellules Bactivées à se différencier en plasmocytes, et dans notre étude la diminution de l’expression de l’IL-6dans la rate pourrait être corrélée à cette déficience en anticorps spécifiques, en particulier quandles deux mycotoxines étaient présentes simultanément.Nous avons également analysé la concentration plasmatique en IgA anti-ovalbumine. Cette classed’anticorps est principalement impliquée dans le système de défense des muqueuses, tel que dans letractus intestinal, respiratoire ou uro-génital. L’élévation des IgA totales ou spécifiques ont étélargement reportée dans la littérature après une exposition au DON (Drochner et al., 2004; Pestkaand Smolinski, 2005; Goyarts et al., 2005; Pinton et al., 2008). Cette immunoactivation pourraitlaisser croire que le DON possède des propriétés bénéfiques en termes de stimulation de l’immunitéet de protection contre certains pathogènes. Or, lors d’expositions répétées en DON chez un modèlemurin, cette induction des IgA semble provoquer une néphropathie avec des atteintes similaires àcelles observées dans le cas de néphropathies à IgA, une des glomérulonéphrites les plus fréquenteschez l’homme (Pestka, 2003). De plus, une sensibilité accrue envers des infections pulmonaires etentériques a pu être associée à cette stimulation en IgA (Li et al., 2005, 2007).Lors de l’association du DON avec la FB, aucun changement n’a été en revanche observé dans lateneur plasmatique des IgA. Nous avons donc supposé que la FB interférait dans le mécanisme induitpar le DON et menant à cette accumulation d’IgA. Il a été montré que ce mécanisme était associé àune augmentation de l’expression de l’IL-6 (Yan et al., 1997; Pestka and Zhou, 2000). Commementionné précédemment, l’IL-6 est une cytokine qui favorise la réponse humorale et ainsi laproduction d’anticorps. Etant donné que la production d’IgA est principalement observée dans lescompartiments muqueux, nous avons admis que l’augmentation des transcrits de l’IL-6 dans l’intestingrêle des animaux exposés au régime mono-contaminé en DON (cf chapitre 1) pouvait êtreresponsable des taux élevés d’IgA spécifiques. De plus, aucune modification de l’expressionintestinale de l’IL-6 a été notée dans les régimes FB et DON+FB, corroborant ainsi les résultats d’IgApour ces régimes. Récemment, il a été montré qu’une supplémentation alimentaire en171

DISCUSSION GENERALEsphingomyéline augmentait la quantité d’IgA dans le gros intestin, suggérant que la sphingomyélinecontrôlait la sécretion d’IgA (Furuya et al., 2008). Comme affirmé précédemment, les FB peuventinhiber la synthèse des sphingolipides complexes dérivés des céramides, tels que la sphingomyéline,grâce à leur mécanisme d’inhibition au niveau de la céramide synthase. Cette inhibition a aussi étéreportée sur l’intestin (Enongene et al., 2000; Loiseau et al., 2007) et donc nous pouvons émettrel’hypothèse que la déplétion en sphingomyéline pourrait être responsable de l’absence d’effets surles IgA quand le DON est associé aux FB.172

DISCUSSION GENERALEMÉCANISME CELLULAIRE DE MODULATION DE L’IMMUNITÉLes nombreux effets physiologiques observés après l’exposition aux mycotoxines ne sont pas dûsà une action directe de ces métabolites sur les cellules ou sur les tissus. En effet, ces altérations dansles fonctions cellulaires sont les conséquences du mode d’action des mycotoxines, qui bien quedifférent selon les toxines, ont en commun la capacité de rapidement perturber les voies designalisation cellulaire, et ainsi d’affecter en aval l’expression de certains gènes. La voie des MAPkinases (MAPK, Mitogen-Activated Protein Kinase) constitue l’une des voies de signalisation les plusimportantes dans la transduction des signaux au niveau cellulaire, et couvre un large éventail defonctions (Cobb, 1999; Dong et al., 2002). Ainsi, plusieurs études ont reporté que les mycotoxinesactivaient rapidement la voie des MAPKs, expliquant ainsi certains effets dans divers processusphysiologiques, tels que la perméabilité intestinale, l’immunité, la carcinogénicité, ou encorel’apoptose (Pinton et al., 2010; Jia et al., 2004; Wattenberg et al., 1996; Shifrin and Anderson, 1999;Chang et al., 2009; Wu et al., 2005). En ce qui concerne les effets additifs observés pour le DON et laFB sur la composante immunitaire, l’activation des MAPKs reportée individuellement pour les deuxtoxines pourrait être la cause de ces effets lorsque les mycotoxines sont associées. Que ce soit pourle DON ou la FB, suite à l’activation des MAPKs une augmentation de la production desprostaglandines et des leucotriènes, liée à la plus forte activité du métabolisme de leur précurseurl’acide arachidonique, a été démontrée (Moon and Pestka, 2002; Pinelli et al., 1999). Ces médiateurslipidiques sont connus pour être impliqués dans l’inflammation, l’immunomodulation, la régulationdes facteurs de transcriptions ou l’activation des protéines kinases (Smith et al., 1996). Dans le casdes FB, il a été montré que la FB1 activait les MAPKs via un mécanisme indépendant del’accumulation des bases sphingoïdes Sa et So (Wattenberg et al., 1996). Dans ce sens, il a étéreporté que la phosphorylation des MAPKs se produisait 5 min après traitement avec la FB1, alorsque 2 heures étaient nécessaires pour observer une augmentation de la concentration en Sa et So(Pinelli et al., 1999).Toutefois, les effets physiologiques observés après ingestion de mycotoxines ne peuvent pas seréduire à cette seule explication. L’activation des MAPKs est un phénomène complexe, rapide ettransitoire, et qui implique de nombreuses étapes et médiateurs dans les multiples voies designalisation (ERK, p38 et JNK). Par exemple, l’élévation des IgA par le DON a été attribuée à lasurexpression de la cyclo-oxygénase-2 (suivie de la modulation de l’IL-6) après activation des MAPKs(Moon and Pestka, 2003). Or, la FB1 qui induit également l’expression de la cyclo-oxygénase-2 (Meliet al., 2000) n’a pas provoquée d’augmentation des IgA dans notre étude que ce soit seule ou encombinaison avec le DON. Par conséquent, l’étude des effets des mycotoxines dans les voies de173

DISCUSSION GENERALEsignalisation cellulaire nécessite d’être prudent dans l’interprétation, et de savoir distinguer entre leseffets spécifiques et non spécifiques en réponse aux stimulis.174

Figure 27 :Vue globale de la muqueuse intestinale, et des compartiments et des acteurs clés du système dedéfense de l’intestin

DISCUSSION GENERALEL’IMMUNITÉ INTESTINALEEn plus de sa fonction principale d’absorption des nutriments, le système digestif constitue lesystème immunitaire le plus important de l’organisme. Pourtant, sa fonction de défense immunitaireest souvent sous-estimée. Trois lignes de défense peuvent être distinguées : La flore intestinale : constituée d’une considérable diversité bactérienne, la flore intestinalereprésente la première barrière de protection contre les intrus indésirables La muqueuse intestinale : constitue la principale interface entre l’environnement extérieur etl’intérieur de l’organisme ; contrôle de façon très sélective le passage des substances del’intestin à la circulation sanguine et lymphatique Le système immunitaire intestinal : abrite le GALT (Gut-Associated Lymphoid Tissue) quicontient à lui seul plus de 70% des cellules immunitaires de l’organisme ; développe sapropre réponse immunitaire de type innée ou adaptative (Figure 27)Comme déjà mentionné dans ce mémoire, bien que l’exposition naturelle des mycotoxines seproduise via l’ingestion d’aliments contaminés, l’effet de ces contaminants sur le tractus digestif estpeu documenté. Toutefois, cet aspect depuis quelques années a semblé susciter l’attention de lacommunauté scientifique, au vu du nombre de recherches croissante dans le domaine. Il est doncimportant de savoir dans quelle mesure la consommation de mycotoxines peut affecter l’immunitéintestinale, et par delà les conséquences en termes de santé humaine et animale. Par exemple, il aété montré dans notre laboratoire que l’ingestion de faibles doses de fumonisines augmentait lacolonisation intestinale chez des porcelets infectés oralement par une souche pathogèneopportuniste d’E. coli (Oswald et al., 2003). Chez ces animaux, une translocation bactérienne accruede la souche pathogène vers les organes extra-intestinaux a également été notée. Cette étudeindique donc que le système immunitaire des porcelets traités avec les mycotoxines était incapablede contrôler l’infection, et était ainsi plus sensible aux pathogènes intestinaux.Nos échantillons prélevés lors des phases animales sur l’intestin (jéjunum et iléon) ontprincipalement fait l’objet d’analyses histologiques et transcriptomiques. Nous avons essayé dedéterminer le profil d’expression de gènes codant pour des cytokines tout au long de l’intestin grêle.Comme précédemment indiqué, les cytokines sont des médiateurs de l’immunité et sont impliquéesdans différents processus, tels que l’inflammation, la réponse immunitaire innée, la réponseadaptative cellulaire et humorale. Au niveau de l’intestin, à l’instar des cellules immunitaires lescellules épithéliales intestinales sont également une importante source de cytokines (Stadnyk, 2002).Il était donc intéressant d’examiner la réponse locale, en particulier au niveau du jéjunum, et celle175

Figure 28 :Réseau qui constitue le système de perméabilitéparacellulaire de l’intestin, impliquant les protéinesde jonction

DISCUSSION GENERALEplus spécifique impliquant le GALT (plaques de Peyer dans l’iléon et ganglions lymphatiquesmésentériques), en réponse à l’ingestion d’aliments ou de solutions contenant des mycotoxines.Dans le chapitre 1, le résultat de l’expression des gènes a révélé une augmentation (significativeou tendance) des transcrits de plusieurs cytokines (IL-1β, TNF-α, IFN-γ, IL-6, IL-2, IL-10) chez lesanimaux traités avec les faibles doses de DON et FB, que ce soit dans le jéjunum ou l’iléon. Cesobservations étaient plutôt intéressantes étant donné que pour la même étude nous avions observéune diminution de certains de ces transcrits dans la rate. Cette régulation positive des cytokinessuggère chez ces porcelets une inflammation chronique de leurs intestins tout au long des cinqsemaines d’exposition. L’augmentation des cytokines pro-inflammatoires a déjà été reportée dans lecas de maladies inflammatoires chroniques intestinales (IBD, Inflammatory Bowel Disease) chezl’humain, telles que la maladie de Crohn’s ou la colite ulcéreuse (Papadakis and Targan, 2000). Deplus, l’observation d’intestins de porc exposés à de faibles concentrations de toxines de Fusariumévoquait une inflammation de la muqueuse de la membrane intestinale, et ce dès 14 joursd’exposition (Obremski et al., 2008). Ainsi, comme il a été récemment souligné, l’ingestion d’alimentscontaminés en mycotoxines pourrait être considérée comme un risque potentiel dans laprédisposition aux inflammations chroniques de l’intestin (Maresca and Fantini, 2010).A noter que l’induction de l’expression de l’IL-6 dans l’intestin grêle des animaux nourris avec leDON seul pourrait être corrélée avec l’élévation des IgA spécifiques que nous avons observés sur lapartie systémique. Cette observation a été mentionnée dans la section précédente et corroboreégalement l’absence d’effets en présence des FB.Outre le rôle des cytokines dans le maintien de l’inflammation chronique, ces médiateursjoueraient également un rôle dans le maintien de la perméabilité des barrières épithéliales ethémato-encéphaliques (Capaldo and Nusrat, 2009). En effet, il a été montré que l’inflammation desmuqueuses lors des IBD était précédée d’une augmentation de la perméabilité paracellulaireintestinale. Ainsi, les manifestations répétées de diarrhées dans le cas des IBD pourraient êtreattribuées à cette perte de fonction de barrière. La fonction de barrière intestinale est maintenue parles jonctions serrées et adhérentes intercellulaires, formant un réseau complexe qui minimisel’espace entre les cellules adjacentes (Figure 28). Notre équipe a récemment reporté que ledéoxynivalénol était capable d’altérer cette fonction de barrière, en agissant sur la protéine dejonction serrée, la claudine-4. Ces résultats ont été mis en évidence dans des modèles in vitro, sur lescellules IPEC-1 (Intestinal Porcine Epithelial Cell line) et Caco-2 (human colon carcinoma cell line), etex vivo et in vivo sur des segments jéjunaux de porc (Pinton et al., 2009, 2010). Dans ces études,l’altération de la perméabilité et l’identification de la claudine-4 ont été confirmées par mesure de larésistance électrique transépithéliale (TEER) et de la perméabilité paracellulaire (au passage de176

DISCUSSION GENERALEdextran et d’E. coli), et par immunofluorescence et immunoblot avec des anticorps spécifiques de laclaudine-4. De plus, l’équipe a dernièrement démontré que cette modification de la perméabilité parle DON était dépendante de l’activation des MAPKs, plus précisément la voie de signalisation ERK ½(Pinton et al., 2010).L’expression et la localisation de ces protéines de jonctions, comprenant la famille des claudines,des cadherines et l’occludine, pourraient être régulées par certaines cytokines pro-inflammatoires,telles que l’IL-1β, le TNF-α et l’IFN-γ (Wisner et al., 2008; Han et al., 2003; Ye et al., 2006; Al-Sadi andMa, 2007). Lors de l’analyse des transcrits dans les tissus intestinaux prélevés sur nos animauxtraités, une augmentation de l’expression de ces trois cytokines a été observée, principalement auniveau de la partie iléale. Par ailleurs, l’étude immunohistochimique sur les coupes d’intestin a révéléune détection plus faible de l’E-cadherine, en particulier au niveau de l’iléon. Ces différentesobservations nous ont donc amené à réaliser des immunoblots sur l’iléon des animaux traités.Comme reporté dans le chapitre 1, l’expression des protéines E-cadherine mais aussi occludine étaitsignificativement réduite, notamment chez les animaux ayant consommé le régime co-contaminé enDON et FB. En revanche, nous avons également cherché à déterminer l’expression de la claudine-4 etmalheureusement, nous n’avons pu confirmer les résultats antérieurs de l’équipe. Nous pouvonssupposer que la nature des échantillons, jéjunum versus iléon, pourrait intervenir dans la dichotomiede ces résultats. De même, on ne peut exclure une internalisation de la structure de la claudine-4,attribuée à la contractilité du complexe actine/myosine induit par les cytokines (Capaldo and Nusrat,2009), délocalisant ainsi la protéine de jonction de la surface apicale des entérocytes sans pourautant modifier son expression.Cette probable perturbation de la perméabilité cellulaire pourrait avoir des conséquencesimportantes, en particulier après exposition à un aliment multi-contaminé, en autorisant le passaged’antigènes ou de substances indésirables de la lumière intestinale vers les compartimentssystémiques ; ce phénomène ayant déjà été observé avec E. coli chez des porcelets traités avec lesfumonisines (Oswald et al., 2003).Un aspect intéressant dans ces travaux de thèse est la réponse intestinale chez les porceletsexposés à une faible ou à une forte dose de fumonisine B1. Dans l’étude sur les co-contaminations, lerégime individuel en fumonisines contenait 4,1 mg FB1/kg d’aliment. Dans l’étude comparative de latoxicité de FB1 et de HFB1, la dose administrée était de 2 mg FB1/kg poids vif/jour, soit à l’âge desporcelets une dose équivalente comprise entre 37 et 44 mg FB1/kg d’aliment. Cet écart d’un facteur10 entre les deux doses utilisées a de manière surprenante mais non inattendue entraîné unedifférence dans le profil d’expression des cytokines dans l’intestin. Néanmoins, il est nécessaire de177

DISCUSSION GENERALEsouligner que les conditions d’intoxication dans les deux phases d’expérimentation animales étaientrelativement différentes. En effet, dans l’étude de la co-contamination les porcelets étaient de sexemasculin, contrairement à l’étude FB1 versus HFB1 où ils étaient de sexe féminin. Le facteur sexe adéjà été reporté dans plusieurs espèces, avec une sensibilité aux mycotoxines plus forte chez lesmâles que chez les femelles (Cote et al., 1985; Marin et al., 2006; Orsi et al., 2007; Tamimi et al.,1997; Domijan et al., 2007). Cette différence de sexe entre les deux études n’était pourtant pas unchoix délibéré, mais simplement une contrainte imposée par notre éleveur local lors de la mise enplace de l’étude FB1/HFB1. Au niveau des autres différences d’expérimentation, la durée et le moded’exposition était dans l’une de cinq semaines et par l’alimentation, et dans l’autre de deux semaineset par gavage.A l’inverse de l’augmentation de l’expression des cytokines décrite dans la section précédente,une diminution de l’expression des mêmes cytokines a été notée chez les animaux traités avec laforte dose de FB1. Cette immunosuppression a été observée dans tous les segments du systèmedigestif examinés. A noter que nous n’avons pas déterminé dans l’étude de la co-contamination laquantité de transcrits dans les ganglions mésentériques. Cette différence d’immunomodulationintestinale est intéressante et soulève ainsi des interrogations quant à la toxicité induite par desconcentrations différentes en mycotoxines. Toutefois, dans les deux cas il semblerait que quelquesoit les doses administrées, l’effet physiologique provoqué affectent de manière considérable lesanimaux. Même si la probabilité de rencontrer des aliments naturellement contaminés avec 40 mgFB1/kg est faible, ces résultats montrent qu’à cette dose les animaux sont probablement plussusceptibles aux infections du fait d’un système immunitaire défectueux. L’effet de cette dose surl’expression des protéines de jonctions n’a pas été évalué mais ce point justifierait quelquesrecherches complémentaires si l’on considère que la diminution de l’expression de ces protéines estliée à l’augmentation de l’expression des cytokines. Néanmoins, il ne serait pas non plus surprenantque la perméabilité cellulaire soit affectée à cette concentration, en partant du principe que la doseest suffisamment élevée pour abolir plusieurs fonctions physiologiques.178

DISCUSSION GENERALE3. Stratégies de lutte pour réduire les effets toxiques du déoxynivalénol et desfumonisines – approches de biotransformationL’EUBACTERIUM SPECIFIQUE DES TRICHOTHECENESDepuis sa découverte en 1997, l’Eubacterium spécifique des trichothécènes (Binder et al., 2000,2001; Fuchs et al., 2002) et nommée BBSH797 en l’honneur de l’équipe de recherche qui l’a isolée(BBSH, Binder Binder Schatzmayr Heidler), a été le premier additif détoxifiant des mycotoxinesformulé à partir d’un micro-organisme vivant.De nombreuses alternatives existent pour réduire l’exposition aux mycotoxines mais peu sontcommercialement envisageables (voir section ci-après la carboxylestérase). En ce qui concerne lestrichothécènes (TCT) et en particulier le déoxynivalénol (DON), les approches d’adsorption sontrelativement inefficaces, et ainsi comme récemment suggéré les méthodes de détoxification par desmicro-organismes semblent être les plus adaptées aux TCT (Awad et al., 2010). L’EubacteriumBBSH797 a la capacité de dégrader le DON en un métabolite beaucoup moins toxique le de-époxy-DON (DOM-1) (Schatzmayr et al., 2006; Sundstøl Eriksen et al., 2004; Zhou et al., 2008). Le groupeépoxyde est considéré comme essentiel dans la toxicité des TCT. L’efficacité de la bactérie dansl’environnement intestinal a déjà été reportée dans un modèle ex vivo porcin traité en DON(Schatzmayr et al., 2006). Dans nos travaux, l’incorporation de cette bactérie a permis de réduiresignificativement certains effets provoqués par l’ingestion du DON. Aucun marqueur d’exposition nenous a permis de confirmer une dégradation efficace du DON dans les intestins des animaux.Néanmoins, nous pouvons affirmer que l’additif bactérien a réduit en partie la biodisponibilité duDON, au vu de la neutralisation partielle des effets provoqués par le DON (hauteur des villosités, IgAspécifiques, prolifération des lymphocytes). Il est à souligner également que la dose de DON utiliséesemblait induire moins de toxicité que la dose choisie pour la FB, ne nous permettant pas ainsi deconclure objectivement sur l’efficacité du procédé.Il serait intéressant de répéter cette expérience avec une multi-contamination en trichothécènesde type A et B à des doses suffisamment élevées pour entraîner des altérations cellulaires ettissulaires.179

DISCUSSION GENERALELA CARBOXYLESTERASE SPECIFIQUE DES FUMONISINESL’un des principaux intérêts de la collaboration de l’industriel BIOMIN portait sur l’évaluation del’efficacité et de la non-toxicité du processus de détoxification des fumonisines par lacarboxylestérase, isolée de Sphingopyxis. Les résultats que nous avons présentés tout au long dumémoire de thèse sont l’aboutissement d’un long travail en amont, réalisé par notre partenaire, dansle screening et l’isolation de micro-organismes capables de dégrader les FB ; dans l’élucidation etl’identification du gène responsable de cette dégradation ; et dans l’expression et la stabilisation dece gène.La contamination naturelle en fumonisines est un problème majeur dans certaines régions de laplanète, telles qu’en Amérique Centrale et du Sud, en Europe du Sud ou encore en Asie. Cettemycotoxine est importante en termes d’ubiquité, de toxicité et de difficulté à être éliminée desdenrées. De plus, l’intérêt grandissant ces dernières années pour le bioéthanol en tant que sourced’énergie renouvelable, a entraîné une inclusion croissante des co-produits de l’industrie de l’éthanoldans les rations alimentaires animales. Or, les mycotoxines pourraient être concentrées jusqu’à troisfois dans ces co-produits (DDGS, Dried Distillers’ Grains and Solubles) en comparaison du grain initial(Wu and Munkvold, 2008). Ainsi, considérant que les fumonisines contaminent naturellement etprincipalement les grains de maïs, et que le bioéthanol est presque entièrement produit à partir descultures de maïs, l’apparition accentuée et à des teneurs plus élevés de cette toxine dansl’alimentation animale pourrait très rapidement devenir un problème en termes de santé animale,d’agronomie et d’économie.De nombreuses stratégies ont été évaluées et/ou développées à des fins commerciales dans lebut de contrer les effets néfastes des fumonisines (cf études bibliographiques). Les méthodesphysiques via un traitement direct sur les grains contaminés ont prouvé leurs efficacités. Néanmoins,l’application commerciale de ces approches reste très limitée. Parmi ces limites, la durée (lors du triou tamisage des grains), le coût (séchage des grains après séparation par flottaison) et les pertesimportantes de matériel (tamisage) lors de ces traitements sont des inconvénients majeurs lorsd’une utilisation industrielle. La dilution des lots contaminés avec ceux non contaminés est désormaisune pratique interdite dans l’UE. Les fumonisines, mais aussi la plupart des mycotoxines sontégalement stables à des températures proches de celles utilisées dans la transformationconventionnelle des aliments. Les méthodes chimiques, bien que relativement efficaces, sonttoutefois proscrites des traitements des aliments ; excepté l’ammoniation des denrées qui esttolérée dans certains pays (principalement sur les arachides). Cette interdiction se justifieprincipalement par le potentiel de toxicité des métabolites obtenus après traitements chimiques180

DISCUSSION GENERALE(réversibilité de la réaction dans les conditions du tractus gastro-intestinal), par les potentiellesinteractions mycotoxines/matrices (mycotoxines « cachées », sous-estimation du contenu entoxines) et par l’impact sur le contenu énergétique et nutritionnel des aliments. Très peu decomposés adsorbants (argiles, parois de levure) se sont montrés efficaces dans l’adsorption desfumonisines, ou alors se sont révélés relativement onéreux pour une application commerciale (e.g.polymères organiques).Cette infructueuse prospection vis-à-vis des fumonisines a motivé le développement de méthodesalternatives et/ou complémentaires telles que la biotransformation de cette toxine. Dans nos études,nous avons pu tester et vérifier deux aspects importants dans la mise en place de ces approches.Premièrement, nous avons évalué le potentiel de toxicité de la fumonisine B1 totalement hydrolysée(HFB1) en comparaison de la molécule mère. L’une des recommandations de la FAO est que leprocessus n’entraîne pas la formation de métabolites toxiques. Par ailleurs, les données de lalittérature sur la toxicité de HFB1 avaient suscité quelques controverses, notamment dans les années90, d’où la nécessité d’avoir réalisé cette étude toxicologique comparative. Nos résultats ontclairement montré le très faible potentiel toxique de ce métabolite en comparaison de la FB1, que cesoit avant ou après absorption intestinale, c’est-à-dire l’impact sur le compartiment intestinal ousystémique. Ces observations convergent dans le sens des dernières recherches sur la toxicité deHFB1, ou plutôt de HFB1 pure, et réfutent également les résultats sur le matériel nixtamalizéaffirmant la HFB1 responsable de la toxicité. Mais comme expliqué dans le chapitre 2, cette toxicitéavait finalement été attribuée à la présence de résidus ou de FB1 « cachée » (fixée à la matrice etnon détectée en HPLC) dans les préparations de matériel nixtamalizé, plutôt qu’à la présence deHFB1 (Park et al., 2004; Burns et al., 2008). Dans ce sens, le groupe de Voss, ayant dans un premiertemps prétendu la dangerosité de HFB1 via la nixtamalization (Voss et al., 1996, 1998), est revenurécemment sur ses positions lors de l’utilisation de HFB1 pure (Voss et al., 2009). De plus, lors duWorld Mycotoxin Forum 2010 organisé aux Pays-Bas, j’ai eu l’occasion de discuter brièvement de cesrésultats avec Ronald Riley, scientifique du groupe de Kenneth Voss, me confirmant ainsi la validitéde nos travaux. D’autres collaborateurs ont également démontré sur les cellules Caco-2, que du faitde la plus faible polarité de HFB1 par rapport à FB1, son passage à travers la barrière intestinale étaitfacilité, augmentant ainsi sa biodisponibilité et suggérant un potentiel danger (Caloni et al., 2002; DeAngelis et al., 2005). Notre collaboration avec BIOMIN nous a permis dans un projet parallèle,d’évaluer les résidus en FB1 et HFB1 dans des échantillons de fèces prélevés lors de la phase animale,et également dans différents échantillons biologiques (plasma, foie, reins, poumons). Ces analysessont pour l’instant en cours mais nous permettront de déterminer le degré d’absorption intestinale181

DISCUSSION GENERALEde HFB1. Toutefois, après ingestion de la solution en HFB1, très peu d’effets ont été observés aprèsdistribution du métabolite dans le compartiment systémique hépatique.Le second aspect dans l’évaluation de cette approche de détoxification biologique, est bienévidemment l’efficacité de la carboxylestérase lors d’ingestion d’aliments contaminés enfumonisines, avec pour objectif d’hydrolyser les fumonisines dans le tractus digestif, et ainsi delimiter la biodisponibilité de ces mycotoxines. Nos travaux, présentés dans le chapitre 3, ont montrésur certains paramètres que la supplémentation en carboxylestérase pouvait significativementneutraliser les effets provoqués par les régimes contaminés. Des améliorations partielles (lésionshépatiques, prolifération des lymphocytes spécifiques, IgG spécifiques) et totales (hématologie etbiochimie, lésions pulmonaires, expression des cytokines de la rate, population et proliférationcellulaire de l’intestin) ont été observées dans les régimes contenant les FB et supplémentés avec lacarboxylestérase. Toutefois, le bénéfice s’est principalement ressenti par rapport aux effets induitspar le régime co-contaminé en DON et en FB. En effet, de faibles doses étaient utilisées dans cetteétude et les effets dus à la co-contamination étaient spécialement plus prononcés et significatifs queles effets individuels. Cependant, il est à préciser que le bénéfice dans le régime co-contaminé estpartagé par la présence des deux agents désactivateurs, spécifiques du DON et de la FB.Cette diminution de la toxicité après ajout de la carboxylestérase est attribuée à sa capacitéd’avoir hydrolysé totalement les FB dans l’intestin des animaux. En effet, au vu des résultats tout aulong de l’expérience du contenu en bases sphingoïdes, l’activité de l’enzyme ne semble pas avoir étéaltérée dans les conditions du tractus gastro-intestinal et aurait permis une hydrolyse totale des FB.Cette approche semble donc prometteuse dans la stratégie de lutte contre les fumonisines. Parailleurs, quelques micro-organismes ont déjà démontré des capacités à détoxifier les fumonisines,mais aucun à notre connaissance n’a pu être développé et proposé aux acteurs de la filière agricole.Cependant, le procédé mériterait quelques études complémentaires avant sa validation pour uneutilisation commerciale. Par exemple, une étude plus approfondie sur l’exposition et la détection desrésidus en FB et HFB dans différentes matrices biologiques après hydrolyse dans l’intestin pourraitapporter des informations complémentaires. De plus, la méthode de pulvérisation utilisée dans notreexpérience n’est pas un procédé commercialement viable, et donc l’efficacité de l’enzyme devra êtrede nouveau prouvée avec une méthode alternative d’incorporation lors de la fabrication desaliments.182

DISCUSSION GENERALELA PRESENCE SIMULTANEE DE LA CARBOXYLESTERASE ET DE L’EUBACTERIUMLa combinaison des deux approches de détoxification a amélioré de manière significative laréponse globale des porcelets suite à la consommation concomitante des deux mycotoxines.Toutefois, dans notre expérimentation une variabilité non négligeable a été observée sur certainsparamètres entre les deux groupes contrôles, supplémentés et non supplémentés. Lesexpérimentations in vivo et les gros animaux génèrent souvent une forte variabilité intra- et interindividuellemais il semblerait néanmoins que ces additifs ou les substances rentrant dans lacomposition de ces additifs aient certains effets non spécifiques. Le fait le plus marquant concerne laprésence de mégalocytoses dans le foie des animaux contrôles supplémentés. Pourtant, le produitlyophilisé contenant l’Eubacterium et incorporé dans nos aliments aurait la particularité de contenirdes extraits d’algues et de plantes aux propriétés immuno- et hépato-protecteurs. La compositionexacte de ce produit est confidentielle mais nos observations justifieraient quelques investigationscomplémentaires, l’utilisation de micro-organismes vivants pouvant provoquer des effets nonsouhaités.183

CONCLUSIONS184

Figure 29 :Effets d’exposition à des mycotoxines chez l’homme et l’animal (adapté de Williams T., XII thInternational IUPAC Symposium on Mycotoxins and Phycotoxins, 20‐25 May 2007)AIGUS(mort)Intensité del’exposition(dose)CHRONIQUES(malnutrition, retard decroissance, immunomodulation,modulation desdéfenses métaboliques, …)Duréed’expositionCUMULATIFS (cancer)Figure 30 :Graphique associé au Tableau 24 sur l’analyse mondiale de denrées agricoles, comprenant2727 produits de nature diverses. Représentation du nombre de mycotoxines détectées dansles échantillons, et par conséquent de la fréquence des multi‐contaminationsNotes : LD, Limite de Détection40%25%35%sous la LD1 mycotoxine> 1 mycotoxine

CONCLUSIONSLes effets d’exposition à des mycotoxines peuvent être résumés en trois catégories (Figure 29) :• De rares épisodes aigus, dus à des accidents de contaminations par de fortes doses detoxines pouvant entraîner la mort ;• Des effets cumulatifs, lors d’expositions à long terme à des faibles doses et en associationavec d’autres facteurs environnementaux ; des études épidémiologiques ont ainsi clairementidentifié le rôle de l’AFB1 en association avec le virus de l’hépatite B dans le développementd’hépato-carcinomes ;• Mais aussi une grande part d’effets chroniques, sans manifestations cliniques précises et quidemeurent aujourd’hui encore mal caractérisés quant à leurs conséquences en termes desanté humaine et animale.Ce dernier point est de nos jours la situation la plus fréquente, où les mycotoxines contaminentnaturellement diverses denrées agricoles à des seuils de contamination relativement bas. Les effetssont très mal connus du fait de l’absence de manifestations cliniques, et aussi du besoin d’utiliser desmarqueurs plus fins et spécifiques pour caractériser ces effets. Ainsi, après stimulation du systèmeimmunitaire via un protocole de vaccination, nous avons montré que la réponse immunitaire mise enplace, impliquant des populations et des médiateurs spécifiques, était significativement altérée avecde faibles concentrations en mycotoxines. Si nous n’avions pas activé le système immunitaire, nousn’aurions pas mis en évidence cette capacité des mycotoxines à altérer à faibles doses descomposantes spécifiques de la réponse immunitaire. Cette sensibilité du système immunitaire auxmycotoxines dans ces conditions de stimulation antigénique, avait déjà été reportée dans des étudesantérieures faites par notre laboratoire. De la même manière, il a été démontré qu’à faibles doses lesmycotoxines avaient la capacité de moduler l’activité d’autres cibles cellulaires associées à desfonctions vitales. Par exemple, des effets sur les fonctions intestinales en réponse à l’ingestion defaibles quantités de ces contaminants ont été reportés dans le travail de thèse et dans d’autresgroupes de recherches. Bien que les données sur la réponse intestinale soient encore limitées, unimpact réel des mycotoxines sur les fonctions d’absorption des nutriments, sur le profil de la florecommensale ou encore sur la prédisposition à des infections entériques a été démontré. Ce champd’investigations suscite un intérêt grandissant dans nos sociétés, où il est de plus en plus considéréqu’un bon état de santé est lié à un bon fonctionnement de l’écosystème intestinal. Parmi les autresfonctions vitales que nous pouvons citer, les fonctions métaboliques du foie peuvent être affectéespar les mycotoxines. Quelques études ont par exemple reporté des effets de faibles doses demycotoxines sur le stress oxydatif et sur la modulation d’enzymes hépatiques impliquées dans ladétoxification et le transport de xénobiotiques.185

CONCLUSIONSAinsi, l’analyse classique des paramètres zootechniques ou biochimiques n’est pas suffisante dansl’analyse du risque mycotoxique. L’altération de ces paramètres est davantage la conséquence à uneexposition aigüe, à des concentrations qui ne reflètent pas forcément les situations de terraincourantes, du moins dans nos pays en Europe.La présence simultanée de plusieurs mycotoxines dans un aliment est également un autre pointsensible dans l’évaluation du risque mycotoxique. Les multi-contaminations sont une réalité deterrain, et sont fréquentes comme le révèle très souvent les échantillonnages des denrées agricoles(Figure 30). Le risque d’être exposé à différentes mycotoxines en même temps est accentué par lefait qu’un produit fini est un mélange de plusieurs matières brutes. Ce potentiel de toxicité desproduits multi-contaminés n’est pas pris en compte dans les réglementations/recommandationsfixées par les instances nationales et européennes. Comme déjà mentionné, ces autorités netiennent compte que des études toxicologiques pour une toxine donnée, et fixent ainsi des seuilsindividuels. Le manque d’expériences in vitro et in vivo sur les effets des multi-contaminations peutexpliquer ce constat en comparaison des expériences réalisées individuellement. Comme montrédans nos travaux et dans notre étude bibliographique, la caractérisation des interactions est en plustrès souvent complexe. En effet, de nombreux facteurs (doses, espèces, sexe et âge, durée et moded’administration) peuvent influencer les réponses, et même lors de l’étude d’une même associationde mycotoxines les résultats peuvent différer. Toutefois, les effets totaux d’une co-contaminationsont généralement nettement supérieurs aux effets totaux individuels, même lorsque l’interactionconduit à des effets moins qu’additifs ou antagonistes. Parmi ces études, il a également étédémontré que la combinaison de mycotoxines à faibles concentrations pouvait affecter les animaux,alors que les concentrations individuelles étaient sous les concentrations reportées commeprovoquant des effets négatifs.Dans nos études, avec des doses reflétant des contaminations naturelles, les animaux ayantingéré le DON et la FB simultanément étaient significativement plus affectés que ceux ayantconsommé « seulement » les toxines individuelles. La réponse vaccinale chez ces animaux étaitapproximativement 40-50% moins efficace que celle des animaux contrôles tout au long del’expérience. Quelle est donc réellement l’impact des multi-contaminations dans les élevages, où laprésence de pathogènes et la vaccination sont des situations fréquentes ?Pour remédier aux effets toxicologiques induits par les mycotoxines, de nombreuses solutions ontété proposées et/ou développées tout au long de la chaîne de production alimentaire, depuis lechamp avec des stratégies de prévention avant toute invasion fongique jusqu’aux élevages avecl’ajout d’additifs pour empêcher la biodisponibilité des toxines. Cette dernière catégorie d’approches186

CONCLUSIONSest utilisée lorsque la prévention de la contamination mycotoxique avant la récolte ou lors de la postrécolteet du stockage n’est pas possible, mais également en tant que stratégie préventive. Par notrecollaboration avec l’industriel BIOMIN, nous avons donc évalué l’efficacité de produits classés danscette catégorie d’additifs. L’activité de notre partenaire réside dans l’identification demicroorganismes et/ou d’enzymes capables de détoxifier les mycotoxines. La diversité structurelledes mycotoxines et la problématique des multi-contaminations impliquent que les stratégies de lutteproposées soient efficaces et spécifiques contre chaque mycotoxine. Nos travaux sur l’évaluation desproduits développés par BIOMIN ont montré une approche de détoxification prometteuse et quipermettrait de limiter considérablement la biodisponibilité, et ainsi les effets néfastes du DON et dela FB chez les animaux.La nécessité de poursuivre dans cette mise en place de stratégies de prévention, dedécontamination, de détoxification mais aussi de modèles de prédiction, est encouragée par ledevenir de ces contaminations fongiques dans les prochaines décennies en conséquence duréchauffement climatique. L’impact du changement du climat (sécheresse, précipitations,catastrophes naturelles) suscite de nombreuses interrogations sur la future distribution et laprévalence des mycotoxines (Bryden, 2009, Paterson and Lima, 2009, Tirado et al., 2010).Dans un contexte général, nos travaux s’inscrivent principalement dans le secteur de laproduction animale, où les dépenses liées à la prophylaxie, aux soins médicaux des animaux etégalement aux pertes attribuées à la baisse de productivité dans les élevages sont des points majeursde la filière agricole. Mais la problématique des mycotoxines est aussi appréhendée de manièredifférente en fonction des zones géographiques. Dans les pays industrialisés, le problème estessentiellement d’ordre économique, affichant de lourdes pertes financières dans les différentssecteurs de la filière agricole. A l’inverse, dû à une précarité et à un manque de moyens deprévention dans les pays en voie de développement, l’exposition aux mycotoxines reste encore unproblème de santé publique dans ces pays. Toutefois, depuis l’intensification des échangesmondiaux, ces pays sont également confrontés au problème économique.En conclusion, la contamination fongique, la production des mycotoxines et les conséquencesliées à ces contaminants sont et resteront un problème d’actualité dans la filière agricole. Cescontaminations étant naturelles et largement tributaires du climat, elles sont difficilementcontrôlables.187

REFERENCESBIBLIOGRAPHIQUES188

REFERENCES BIBLIOGRAPHIQUESAABDEL-WAHHAB, M. A., NADA, S. A. & AMRA, H. A. 1999. Effect of aluminosilicates and bentonite onaflatoxin-induced developmental toxicity in rat. Journal of Applied Toxicology, 19, 199-204.ABDEL-WAHHAB, M. A., NADA, S. A. & KHALIL, F. A. 2002. Physiological and toxicological responses inrats fed aflatoxin-contaminated diet with or without sorbent materials. Animal Feed Scienceand Technology, 97, 209-219.ABOU-KARAM, M., ABBAS, H. K. & SHIER, W. T. 2004. N-fatty acylation of hydrolyzed fumonisin B-1,but not of intact fumonisin B-1, strongly enhances in vitro mammalian toxicity. Journal ofToxicology-Toxin Reviews, 23, 123-151.ACCENSI, F., PINTON, P., CALLU, P., ABELLA-BOURGES, N., GUELFI, J. F., GROSJEAN, F. & OSWALD, I. P.2006. Ingestion of low doses of deoxynivalenol does not affect hematological, biochemical,or immune responses of piglets. J Anim Sci, 84, 1935-42.ADAMS, C. A. 2001. Virtues of cleanliness: feed quality and hygiene in total nutrition. FeedingAnimals for Health and Growth. Nottingham UK: Nottingham University Press.ADAMS, C. A. 2004. Food safety and quality: the effect of animal feeds. Nutrition Abstracts andReviews, Series A 74, 41-51.AFSSA, Agence Française de Sécurité Sanitaire des Aliments, 2009. Évaluation des risques liés à laprésence de mycotoxines dans les chaînes alimentaires humaine et animale.AHAMAD, D. B., VAIRAMUTHU, S., BALASUBRAMANIAM, G. A., MANOHAR, B. M. & BALACHANDRAN,C. 2006. Individual and combined effects of citrinin and aflatoxin B1 in broiler chickens: Aclinico-pathology study. Indian Journal of Veterinary Pathology, 30.AL-SADI, R., YE, D. M., DOKLADNY, K. & MA, T. Y. 2008. Mechanism of IL-1 beta-induced increase inintestinal epithelial tight junction permeability. Journal of Immunology, 180, 5653-5661.AL-SADI, R. M. & MA, T. Y. 2007. IL-1 beta causes an increase in intestinal epithelial tight junctionpermeability. Journal of Immunology, 178, 4641-4649.ALLAMEH, A., SAFAMEHR, A., MIRHADI, S. A., SHIVAZAD, M., RAZZAGHI-ABYANEH, M. & AFSHAR-NADERI, A. 2005. Evaluation of biochemical and production parameters of broiler chicks fedammonia treated aflatoxin contaminated maize grains. Animal Feed Science and Technology,122, 289-301.ALMOND, G. W. 1996. Research applications using pigs. Veterinary Clinics of North America-FoodAnimal Practice, 12, 707-&.AMUZIE, C. J., HARKEMA, J. R. & PESTKA, J. J. 2008. Tissue distribution and proinflammatory cytokineinduction by the trichothecene deoxynivalenol in the mouse: Comparison of nasal vs. oralexposure. Toxicology, 248, 39-44.AOUDIA, N., CALLU, P., GROSJEAN, F. & LARONDELLE, Y. 2009. Effectiveness of mycotoxinsequestration activity of micronized wheat fibres on distribution of ochratoxin A in plasma,liver and kidney of piglets fed a naturally contaminated diet. Food and Chemical Toxicology,47, 1485-1489.ARAVIND, K. L., PATIL, V. S., DEVEGOWDA, G., UMAKANTHA, B. & GANPULE, S. P. 2003. Efficacy ofesterified glucomannan to counteract mycotoxicosis in naturally contaminated feed onperformance and serum biochemical and hematological parameters in broilers. PoultryScience, 82, 571-576.AVANTAGGIATO, G., HAVENAAR, R. & VISCONTI, A. 2003. Assessing the zearalenone-binding activityof adsorbent materials during passage through a dynamic in vitro gastrointestinal model.Food and Chemical Toxicology, 41, 1283-1290.AVANTAGGIATO, G., HAVENAAR, R. & VISCONTI, A. 2004. Evaluation of the intestinal absorption ofdeoxynivalenol and nivalenol by an in vitro gastrointestinal model, and the binding efficacy ofactivated carbon and other adsorbent materials. Food and Chemical Toxicology, 42, 817-824.189

REFERENCES BIBLIOGRAPHIQUESAVANTAGGIATO, G., SOLFRIZZO, M. & VISCONTI, A. 2005. Recent advances on the use of adsorbentmaterials for detoxification of Fusarium mycotoxins. Food Additives and Contaminants, 22,379-388.AWAD, W. A., BOHM, J., RAZZAZI-FAZELI, E. & ZENTEK, J. 2006. Effects of feeding deoxynivalenolcontaminated wheat on growth performance, organ weights and histological parameters ofthe intestine of broiler chickens. Journal of Animal Physiology and Animal Nutrition, 90, 32-37.AWAD, W. A., GHAREEB, K., BOHM, J. & ZENTEK, J. 2010. Decontamination and detoxificationstrategies for the Fusarium mycotoxin deoxynivalenol in animal feed and the effectiveness ofmicrobial biodegradation. Food Additives and Contaminants Part a-Chemistry AnalysisControl Exposure & Risk Assessment, 27, 510-520.AYRAL, A. M., DUBECH, N., LEBARS, J. & ESCOULA, L. 1992. Invitro Effect of Diacetoxyscirpenol andDeoxynivalenol on Microbicidal Activity of Murine Peritoneal-Macrophages. Mycopathologia,120, 121-127.AZCONAOLIVERA, J. I., OUYANG, Y., MURTHA, J., CHU, F. S. & PESTKA, J. J. 1995. Induction of CytokineMessenger-Rnas in Mice after Oral-Exposure to the Trichothecene Vomitoxin(Deoxynivalenol) - Relationship to Toxin Distribution and Protein-Synthesis Inhibition.Toxicology and Applied Pharmacology, 133, 109-120.AZIZ, N. H. & YOUSSEF, B. M. 2002. Inactivation of Naturally Occurring of Mycotoxins in someEgyptian Foods and Agricultural Commodities by Gamma-Irradiation. Egyptian Journal ofFood Science, 30, 167-177.BBACON, C. W., PORTER, J. K. & NORRED, W. P. 1995. Toxic Interaction of Fumonisin B-1 and FusaricAcid Measured by Injection into Fertile Chicken Egg. Mycopathologia, 129, 29-35.BATA, A. & LASZTITY, R. 1999. Detoxification of mycotoxin-contaminated food and feed bymicroorganisms. Trends in Food Science & Technology, 10, 223-228.BAUER, J. 1994. Methods for Detoxification of Mycotoxins in Feedstuffs. Monatshefte FurVeterinarmedizin, 49, 175-181.BEAVER, R. W., WILSON, D. M., JAMES, M. A., HAYDON, K. D., COLVIN, B. M., SANGSTER, L. T., PIKUL,A. H. & GROOPMAN, J. D. 1990. Distribution of Aflatoxins in Tissues of Growing-Pigs Fed anAflatoxin-Contaminated Diet Amended with a High-Affinity Aluminosilicate Sorbent.Veterinary and Human Toxicology, 32, 16-18.BENNETT, G. A. & ANDERSON, R. A. 1978. Distribution of Aflatoxin and-or Zearalenone in Wet-MilledCorn Products - Review. Journal of Agricultural and Food Chemistry, 26, 1055-1060.BENNETT, G. A., PEPLINSKI, A. J., BREKKE, O. L., JACKSON, L. K. & WICHSER, W. R. 1976. Zearalenone -Distribution in Dry-Milled Fractions of Contaminated Corn. Cereal Chemistry, 53, 299-307.BENNETT, G. A. & RICHARD, J. L. 1996. Influence of processing on Fusarium mycotoxins incontaminated grains. Food Technology, 50, 235-238.BENNETT, G. A., SHOTWELL, O. L. & HESSELTINE, C. W. 1980. Destruction of Zearalenone inContaminated Corn. Journal of the American Oil Chemists Society, 57, 245-247.BERGSJO, B., LANGSETH, W., NAFSTAD, I., JANSEN, J. H. & LARSEN, H. J. S. 1993. The Effects ofNaturally Deoxynivalenol-Contaminated Oats on the Clinical Condition, Blood Parameters,Performance and Carcass Composition of Growing Pigs. Veterinary ResearchCommunications, 17, 283-294.BERMUDEZ, A. J., LEDOUX, D. R., ROTTINGHAUS, G. E. & BENNETT, G. A. 1997. The individual andcombined. Effects of the fusarium mycotoxins moniliformin and fumonisin B-1 in turkeys.Avian Diseases, 41, 304-311.190

REFERENCES BIBLIOGRAPHIQUESBHANDARI, N., BROWN, C. C. & SHARMA, R. P. 2002. Fumonisin B-1-induced localized activation ofcytokine network in mouse liver. Food and Chemical Toxicology, 40, 1483-1491.BHAT, R. V., SHETTY, P. H., RAO, P. A. & RAO, V. S. 1997. A foodborne disease outbreak due to theconsumption of moldy sorghum and maize containing fumonisin mycotoxins. Journal ofToxicology-Clinical Toxicology, 35, 249-255.BIMCZOK, D., DOLL, S., RAU, H., GOYARTS, T., WUNDRACK, N., NAUMANN, M., DANICKE, S. &ROTHKOTTER, H. J. 2007. The Fusarium toxin deoxynivalenol disrupts phenotype andfunction of monocyte-derived dendritic cells in vivo and in vitro. Immunobiology, 212, 655-66.BINDER, E. M. 2007. Managing the risk of mycotoxins in modern feed production. 133, 149-166.BINDER, E. M., HEIDLER, D., SCHATZMAYR, G., THIMM, N., FUCHS, E., SCHUH, M., KRSKA, R. &BINDER, J. Microbial detoxification of mycotoxins in animal feed. Proceedings of the 10thInternational IUPAC Symposium on the Mycotoxins and Phycotoxins, 2000 Brazil.BINDER, E. M., HEIDLER, D., SCHATZMAYR, G., THIMM, N., FUCHS, E., SCHUH, M., KRSKA, R. &BINDER, J. Microbial detoxification of mycotoxins. Proceedings of the World MycotoxinForum, 2001 Netherlands.BINDER, E. M., TAN, L. M., CHIN, L. J., HANDL, J. & RICHARD, J. 2007. Worldwide occurrence ofmycotoxins in commodities, feeds and feed ingredients. Animal Feed Science andTechnology, 137, 265-282.BLACKWELL, B. A., GILLIAM, J. T., SAVARD, M. E., MILLER, J. D. & DUVICK, J. P. 1999. Oxidativedeamination of hydrolyzed fumonisin B-1 (AP(1)) by cultures of Exophiala spinifera. NaturalToxins, 7, 31-38.BOEIRA, L. S., BRYCE, J. H., STEWART, G. G. & FLANNIGAN, B. 2000. The effect of combinations ofFusarium mycotoxins (deoxynivalenol, zearalenone and fumonisin B1) on growth of brewingyeasts. J Appl Microbiol, 88, 388-403.BÖHM, J., GRAJEWSKI, J., ASPERGER, H., CECON, B., RABUS, B. & RAZZAZI, E. 2000. Study onbiodegradation of some A- and B-trichothecenes and ochratoxin A by use of probioticmicroorganisms. Mycotoxin Research, 16, 70-74.BONNA, R. J., AULERICH, R. J., BURSIAN, S. J., POPPENGA, R. H., BRASELTON, W. E. & WATSON, G. L.1991. Efficacy of Hydrated Sodium Calcium Aluminosilicate and Activated-Charcoal inReducing the Toxicity of Dietary Aflatoxin to Mink. Archives of Environmental Contaminationand Toxicology, 20, 441-447.BOTHAST, R. J. & HESSELTINE, C. W. 1975. Bright Greenish-Yellow Fluorescence and Aflatoxin inAgricultural Commodities. Applied Microbiology, 30, 337-338.BOUHET, S., HOURCADE, E., LOISEAU, N., FIKRY, A., MARTINEZ, S., ROSELLI, M., GALTIER, P.,MENGHERI, E. & OSWALD, I. P. 2004. The mycotoxin fumonisin B-1 alters the proliferationand the barrier function of porcine intestinal epithelial cells. Toxicological Sciences, 77, 165-171.BOUHET, S., LE DORZE, E., PERES, S., FAIRBROTHER, J. M. & OSWALD, I. P. 2006. Mycotoxin fumonisinB-1 selectively down-regulates the basal IL-8 expression in pig intestine: in vivo and in vitrostudies. Food and Chemical Toxicology, 44, 1768-1773.BOUHET, S. & OSWALD, I. P. 2005. The effects of mycotoxins, fungal food contaminants, on theintestinal epithelial cell-derived innate immune response. Veterinary Immunology andImmunopathology, 108, 199-209.BOUHET, S. & OSWALD, I. P. 2007. The intestine as a possible target for fumonisin toxicity. Mol NutrFood Res, 51, 925-31.BRATICH, P. M., BUCK, W. B. & HASCHEK, W. M. 1990. Prevention of T-2 Toxin-Induced MorphologicEffects in the Rat by Highly Activated-Charcoal. Archives of Toxicology, 64, 251-253.191

REFERENCES BIBLIOGRAPHIQUESBREINHOLT, V., HENDRICKS, J., PEREIRA, C., ARBOGAST, D. & BAILEY, G. 1995. Dietary Chlorophyllin Isa Potent Inhibitor of Aflatoxin B1 Hepatocarcinogenesis in Rainbow Trout. Cancer Research,55, 57-62.BROOMHEAD, J. N., LEDOUX, D. R., BERMUDEZ, A. J. & ROTTINGHAUS, G. E. 2002. Chronic effects offumonisin B-1 in broilers and turkeys fed dietary treatments to market age. Poultry Science,81, 56-61.BROWN, T. P., MANNING, R. O., FLETCHER, O. J. & WYATT, R. D. 1986. The Individual and CombinedEffects of Citrinin-a and Ochratoxin-a on Renal Ultrastructure in Layer Chicks. Avian Diseases,30, 191-198.BROWN, T. P., ROTTINGHAUS, G. E. & WILLIAMS, M. E. 1992. Fumonisin Mycotoxicosis in Broilers -Performance and Pathology. Avian Diseases, 36, 450-454.BRUININK, A., RASONYI, T. & SIDLER, C. 1998. Differences in neurotoxic effects of ochratoxin A,ochracin and ochratoxin-αin vitro. Natural Toxins, 6, 173-177.BRYDEN, W. L. 2009. Mycotoxins and Mycotoxicoses: Significance, Occurrence and Mitigation in theFood Chain, John Wiley & Sons, Ltd.BULLERMAN, L. B. & BIANCHINI, A. 2007. Stability of mycotoxins during food processing.International Journal of Food Microbiology, 119, 140-146.BULLERMAN, L. B., RYU, D. & BIANCHINI, A. 2007. Biological Evaluation of Reduction of Fumonisin B1Toxicity in Corn Grits by Extrusion Processing. Final Report to The Andersons Research GrantProgram of NC 213. Management of Grain Quality and Security in World Markets.BURNS, T. D., SNOOK, M. E., RILEY, R. T. & VOSS, K. A. 2008. Fumonisin concentrations and in vivotoxicity of nixtamalized Fusarium verticillioides culture material: Evidence for fumonisinmatrixinteractions. Food and Chemical Toxicology, 46, 2841-2848.BURSIAN, S. J., AULERICH, R. J., CAMERON, J. K., AMES, N. K. & STEFICEK, B. A. 1992. Efficacy ofHydrated Sodium Calcium Aluminosilicate in Reducing the Toxicity of Dietary Zearalenone toMink. Journal of Applied Toxicology, 12, 85-90.BURSIAN, S. J., MITCHELL, R. R., YAMINI, B., FITZGERALD, S. D., MURPHY, P. A., FERNADEZ, G.,ROTTINGHAUS, G. E., MORAN, L., LEEFERS, K. & CHOI, I. 2004. Efficacy of a commercialmycotoxin binder in alleviating effects of ochratoxin A, fumonisin B-1, moniliformin andzearalenone in adult mink. Veterinary and Human Toxicology, 46, 122-129.CCALONI, F., SPOTTI, M., POMPA, G., ZUCCO, F., STAMMATI, A. & DE ANGELIS, I. 2002. Evaluation ofFumonisin B-1 and its metabolites absorption and toxicity on intestinal cells line Caco-2.Toxicon, 40, 1181-1188.CAMILO, S. B., ONO, C. J., UENO, Y. & HIROOKA, E. Y. 2000. Anti-Fusarium moniliforme activity andfumonisin biodegradation by corn and silage microflora. Brazilian Archives of Biology andTechnology, 43, 159-164.CANELA, R., PUJOL, R., SALA, N. & SANCHIS, V. 1996. Fate of fumonisins B-1 and B-2 in steeped cornkernels. Food Additives and Contaminants, 13, 511-517.CAPALDO, C. T. & NUSRAT, A. 2009. Cytokine regulation of tight junctions. Biochimica Et BiophysicaActa-Biomembranes, 1788, 864-871.CARLSON, D. B., WILLIAMS, D. E., SPITSBERGEN, J. M., ROSS, P. F., BACON, C. W., MEREDITH, F. I. &RILEY, R. T. 2001. Fumonisin B-1 promotes aflatoxin B-1 and N-methyl-N '-nitronitrosoguanidine-initiatedliver tumors in rainbow trout. Toxicology and AppliedPharmacology, 172, 29-36.CARSON, M. S. & SMITH, T. K. 1983. Role of Bentonite in Prevention of T-2 Toxicosis in Rats. Journalof Animal Science, 57, 1498-1506.192

REFERENCES BIBLIOGRAPHIQUESCASADO, J. M., THEUMER, M., MASIH, D. T., CHULZE, S. & RUBINSTEIN, H. R. 2001. Experimentalsubchronic mycotoxicoses in mice: individual and combined effects of dietary exposure tofumonisins and aflatoxin B1. Food and Chemical Toxicology, 39, 579-586.CAST 2003. Mycotoxins: risks in plant, animal, and human system. Report 139. 199 p. J.L. Richard andG.A. Payne, ed. CAST, Ames, USA.CASTELO, M. M. 1999. Stability of Mycotoxins in Thermally Processed Corn Products. Section IV. Lossof Fumonisin B1 during the Corn Flake Process with and without Sugars.CASTELO, M. M., JACKSON, L. S., HANNA, M. A., REYNOLDS, B. H. & BULLERMAN, L. B. 2001. Loss offuminosin B-1 in extruded and baked corn-based foods with sugars. Journal of Food Science,66, 416-421.CASTELO, M. M., KATTA, S. K., SUMNER, S. S., HANNA, M. A. & BULLERMAN, L. B. 1998a. Extrusioncooking reduces recoverability of fumonisin B-1 from extruded corn grits. Journal of FoodScience, 63, 696-698.CASTELO, M. M., SUMNER, S. S. & BULLERMAN, L. B. 1998b. Stability of fumonisins in thermallyprocessed corn products. Journal of Food Protection, 61, 1030-1033.CAZZANIGA, D., BASILICO, J. C., GONZALEZ, R. J., TORRES, R. L. & DE GREEF, D. M. 2001. Mycotoxinsinactivation by extrusion cooking of corn flour. Letters in Applied Microbiology, 33, 144-147.CELIK, I., DEMET, O., DONMEZ, H. H., OGUZ, H. & BOYDAK, M. 1996. Determination of phagocytic andcandidacidal activities of peritoneal macrophages isolated from chickens fed with aflatoxinand an aflatoxin adsorbing agent, polyvinylpolypyrrolidone. Veteriner Bilimleri Dergisi, 12,145-151.CELIK, I., OGUZ, H., DEMET, O., DONMEZ, H. H., BOYDAK, M. & SUR, E. 2000. Efficacy ofpolyvinylpolypyrrolidone in reducing the immunotoxicity of aflatoxin in growing broilers.British Poultry Science, 41, 430-439.CETIN, Y. & BULLERMAN, L. B. 2005. Evaluation of reduced toxicity of zearalenone by extrusionprocessing as measured by the MTT cell proliferation assay. Journal of Agricultural and FoodChemistry, 53, 6558-6563.CETIN, Y. & BULLERMAN, L. B. 2006. Confirmation of reduced toxicity of deoxynivalenol in extrusionprocessedcorn grits by the MTT bioassay. Journal of Agricultural and Food Chemistry, 54,1949-1955.CHANG, C. H., YU, F. Y., WANG, L. T., LIN, Y. S. & LIU, B. H. 2009. Activation of ERK and JNK signalingpathways by mycotoxin citrinin in human cells. Toxicology and Applied Pharmacology, 237,281-287.CHATTERJEE, D. & MUKHERJEE, S. K. 1993. Destruction of Phagocytosis-Suppressing Activity ofAflatoxin-B(1) by Ozone. Letters in Applied Microbiology, 17, 52-54.CHELKOWSKI, J., GOLINSKI, P., GODLEWSKA, B., RADOMYSKA, W., SZEBIOTKO, K. & WIEWIOROWSKA,M. 1981. Myco-Toxins in Cereal Grain .4. Inactivation of Ochratoxin-a and Other Myco-ToxinsDuring Ammoniation. Nahrung-Food, 25, 631-637.CHEN, F., MA, Y. L., XUE, C. Y., MA, J. Y., XIE, Q. M., BI, Y. Z. & CAO, Y. C. 2008. The combination ofdeoxynivalenol and zearalenone at permitted feed concentrations causes seriousphysiological effects in young pigs. Journal of Veterinary Science, 9, 39-44.CHENG, Y. H., WENG, C. F., CHEN, B. J. & CHANG, M. H. 2006. Toxicity of different Fusariummycotoxins on growth performance, immune responses and efficacy of a mycotoxindegrading enzyme in pigs. Animal Research, 55, 579-590.CHUNG, T. K. & BAKER, D. H. 1990. Phosphorus Utilization in Chicks Fed Hydrated Sodium CalciumAluminosilicate. Journal of Animal Science, 68, 1992-1998.CIEGLER, A., LILLEHOJ, E. B., PETERSON, R. E. & HALL, H. H. 1966. Microbial detoxification of aflatoxin.Applied microbiology, 14, 934-939.193

REFERENCES BIBLIOGRAPHIQUESCINAR, M., YILDIRIM, E., YALCINKAYA, I. & ERASLAN, G. 2008. The Effects of Yeast Glucomannan(Mycosorb) on Lipid Peroxidation and Non-Enzymatic Antioxidative Status in ExperimentallyInduced Aflatoxicosis in Broilers. Journal of Animal and Veterinary Advances, 7, 539-544.COBB, M. H. 1999. MAP kinase pathways. Progress in Biophysics and Molecular Biology, 71, 479-500.COENEN, M. Capacity of zeolithe to depress the oestrogenic effects of zearalenone. Proceedings ofthe Society of Nutrition and Physiology, 2001. 177.COLLINS, G. J. & ROSEN, J. D. 1981. Distribution of T-2 Toxin in Wet-Milled Corn Products. Journal ofFood Science, 46, 877-879.COLLINS, T. F. X., SPRANDO, R. L., BLACK, T. N., OLEJNIK, N., EPPLEY, R. M., SHACKELFORD, M. E.,HOWARD, P. C., RORIE, J. I., BRYANT, M. & RUGGLES, D. I. 2006. Effects of aminopentol on inutero development in rats. Food and Chemical Toxicology, 44, 161-169.COMMISSION REGULATION (EC) NO 466/2001 OF 8 MARCH 2001 SETTING MAXIMUM LEVELS FORCERTAIN CONTAMINANTS IN FOODSTUFFS, B.CONWAY, H. F., ANDERSON, R. A. & BAGLEY, E. B. 1978. Detoxification of Aflatoxin-ContaminatedCorn by Roasting. Cereal Chemistry, 55, 115-117.CORTEZ-ROCHA, M. O., TRIGO-STOCKLI, D. M., WETZEL, D. L. & REED, C. R. 2002. Effect of extrusionprocessing on fumonisin B-1 and hydrolyzed fumonisin B-1 in contaminated alkali-cookedcorn. Bulletin of Environmental Contamination and Toxicology, 69, 471-478.COTE, L. M., BEASLEY, V. R., BRATICH, P. M., SWANSON, S. P., SHIVAPRASAD, H. L. & BUCK, W. B.1985. Sex-Related Reduced Weight Gains in Growing Swine Fed Diets ContainingDeoxynivalenol. Journal of Animal Science, 61, 942-950.COULOMBE, R. A., GUARISCO, J. A., KLEIN, P. J. & HALL, J. O. 2005. Chemoprevention of aflatoxicosisin poultry by dietary butylated hydroxytoluene. Animal Feed Science and Technology, 121,217-225.CREPPY, E. E. 2002. Update of survey, regulation and toxic effects of mycotoxins in Europe.Toxicology Letters, 127, 19-28.CUI, W., LI, L. X., SUN, C. M., WEN, Y., ZHOU, Y., DONG, Y. L. & LIU, P. 2010. Tumor necrosis factoralpha increases epithelial barrier permeability by disrupting tight junctions in Caco-2 cells.Brazilian Journal of Medical and Biological Research, 43, 330-337.DDAKOVIC, A., TOMASEVIC-CANOVIC, M., DONDUR, V., ROTTINGHAUS, G. E., MEDAKOVIC, V. & ZARIC,S. 2005. Adsorption of mycotoxins by organozeolites. Colloids and Surfaces B-Biointerfaces,46, 20-25.DALVI, R. R. & ADEMOYERO, A. A. 1984. Toxic Effects of Aflatoxin-B1 in Chickens Given FeedContaminated with Aspergillus-Flavus and Reduction of the Toxicity by Activated-Charcoaland Some Chemical-Agents. Avian Diseases, 28, 61-69.DALVI, R. R. & MCGOWAN, C. 1984. Experimental Induction of Chronic Aflatoxicosis in Chickens byPurified Aflatoxin-B1 and Its Reversal by Activated-Charcoal, Phenobarbital, and ReducedGlutathione. Poultry Science, 63, 485-491.DANICKE, S., GOYARTS, T. & VALENTA, H. 2007. On the specific and unspecific effects of a polymericglucomannan mycotoxin adsorbent on piglets when fed with uncontaminated or withFusarium toxins contaminated diets. Archives of Animal Nutrition, 61, 266-275.DE ANGELIS, I., FRIGGE, G., RAIMONDI, F., STAMMATI, A., ZUCCO, F. & CALONI, F. 2005. Absorptionof Fumonisin B-1 and aminopentol on an in vitro model of intestinal epithelium; the role of P-glycoprotein. Toxicon, 45, 285-291.DE VRIES, J. E. 1995. Immunosuppressive and Anti-inflammatory Properties of Interleukin 10. Annalsof Medicine, 27, 537-541.194

REFERENCES BIBLIOGRAPHIQUESDEGIRMENCIOGLU, N., ESECALI, H., COKAL, Y. & BILGIC, M. 2005. From safety feed to safety food: theapplication of HACCP in mycotoxin control. Archive Zootechnic, 8, 19-32.DEL BIANCHI, M., OLIVEIRA, C. A. F., ALBUQUERQUE, R., GUERRA, J. L. & CORREA, B. 2005. Effects ofprolonged oral administration of aflatoxin B-1 and fumonisin B-1 in broiler chickens. PoultryScience, 84, 1835-1840.DENLI, M., BLANDON, J. C., GUYNOT, M. E., SALADO, S. & PEREZ, J. F. 2008. Efficacy of a NewOchratoxin-Binding Agent (OcraTox) to Counteract the Deleterious Effects of Ochratoxin A inLaying Hens. Poultry Science, 87, 2266-2272.DENLI, M., BLANDON, J. C., GUYNOT, M. E., SALADO, S. & PEREZ, J. F. 2009. Effects of dietaryAflaDetox on performance, serum biochemistry, histopathological changes, and aflatoxinresidues in broilers exposed to aflatoxin B-1. Poultry Science, 88, 1444-1451.DESAI, K., SULLARDS, M. C., ALLEGOOD, J., WANG, E., SCHMELZ, E. M., HARTL, M., HUMPF, H. U.,LIOTTA, D. C., PENG, Q. & MERRILL, A. H. 2002. Fumonisins and fumonisin analogs asinhibitors of ceramide synthase and inducers of apoptosis. Biochimica Et Biophysica Acta-Molecular and Cell Biology of Lipids, 1585, 188-192.DEVRIENDT, B., GALLOIS, M., VERDONCK, F., WACHE, Y., BIMCZOK, D., OSWALD, I. P., GODDEERIS, B.M. & COX, E. 2009. The food contaminant fumonisin B-1 reduces the maturation of porcineCD11R1(+) intestinal antigen presenting cells and antigen-specific immune responses, leadingto a prolonged intestinal ETEC infection. Veterinary Research, 40.DIAZ-LLANO, G. & SMITH, T. K. 2006. Effects of feeding grains naturally contaminated with Fusariummycotoxins with and without a polymeric glucomannan mycotoxin adsorbent onreproductive performance and serum chemistry of pregnant gilts. Journal of Animal Science,84, 2361-2366.DIAZ, G. J. 2002. Evaluation of the efficacy of a feed additive to ameliorate the toxic effects of 4,15-diacetoxiscirpenol in growing chicks. Poultry Science, 81, 1492-1495.DIAZ, G. J., CORTES, A. & ROLDAN, L. 2005. Evaluation of the efficacy of four feed additives againstthe adverse effects of T-2 toxin in growing broiler chickens. Journal of Applied PoultryResearch, 14, 226-231.DIAZ, G. J., SQUIRES, E. J., JULIAN, R. J. & BOERMANS, H. J. 1994. Individual and Combined Effects ofT-2 Toxin and Das in Laying Hens. British Poultry Science, 35, 393-405.DILKIN, P., HASSEGAWA, R., REIS, T. A., MALLMANN, C. A. & CORREA, B. 2004. Intoxicaçaoexperimental de suinos por fumonisinas. Ciência Rural, 34, 175-181.DILKIN, P., ZORZETE, P., MALLMANN, C. A., GOMES, J. D. F., UTIYAMA, C. E., OETTING, L. L. & CORREA,B. 2003. Toxicological effects of chronic low doses of aflatoxin B-1 and fumonisin B-1-containing Fusarium mondiforme culture material in weaned piglets. Food and ChemicalToxicology, 41, 1345-1353.DIXON, J. B., KANNEWISCHER, I., ARVIDE, M. G. T. & VELAZQUEZ, A. L. B. 2008. Aflatoxinsequestration in animal feeds by quality-labeled smectite clays: An introductory plan. AppliedClay Science, 40, 201-208.DOLLEAR, F. G. 1969. Detoxification of Aflatoxins in Foods and Feeds. Aflatoxin-Scientific Background,Control, and Implications, L.A. Goldblatt (Ed.), Academic Press, New York, NY, 359-391.DOMBRINK-KURTZMAN, M. A., DVORAK, T. J., BARRON, M. E. & ROONEY, L. W. 2000. Effect ofnixtamalization (alkaline cooking) on fumonisin-contaminated corn for production of masaand tortillas. Journal of Agricultural and Food Chemistry, 48, 5781-5786.DOMIJAN, A. M., PERAICA, M., VRDOLJAK, A. L., RADIC, B., ZLENDER, V. & FUCHS, R. 2007. Theinvolvement of oxidative stress in ochratoxin A and fumonisin B-1 toxicity in rats. MolecularNutrition & Food Research, 51, 1147-1151.DONG, C., DAVIS, R. J. & FLAVELL, R. A. 2002. MAP kinases in the immune response. Annual Review ofImmunology, 20, 55-72.195

REFERENCES BIBLIOGRAPHIQUESDOYLE, M. P., APPLEBAUM, R. S., BRACKETT, R. E. & MARTH, E. H. 1982. Physical, Chemical andBiological Degradation of Myco-Toxins in Foods and Agricultural Commodities. Journal ofFood Protection, 45, 964-971.DRESDEN-OSBORNE, C. & NOBLET, G. P. 2002. Fumonisin B-1 affects viability and alters nitric oxideproduction of a murine macrophage cell line. International Immunopharmacology, 2, 1087-1093.DROCHNER, W., SCHOLLENBERGER, M., PIEPHO, H. P., GOTZ, S., LAUBER, U., TAFAJ, M., KLOBASA, F.,WEILER, U., CLAUS, R. & STEFFL, M. 2004. Serum IgA-promoting effects induced by feed loadscontaining isolated deoxynivalenol (DON) in growing piglets. Journal of Toxicology andEnvironmental Health-Part a-Current Issues, 67, 1051-1067.DUGYALA, R. R., SHARMA, R. P., TSUNODA, M. & RILEY, R. T. 1998. Tumor necrosis factor-alpha as acontributor in fumonisin B-1 toxicity. Journal of Pharmacology and ExperimentalTherapeutics, 285, 317-324.DUTTON, M. F., WESTLAKE, K. & ANDERSON, M. S. 1984. The Interaction between Additives, Yeastsand Patulin Production in Grass-Silage. Mycopathologia, 87, 29-33.DUVICK, J. 2001. Prospects for reducing fumonisin contamination of maize through geneticmodification. Environmental Health Perspectives, 109, 337-342.DVORSKA, J. E., PAPPAS, A. C., KARADAS, F., SPEAKE, B. K. & SURAI, P. F. 2007. Protective effect ofmodified glucomannans and organic selenium against antioxidant depletion in the chickenliver due to T-2 toxin-contaminated feed consumption. Comparative Biochemistry andPhysiology C-Toxicology & Pharmacology, 145, 582-587.EEDRINGTON, T. S., HARVEY, R. B. & KUBENA, L. F. 1995. Toxic Effects of Aflatoxin-B-1 and Ochratoxina,Alone and in Combination, on Chicken Embryos. Bulletin of Environmental Contaminationand Toxicology, 54, 331-336.EDRINGTON, T. S., KUBENA, L. F., HARVEY, R. B. & ROTTINGHAUS, G. E. 1997. Influence of asuperactivated charcoal on the toxic effects of aflatoxin or T-2 toxin in growing broilers.Poultry Science, 76, 1205-1211.EFSA, European Food Safety Authority, 2005. Opinion of the Scientific Panel on Contaminants in FoodChain on a request from the Commission related to fumonisins as undesirable substances inanimal feed. The EFSA Journal, 235, 1-32.EL-NEZAMI, H., MYKKANEN, H., KANKAANPAA, P., SALMINEN, S. & AHOKAS, J. 2000. Ability ofLactobacillus and Propionibacterium strains to remove aflatoxin B-1 from the chickenduodenum. Journal of Food Protection, 63, 549-552.EL-NEZAMI, H., POLYCHRONAKI, N., SALMINEN, S. & MYKKANEN, H. 2002a. Binding rather thanmetabolism may explain the interaction of two food-grade Lactobacillus strains withzearalenone and its derivative alpha-zearalenol. Applied and Environmental Microbiology, 68,3545-3549.EL-NEZAMI, H. S., CHREVATIDIS, A., AURIOLA, S., SALMINEN, S. & MYKKANEN, H. 2002b. Removal ofcommon Fusarium toxins in vitro by strains of Lactobacillus and Propionibacterium. FoodAdditives and Contaminants, 19, 680-686.ELSHARKAWY, S. & ABULHAJJ, Y. 1987. Microbial Transformation of Zearalenone .1. Formation ofZearalenone-4-O-Beta-Glucoside. Journal of Natural Products, 50, 520-521.ELSHARKAWY, S. & ABULHAJJ, Y. J. 1988. Microbial Cleavage of Zearalenone. Xenobiotica, 18, 365-371.ENONGENE, E. N., SHARMA, R. P., BHANDARI, N., VOSS, K. A. & RILEY, R. T. 2000. Disruption ofsphingolipid metabolism in small intestines, liver and kidney of mice dosed subcutaneouslywith fumonisin B-1. Food and Chemical Toxicology, 38, 793-799.196

REFERENCES BIBLIOGRAPHIQUESETIENNE, M. & WACHE, Y. 2008. Biological and physiological effects of deoxynivalenol (DON) in thepig. In: Oswald, I.P., Taranu, I. (Eds.), Mycotoxins in Farm Animals, Research Signpost, Kerala,113-130.FFARAG, R. S., RASHED, M. M., HUSSEIN, A. A. & ABO-HAGAR, A. 1995. Effect of Gamma Radiation onthe Infected Yellow Corn and Peanuts by Aspergillus flavus. Chemie MikrobiologieTechnologie der Lebensmittel, 17, 93-98.FERNANDEZ-SURUMAY, G., OSWEILER, G. D., YAEGER, M. J., HAUCK, C. C., HENDRICH, S. & MURPHY,P. A. 2004. Glucose reaction with fumonisin B-1 partially reduces its toxicity in swine. Journalof Agricultural and Food Chemistry, 52, 7732-7739.FERNANDEZ-SURUMAY, G., OSWEILER, G. D., YAEGER, M. J., ROTTINGHAUS, G. E., HENDRICH, S.,BUCKLEY, L. K. & MURPHY, P. A. 2005. Fumonisin B-glucose reaction products are less toxicwhen fed to swine. Journal of Agricultural and Food Chemistry, 53, 4264-4271.FIERS, W. 1991. Tumor-Necrosis-Factor - Characterization at the Molecular, Cellular and Invivo Level.Febs Letters, 285, 199-212.FINK-GREMMELS, J. 1999. Mycotoxins: Their implications for human and animal health. VeterinaryQuarterly, 21, 115-120.FLYNN, T. J., STACK, M. E., TROY, A. L. & CHIRTEL, S. J. 1997. Assessment of the embryotoxic potentialof the total hydrolysis product of fumonisin B-1 using cultured organogenesis-staged ratembryos. Food and Chemical Toxicology, 35, 1135-1141.FORSELL, J. H., WITT, M. F., TAI, J. H., JENSEN, R. & PESTKA, J. J. 1986. Effects of 8-Week Exposure ofthe B6c3f1 Mouse to Dietary Deoxynivalenol (Vomitoxin) and Zearalenone. Food andChemical Toxicology, 24, 213-219.FREMY, J. M., GAUTIER, J. P., HERRY, M. P., TERRIER, C. & CALET, C. 1988. Effects of Ammoniation onthe Carry-over of Aflatoxins into Bovine-Milk. Food Additives and Contaminants, 5, 39-44.FRICKE, R. F. & JORGE, J. 1990. Assessment of Efficacy of Activated-Charcoal for Treatment of AcuteT-2 Toxin Poisoning. Journal of Toxicology-Clinical Toxicology, 28, 421-431.FRIEND, D. W., THOMPSON, B. K., TRENHOLM, H. L., BOERMANS, H. J., HARTIN, K. E. & PANICH, P. L.1992. Toxicity of T-2 Toxin and Its Interaction with Deoxynivalenol When Fed to Young-Pigs.Canadian Journal of Animal Science, 72, 703-711.FRIEND, D. W., TRENHOLM, H. L., YOUNG, J. C., THOMPSON, B. K. & HARTIN, K. E. 1984. Effect ofAdding Potential Vomitoxin (Deoxynivalenol) Detoxicants or a F-Graminearum InoculatedCorn Supplement to Wheat Diets Fed to Pigs. Canadian Journal of Animal Science, 64, 733-741.FUCHS, E., BINDER, E. M., HEIDLER, D. & KRSKA, R. Charakterisierung von metaboliten nach denbakteriellen abbau von A- und B-Trichothecenen durch BBSH 797. Proceeding of the 22ndMykotoxin-Workshop, 2000 Germany.FUCHS, E., BINDER, E. M., HEIDLER, D. & KRSKA, R. 2002. Structural characterization of metabolitesafter the microbial degradation of type A trichothecenes by the bacterial strain BBSH 797.Food Additives and Contaminants, 19, 379-386.FURUYA, H., OHKAWARA, S., NAGASHIMA, K., ASANUMA, N. & HINO, T. 2008. Dietary sphingomyelinalleviates experimental inflammatory bowel disease in mice. International Journal forVitamin and Nutrition Research, 78, 41-49.GGABAL, M. A. & AZZAM, A. H. 1998. Interaction of aflatoxin in the feed and immunization againstselected infectious diseases in poultry. II. Effect on one-day-old layer chicks simultaneously197

REFERENCES BIBLIOGRAPHIQUESvaccinated against Newcastle disease, infectious bronchitis and infectious bursal disease.Avian Pathology, 27, 290-295.GALEY, F. D., LAMBERT, R. J., BUSSE, M. & BUCK, W. B. 1987. Therapeutic Efficacy of SuperactiveCharcoal in Rats Exposed to Oral Lethal Doses of T-2 Toxin. Toxicon, 25, 493-499.GALVANO, F., RITIENI, A., PIVA, G. & PIETRI, A. 2005. Mycotoxins in the human food chain. In: DIAZ,D. E. (ed.) Mycotoxin Bluebook. Nottingham UK: Nottingham University Press.GAOU, I., DUBOIS, M., PFOHL-LESZKOWICZ, A., COSTE, C., DE JOUFFREY, S. & PARENT-MASSIN, D.2005. Safety of Oxygreen (R), an ozone treatment on wheat grains. Part 1. A four-weektoxicity study in rats by dietary administration of treated wheat. Food Additives andContaminants, 22, 1113-1119.GARCIA, A. R., AVILA, E., ROSILES, R. & PETRONE, V. M. 2003. Evaluation of two mycotoxin binders toreduce toxicity of broiler diets containing ochratoxin A and T-2 toxin contaminated grain.Avian Diseases, 47, 691-699.GAUMY, J. L., BAILLY, J. D., BURGAT, V. & GUERRE, P. 2001. Zéaralénone : propriétés et toxicitéexpérimentale. Revue Médecine Vétérinaire, 152, 219-234.GELDERBLOM, W. C. A., MARASAS, W. F. O., LEBEPE-MAZUR, S., SWANEVELDER, S., VESSEY, C. J. &HALL, P. D. 2002. Interaction of fumonisin B-1 and aflatoxin B-1 in a short-termcarcinogenesis model in rat liver. Toxicology, 171, 161-173.GENTLES, A., SMITH, E. E., KUBENA, L. F., DUFFUS, E., JOHNSON, P., THOMPSON, J., HARVEY, R. B. &EDRINGTON, T. S. 1999. Toxicological evaluations of cyclopiazonic acid and ochratoxin a inbroilers. Poultry Science, 78, 1380-1384.GIRISH, C. K. & DEVEGOWDA, C. 2006. Efficacy of glucomannan-containing yeast product (Mycosorb(R)) and hydrated sodium calcium aluminosilicate in preventing the individual and combinedtoxicity of aflatoxin and T-2 toxin in commercial broilers. Asian-Australasian Journal ofAnimal Sciences, 19, 877-883.GLAHN, R. P., WIDEMAN, R. F., EVANGELISTI, J. W. & HUFF, W. E. 1988. Effects of Ochratoxin-a Aloneand in Combination with Citrinin on Kidney-Function of Single Comb White Leghorn Pullets.Poultry Science, 67, 1034-1042.GOPEE, N. V., HE, Q. R. & SHARMA, R. P. 2003. Fumonisin B-1-induced apoptosis is associated withdelayed inhibition of protein kinase C, nuclear factor-kappa B and tumor necrosis factoralpha in LLC-PK1 cells. Chemico-Biological Interactions, 146, 131-145.GOUZE, M. E., LAFFITTE, J., DEDIEU, G., GALINIER, A., THOUVENOT, J. P., OSWALD, I. P. & GALTIER, P.2005. Individual and combined effects of low oral doses of deoxynivalenol and nivalenol inmice. Cellular and Molecular Biology, 51, OL809-OL817.GOWDA, N. K. S., LEDOUX, D. R., ROTTINGHAUS, G. E., BERMUDEZ, A. J. & CHEN, Y. C. 2008. Efficacyof turmeric (Curcuma longa), containing a known level of curcumin, and a hydrated sodiumcalcium aluminosilicate to ameliorate the adverse effects of aflatoxin in broiler chicks.Poultry Science, 87, 1125-1130.GOYARTS, T., DANICKE, S., ROTHKOTTER, H. J., SPILKE, J., TIEMANN, U. & SCHOLLENBERGER, M. 2005.On the effects of a chronic deoxynivalenol intoxication on performance, haematological andserum parameters of pigs when diets are offered either for ad libitum consumption or fedrestrictively. J Vet Med A Physiol Pathol Clin Med, 52, 305-14.GRANT, P. G. & PHILLIPS, T. D. 1998. Isothermal adsorption of aflatoxin B-1 on HSCAS clay. Journal ofAgricultural and Food Chemistry, 46, 599-605.GRATZ, S. 2007. Aflatoxin binding by probiotics experimental studies on intestinal aflatoxin transport,metabolism and toxicity. Doctoral dissertation, Kuopio University.GRATZ, S., TAUBEL, M., JUVONEN, R. O., VILUKSELA, M., TURNER, P. C., MYKKANEN, H. & EL-NEZAMI,H. 2006. Lactobacillus rhamnosus strain GG modulates intestinal absorption, fecal excretion,and toxicity of aflatoxin B-1 in rats. Applied and Environmental Microbiology, 72, 7398-7400.198

REFERENCES BIBLIOGRAPHIQUESGRENIER, B., LOUREIRO-BRACARENSE, A. P., LUCIOLI, J., PACHECO, G., COSSALTER, A. M., MOLL, W.D., SCHATZMAYR, G. & OSWALD, I. P. 2011. Individual and combined effects of subclinicaldoses of deoxynivalenol and fumonisins in piglets. Molecular Nutrition & Food Research, inpress. DOI 10.1002/mnfr.201000402.GUARISCO, J. A., HALL, J. O. & COULOMBE, R. A. 2008. Butylated hydroxytoluene chemoprevention ofaflatoxicosis - Effects on aflatoxin B-1 bioavailability, hepatic DNA adduct formation, andbiliary excretion. Food and Chemical Toxicology, 46, 3727-3731.GUERRE, P. 2000. Interest of the treatments of raw materials and usage of adsorbents todecontaminate animal food containing mycotoxins. Revue De Medecine Veterinaire, 151,1095-1106.GUILLOTEAU, P., ZABIELSKI, R., HAMMON, H. M. & METGES, C. C. 2010. Nutritional programming ofgastrointestinal tract development. Is the pig a good model for man? Nutrition ResearchReviews, 23, 4-22.GUPTA, G. & GARDNER, W. 2005. Use of clay mineral (montmorillonite) for reducing poultry litterleachate toxicity (EC50). Journal of Hazardous Materials, 118, 81-83.GWIAZDOWSKA, D., FILIPIAK, M., CZACZYK, K. & GWIAZDOWSKI, R. 2006. Removal of zearalenonefrom medium by different species of propionibacterium, lactobacillus and bifidobacterium.HHALABI, K. S., NATOUR, R. M. & TAMIMI, S. O. 1998. Individual and combined effects of chronicochratoxin a and zearalenone mycotoxins on rat liver and kidney. Arab Gulf Journal ofScientific Research, 16, 379-392.HALLOY, D. J., GUSTIN, P. G., BOUHET, S. & OSWALD, I. P. 2005. Oral exposure to culture materialextract containing fumonisins predisposes swine to the development of pneumonitis causedby Pasteurella multocida. Toxicology, 213, 34-44.HAMEED, H. G. 1993. Extrusion and Chemical Treatments for Destruction of Aflatoxin in Naturally-Contaminated Corn. University of Arizona.HAN, X. N., FINK, M. P. & DELUDE, R. L. 2003. Proinflammatory cytokines cause no center dotdependentand -independent changes in expression and localization of tight junctionproteins in intestinal epithelial cells. Shock, 19, 229-237.HANIF, N. Q., MUHAMMAD, G., SIDDIQUE, M., KHANUM, A., AHMED, T., GADAHAI, J. A. & KAUKAB,G. 2008. Clinico-pathomorphological, serum biochemical and histological studies in broilersfed ochratoxin A and a toxin deactivator (Mycofix (R) Plus). British Poultry Science, 49, 632-642.HARTINGER, D., HEINL, S., SCHWARTZ, H. E., GRABHERR, R., SCHATZMAYR, G., HALTRICH, D. & MOLL,W. D. 2010. Enhancement of solubility in Escherichia coli and purification of anaminotransferase from Sphingopyxis sp MTA144 for deamination of hydrolyzed fumonisin B-1. Microbial Cell Factories, 9.HARVEY, R. B., EDRINGTON, T. S., KUBENA, L. F., ELISSALDE, M. H., CASPER, H. H., ROTTINGHAUS, G.E. & TURK, J. R. 1996. Effects of dietary fumonisin B-1-containing culture material,deoxynivalenol-contaminated wheat, or their combination on growing barrows. AmericanJournal of Veterinary Research, 57, 1790-1794.HARVEY, R. B., EDRINGTON, T. S., KUBENA, L. F., ELISSALDE, M. H., CORRIER, D. E. & ROTTINGHAUS,G. E. 1995a. Effect of Aflatoxin and Diacetoxyscirpenol in Ewe Lambs. Bulletin ofEnvironmental Contamination and Toxicology, 54, 325-330.HARVEY, R. B., EDRINGTON, T. S., KUBENA, L. F., ELISSALDE, M. H. & ROTTINGHAUS, G. E. 1995b.Influence of Aflatoxin and Fumonisin B-1-Containing Culture Material on Growing Barrows.American Journal of Veterinary Research, 56, 1668-1672.199

REFERENCES BIBLIOGRAPHIQUESHARVEY, R. B., EDRINGTON, T. S., KUBENA, L. F., ROTTINGHAUS, G. E., TURK, J. R., GENOVESE, K. J.,ZIPRIN, R. L. & NISBET, D. J. 2002. Toxicity of fumonisin from Fusarium verticillioides culturematerial and moniliformin from Fusarium fujikuroi culture material when fed singly and incombination to growing barrows. Journal of Food Protection, 65, 373-377.HARVEY, R. B., HUFF, W. E., KUBENA, L. F. & PHILLIPS, T. D. 1989a. Evaluation of DietsCocontaminated with Aflatoxin and Ochratoxin Fed to Growing-Pigs. American Journal ofVeterinary Research, 50, 1400-1405.HARVEY, R. B., KUBENA, L. F., ELISSALDE, M. H., CORRIER, D. E., HUFF, W. E., ROTTINGHAUS, G. E. &CLEMENT, B. A. 1991a. Cocontamination of swine diets by aflatoxin and diacetoxyscirpenol. JVet Diagn Invest, 3, 155-160.HARVEY, R. B., KUBENA, L. F., ELISSALDE, M. H., CORRIER, D. E. & PHILLIPS, T. D. 1994a. Comparisonof 2 Hydrated Sodium-Calcium Aluminosilicate Compounds to Experimentally ProtectGrowing Barrows from Aflatoxicosis. Journal of Veterinary Diagnostic Investigation, 6, 88-92.HARVEY, R. B., KUBENA, L. F., ELISSALDE, M. H. & PHILLIPS, T. D. 1993. Efficacy of Zeolitic OreCompounds on the Toxicity of Aflatoxin to Growing Broiler-Chickens. Avian Diseases, 37, 67-73.HARVEY, R. B., KUBENA, L. F., ELISSALDE, M. H., ROTTINGHAUS, G. E. & CORRIER, D. E. 1994b.Administration of Ochratoxin-a and Ochratoxin-T-2 Toxin to Growing Swine. AmericanJournal of Veterinary Research, 55, 1757-1761.HARVEY, R. B., KUBENA, L. F., HUFF, W. E., CORRIER, D. E., CLARK, D. E. & PHILLIPS, T. D. 1989b.Effects of Aflatoxin, Deoxynivalenol, and Their Combinations in the Diets of Growing-Pigs.American Journal of Veterinary Research, 50, 602-607.HARVEY, R. B., KUBENA, L. F., HUFF, W. E., CORRIER, D. E., ROTTINGHAUS, G. E. & PHILLIPS, T. D.1990. Effects of Treatment of Growing Swine with Aflatoxin and T-2 Toxin. American Journalof Veterinary Research, 51, 1688-1693.HARVEY, R. B., KUBENA, L. F., PHILLIPS, T. D., CORRIER, D. E., ELISSALDE, M. H. & HUFF, W. E. 1991b.Diminution of Aflatoxin Toxicity to Growing Lambs by Dietary Supplementation withHydrated Sodium Calcium Aluminosilicate. American Journal of Veterinary Research, 52, 152-156.HARVEY, R. B., KUBENA, L. F., ROTTINGHAUS, G. E., TURK, J. R., CASPER, H. H. & BUCKLEY, S. A. 1997.Moniliformin from Fusarium fujikuroi culture material and deoxynivalenol from naturallycontaminated wheat incorporated into diets of broiler chicks. Avian Diseases, 41, 957-963.HASCHEK, W. M., GUMPRECHT, L. A., SMITH, G., TUMBLESON, M. E. & CONSTABLE, P. D. 2001.Fumonisin toxicosis in swine: An overview of porcine pulmonary edema and currentperspectives. Environmental Health Perspectives, 109, 251-257.HASKARD, C. A., EL-NEZAMI, H. S., KANKAANPAA, P. E., SALMINEN, S. & AHOKAS, J. T. 2001. Surfacebinding of aflatoxin B-1 by lactic acid bacteria. Applied and Environmental Microbiology, 67,3086-3091.HAYES, A. W., CAIN, J. A. & MOORE, B. G. 1977. Effect of Aflatoxin-B1, Ochratoxin-a and Rubratoxin-Bon Infant Rats. Food and Cosmetics Toxicology, 15, 23-27.HE, P., YOUNG, L. G. & FORSBERG, C. 1992. Microbial Transformation of Deoxynivalenol (Vomitoxin).Applied and Environmental Microbiology, 58, 3857-3863.HEINL, S., HARTINGER, D., THAMHESL, M., VEKIRU, E., KRSKA, R., SCHATZMAYR, G., MOLL, W. D. &GRABHERR, R. 2010. Degradation of fumonisin B-1 by the consecutive action of two bacterialenzymes. Journal of Biotechnology, 145, 120-129.HENDRICH, S., MILLER, K. A., WILSON, T. M. & MURPHY, P. A. 1993. Toxicity of Fusarium-Proliferatum-Fermented Nixtamalized Corn-Based Diets Fed to Rats - Effect of Nutritional-Status. Journal of Agricultural and Food Chemistry, 41, 1649-1654.HERSHBERG, R. M. & MAYER, L. F. 2000. Antigen processing and presentation by intestinal epithelialcells - polarity and complexity. Immunology Today, 21, 123-128.200

REFERENCES BIBLIOGRAPHIQUESHERZALLAH, S., ALSHAWABKEH, K. & AL FATAFTAH, A. 2008. Aflatoxin Decontamination of ArtificiallyContaminated Feeds by Sunlight, gamma-Radiation, and Microwave Heating. Journal ofApplied Poultry Research, 17, 515-521.HOOGENBOOM, L. A. P., TULLIEZ, J., GAUTIER, J. P., COKER, R. D., MELCION, J. P., NAGLER, M. J.,POLMAN, T. H. G. & DELORT-LAVAL, J. 2001. Absorption, distribution and excretion ofaflatoxin-derived ammoniation products in lactating cows. Food Additives and Contaminants,18, 47-58.HOPKINS, A., SMITH, G. W. & WHITLOW, L. W. 2008. Changes in milk aflatoxin concentrations inresponse to investigational sequestering agents added to aflatoxin-contaminated diets fed tolactating Holstein cows. Journal of Animal Science, 86, 268.HOWARD, P. C., COUCH, L. H., PATTON, R. E., EPPLEY, R. M., DOERGE, D. R., CHURCHWELL, M. I.,MARQUES, M. M. & OKERBERG, C. V. 2002. Comparison of the toxicity of several fumonisinderivatives in a 28-day feeding study with female B6C3F(1) mice. Toxicology and AppliedPharmacology, 185, 153-165.HUFF, W. E. 1980. A Physical Method for the Segregation of Aflatoxin-Contaminated Corn. CerealChemistry, 57, 236-238.HUFF, W. E. & DOERR, J. A. 1981. Synergism between Aflatoxin and Ochratoxin-a in Broiler-Chickens.Poultry Science, 60, 550-555.HUFF, W. E., DOERR, J. A., WABECK, C. J., CHALOUPKA, G. W., MAY, J. D. & MERKLEY, J. W. 1983.Individual and Combined Effects of Aflatoxin and Ochratoxin-a on Bruising in Broiler-Chickens. Poultry Science, 62, 1764-1771.HUFF, W. E., DOERR, J. A., WABECK, C. J., CHALOUPKA, G. W., MAY, J. D. & MERKLEY, J. W. 1984. TheIndividual and Combined Effects of Aflatoxin and Ochratoxin a on Various ProcessingParameters of Broiler-Chickens. Poultry Science, 63, 2153-2161.HUFF, W. E. & HAGLER, W. M. 1982. Evaluation of Density Segregation as a Means to Estimate theDegree of Aflatoxin Contamination of Corn. Cereal Chemistry, 59, 152-153.HUFF, W. E. & HAGLER, W. M. 1985. Density Segregation of Corn and Wheat Naturally Contaminatedwith Aflatoxin, Deoxynivalenol and Zearalenone. Journal of Food Protection, 48, 416-420.HUFF, W. E., HARVEY, R. B., KUBENA, L. F. & ROTTINGHAUS, G. E. 1988. Toxic Synergism betweenAflatoxin and T-2 Toxin in Broiler-Chickens. Poultry Science, 67, 1418-1423.HUFF, W. E., KUBENA, L. F., HARVEY, R. B., HAGLER, W. M., SWANSON, S. P., PHILLIPS, T. D. &CREGER, C. R. 1986. Individual and Combined Effects of Aflatoxin and Deoxynivalenol (Don,Vomitoxin) in Broiler-Chickens. Poultry Science, 65, 1291-1298.HUFF, W. E., KUBENA, L. F., HARVEY, R. B. & PHILLIPS, T. D. 1992. Efficacy of Hydrated SodiumCalcium Aluminosilicate to Reduce the Individual and Combined Toxicity of Aflatoxin andOchratoxin-A. Poultry Science, 71, 64-69.HULT, K., TEILING, A. & GATENBECK, S. 1976. Degradation of Ochratoxin-a by a Ruminant. Appliedand Environmental Microbiology, 32, 443-444.HUMPF, H. U., SCHMELZ, E. M., MEREDITH, F. I., VESPER, H., VALES, T. R., WANG, E., MENALDINO, D.S., LIOTTA, D. C. & MERRILL, A. H. 1998. Acylation of naturally occurring and synthetic 1-deoxysphinganines by ceramide synthase - Formation of N-palmitoyl-aminopentol producesa toxic metabolite of hydrolyzed fumonisin, AP(1), and a new category of ceramide synthaseinhibitor. Journal of Biological Chemistry, 273, 19060-19064.HUMPF, H. U. & VOSS, K. A. 2004. Effects of thermal food processing on the chemical structure andtoxicity of fumonisin mycotoxins. Molecular Nutrition & Food Research, 48, 255-269.HUWIG, A., FREIMUND, S., KAPPELI, O. & DUTLER, H. 2001. Mycotoxin detoxication of animal feed bydifferent adsorbents. Toxicology Letters, 122, 179-188.HYMERY, N., SIBIRIL, Y. & PARENT-MASSIN, D. 2006. In vitro effects of trichothecenes on humandendritic cells. Toxicology in Vitro, 20, 899-909.201

REFERENCES BIBLIOGRAPHIQUESIIDZKO, M., PANTHER, E., CORINTI, S., MORELLI, A., FERRARI, D., HEROUY, Y., DICHMANN, S.,MOCKENHAUPT, M., GEBICKE-HAERTER, P., DI VIRGILIO, F., GIROLOMONI, G. & NORGAUER,J. 2002. Sphingosine 1-phosphate induces chemotaxis of immature dendritic cells andmodulates cytokine-release in mature human dendritic cells for emergence of Th2 immuneresponses. Faseb Journal, 16, 625-+.JJAMES, L. J. & SMITH, T. K. 1982. Effect of Dietary Alfalfa on Zearalenone Toxicity and Metabolism inRats and Swine. Journal of Animal Science, 55, 110-118.JAVED, T., BUNTE, R. M., DOMBRINK-KURTZMAN, M. A., RICHARD, J. L., BENNETT, G. A., COTE, L. M.& BUCK, W. B. 2005. Comparative pathologic changes in broiler chicks on feed amended withFusarium proliferatum culture material or purified fumonisin B-1 and moniliformin.Mycopathologia, 159, 553-564.JIA, Q. S., ZHOU, H. R., BENNINK, M. & PESTKA, J. J. 2004. Docosahexaenoic acid attenuatesmycotoxin-induced immunoglobulin a nephropathy, interleukin-6 transcription, andmitogen-activated protein kinase phosphorylation in mice. Journal of Nutrition, 134, 3343-3349.JIN, Y. X., KNUDSEN, E., WANG, L., BRYCESON, Y., DAMAJ, B., GESSANI, S. & MAGHAZACHI, A. A. 2003.Sphingosine 1-phosphate is a novel inhibitor of T-cell proliferation. Blood, 101, 4909-4915.JINDAL, N., MAHIPAL, S. K. & MAHAJAN, N. K. 1993. Effect of some compounds on distribution ofaflatoxin tissues of broilers. Journal of Animal Science, 8, 85-88.JOUANY, J. P. 2007. Methods for preventing, decontaminating and minimizing the toxicity ofmycotoxins in feeds. Animal Feed Science and Technology, 137, 342-362.KKABAK, B., DOBSON, A. D. W. & VAR, I. 2006. Strategies to prevent mycotoxin contamination of foodand animal feed: A review. Critical Reviews in Food Science and Nutrition.KALOREY, D. R., KURKURE, N. V., RAMGAONKAR, J. S., SAKHARE, P. S., WARKE, S. & NIGOT, N. K.2005. Effect of polyherbal feed supplement "Growell" during induced aflatoxicosis,ochratoxicosis and combined mycotoxicoses in broilers. Asian-Australasian Journal of AnimalSciences, 18, 375-383.KAMALAVENKATESH, P., VAIRAMUTHU, S., BALACHANDRAN, C., MANOHAR, B. M. & RAJ, G. D. 2005.Immunopathological effect of the mycotoxins cyclopiazonic acid and T-2 toxin on broilerchicken. Mycopathologia, 159, 273-279.KAMIMURA, H. 1986. Conversion of Zearalenone to Zearalenone Glycoside by Rhizopus Sp. Appliedand Environmental Microbiology, 52, 515-519.KARAMAN, M., BASMACIOGLU, H., ORTATATLI, M. & OGUZ, H. 2005. Evaluation of the detoxifyingeffect of yeast glucomannan on aflatoxicosis in broilers as assessed by gross examination andhistopathology. British Poultry Science, 46, 394-400.KARLOVSKY, P. 1999. Biological detoxification of fungal toxins and its use in plant breeding, feed andfood production. Natural Toxins, 7, 1-23.KATTA, S. K., CAGAMPANG, A. E., JACKSON, L. S. & BULLERMAN, L. B. 1997. Distribution of Fusariummolds and fumonisins in dry-milled corn fractions. Cereal Chemistry, 74, 858-863.KATTA, S. K., JACKSON, L. S., SUMNER, S. S., HANNA, M. A. & BULLERMAN, L. B. 1999. Effect oftemperature and screw speed on stability of fumonisin B-1 in extrusion-cooked corn grits.Cereal Chemistry, 76, 16-20.202

REFERENCES BIBLIOGRAPHIQUESKERKADI, A., BARRIAULT, C., TUCHWEBER, B., FROHLICH, A. A., MARQUARDT, R. R., BOUCHARD, G. &YOUSEF, I. M. 1998. Dietary cholestyramine reduces ochratoxin A-induced nephrotoxicity inthe rat by decreasing plasma levels and enhancing fecal excretion of the toxin. Journal ofToxicology and Environmental Health-Part a-Current Issues, 53, 231-250.KIDD, M. T., HAGLER, W. M. & QURESHI, M. A. 1995. Trichothecene Mycotoxins Depress theMononuclear-Phagocytic System of Young Turkeys. Immunopharmacology andImmunotoxicology, 17, 385-398.KIESSLING, K. H., PETTERSSON, H., SANDHOLM, K. & OLSEN, M. 1984. Metabolism of Aflatoxin,Ochratoxin, Zearalenone, and 3 Trichothecenes by Intact Rumen Fluid, Rumen Protozoa, andRumen Bacteria. Applied and Environmental Microbiology, 47, 1070-1073.KIM, E. K., SCOTT, P. M. & LAU, B. P. Y. 2003. Hidden fumonisin in corn flakes. Food Additives andContaminants Part a-Chemistry Analysis Control Exposure & Risk Assessment, 20, 161-169.KING, R. R., MCQUEEN, R. E., LEVESQUE, D. & GREENHALGH, R. 1984. Transformation ofDeoxynivalenol (Vomitoxin) by Rumen Microorganisms. Journal of Agricultural and FoodChemistry, 32, 1181-1183.KINSER, S., JIA, Q. S., LI, M. X., LAUGHTER, A., CORNWELL, P. D., CORTON, J. C. & PESTKA, J. J. 2004.Gene expression profiling in spleens of deoxynivalenol-exposed mice: Immediate early genesas primary targets. Journal of Toxicology and Environmental Health-Part a-Current Issues, 67,1423-1441.KIRAN, M. M., DEMET, O., ORTATATH, M. & OGUZ, H. 1998. The preventive effect ofpolyvinylpolypyrrolidone on aflatoxicosis in broilers. Avian Pathology, 27, 250-255.KITCHEN, D. N., CARLTON, W. W. & TUITE, J. 1977a. Ochratoxin-a and Citrinin Induced Nephrosis inBeagle Dogs .1. Clinical and Clinicopathological Features. Veterinary Pathology, 14, 154-172.KITCHEN, D. N., CARLTON, W. W. & TUITE, J. 1977b. Ochratoxin-a and Citrinin Induced Nephrosis inBeagle Dogs .2. Pathology. Veterinary Pathology, 14, 261-272.KNOL, W., BOL, J. & HUIS IN'T VELD, J. H. J. 1990. Detoxification of aflatoxin B1 in feeds by Rhizopusoryzae in solid state. In: ZEUTHEN, P., CHEF TEL, J. C., ERIKSSON, C., GORMLEY, T. R., LINKO, P.& PAULUS, K. (eds.) Food Biotechnology. London: Elsevier Applied Science.KOLF-CLAUW, M., CASTELLOTE, J., JOLY, B., BOURGES-ABELLA, N., RAYMOND-LETRON, I., PINTON, P.& OSWALD, I. P. 2009. Development of a pig jejunal explant culture for studying thegastrointestinal toxicity of the mycotoxin deoxynivalenol: Histopathological analysis.Toxicology in Vitro, 23, 1580-1584.KOLLARCZIK, B., GAREIS, M. & HANELT, M. 1994. In vitro transformation of the Fusarium mycotoxinsdeoxynivalenol and zearalenone by the normal gut microflora of pigs. Natural Toxins, 2, 105-110.KOLOSSOVA, A., STROKA, J., BREIDBACH, A., KROEGER, K., AMBROSIO, M., BOUTEN, K. & ULBERTH, F.2009. Evaluation of the effect of mycotoxin binders in animal feed on the analyticalperformance of standardised methods for the determination of mycotoxins in feed. JCRscientific and technical reports, 6-12.KOUADIO, J. H., DANO, S. D., MOUKHA, S., MOBIO, T. A. & CREPPY, E. E. 2007. Effects ofcombinations of Fusarium mycotoxins on the inhibition of macromolecular synthesis,malondialdehyde levels, DNA methylation and fragmentation, and viability in Caco-2 cells.Toxicon, 49, 306-317.KUBENA, L. F., EDRINGTON, T. S., HARVEY, R. B., BUCKLEY, S. A., PHILLIPS, T. D., ROTTINGHAUS, G. E.& CASPER, H. H. 1997a. Individual and combined effects of fumonisin B1 present in Fusariummoniliforme culture material and T-2 toxin or deoxynivalenol in broiler chicks. Poult Sci, 76,1239-47.KUBENA, L. F., EDRINGTON, T. S., HARVEY, R. B., PHILLIPS, T. D., SARR, A. B. & ROTTINGHAUS, G. E.1997b. Individual and combined effects of fumonisin B-1 present in Fusarium moniliforme203

REFERENCES BIBLIOGRAPHIQUESculture material and diacetoxyscirpenol or ochratoxin A in turkey poults. Poultry Science, 76,256-264.KUBENA, L. F., EDRINGTON, T. S., KAMPS-HOLTZAPPLE, C., HARVEY, R. B., ELISSALDE, M. H. &ROTTINGHAUS, G. E. 1995a. Influence of fumonisin B1, present in Fusarium moniliformeculture material, and T-2 toxin on turkey poults. Poult Sci, 74, 306-13.KUBENA, L. F., EDRINGTON, T. S., KAMPSHOLTZAPPLE, C., HARVEY, R. B., ELISSALDE, M. H. &ROTTINGHAUS, G. E. 1995b. Effects of Feeding Fumonisin B-1 Present in Fusarium-Moniliforme Culture Material and Aflatoxin Singly and in Combination to Turkey Poults.Poultry Science, 74, 1295-1303.KUBENA, L. F., HARVEY, R. B., BAILEY, R. H., BUCKLEY, S. A. & ROTTINGHAUS, G. E. 1998. Effects of ahydrated sodium calcium aluminosilicate (T-Bind (TM)) on mycotoxicosis in young broilerchickens. Poultry Science, 77, 1502-1509.KUBENA, L. F., HARVEY, R. B., BUCKLEY, S. A., BAILEY, R. H. & ROTTINGHAUS, G. E. 1999. Effects oflong-term feeding of diets containing moniliformin, supplied by Fusarium fujikuroi culturematerial, and fumonisin, supplied by Fusarium moniliforme culture material, to laying hens.Poultry Science, 78, 1499-1505.KUBENA, L. F., HARVEY, R. B., BUCKLEY, S. A., EDRINGTON, T. S. & ROTTINGHAUS, G. E. 1997c.Individual and combined effects of moniliformin present in Fusarium fujikuroi culturematerial and aflatoxin in broiler chicks. Poultry Science, 76, 265-270.KUBENA, L. F., HARVEY, R. B., EDRINGTON, T. S. & ROTTINGHAUS, G. E. 1994a. Influence ofOchratoxin-a and Diacetoxyscirpenol Singly and in Combination on Broiler-Chickens. PoultryScience, 73, 408-415.KUBENA, L. F., HARVEY, R. B., HUFF, W. E., CORRIER, D. E., PHILLIPS, T. D. & ROTTINGHAUS, G. E.1989a. Influence of Ochratoxin-a and T-2 Toxin Singly and in Combination on Broiler-Chickens. Poultry Science, 68, 867-872.KUBENA, L. F., HARVEY, R. B., HUFF, W. E., CORRIER, D. E., PHILLIPS, T. D. & ROTTINGHAUS, G. E.1990. Efficacy of a Hydrated Sodium Calcium Aluminosilicate to Reduce the Toxicity ofAflatoxin and T-2 Toxin. Poultry Science, 69, 1078-1086.KUBENA, L. F., HARVEY, R. B., HUFF, W. E., ELISSALDE, M. H., YERSIN, A. G., PHILLIPS, T. D. &ROTTINGHAUS, G. E. 1993a. Efficacy of a Hydrated Sodium Calcium Aluminosilicate to Reducethe Toxicity of Aflatoxin and Diacetoxyscirpenol. Poultry Science, 72, 51-59.KUBENA, L. F., HARVEY, R. B., PHILLIPS, T. D. & CLEMENT, B. A. 1993b. Effect of Hydrated SodiumCalcium Aluminosilicates on Aflatoxicosis in Broiler Chicks. Poultry Science, 72, 651-657.KUBENA, L. F., HUFF, W. E., HARVEY, R. B., CORRIER, D. E., PHILLIPS, T. D. & CREGER, C. R. 1988.Influence of ochratoxin A and deoxynivalenol on growing broiler chicks. Poult Sci, 67, 253-60.KUBENA, L. F., HUFF, W. E., HARVEY, R. B., PHILLIPS, T. D. & ROTTINGHAUS, G. E. 1989b. Individualand combined toxicity of deoxynivalenol and T-2 toxin in broiler chicks. Poult Sci, 68, 622-6.KUBENA, L. F., HUFF, W. E., HARVEY, R. B., YERSIN, A. G., ELISSALDE, M. H., WITZEL, D. A., GIROIR, L.E., PHILLIPS, T. D. & PETERSEN, H. D. 1991. Effects of a Hydrated Sodium CalciumAluminosilicate on Growing Turkey Poults During Aflatoxicosis. Poultry Science, 70, 1823-1830.KUBENA, L. F., PHILLIPS, T. D., WITZEL, D. A. & HEIDELBAUGH, N. D. 1984. Toxicity of Ochratoxin-aand Penicillic Acid to Chicks. Bulletin of Environmental Contamination and Toxicology, 32,711-716.KUBENA, L. F., SMITH, E. E., GENTLES, A., HARVEY, R. B., EDRINGTON, T. S., PHILLIPS, T. D. &ROTTINGHAUS, G. E. 1994b. Individual and Combined Toxicity of T-2 Toxin and CyclopiazonicAcid in Broiler Chicks. Poultry Science, 73, 1390-1397.KUMAR, M., DWIVEDI, P., SHARMA, A., SINGH, N. & PATIL, R. 2007. Ochratoxin A and citrininnephrotoxicity in New Zealand White rabbits: an ultrastructural assessment.Mycopathologia, 163, 21-30.204

REFERENCES BIBLIOGRAPHIQUESKUMAR, M., DWIVEDI, P., SHARMA, A. K., TELANG, A. G., PATIL, R. D. & SINGH, N. D. 2008.Immunotoxicity of ochratoxin A and citrinin in New Zealand White rabbits. World RabbitScience, 16, 7-12.KUMAR, R. & BALACHANDRAN, C. 2005. Haematological and biochemical alterations in broilerchicken fed aflatoxin and cyclopiazonic acid. Indian Veterinary Journal, 82, 1255-1257.KUMAR, R. & BALACHANDRAN, C. 2009. Histopathological changes in broiler chickens fed aflatoxinand cyclopiazonic acid. Veterinarski Arhiv, 79, 31-40.KUTZ, R. E., SAMPSON, J. D., POMPEU, L. B., LEDOUX, D. R., SPAIN, J. N., VAZQUEZ-ANON, M. &ROTTINGHAUS, G. E. 2009. Efficacy of Solis, NovasilPlus, and MTB-100 to reduce aflatoxin M-1 levels in milk of early to mid lactation dairy cows fed aflatoxin B-1. Journal of Dairy Science,92, 3959-3963.LLALLES, J. P., LESSARD, M., OSWALD, I. P. & DAVID, J. C. 2010. Consumption of fumonisin B-1 for 9days induces stress proteins along the gastrointestinal tract of pigs. Toxicon, 55, 244-249.LE GALL, M., GALLOIS, M., SEVE, B., LOUVEAU, I., HOLST, J. J., OSWALD, I. P., LALLES, J. P. &GUILLOTEAU, P. 2009. Comparative effect of orally administered sodium butyrate before orafter weaning on growth and several indices of gastrointestinal biology of piglets. BritishJournal of Nutrition, 102, 1285-1296.LEDOUX, D. R., BROOMHEAD, J. N., BERMUDEZ, A. J. & ROTTINGHAUS, G. E. 2003. Individual andcombined effects of the fusarium Mycotoxins fumonisin B-1 and moniliformin in broilerchicks. Avian Diseases, 47, 1368-1375.LEDOUX, D. R., ROTTINGHAUS, G. E., BERMUDEZ, A. J. & ALONSO-DEBOLT, M. 1999. Efficacy of ahydrated sodium calcium aluminosilicate to ameliorate the toxic effects of aflatoxin in broilerchicks. Poultry Science, 78, 204-210.LEESON, S., DIAZ, G. & SUMMERS, J. D. 1995. Poultry metabolic disorders and mycotoxins. UniversityBooks, Guelph, Canada. p352.LEMKE, S. L., MAYURA, K., OTTINGER, S. E., MCKENZIE, K. S., WANG, N., FICKEY, C., KUBENA, L. F. &PHILLIPS, T. D. 1999. Assessment of the estrogenic effects of zearalenone after treatmentwith ozone utilizing the mouse uterine weight bioassay. Journal of Toxicology andEnvironmental Health-Part a-Current Issues, 56, 283-295.LESLIE, J. F., PLATTNER, R. D., DESJARDINS, A. E. & KLITTICH, C. J. R. 1992. Fumonisin B1 Production byStrains from Different Mating Populations of Gibberella-Fujikuroi (Fusarium Section Liseola).Phytopathology, 82, 341-345.LI, F. Q., YOSHIZAWA, T., KAWAMURA, O., LUO, X. Y. & LI, Y. W. 2001. Aflatoxins and fumonisins incorn from the high-incidence area for human hepatocellular carcinoma in Guangxi, China.Journal of Agricultural and Food Chemistry, 49, 4122-4126.LI, M. X., CUFF, C. F. & PESTKA, J. 2005. Modulation of murine host response to enteric reovirusinfection by the trichothecene deoxynivalenol. Toxicological Sciences, 87, 134-145.LI, M. X., HARKEMA, J. R., CUFF, C. F. & PESTKA, J. J. 2007. Deoxynivalenol exacerbates viralbronchopneumonia induced by respiratory reovirus infection. Toxicological Sciences, 95, 412-426.LI, Y. C., LEDOUX, D. R., BERMUDEZ, A. J., FRITSCHE, K. L. & ROTTINGHAUS, G. E. 2000. The individualand combined effects of fumonisin B-1 and moniliformin on performance and selectedimmune parameters in turkey poults. Poultry Science, 79, 871-878.LINDEMANN, M. D., BLODGETT, D. J., KORNEGAY, E. T. & SCHURIG, G. G. 1993. Potential Amelioratorsof Aflatoxicosis in Weanling Growing Swine. Journal of Animal Science, 71, 171-178.LIU, H. J., LU, Y., HAYNES, J. S., CUNNICK, J. E., MURPHY, P. & HENDRICH, S. 2001. Reaction offumonisin with glucose prevents promotion of hepatocarcinogenesis in female F344/N rats205

REFERENCES BIBLIOGRAPHIQUESwhile maintaining normal hepatic sphinganine/sphingosine ratios. Journal of Agricultural andFood Chemistry, 49, 4113-4121.LOISEAU, N., DEBRAUWER, L., SAMBOU, T., BOUHET, S., MILLER, J. D., MARTIN, P. G., VIADERE, J. L.,PINTON, P., PUEL, O., PINEAU, T., TULLIEZ, J., GALTIER, P. & OSWALD, I. P. 2007. Fumonisin B-1 exposure and its selective effect on porcine jejunal segment: Sphingolipids, glycolipids andtrans-epithelial passage disturbance. Biochemical Pharmacology, 74, 144-152.LOPEZ-GARCIA, R., PARK, D. L. & PHILLIPS, T. D. 1999. Integrated Mycotoxin Management Systems.Food, Nutrition and Agriculture, 23, 38-48.LU, Z., DANTZER, W. R., HOPMANS, E. C., PRISK, V., CUNNICK, J. E., MURPHY, P. A. & HENDRICH, S.1997. Reaction with fructose detoxifies fumonisin B-1 while stimulating liver-associatednatural killer cell activity in rats. Journal of Agricultural and Food Chemistry, 45, 803-809.MMADHESWARAN, R., BALACHANDRAN, C. & MANOHAR, B. M. 2004. Influence of dietary culturematerial containing aflatoxin and T-2 toxin on certain serum biochemical constituents inJapanese quail. Mycopathologia, 158, 337-341.MADHESWARAN, R., BALACHANDRAN, C. & MANOHAR, B. M. 2005. Effect of feeding aflatoxin and T-2 toxin on the growth rate and haematology of Japanese quail. Indian Veterinary Journal, 82,597-600.MADHESWARAN, R., BALACHANDRAN, C., MANOHAR, B. M., VAIRAMUTHU, S. & CHANDRAMOHAN,A. 2006. Ultrastructural changes in liver of Japanese quail fed aflatoxin and T-2 toxin. IndianVeterinary Journal, 83, 1125-1126.MADHYASTHA, M. S., FROHLICH, A. A. & MARQUARDT, R. R. 1992. Effect of Dietary Cholestyramineon the Elimination Pattern of Ochratoxin-a in Rats. Food and Chemical Toxicology, 30, 709-714.MADSEN, A., HALD, B. & MORTENSEN, H. P. 1983. Feeding Experiments with Ochratoxin-aContaminated Barley for Bacon Pigs .3. Detoxification by Ammoniation Heating + Naoh, orAutoclaving. Acta Agriculturae Scandinavica, 33, 171-175.MAKINO, H., TOGO, S., KUBOTA, T., MORIOKA, D., MORITA, T., KOBAYASHI, T., TANAKA, K., SHIMIZU,T., MATSUO, K., NAGASHIMA, Y. & SHIMADA, H. 2005. A good model of hepatic failure afterexcessive hepatectomy in mice. Journal of Surgical Research, 127, 171-176.MANNING, R. O., BROWN, T. P., WYATT, R. D. & FLETCHER, O. J. 1985. The Individual and CombinedEffects of Citrinin and Ochratoxin-a in Broiler Chicks. Avian Diseases, 29, 986-997.MARASAS, W. F. O. 2001. Discovery and occurrence of the fumonisins: A historical perspective.Environmental Health Perspectives, 109, 239-243.MARESCA, M. & FANTINI, J. 2010. Some food-associated mycotoxins as potential risk factors inhumans predisposed to chronic intestinal inflammatory diseases. Toxicon, 56, 282-294.MARIN, D. E., GOUZE, M. E., TARANU, I. & OSWALD, I. P. 2007. Fumonisin B1 alters cell cycleprogression and interleukin-2 synthesis in swine peripheral blood mononuclear cells.Molecular Nutrition & Food Research, 51, 1406-1412.MARIN, D. E., TARANU, I., PASCALE, F., LIONIDE, A., BURLACU, R., BAILLY, J. D. & OSWALD, I. P. 2006.Sex-related differences in the immune response of weanling piglets exposed to low doses offumonisin extract. British Journal of Nutrition, 95, 1185-1192.MARTINO, A., VOLPE, E., AURICCHIO, G., IZZI, V., POCCIA, F., MARIANI, F., COLIZZI, V. & BALDINI, P.M. 2007. Sphingosine 1-phosphate interferes on the differentiation of human monocytesinto competent dendritic cells. Scandinavian Journal of Immunology, 65, 84-91.MARTINOVA, E. A. & MERRILL, A. H. 1995. Fumonisin B-1 Alters Sphingolipid Metabolism andImmune Function in Balb/C Mice - Immunological Responses to Fumonisin B-1.Mycopathologia, 130, 163-170.206

REFERENCES BIBLIOGRAPHIQUESMARZOCCO, S., RUSSO, R., BIANCO, G., AUTORE, G. & SEVERINO, L. 2009. Pro-apoptotic effects ofnivalenol and deoxynivalenol trichothecenes in J774A.1 murine macrophages. ToxicologyLetters, 189, 21-26.MAYURA, K., ABDEL-WAHHAB, M. A., MCKENZIE, K. S., SARR, A. B., EDWARDS, J. F., NAGUIB, K. &PHILLIPS, T. D. 1998. Prevention of maternal and developmental toxicity in rats via dietaryinclusion of common aflatoxin sorbents: Potential for hidden risks. Toxicological Sciences, 41,175-182.MAYURA, K., PARKER, R., BERNDT, W. O. & PHILLIPS, T. D. 1984. Effect of Simultaneous PrenatalExposure to Ochratoxin-a and Citrinin in the Rat. Journal of Toxicology and EnvironmentalHealth, 13, 553-561.MCKENZIE, K. S., KUBENA, L. F., DENVIR, A. J., ROGERS, T. D., HITCHENS, G. D., BAILEY, R. H., HARVEY,R. B., BUCKLEY, S. A. & PHILLIPS, T. D. 1998. Aflatoxicosis in turkey poults is prevented bytreatment of naturally contaminated corn with ozone generated by electrolysis. PoultryScience, 77, 1094-1102.MCKENZIE, K. S., SARR, A. B., MAYURA, K., BAILEY, R. H., MILLER, D. R., ROGERS, T. D., NORRED, W. P.,VOSS, K. A., PLATTNER, R. D., KUBENA, L. F. & PHILLIPS, T. D. 1997. Oxidative degradation anddetoxification of mycotoxins using a novel source of ozone. Food and Chemical Toxicology,35, 807-820.MEISSONNIER, G. M., OSWALD, I. P. & GALTIER, P. 2005. Aflatoxicoses chez le porc - Étudebibliographique de données cliniques et expérimentales. Revue Médecine Vétérinaire, 156,591-605.MEISSONNIER, G. M., LAFFITTE, J., RAYMOND, I., BENOIT, E., COSSALTER, A. M., PINTON, P., BERTIN,G., OSWALD, I. P. & GALTIER, P. 2008a. Subclinical doses of T-2 toxin impair acquired immuneresponse and liver cytochrome P450 in pigs. Toxicology, 247, 46-54.MEISSONNIER, G. M., PINTON, P., LAFFITTE, J., COSSALTER, A. M., GONG, Y. Y., WILD, C. P., BERTIN,G., GALTIER, P. & OSWALD, I. P. 2008b. Immunotoxicity of aflatoxin B1: Impairment of thecell-mediated response to vaccine antigen and modulation of cytokine expression. Toxicologyand Applied Pharmacology, 231, 142-149.MEISSONNIER, G. M., RAYMOND, I., LAFFITTE, J., COSSALTER, A. M., PINTON, P., BENOIT, E., BERTIN,G., GALTIER, P. & OSWALD, I. P. 2009. Dietary glucomannan improves the vaccinal responsein pigs exposed to aflatoxin B-1 or T-2 toxin. World Mycotoxin Journal, 2, 161-172.MELI, R., FERRANTE, M. C., RASO, G. M., CAVALIERE, M., DI CARLO, R. & LUCISANO, A. 2000. Effect offumonisin B1 on inducible nitric oxide synthase and cyclooxygenase-2 in LPS-stimulatedJ774A.1 cells. Life Sciences, 67, 2845-2853.MENDEZ-ALBORES, A., ARAMBULA-VILLA, G., LOAREA-PINA, M. G. F., CASTANO-TOSTADO, E. &MORENO-MARTINEZ, E. 2005. Safety and efficacy evaluation of aqueous citric acid todegrade B-aflatoxins in maize. Food and Chemical Toxicology, 43, 233-238.MENDEZ-ALBORES, A., DEL RIO-GARCIA, J. C. & MORENO-MARTINEZ, E. 2007. Decontamination ofaflatoxin duckling feed with aqueous citric acid treatment. Animal Feed Science andTechnology, 135, 249-262.MENDEZ-ALBORES, J. A., VILLA, G. A., DEL RIO-GARCIA, J. & MARTINEZ, E. M. 2004. Aflatoxindetoxificationachieved with Mexican traditional nixtamalization process (MTNP) isreversible. Journal of the Science of Food and Agriculture, 84, 1611-1614.MERRILL, A. H., SCHMELZ, E. M., DILLEHAY, D. L., SPIEGEL, S., SHAYMAN, J. A., SCHROEDER, J. J.,RILEY, R. T., VOSS, K. A. & WANG, E. 1997. Sphingolipids - The enigmatic lipid class:Biochemistry, physiology, and pathophysiology. Toxicology and Applied Pharmacology, 142,208-225.MERRILL, A. H., SULLARDS, M. C., WANG, E., VOSS, K. A. & RILEY, R. T. 2001. Sphingolipid metabolism:Roles in signal transduction and disruption by fumonisins. Environmental Health Perspectives,109, 283-289.207

REFERENCES BIBLIOGRAPHIQUESMIAZZO, R., PERALTA, M. F., MAGNOLI, C., SALVANO, M., FERRERO, S., CHIACCHIERA, S. M.,CARVALHO, E. C. Q., ROSA, C. A. R. & DALCERO, A. 2005. Efficacy of sodium bentonite as adetoxifier of broiler feed contaminated with aflatoxin and fumonisin. Poultry Science, 84, 1-8.MILLER, E. R. & ULLREY, D. E. 1987. The Pig as a Model for Human-Nutrition. Annual Review ofNutrition, 7, 361-382.MOLL, W. D., HEINL, S., HARTINGER, D., GRIESSLER, K., KLINGENBRUNNER, V., VEKIRU, E., SCHWARTZ,H., KRSKA, R., GRABHERR, R. & SCHATZMAYR, G. Biodegradation of fumonisins. Proceedingsof the International Society for Mycotoxicology, 2009 Austria.MOLNAR, O., SCHATZMAYR, G., FUCHS, E. & PRILLINGER, H. 2004. Trichosporon mycotoxinivorans spnov., a new yeast species useful in biological detoxification of various mycotoxins. Systematicand Applied Microbiology, 27, 661-671.MONBALIU, S., VAN POUCKE, C., DETAVERNIER, C., DUMOULIN, F., VAN DE VELDE, M., SCHOETERS,E., VAN DYCK, S., AVERKIEVA, O., VAN PETEGHEM, C. & DE SAEGER, S. 2010. Occurrence ofMycotoxins in Feed as Analyzed by a Multi-Mycotoxin LC-MS/MS Method. Journal ofAgricultural and Food Chemistry, 58, 66-71.MOON, Y. & PESTKA, J. J. 2002. Vomitoxin-induced cyclooxygenase-2 gene expression inmacrophages mediated by activation of ERK and p38 but not JNK mitogen-activated proteinkinases. Toxicological Sciences, 69, 373-382.MOON, Y. & PESTKA, J. J. 2003. Cyclooxygenase-2 mediates interleukin-6 upregulation by vomitoxin(deoxynivalenol) in vitro and in vivo. Toxicology and Applied Pharmacology, 187, 80-88.MORRIS, C. M., LI, Y. C., LEDOUX, D. R., BERMUDEZ, A. J. & ROTTINGHAUS, G. E. 1999. The individualand combined effects of feeding moniliformin, supplied by Fusarium fujikuroi culturematerial, and deoxynivalenol in young turkey poults. Poultry Science, 78, 1110-1115.MORRISSEY, R. E., NORRED, W. P., HINTON, D. M., COLE, R. J. & DORNER, J. W. 1987. CombinedEffects of the Mycotoxins Aflatoxin-B1 and Cyclopiazonic Acid on Sprague-Dawley Rats. Foodand Chemical Toxicology, 25, 837-842.MURPHY, P. A., RICE, L. G. & ROSS, P. F. 1993. Fumonisin-B1, Fumonisin-B2, and Fumonisin-B3Content of Iowa, Wisconsin, and Illinois Corn and Corn Screenings. Journal of Agricultural andFood Chemistry, 41, 263-266.MUTUNGI, C., LAMUKA, P., ARIMI, S., GATHUMBI, J. & ONYANGO, C. 2008. The fate of aflatoxinsduring processing of maize into muthokoi - A traditional Kenyan food. Food Control, 19, 714-721.NNORRED, W. P. & MORRISSEY, R. E. 1983. Effects of Long-Term Feeding of Ammoniated, Aflatoxin-Contaminated Corn to Fischer 344 Rats. Toxicology and Applied Pharmacology, 70, 96-104.NORRED, W. P., VOSS, K. A., BACON, C. W. & RILEY, R. T. 1991. Effectiveness of Ammonia Treatmentin Detoxification of Fumonisin Contaminated Corn. Food and Chemical Toxicology, 29, 815-819.NORRED, W. P., WANG, E., YOO, H., RILEY, R. T. & MERRILL, A. H. 1992. Invitro Toxicology ofFumonisins and the Mechanistic Implications. Mycopathologia, 117, 73-78.OOBREMSKI, K., ZIELONKA, L., GAJECKA, M., JAKIMIUK, E., BAKULA, T., BARANOWSKI, M. & GAJECKI,M. 2008. Histological estimation of the small intestine wall after administration of feedcontaining deoxynivalenol, T-2 toxin and zearalenone in the pig. Polish Journal of VeterinarySciences, 11, 339-345.208

REFERENCES BIBLIOGRAPHIQUESODHAV, B., ADAM, J. K. & BHOOLA, K. D. 2008. Modulating effects of fumonisin B1 and ochratoxin Aon leukocytes and messenger cytokines of the human immune system. InternationalImmunopharmacology, 8, 799-809.OGIDO, R., OLIVEIRA, C. A. F., LEDOUX, D. R., ROTTINGHAUS, G. E., CORREA, B., BUTKERAITIS, P., REIS,T. A., GONCALES, E. & ALBUQUERQUE, R. 2004. Effects of prolonged administration ofaflatoxin B-1 and fumonisin B-1 in laying Japanese quail. Poultry Science, 83, 1953-1958.OGUZ, H., HADIMLI, H. H., KURTOGLU, V. & ERGANIS, O. 2003. Evaluation of humoral immunity ofbroilers during chronic aflatoxin (50 and 100 ppb) and clinoptilolite exposure. Revue DeMedecine Veterinaire, 154, 483-486.OGUZ, H. & KURTOGLU, V. 2000. Effect of clinoptilolite on performance of broiler chickens duringexperimental aflatoxicosis. British Poultry Science, 41, 512-517.OGUZ, H., KURTOGLU, V. & COSKUN, B. 2000. Preventive efficacy of clinoptilolite in broilers duringchronic aflatoxin (50 and 100 ppb) exposure. Research in Veterinary Science, 69, 197-201.ORSI, R. B., OLIVEIRA, C. A. F., DILKIN, P., XAVIER, J. G., DIREITO, G. M. & CORREA, B. 2007. Effects oforal administration of aflatoxin B-1 and fumonisin B-1 in rabbits (Oryctolagus cuniculus).Chemico-Biological Interactions, 170, 201-208.ORTATATLI, M. & OGUZ, H. 2001. Ameliorative effects of dietary clinoptilolite on pathologicalchanges in broiler chickens during aflatoxicosis. Research in Veterinary Science, 71, 59-66.ORTATATLI, M., OGUZ, H., HATIPOGLU, F. & KARAMAN, M. 2005. Evaluation of pathological changesin broilers during chronic aflatoxin (50 and 100 ppb) and clinoptilolite exposure. Research inVeterinary Science, 78, 61-68.OSWALD, I. P. & COMERA, C. 1998. Immunotoxicity of mycotoxins. Revue De Medecine Veterinaire,149, 585-590.OSWALD, I. P., DESAUTELS, C., LAFFITTE, J., FOURNOUT, S., PERES, S. Y., ODIN, M., LE BARS, P., LEBARS, J. & FAIRBROTHER, J. M. 2003. Mycotoxin fumonisin B-1 increases intestinalcolonization by pathogenic Escherichia coli in pigs. Applied and Environmental Microbiology,69, 5870-5874.OSWALD, I. P., DOZOIS, C. M., BARLAGNE, R., FOURNOUT, S., JOHANSEN, M. V. & BOGH, H. O. 2001.Cytokine mRNA expression in pigs infected with Schistosoma japonicum. Parasitology, 122,299-307.PPALENCIA, E., TORRES, O., HAGLER, W., MEREDITH, F. I., WILLIAMS, L. D. & RILEY, R. T. 2003. Totalfumonisins are reduced in tortillas using the traditional nixtamalization method of mayancommunities. Journal of Nutrition, 133, 3200-3203.PAPADAKIS, K. A. & TARGAN, S. R. 2000. Role of cytokines in the pathogenesis of inflammatory boweldisease. Annual Review of Medicine, 51, 289-298.PAPAIOANNOU, D. S., KYRIAKIS, S. C., PAPASTERIADIS, A., ROUMBIES, N., YANNAKOPOULOS, A. &ALEXOPOULOS, C. 2002. Effect of in-feed inclusion of a natural zeolite (clinoptilolite) oncertain vitamin, macro and trace element concentrations in the blood, liver and kidneytissues of sows. Research in Veterinary Science, 72, 61-68.PARK, D. L. 1993. Perspectives on Mycotoxin Decontamination Procedures. Food Additives andContaminants, 10, 49-60.PARK, D. L., LEE, L. S., PRICE, R. L. & POHLAND, A. E. 1988. Review of the Decontamination ofAflatoxins by Ammoniation - Current Status and Regulation. Journal of the Association ofOfficial Analytical Chemists, 71, 685-703.PARK, D. L., LOPEZGARCIA, R., TRUJILLO, S. & PRICE, R. L. 1995. Reduction of Risks Associated withFumonisin Contamination in Corn. Abstracts of Papers of the American Chemical Society, 209,132-AGFD.209

REFERENCES BIBLIOGRAPHIQUESPARK, D. L., RUA, S. M., MIROCHA, C. J., ABDALLA, E. & CONG, Y. W. 1992. Mutagenic Potentials ofFumonisin Contaminated Corn Following Ammonia Decontamination Procedure.Mycopathologia, 117, 105-108.PARK, J. W., SCOTT, P. M., LAU, B. P. Y. & LEWIS, D. A. 2004. Analysis of heat-processed corn foods forfumonisins and bound fumonisins. Food Additives and Contaminants Part a-ChemistryAnalysis Control Exposure & Risk Assessment, 21, 1168-1178.PARLAT, S. S., OZCAN, M. & OGUZ, H. 2001. Biological suppression of aflatoxicosis in Japanese quail(Coturnix coturnix japonica) by dietary addition of yeast (Saccharomyces cerevisiae).Research in Veterinary Science, 71, 207-211.PARLAT, S. S., YILDIZ, A. O. & OGUZ, H. 1999. Effect of clinoptilolite on performance of Japanese quail(Coturnix coturnix japonica) during experimental aflatoxicosis. British Poultry Science, 40,495-500.PASTEINER, S. 1994. Mycotoxins in animal husbandry. In: Biomin GTI Ges St. Pölten.PASTER, N., BARKAIGOLAN, R. & PADOVA, R. 1985. Effect of Gamma-Radiation on OchratoxinProduction by the Fungus Aspergillus-Ochraceus. Journal of the Science of Food andAgriculture, 36, 445-449.PATERSON, R. R. M. & LIMA, N. 2009. How will climate change affect mycotoxins in food? FoodResearch International, 43, 1902-1914.PATTERSON, D. S. P., SHREEVE, B. J., ROBERTS, B. A., BERRETT, S., BRUSH, P. J., GLANCY, E. M. &KROGH, P. 1981. Effect on Calves of Barley Naturally Contaminated with Ochratoxin-a andGroundnut Meal Contaminated with Low Concentrations of Aflatoxin-B-1. Research inVeterinary Science, 31, 213-218.PEARSON, T. C., WICKLOW, D. T. & PASIKATAN, M. C. 2004. Reduction of aflatoxin and fumonisincontamination in yellow corn by high-speed dual-wavelength sorting. Cereal Chemistry, 81,490-498.PEIRSON, S. N., BUTLER, J. N. & FOSTER, R. G. 2003. Experimental validation of novel andconventional approaches to quantitative real-time PCR data analysis. Nucleic Acids Research,31.PELTONEN, K., EL-NEZAMI, H., HASKARD, C., AHOKAS, J. & SALMINEN, S. 2001. Aflatoxin B-1 bindingby dairy strains of lactic acid bacteria and bifidobacteria. Journal of Dairy Science, 84, 2152-2156.PESTKA, J. J. 2003. Deoxynivalenol-induced IgA production and IgA nephropathy-aberrant mucosalimmune response with systemic repercussions. Toxicology Letters, 140, 287-295.PESTKA, J. J. 2010. Deoxynivalenol-Induced Proinflammatory Gene Expression: Mechanisms andPathological Sequelae. Toxins, 2, 1300-1317.PESTKA, J. J. & SMOLINSKI, A. T. 2005. Deoxynivalenol: Toxicology and potential effects on humans.Journal of Toxicology and Environmental Health-Part B-Critical Reviews, 8, 39-69.PESTKA, J. J., TAI, J. H., WITT, M. F., DIXON, D. E. & FORSELL, J. H. 1987. Suppression of Immune-Response in the B6c3f1 Mouse after Dietary Exposure to the Fusarium MycotoxinsDeoxynivalenol (Vomitoxin) and Zearalenone. Food and Chemical Toxicology, 25, 297-304.PESTKA, J. J. & ZHOU, H. R. 2000. Interleukin-6-deficient mice refractory to IgA dysregulation but notanorexia induction by vomitoxin (d eoxynivalenol) ingestion. Food and Chemical Toxicology,38, 565-575.PESTKA, J. J., ZHOU, H. R., MOON, Y. & CHUNG, Y. J. 2004. Cellular and molecular mechanisms forimmune modulation by deoxynivalenol and other trichothecenes: unraveling a paradox.Toxicology Letters, 153, 61-73.PETTERSSON, H. 2004. Controlling mycotoxins in animal feed. In: MAGEN, N. & OLSEN, M. (eds.)Mycotoxins in food, detection and control. Woodhead publishing Ltd and CRC press LLC.PFOHL-LESZKOWICZ, A. & MANDERVILLE, R. A. 2007. Ochratoxin A : an overview on toxicity andcarcinogenicity in animals and humans. Molecular Nutrition and Food Research, 51, 61-99.210

REFERENCES BIBLIOGRAPHIQUESPFOHL-LESZKOWICZ, A., MOLINIE, A., TOZLOVANU, M. & MANDERVILLE, R. A. 2008. Combined ToxicEffects of Ochratoxin A and Citrinin, In Vivo and In Vitro. In: SIANTAR, D. P., TRUCKSESS, M.W., SCOTT, P. M. & HERMAN, E. M. (eds.) Food Contaminants: Mycotoxins and FoodAllergens. Washington: Amer Chemical Soc.PHILLIPS, T. D. 1999. Dietary clay in the chemoprevention of aflatoxin-induced disease. ToxicologicalSciences, 52, 118-126.PHILLIPS, T. D., CLEMENT, B. A., KUBENA, L. F. & HARVEY, R. B. 1990. Detection and Detoxification ofAflatoxins - Prevention of Aflatoxicosis and Aflatoxin Residues with Hydrated Sodium CalciumAluminosilicate. Veterinary and Human Toxicology, 32, 15-19.PHILLIPS, T. D., CLEMENT, B. A. & PARK, D. L. 1994. Approaches to reduction of aflatoxin in foods andfeeds. In: EATON, L. D. & GROOPMAN, J. D. (eds.) The Toxicology of Aflatoxins: HumanHealth, Veterinary Agricultural Significance. Academic Press, New York.PHILLIPS, T. D., KUBENA, L. F., HARVEY, R. B., TAYLOR, D. R. & HEIDELBAUGH, N. D. 1988. HydratedSodium Calcium Aluminosilicate - a High-Affinity Sorbent for Aflatoxin. Poultry Science, 67,243-247.PHILLIPS, T. D., SARR, A. B., CLEMENT, B. A., KUBENA, L. F. & HARVEY, R. B. 1991. Prevention ofaflatoxicosis in farm animals via selective chemisorption of aflatoxin. In: BRAY, G. A. & RYAN,D. H. (eds.) Mycotoxins, Cancer and Health. Louisiana State University Press, Baton Rouge.PIER, A. C., BELDEN, E. L., ELLIS, J. A., NELSON, E. W. & MAKI, L. R. 1989. Effects of Cyclopiazonic Acidand Aflatoxin Singly and in Combination on Selected Clinical, Pathological and ImmunologicalResponses of Guinea-Pigs. Mycopathologia, 105, 135-142.PIERIDES, M., EL-NEZAMI, H., PELTONEN, K., SALMINEN, S. & AHOKAS, J. 2000. Ability of dairy strainsof lactic acid bacteria to bind aflatoxin M-1 in a food model. Journal of Food Protection, 63,645-650.PIETRI, A., BERTUZZI, T., PIVA, G., BINDER, E. M., SCHATZMAYR, G. & RODRIGUES, I. 2009. Aflatoxintransfer from naturally contaminated feed to milk of dairy cows and the efficiency of amycotoxin deactivating product. International Journal of Dairy Science, 4, 34-42.PINELLI, E., POUX, N., GARREN, L., PIPY, B., CASTEGNARO, M., MILLER, D. J. & PFOHL-LESZKOWICZ, A.1999. Activation of mitogen-activated protein kinase by fumonisin B-1 stimulates cPLA(2)phosphorylation, the arachidonic acid cascade and cAMP production. Carcinogenesis, 20,1683-1688.PINTON, P., ACCENSI, F., BEAUCHAMP, E., COSSALTER, A. M., CALLU, P., GROSJEAN, F. & OSWALD, I.P. 2008. Ingestion of deoxynivalenol (DON) contaminated feed alters the pig vaccinalimmune responses. Toxicol Lett, 177, 215-22.PINTON, P., BRAICU, C., NOUGAYREDE, J. P., LAFFITTE, J., TARANU, I. & OSWALD, I. P. 2010.Deoxynivalenol impairs porcine intestinal barrier function and decreases the proteinexpression of claudin-4 through a Mitogen Activated Protein Kinase dependent mechanism.Journal of Nutrition, In Press.PINTON, P., NOUGAYREDE, J. P., DEL RIO, J. C., MORENO, C., MARIN, D. E., FERRIER, L., BRACARENSE,A. P., KOLF-CLAUW, M. & OSWALD, I. P. 2009. The food contaminant deoxynivalenol,decreases intestinal barrier permeability and reduces claudin expression. Toxicology andApplied Pharmacology, 237, 41-48.PIVA, A., CASADEI, G., PAGLIUCA, G., CABASSI, E., GALVANO, F., SOLFRIZZO, M., RILEY, R. T. & DIAZ, D.E. 2005. Activated carbon does not prevent the toxicity of culture material containingfumonisin B-1 when fed to weanling piglets. Journal of Animal Science, 83, 1939-1947.PIVA, G., GALVANO, F., PIETRI, A. & PIVA, A. 1995. Detoxification Methods of Aflatoxins - a Review.Nutrition Research, 15, 767-776.PLACINTA, C. M., D'MELLO, J. P. F. & MACDONALD, A. M. C. 1999. A review of worldwidecontamination of cereal grains and animal feed with Fusarium mycotoxins. Animal FeedScience and Technology, 78, 21-37.211

REFERENCES BIBLIOGRAPHIQUESPLANK, G., BAUER, J., GRÜNKERNEIER, A., FISCHER, S., GEDEK, B. & BERNER, H. 1990.Untersucgungen zur protektiven wirkung von adsorbentien gegenüber ochratoxin A beimschwein. Tieräztl. Prax., 18, 483-489.PLANK, G., SCHUH, M. & BINDER, E. M. 2009. Untersuchungen über die detoxifizierende wirkung derFuttermittelzusätze Biomin BBSH 797 und Mycofix Plus 3.E in bezug auf DON imFerkelaufzuchtfutter. Wiener Tierärztliche Monatszeitschrift - VeterinärmedizinischeUniversität, 96, 55-71.POLITIS, I., FEGEROS, K., NITSCH, S., SCHATZMAYR, G. & KANTAS, D. 2005. Use of Trichosporonmycotoxinivorans to suppress the effects of ochratoxicosis on the immune system of broilerchicks. British Poultry Science, 46, 58-65.POZZI, C. R., CORREA, B., XAVIER, J. G., DIREITO, G. M., ORSI, R. B. & MATARAZZO, S. V. 2001. Effectsof prolonged oral administration of fumonisin B-1 and aflatoxin B-1 in rats. Mycopathologia,151, 21-27.PRICE, R. L. & JORGENSEN, K. V. 1985. Effects of Processing on Aflatoxin Levels and on MutagenicPotential of Tortillas Made from Naturally Contaminated Corn. Journal of Food Science, 50,347-&.PRUDENTE JR, A. D. & KING, J. M. 2002. Efficacy and safety evaluation of ozonation to degradeaflatoxin in corn. Journal of Food Science, 67, 2866-2872.PUJOL, R., TORRES, M., SANCHIS, V. & CANELA, R. 1999. Fate of fumonisin B-1 in corn kernel steepingwater containing SO2. Journal of Agricultural and Food Chemistry, 47, 276-278.QQURESHI, M. A. & HAGLER, W. M. 1992. Effect of Fumonisin-B1 Exposure on Chicken MacrophageFunctions Invitro. Poultry Science, 71, 104-112.RRAJMON, R., SEDMIKOVA, M., JILEK, F., KOUBKOVA, M., HARTLOVA, H., BARTA, I. & SMERAK, P. 2001.Combined effects of repeated low doses of aflatoxin B-1 and T-2 toxin on the Chinesehamster. Veterinarni Medicina, 46, 301-307.RAJU, M. & DEVEGOWDA, G. 2000. Influence of esterified-glucomannan on performance and organmorphology, serum biochemistry and haematology in broilers exposed to individual andcombined mycotoxicosis (aflatoxin, ochratoxin and T-2 toxin). British Poultry Science, 41, 640-650.RAJU, M. & DEVEGOWDA, G. 2002. Esterified-glucomannan in broiler chicken diets-contaminatedwith aflatoxin, ochratoxin and T-2 toxin: Evaluation of its binding ability (in vitro) and efficacyas immunomodulator. Asian-Australasian Journal of Animal Sciences, 15, 1051-1056.RAMIRO-PUIG, E., PEREZ-CANO, F. J., CASTELLOTE, C., FRANCH, A. & CASTELL, M. 2008. The bowel: Akey component of the immune system. Revista Espanola De Enfermedades Digestivas, 100,29-34.RAMOS, A. J. 1994. Prevencion de los efectos cancerigenos de las aflatoxinas y la zearalenonamediante el empleo de compuestos adsorbentes no nutritivos. Estudios in vitro. PhD Thesis.Polytechnic Univeristy of Valencia, Spain.RAMOS, A. J., FINKGREMMELS, J. & HERNANDEZ, E. 1996. Prevention of toxic effects of mycotoxinsby means of nonnutritive adsorbent compounds. Journal of Food Protection, 59, 631-641.RAMOS, A. J. & HERNANDEZ, E. 1996. In vitro aflatoxin adsorption by means of a montmorillonitesilicate. A study of adsorption isotherms. Animal Feed Science and Technology, 62, 263-269.REDDY, N. B., DEVEGOWDA, G. & GIRISH, C. K. 2003. Modified glucomannan: a promising solution tobind T-2 toxin in broilers. Feed Technology, 8, 23-24.212

REFERENCES BIBLIOGRAPHIQUESREFAI, M. K., AZIZ, N. H., ELFAR, F. & HASSAN, A. A. 1996. Detection of ochratoxin produced by A-ochraceus in feedstuffs and its control by gamma radiation. Applied Radiation and Isotopes,47, 617-621.RICHARD, J. L., THURSTON, J. R., DEYOE, B. L. & BOOTH, G. D. 1975. Effect of Ochratoxin and Aflatoxinon Serum-Proteins, Complement Activity, and Antibody-Production to Brucella-Abortus inGuinea-Pigs. Applied Microbiology, 29, 27-29.RICHARD, J. L., THURSTON, J. R., LILLEHOJ, E. B., CYSEWSKI, S. J. & BOOTH, G. D. 1978. ComplementActivity, Serum-Protein, and Hepatic Changes in Guinea-Pigs Given Sterigmatocystin orAflatoxin, Alone or in Combination. American Journal of Veterinary Research, 39, 163-166.RILEY, R. T. 1998. Mechanistic interactions of mycotoxins: Theoretical considerations. In: SINHA, K. K.& BHATNAGAR, D. (eds.) Mycotoxins in Agriculture and Food Safety. New York: MarcelDekker.RILEY, R. T., AN, N. H., SHOWKER, J. L., YOO, H. S., NORRED, W. P., CHAMBERLAIN, W. J., WANG, E.,MERRILL, A. H., MOTELIN, G., BEASLEY, V. R. & HASCHEK, W. M. 1993. Alteration of Tissueand Serum Sphinganine to Sphingosine Ratio - an Early Biomarker of Exposure to Fumonisin-Containing Feeds in Pigs. Toxicology and Applied Pharmacology, 118, 105-112.RILEY, R. T., ENONGENE, E., VOSS, K. A., NORRED, W. P., MEREDITH, F. I., SHARMA, R. P.,SPITSBERGEN, J., WILLIAMS, D. E., CARLSON, D. B. & MERRILL, A. H. 2001. Sphingolipidperturbations as mechanisms for fumonisin carcinogenesis. Environmental HealthPerspectives, 109, 301-308.RILEY, R. T., HINTON, D. M., CHAMBERLAIN, W. J., BACON, C. W., WANG, E., MERRILL, A. H. & VOSS,K. A. 1994a. Dietary Fumonisin B-1 Induces Disruption of Sphingolipid Metabolism inSprague-Dawley Rats - a New Mechanism of Nephrotoxicity. Journal of Nutrition, 124, 594-603.RILEY, R. T., WANG, E. & MERRILL, A. H. 1994b. Liquid-Chromatographic Determination ofSphinganine and Sphingosine - Use of the Free Sphinganine-to-Sphingosine Ratio as aBiomarker for Consumption of Fumonisins. Journal of Aoac International, 77, 533-540.ROCHA, O., ANSARI, K. & DOOHAN, F. M. 2005. Effects of trichothecene mycotoxins on eukaryoticcells: A review. Food Additives and Contaminants, 22, 369-378.RODRIGUES, I. & GRIESSLER, K. 2010. BIOMIN's Mycotoxin Survey - Annual Report 2009.ROSA, C. A. R., MIAZZO, R., MAGNOLI, C., SALVANO, M., CHIACCHIERA, S. M., FERRERO, S., SAENZ,M., CARVALHO, E. C. Q. & DALCERO, A. 2001. Evaluation of the efficacy of bentonite from thesouth of Argentina to ameliorate the toxic effects of aflatoxin in broilers. Poultry Science, 80,139-144.ROSCOE, V., LOMBAERT, G. A., HUZEL, V., NEUMANN, G., MELIETIO, J., KITCHEN, D., KOTELLO, S.,KRAKALOVICH, T., TRELKA, R. & SCOTT, P. M. 2008. Mycotoxins in breakfast cereals from theCanadian retail market: A 3-year survey. Food Additives and Contaminants Part a-ChemistryAnalysis Control Exposure & Risk Assessment, 25, 347-355.ROTH, A., CHAKOR, K., CREPPY, E. E., KANE, A., ROSCHENTHALER, R. & DIRHEIMER, G. 1988. Evidencefor an Enterohepatic Circulation of Ochratoxin-a in Mice. Toxicology, 48, 293-308.ROTHKOTTER, H. J., SOWA, E. & PABST, R. 2002. The pig as a model of developmental immunology.Human & Experimental Toxicology, 21, 533-536.ROTTER, B. A., PRELUSKY, D. B. & PESTKA, J. J. 1996a. Toxicology of deoxynivalenol (vomitoxin).Journal of Toxicology and Environmental Health, 48, 1-34.ROTTER, B. A., THOMPSON, B. K. & LESSARD, M. 1995a. Effects of Deoxynivalenol-Contaminated Dieton Performance and Blood Parameters in Growing Swine. Canadian Journal of AnimalScience, 75, 297-302.ROTTER, B. A., THOMPSON, B. K., PRELUSKY, D. B., TRENHOLM, H. L., STEWART, B., MILLER, J. D. &SAVARD, M. E. 1996b. Response of growing swine to dietary exposure to pure fumonisin B1during an eight-week period: Growth and clinical parameters. Natural Toxins, 4, 42-50.213

REFERENCES BIBLIOGRAPHIQUESROTTER, R. G., MARQUARDT, R. R., FROHLICH, A. A. & ABRAMSON, D. 1990. Ensiling as a Means ofReducing Ochratoxin-a Concentrations in Contaminated Barley. Journal of the Science of Foodand Agriculture, 50, 155-166.ROTTER, R. G., ROTTER, B. A., THOMPSON, B. K., PRELUSKY, D. B. & TRENHOLM, H. L. 1995b.Effectiveness of Density Segregation and Sodium-Carbonate Treatment on the Detoxificationof Fusarium-Contaminated Corn Fed to Growing Pigs. Journal of the Science of Food andAgriculture, 68, 331-336.RYU, D., HANNA, M. A. & BULLERMAN, L. B. 1999. Stability of zearalenone during extrusion of corngrits. Journal of Food Protection, 62, 1482-1484.SSAKHARE, P. S., HARNE, S. D., KALOREY, D. R., WARKE, S. R., BHANDARKAR, A. G. & KURKURE, N. V.2007. Effect of Toxiroak® polyherbal feed supplement during induced aflatoxicosis,ochratoxicosis and combined mycotoxicoses in broilers. Veterinarski Arhiv, 2, 129-146.SANTIN, E. 2005. Mould growth and mycotoxin production. In: DIAZ, D. E. (ed.) The MycotoxinBluebook. Nottingham: Nottingham University Press.SANTIN, E., MAIORKA, A., KRABBE, E. L., PAULILLO, A. C. & ALESSI, A. C. 2002a. Effect of hydratedsodium calcium aluminosilicate on the prevention of the toxic effects of ochratoxin. Journalof Applied Poultry Research, 11, 22-28.SANTIN, E., PAULILLO, A. C., MAIORKA, P. C., ALESSI, A. C., KRABBE, E. L. & MAIORKA, A. 2002b. Theeffects of ochratoxin/aluminosilicate interaction on the tissues and humoral immuneresponse of broilers. Avian Pathology, 31, 73-79.SCHAFFNER, W. & WEISSMANN, C. 1973. A rapid, sensitive, and specific method for thedetermination of protein in dilute solution. Analytical Biochemistry, 56, 502-514.SCHATZMAYR, G., HEIDLER, D., FUCHS, E., NITSCH, S., MOHNL, M., TÄUBEL, M., LOIBNER, A., BRAUN,R. & BINDER, E. 2003. Investigation of different yeast strains for the detoxification ofochratoxin A. Mycotoxin Research, 19, 124-128.SCHATZMAYR, G., ZEHNER, F., TAUBEL, M., SCHATZMAYR, D., KLIMITSCH, A., LOIBNER, A. P. &BINDER, E. M. 2006. Microbiologicals for deactivating mycotoxins. Molecular Nutrition &Food Research, 50, 543-551.SCHEIDELER, S. E. 1993. Effects of Various Types of Aluminosilicates and Aflatoxin-B1 on AflatoxinToxicity, Chick Performance, and Mineral Status. Poultry Science, 72, 282-288.SCHELL, T. C., LINDEMANN, M. D., KORNEGAY, E. T. & BLODGETT, D. J. 1993a. Effects of FeedingAflatoxin-Contaminated Diets with and without Clay to Weanling and Growing Pigs onPerformance, Liver-Function, and Mineral Metabolism. Journal of Animal Science, 71, 1209-1218.SCHELL, T. C., LINDEMANN, M. D., KORNEGAY, E. T., BLODGETT, D. J. & DOERR, J. A. 1993b.Effectiveness of Different Types of Clay for Reducing the Detrimental Effects of Aflatoxin-Contaminated Diets on Performance and Serum Profiles of Weanling Pigs. Journal of AnimalScience, 71, 1226-1231.SCHMELZ, E. M., DOMBRINK-KURTZMAN, M. A., ROBERTS, P. C., KOZUTSUMI, Y., KAWASAKI, T. &MERRILL, A. H. 1998. Induction of apoptosis by fumonisin B-1 in HT29 cells is mediated by theaccumulation of endogenous free sphingoid bases. Toxicology and Applied Pharmacology,148, 252-260.SCHOTHORST, R. C. & VAN EGMOND, H. P. 2004. Report from SCOOP task 3.2.10 "collection ofoccurrence data of Fusarium toxins in food and assessment of dietary intake by thepopulation of EU member states" - Subtask: trichothecenes. Toxicology Letters, 153, 133-143.214

REFERENCES BIBLIOGRAPHIQUESSCHWARTZ, H. E., MOLL, W. D., SCHATZMAYR, G. & KRSKA, R. An accelerated method fordetermining sphingoid bases in serum of pigs. 31st Mycotoxin Workshop, 2009 Münster(Germany). 49.SCOTT, P. M. 1993. Fumonisins. International Journal of Food Microbiology, 18, 257-270.SCOTT, P. M. 1998. Industrial and farm detoxification processes for mycotoxins. Revue De MedecineVeterinaire, 149, 543-548.SCOTT, P. M. & LAWRENCE, G. A. 1994. Stability and Problems in Recovery of Fumonisins Added toCorn-Based Foods. Journal of Aoac International, 77, 541-545.SCUDAMORE, K. A. 2005. Prevention of ochratoxin A in commodities and likely effects of processingfractionation and animal feeds. Food Additives and Contaminants, 22, 17-25.SCUDAMORE, K. A. & PATEL, S. 2009. Occurrence of Fusarium mycotoxins in maize imported into theUK, 2004-2007. Food Additives and Contaminants Part a-Chemistry Analysis Control Exposure& Risk Assessment, 26, 363-371.SEIFERLEIN, M., HUMPF, H. U., VOSS, K. A., SULLARDS, M. C., ALLEGOOD, J. C., WANG, E. & MERRILL,A. H. 2007. Hydrolyzed fumonisins HFB1 and HFB2 are acylated in vitro and in vivo byceramide synthase to form cytotoxic N-acyl-metabolites. Molecular Nutrition & FoodResearch, 51, 1120-1130.SHARMA, D., ASRANI, R. K., LEDOUX, D. R., JINDAL, N., ROTTINGHAUS, G. E. & GUPTA, V. K. 2008.Individual and combined effects of fumonisin B-1 and moniliformin on clinicopathologicaland cell-mediated immune response in Japanese quail. Poultry Science, 87, 1039-1051.SHEPHARD, G. S. 2008. Determination of mycotoxins in human foods. Chemical Society Reviews, 37,2468-2477.SHEPHARD, G. S., THIEL, P. G., SYDENHAM, E. W., VLEGGAAR, R. & ALBERTS, J. F. 1994. Determinationof the Mycotoxin Fumonisin B-1 and Identification of Its Partially Hydrolyzed Metabolites inthe Feces of Nonhuman-Primates. Food and Chemical Toxicology, 32, 23-29.SHEPHERD, E. C., PHILLIPS, T. D., JOINER, G. N., KUBENA, L. F. & HEIDELBAUGH, N. D. 1981.Ochratoxin-a and Penicillic Acid Interaction in Mice. Journal of Environmental Science andHealth Part B-Pesticides Food Contaminants and Agricultural Wastes, 16, 557-573.SHETTY, P. H. & BHAT, R. V. 1999. A physical method for segregation of fumonisin-contaminatedmaize. Food Chemistry, 66, 371-374.SHI, Y. H., XU, Z. R., FENG, J. L. & WANG, C. Z. 2006. Efficacy of modified montmorillonitenanocomposite to reduce the toxicity of aflatoxin in broiler chicks. Animal Feed Science andTechnology, 129, 138-148.SHIFRIN, V. I. & ANDERSON, P. 1999. Trichothecene mycotoxins trigger a ribotoxic stress responsethat activates c-Jun N-terminal kinase and p38 mitogen-activated protein kinase and inducesapoptosis. Journal of Biological Chemistry, 274, 13985-13992.SHIMA, J., TAKASE, S., TAKAHASHI, Y., IWAI, Y., FUJIMOTO, H., YAMAZAKI, M. & OCHI, K. 1997. Noveldetoxification of the trichothecene mycotoxin deoxynivalenol by a soil bacterium isolated byenrichment culture. Applied and Environmental Microbiology, 63, 3825-3830.SIRAJ, M. Y., PHILLIPS, T. D. & HAYES, A. W. 1981. Effects of the Myco-Toxins Citrinin and Ochratoxinaon Hepatic Mixed-Function Oxidase and Adenosine-Triphosphatase in Neonatal Rats.Journal of Toxicology and Environmental Health, 8, 131-140.SIVAKUMAR, G., DWIVEDI, P., SHARMA, A. K., KUMAR, M. & NIMALESAN, S. 2009. Fumonisin B1 andochratoxin A induced biochemical changes in young male New Zealand White rabbits. IndianJournal of Veterinary Pathology, 33.SIWELA, A. H., SIWELA, M., MATINDI, G., DUBE, S. & NZIRAMASANGA, N. 2005. Decontamination ofaflatoxin-contaminated maize by dehulling. Journal of the Science of Food and Agriculture,85, 2535-2538.215

REFERENCES BIBLIOGRAPHIQUESSMITH, E. E., KUBENA, L. F., BRAITHWAITE, C. E., HARVEY, R. B., PHILLIPS, T. D. & REINE, A. H. 1992.Toxicological Evaluation of Aflatoxin and Cyclopiazonic Acid in Broiler-Chickens. PoultryScience, 71, 1136-1144.SMITH, T. K. 1980. Influence of Dietary Fiber, Protein and Zeolite on Zearalenone Toxicosis in Ratsand Swine. Journal of Animal Science, 50, 278-285.SMITH, T. K., MCMILLAN, E. G. & CASTILLO, J. B. 1997. Effect of feeding blends of Fusariummycotoxin-contaminated grains containing deoxynivalenol and fusaric acid on growth andfeed consumption of immature swine. Journal of Animal Science, 75, 2184-2191.SMITH, W. L., FITZPATRICK, F. A. & DENNIS, E. V. A. J. E. V. 1996. Chapter 11 The eicosanoids:cyclooxygenase, lipoxygenase, and epoxygenase pathways. New ComprehensiveBiochemistry. Elsevier.SODERHOLM, J. D., OLAISON, G., PETERSON, K. H., FRANZEN, L. E., LINDMARK, T., WIREN, M.,TAGESSON, C. & SJODAHL, R. 2002. Augmented increase in tight junction permeability byluminal stimuli in the non-inflamed ileum of Crohn's disease. Gut, 50, 307-313.SOLFRIZZO, M., VISCONTI, A., AVANTAGGIATO, G., TORRES, A. & CHULZE, S. 2001. In vitro and in vivostudies to assess the effectiveness of cholestyramine as a binding agent for fumonisins.Mycopathologia, 151, 147-153.SORIANO, J. M., GONZALEZ, L. & CATALA, A. I. 2005. Mechanism of action of sphingolipids and theirmetabolites in the toxicity of fumonisin B1. Progress in Lipid Research, 44, 345-356.SOVA, Z., POHUNKOVA, H., REISNEROVA, H., SLAMOVA, A. & HAISL, K. 1991. Hematological andHistological Response to the Diet Containing Aflatoxin B-1 and Zeolite in Broilers of Domestic-Fowl. Acta Veterinaria Brno, 60, 31-40.SPEIJERS, G. J. A. & SPEIJERS, M. H. M. 2004. Combined toxic effects of mycotoxins. ToxicologyLetters, 153, 91-98.STADNYK, A. W. 2002. Intestinal epithelial cells as a source of inflammatory cytokines andchemokines. Canadian Journal of Gastroenterology, 16, 241-246.STANLEY, V. G., OJO, R., WOLDESENBET, S., HUTCHINSON, D. H. & KUBENA, L. F. 1993. The Use ofSaccharomyces-Cerevisiae to Suppress the Effects of Aflatoxicosis in Broiler Chicks. PoultryScience, 72, 1867-1872.STANLEY, V. G., WINSMAN, M., DUNKLEY, C., OGUNLEYE, T., DALEY, M., KRUEGER, W. F., SEFTON, A.E. & HINTON, A. 2004. The impact of yeast culture residue on the suppression of dietaryaflatoxin on the performance of broiler breeder hens. Journal of Applied Poultry Research,13, 533-539.STOEV, S. D., GOUNDASHEVA, D., MIRTCHEVA, T. & MANTLE, P. G. 2000. Susceptibility to secondarybacterial infections in growing pigs as an early response in ochratoxicosis. Experimental andToxicologic Pathology, 52, 287-296.STROUD, J. S., ENGLISH, E., DAVIDSON, S., HOPKINS, B. A., LATIMER, G., HAGLER, W. M., BROWNIE, C.& WHITLOW, L. W. 2006. Dietary additives to reduce aflatoxin transmission to milk of dairycows. Journal of Dairy Science, 89, 129.ŠTYRIAK, I., CONKOVÁ, E., KMEC, V., BÖHM, J. & RAZZAZI, E. 2001. The use of yeast for microbialdegradation of some selected mycotoxins. Mycotoxin Research, 17, 24-27.SULYOK, M., KRSKA, R. & SCHUHMACHER, R. 2007. A liquid chromatography/tandem massspectrometric multi-mycotoxin method for the quantification of 87 analytes and itsapplication to semi-quantitative screening of moldy food samples. Analytical andBioanalytical Chemistry, 389, 1505-1523.SUNDSTØL ERIKSEN, G., PETTERSSON, H. & LUNDH, T. 2004. Comparative cytotoxicity ofdeoxynivalenol, nivalenol, their acetylated derivatives and de-epoxy metabolites. Food andChemical Toxicology, 42, 619-624.216

REFERENCES BIBLIOGRAPHIQUESSWANSON, S. P., HELASZEK, C., BUCK, W. B., ROOD, H. D. & HASCHEK, W. M. 1988. The Role ofIntestinal Microflora in the Metabolism of Trichothecene Mycotoxins. Food and ChemicalToxicology, 26, 823-829.SWANSON, S. P., ROOD, H. D., BEHRENS, J. C. & SANDERS, P. E. 1987. Preparation andCharacterization of the Deepoxy Trichothecenes - Deepoxy Ht-2, Deepoxy T-2 Triol, DeepoxyT-2 Tetraol, Deepoxy 15-Monoacetoxyscirpenol, and Deepoxy Scirpentriol. Applied andEnvironmental Microbiology, 53, 2821-2826.SYDENHAM, E. W., STOCKENSTROM, S., THIEL, P. G., SHEPHARD, G. S., KOCH, K. R. & MARASAS, W. F.O. 1995. Potential of Alkaline-Hydrolysis for the Removal of Fumonisins from ContaminatedCorn. Journal of Agricultural and Food Chemistry, 43, 1198-1201.SYDENHAM, E. W., VANDERWESTHUIZEN, L., STOCKENSTROM, S., SHEPHARD, G. S. & THIEL, P. G.1994. Fumonisin-Contaminated Maize - Physical Treatment for the Partial Decontaminationof Bulk Shipments. Food Additives and Contaminants, 11, 25-32.TTABUC, C., MARIN, D., GUERRE, P., SESAN, T. & BAILLY, J. D. 2009. Molds and Mycotoxin Content ofCereals in Southeastern Romania. Journal of Food Protection, 72, 662-665.TAMIMI, S. O., NATOUR, R. M. & HALABI, K. S. 1997. Individual and combined effects of chronic T-2toxin and aflatoxin B1 mycotoxins on rat lives and kidney. Arab Gulf Journal of ScientificResearch, 15, 717-732.TAPIA, M. O. & SEAWRIGHT, A. A. 1985. Experimental combined aflatoxin B1 and ochratoxin Aintoxication in pigs. Australian Veterinary Journal, 62, 33-37.TARANU, I., MARIN, D. E., BOUHET, S., PASCALE, F., BAILLY, J. D., MILLER, J. D., PINTON, P. &OSWALD, I. P. 2005. Mycotoxin fumonisin B-1 alters the cytokine profile and decreases thevaccinal antibody titer in pigs. Toxicological Sciences, 84, 301-307.TESSARI, E. N. C., KOBASHIGAWA, E., CARDOSO, A., LEDOUX, D. R., ROTTINGHAUS, G. E. & OLIVEIRA,C. A. F. 2010. Effect of aflatoxin B1 and Fumonisin B1 on blood biochemical parameteres inbroilers. Toxins, 2, 453-460.TESSARI, E. N. C., OLIVEIRA, C. A. F., CARDOSO, A., LEDOUX, D. R. & ROTTINGHAUS, G. E. 2006. Effectsof aflatoxin B-1 and fumonisin B-1 on body weight, antibody titres and histology of broilerchicks. British Poultry Science, 47, 357-364.THEUMER, M. G., CANEPA, M. C., LOPEZ, A. G., MARY, V. S., DAMBOLENA, J. S. & RUBINSTEIN, H. R.2010. Subchronic mycotoxicoses in Wistar rats: Assessment of the in vivo and in vitrogenotoxicity induced by fumonisins and aflatoxin B-1, and oxidative stress biomarkers status.Toxicology, 268, 104-110.THEUMER, M. G., LOPEZ, A. G., AOKI, M. P., CANEPA, M. C. & RUBINSTEIN, H. R. 2008. Subehronicmycotoxicoses in rats. Histopathological changes and modulation of the sphinganine tosphingosine (Sa/So) ratio imbalance induced by Fusarium verticillioides culture material, dueto the coexistence of aflatoxin B1 in the diet. Food and Chemical Toxicology, 46, 967-977.THEUMER, M. G., LOPEZ, A. G., MASIH, D. T., CHULZE, S. N. & RUBINSTEIN, H. R. 2002.Immunobiological effects of fumonisin B1 in experimental subchronic mycotoxicoses in rats.Clinical and Diagnostic Laboratory Immunology, 9, 149-155.THURSTON, J. R., SACKS, J. M., RICHARD, J. L., PEDEN, W. M. & DRIFTMIER, K. 1989. Complement,Bacteriostatic, and Enzymatic-Activities in Sera from Guinea-Pigs Given Aflatoxin and orRubratoxin. American Journal of Veterinary Research, 50, 356-358.TIEMANN, U., BRUSSOW, K. P., KUCHENMEISTER, U., JONAS, L., KOHLSCHEIN, P., POHLAND, R. &DANICKE, S. 2006. Influence of diets with cereal grains contaminated by graded levels of twoFusarium toxins on selected enzymatic and histological parameters of liver in gilts. Food andChemical Toxicology, 44, 1228-1235.217

REFERENCES BIBLIOGRAPHIQUESTIRADO, M. C., CLARKE, R., JAYKUS, L. A., MCQUATTERS-GOLLOP, A. & FRANK, J. M. 2010. Climatechange and food safety: A review. Food Research International, 43, 1745-1765.TOMASEVIC-CANOVIC, M., DAKOVIC, A., ROTTINGHAUS, G., MATIJASEVIC, S. & DURICIC, M. 2003.Surfactant modified zeolites - new efficient adsorbents for mycotoxins. Microporous andMesoporous Materials, 61, 173-180.TRAN, S. T., TARDIEU, D., AUVERGNE, A., BAILLY, J. D., BABILÉ, R., DURAND, S., BENARD, G. &GUERRE, P. 2006. Serum sphinganine and the sphinganine to sphingosine ratio as abiomarker of dietary fumonisins during chronic exposure in ducks. Chemico-BiologicalInteractions, 160, 41-50.TRENHOLM, H. L., CHARMLEY, L. L., PRELUSKY, D. B. & WARNER, R. M. 1991. 2 Physical Methods forthe Decontamination of 4 Cereals Contaminated with Deoxynivalenol and Zearalenone.Journal of Agricultural and Food Chemistry, 39, 356-360.TRENHOLM, H. L., CHARMLEY, L. L., PRELUSKY, D. B. & WARNER, R. M. 1992. Washing ProceduresUsing Water or Sodium-Carbonate Solutions for the Decontamination of 3 CerealsContaminated with Deoxynivalenol and Zearalenone. Journal of Agricultural and FoodChemistry, 40, 2147-2151.TRINCHIERI, G. 2003. Interleukin-12 and the regulation of innate resistance and adaptive immunity.Nature Reviews Immunology, 3, 133-146.UUENO, Y., NAKAYAMA, K., ISHII, K., TASHIRO, F., MINODA, Y., OMORI, T. & KOMAGATA, K. 1983.Metabolism of T-2-Toxin in Curtobacterium Sp Strain-114-2. Applied and EnvironmentalMicrobiology, 46, 120-127.UNDERHILL, K. L., ROTTER, B. A., THOMPSON, B. K., PRELUSKY, D. B. & TRENHOLM, H. L. 1995.Effectiveness of Cholestyramine in the Detoxification of Zearalenone as Determined in Mice.Bulletin of Environmental Contamination and Toxicology, 54, 128-134.USFDA 2001. United State Food and Drug Administration, Center for Veterinary MedicineBackground Paper in Support of Fumonisin Levels in Animal Feed: Executive summary of thisscientific support document. Guidance for industry: Fumonisin levels in human foods andanimal feeds.VVAN CRUCHTEN, S. & VAN DEN BROECK, W. 2002. Morphological and Biochemical Aspects ofApoptosis, Oncosis and Necrosis. Anatomia, Histologia, Embryologia, 31, 214-223.VAN DE WALLE, J., SERGENT, T., PIRONT, N., TOUSSAINT, O., SCHNEIDER, Y. J. & LARONDELLE, Y.2010. Deoxynivalenol affects in vitro intestinal epithelial cell barrier integrity throughinhibition of protein synthesis. Toxicology and Applied Pharmacology, 245, 291-298.VAN DER WESTHUIZEN, L., SHEPHARD, G. S., SNYMAN, S. D., ABEL, S., SWANEVELDER, S. &GELDERBLOM, W. C. A. 1998. Inhibition of sphingolipid biosynthesis in rat primaryhepatocyte cultures by Fumonisin, B-1 and other structurally related compounds. Food andChemical Toxicology, 36, 497-503.VAN RENSBURG, C. J., VAN RENSBURG, C. E. J., VAN RYSSEN, J. B. J., CASEY, N. H. & ROTTINGHAUS, G.E. 2006. In vitro and in vivo assessment of humic acid as an aflatoxin binder in broilerchickens. Poultry Science, 85, 1576-1583.VARGA, I. & VANYI, A. 1992. Interaction of T-2-Fusariotoxin with Anticoccidial Efficacy of Lasalocid inChickens. International Journal for Parasitology, 22, 523-525.218

REFERENCES BIBLIOGRAPHIQUESVEKIRU, E., FRUHAUF, S., SAHIN, M., OTTNER, F., SCHATZMAYR, G. & KRSKA, R. 2007. Investigation ofvarious adsorbents for their ability to bind aflatoxin B1</sub&gt. MycotoxinResearch, 23, 27-33.VELDMAN, A. 1992. Effect of sorbentia on carry-over of aflatoxin from cow feed to milk.Milchwissenschaft, 47, 777-780.VENTURINI, M. C., QUIROGA, M. A., RISSO, M. A., DILORENZO, C., OMATA, Y., VENTURINI, L. &GODOY, H. 1996. Mycotoxin T-2 and aflatoxin B-1 as immunosuppressors in mice chronicallyinfected with Toxoplasma gondii. Journal of Comparative Pathology, 115, 229-237.VERMA, J., JOHRI, T. S. & SWAIN, B. K. 2003. Effect of varying levels of aflatoxin, ochratoxin and theircombinations on the performance and egg quality characteristics in laying hens. Asian-Australasian Journal of Animal Sciences, 16, 1015-1019.VERMA, J., JOHRI, T. S. & SWAIN, B. K. 2007. Effect of aflatoxin, ochratoxin and their combination onprotein and energy utilisation in white leghorn laying hens. Journal of the Science of Food andAgriculture, 87, 760-764.VERMA, J., JOHRI, T. S., SWAIN, B. K. & AMEENA, S. 2004. Effect of graded levels of aflatoxin,ochratoxin and their combinations on the performance and immune response of broilers.British Poultry Science, 45, 512-518.VERMA, N., RETTENMEIER, A. W. & SCHMITZ-SPANKE, S. 2011. Recent advances in the use of Susscrofa (pig) as a model system for proteomic studies. Proteomics, 11, 1-18.VESELA, D., VESELY, D. & JELINEK, R. 1983. Toxic Effects of Ochratoxin-a and Citrinin, Alone and inCombination, on Chicken Embryos. Applied and Environmental Microbiology, 45, 91-93.VISCONTI, A., SOLFRIZZO, M., DOKO, M. B., BOENKE, A. & PASCALE, M. 1996. Stability of fumonisinsat different storage periods and temperatures in gamma-irradiated maize. Food Additivesand Contaminants, 13, 929-938.VOSS, K. A., BACON, C. W., MEREDITH, F. I. & NORRED, W. P. 1996. Comparative subchronic toxicitystudies of nixtamalized and water-extracted Fusarium moniliforme culture material. Foodand Chemical Toxicology, 34, 623-&.VOSS, K. A., RILEY, R. T., BACON, C. W., MEREDITH, F. I. & NORRED, W. P. 1998. Toxicity andsphinganine levels are correlated in rats fed fumonisin B-1 (FB1) or hydrolyzed FB1.Environmental Toxicology and Pharmacology, 5, 101-104.VOSS, K. A., RILEY, R. T., NORRED, W. P., BACON, C. W., MEREDITH, F. I., HOWARD, P. C., PLATTNER,R. D., COLLINS, T. F. X., HANSEN, D. K. & PORTER, J. K. 2001. An overview of rodent toxicities:Liver and kidney effects of fumonisins and Fusarium moniliforme. Environmental HealthPerspectives, 109, 259-266.VOSS, K. A., RILEY, R. T., SNOOK, M. E. & GELINEAU-VAN WAES, J. 2009. Reproductive andSphingolipid Metabolic Effects of Fumonisin B-1 and its Alkaline Hydrolysis Product in LM/BcMice: Hydrolyzed Fumonisin B-1 Did Not Cause Neural Tube Defects. Toxicological Sciences,112, 459-467.VOSS, K. A., SMITH, G. W. & HASCHEK, W. M. 2007. Fumonisins: Toxicokinetics, mechanism of actionand toxicity. Animal Feed Science and Technology, 137, 299-325.WWACHE, Y. J., HBABI-HADDIOUI, L., GUZYLACK-PIRIOU, L., BELKHELFA, H., ROQUES, C. & OSWALD, I. P.2009. The mycotoxin Deoxynivalenol inhibits the cell surface expression of activationmarkers in human macrophages. Toxicology, 262, 239-244.WAES, J. G., STARR, L., MADDOX, J., ALEMAN, F., VOSS, K. A., WILBERDING, J. & RILEY, R. T. 2005.Maternal fumonisin exposure and risk for neural tube defects: Mechanisms in an in vivomouse model. Birth Defects Research Part a-Clinical and Molecular Teratology, 73, 487-497.219

REFERENCES BIBLIOGRAPHIQUESWANG, E., NORRED, W. P., BACON, C. W., RILEY, R. T. & MERRILL, A. H. 1991. Inhibition ofSphingolipid Biosynthesis by Fumonisins - Implications for Diseases Associated withFusarium-Moniliforme. Journal of Biological Chemistry, 266, 14486-14490.WANGIKAR, P. B., DWIVEDI, P., SHARMA, A. K. & SINHA, N. 2004a. Effect in rats of simultaneousprenatal exposure to ochratoxin A and aflatoxin B-1. II. Histopathological features ofteratological anomalies induced in fetuses. Birth Defects Research Part B-Developmental andReproductive Toxicology, 71, 352-358.WANGIKAR, P. B., DWIVEDI, P. & SINHA, N. 2004b. Effect in rats of simultaneous prenatal exposure toochratoxin A and aflatoxin B-1. I. Maternal toxicity and fetal malformations. Birth DefectsResearch Part B-Developmental and Reproductive Toxicology, 71, 343-351.WANGIKAR, P. B., DWIVEDI, P., SINHA, N., SHARMA, A. K. & TELANG, A. G. 2005. Teratogenic effectsin rabbits of simultaneous exposure to ochratoxin A and aflatoxin B1 with special referenceto microscopic effects. Toxicology, 215, 37-47.WATTENBERG, E. V., BADRIA, F. A. & SHIER, W. T. 1996. Activation of mitogen-activated proteinkinase by the carcinogenic mycotoxin fumonisin B-1. Biochemical and Biophysical ResearchCommunications, 227, 622-627.WEGST, W. & LINGENS, F. 1983. Bacterial-Degradation of Ochratoxin-A. Fems Microbiology Letters,17, 341-344.WEIBKING, T. S., LEDOUX, D. R., BERMUDEZ, A. J. & ROTTINGHAUS, G. E. 1994. Individual andCombined Effects of Feeding Fusarium-Moniliforme Culture Material, Containing KnownLevels of Fumonisin-B1, and Aflatoxin-B1 in the Young Turkey Poult. Poultry Science, 73,1517-1525.WILLIAMS, K. C., BLANEY, B. J. & PETERS, R. T. 1994. Pigs Fed Fusarium-Infected Maize ContainingZearalenone and Nivalenol with Sweeteners and Bentonite. Livestock Production Science, 39,275-281.WISNER, D. M., HARRIS, L. R., GREEN, C. L. & PORITZ, L. S. 2008. Opposing regulation of the tightjunction protein claudin-2 by interferon-gamma and interleukin-4. Journal of SurgicalResearch, 144, 1-7.WOOD, G. M. 1982. Effects of Processing on Myco-Toxins in Maize. Chemistry & Industry, 972-974.WU, F. 2004. Mycotoxin risk assessment for the purpose of setting international regulatorystandards. Environmental Science & Technology, 38, 4049-4055.WU, F. & MUNKVOLD, G. P. 2008. Mycotoxins in ethanol co-products: Modeling economic impacts onthe livestock industry and management strategies. Journal of Agricultural and FoodChemistry, 56, 3900-3911.WU, T. S., YU, F. Y., SU, C. C., KAN, J. C., CHUNG, C. P. & LIU, B. H. 2005. Activation of ERK mitogenactivatedprotein kinase in human cells by the mycotoxin patulin. Toxicology and AppliedPharmacology, 207, 103-111.WYATT, R. D., TUNG, H. T. & HAMILTON, P. B. 1973. Effect of simultaneous feeding of aflatoxin andrubratoxin to chickens. Poult Sci, 52, 395-397.XXING, Z., GAULDIE, J., COX, G., BAUMANN, H., JORDANA, M., LEI, X. F. & ACHONG, M. K. 1998. IL-6 isan antiinflammatory cytokine required for controlling local or systemic acute inflammatoryresponses. Journal of Clinical Investigation, 101, 311-320.Y220

REFERENCES BIBLIOGRAPHIQUESYAN, D., ZHOU, H. R., BROOKS, K. H. & PESTKA, J. J. 1997. Potential role for IL-5 and IL-6 in enhancedIgA secretion by Peyer's patch cells isolated from mice acutely exposed to vomitoxin.Toxicology, 122, 145-158.YE, D. M., MA, I. & MA, T. Y. 2006. Molecular mechanism of tumor necrosis factor-alpha modulationof intestinal epithelial tight junction barrier. American Journal of Physiology-Gastrointestinaland Liver Physiology, 290, G496-G504.YIANNIKOURIS, A., ANDRE, G., POUGHON, L., FRANCOIS, J., DUSSAP, C. G., JEMINET, G., BERTIN, G. &JOUANY, J. P. 2006. Chemical and conformational study of the interactions involved inmycotoxin complexation with beta-D-glucans. Biomacromolecules, 7, 1147-1155.YIANNIKOURIS, A., BERTIN, G. & JOUANY, J. P. 2007. Reducing mycotoxin impact: the science behindMycosorb. http://en.engormix.com/MA-mycotoxins/prevention/articles/reducing-mycotoxinimpact-science_636.htm.YIANNIKOURIS, A., FRANCOIS, J., POUGHON, L., DUSSAP, C. G., BERTIN, G., JEMINET, G. & JOUANY, J.P. 2004. Adsorption of zearalenone by beta-D-glucans in the Saccharomyces cerevisiae cellwall. Journal of Food Protection, 67, 1195-1200.YILDIRIM, M., MANNING, B. B., LOVELL, R. T., GRIZZLE, J. M. & ROTTINGHAUS, G. E. 2000. Toxicity ofmoniliformin and fumonisin B-1 fed singly and in combination in diets for young channelcatfish Ictalurus punctatus. Journal of the World Aquaculture Society, 31, 599-608.YOO, H. S., NORRED, W. P. & RILEY, R. T. 1996. A rapid method for quantifying free sphingoid basesand complex sphingolipids in microgram amounts of cells following exposure to fumonisin B-1. Toxicology in Vitro, 10, 77-84.YOSHIZAWA, T., TAKEDA, H. & OHI, T. 1983. Structure of a Novel Metabolite from Deoxynivalenol, aTrichothecene Myco-Toxin, in Animals. Agricultural and Biological Chemistry, 47, 2133-2135.YOUNG, J. C. 1986. Reduction in Levels of Deoxynivalenol in Contaminated Corn by Chemical andPhysical Treatment. Journal of Agricultural and Food Chemistry, 34, 465-467.YOUNG, J. C., TRENHOLM, H. L., FRIEND, D. W. & PRELUSKY, D. B. 1987. Detoxification ofDeoxynivalenol with Sodium Bisulfite and Evaluation of the Effects When Pure Mycotoxin orContaminated Corn Was Treated and Given to Pigs. Journal of Agricultural and FoodChemistry, 35, 259-261.YOUNG, J. C., ZHU, H. H. & ZHOU, T. 2006. Degradation of trichothecene mycotoxins by aqueousozone. Food and Chemical Toxicology, 44, 417-424.YU, D. Y., LI, X. L. & LI, W. F. 2008. Effect of Montmorillonite Superfine Composite on GrowthPerformance and Tissue Lead Level in Pigs. Biological Trace Element Research, 125, 229-235.ZZHANG, H., NAGASHIMA, H. & GOTO, T. 1997. Natural occurrence of mycotoxins in corn, samplesfrom high and low risk areas for human esophageal cancer in China. Mycotoxins, 44, 29-35.ZHOU, B., LI, Y., GILLESPIE, J., HE, G. Q., HORSLEY, R. & SCHWARZ, P. 2007. Doehlert matrix design foroptimization of the determination of bound deoxynivalenol in barley grain withtrifluoroacetic acid (TFA). Journal of Agricultural and Food Chemistry, 55, 10141-10149.ZHOU, H. R., ISLAM, Z. & PESTKA, J. J. 2003. Rapid, sequential activation of mitogen-activated proteinkinases and transcription factors precedes proinflammatory cytokine mRNA expression inspleens of mice exposed to the trichothecene vomitoxin. Toxicological Sciences, 72, 130-142.ZHOU, H. R., YAN, D. & PESTKA, J. J. 1997. Differential cytokine mRNA expression in mice after oralexposure to the trichothecene vomitoxin (deoxynivalenol): Dose response and time course.Toxicology and Applied Pharmacology, 144, 294-305.ZHOU, T., HE, J. & GONG, J. 2008. Microbial transformation of trichothecene mycotoxins. WorldMycotoxin Journal, 1, 23-30.221

REFERENCES BIBLIOGRAPHIQUESZIELONKA, L., WISNIEWSKA, M., GAJECKA, M., OBREMSKI, K. & GAJECKI, M. 2009. Influence of lowdoses of deoxynivalenol on histopathology of selected organs of pigs. Polish Journal ofVeterinary Sciences, 12, 89-95.222

Effet chez le porcelet d'une exposition Ã un rÃ©gime co-contaminÃ© en ...

Create successful ePaper yourself

Delete template?

Save as template?