Microbes Environ. Vol. 21, No. 3, 189–199, 2006 http://wwwsoc.nii.ac.jp/jsme2/ Enrichment and Phylogenetic Analysis of Moderately Thermophilic Myxobacteria from Hot Springs in Japan TAKASHI IIZUKA1*, MITSUNORI TOKURA1, YASUKO JOJIMA1, AKIRA HIRAISHI2, SHIGERU YAMANAKA1,3 and RYOSUKE FUDOU1 1 Microbiology Group, Institute of Life Sciences, Ajinomoto Co., Inc., 1–1 Suzuki-cho, Kawasaki-ku, Kawasaki, 210–8681, Japan 2 Department of Ecological Engineering, Toyohashi University of Technology, Toyohashi, 441–8580, Japan 3 Experimental Farm, Faculty of Textile Science and Technology, Shinshu University, Ueda, 386–8567, Japan (Received April 4, 2006—Accepted August 11, 2006) During the search for thermophilic myxobacteria in geothermal environments, four myxobacteria-like cultures that grew at temperatures of up to 50°C and optimally at 45–49°C were enriched from various hot springs in Japan. Three of the cultures were derived from freshwater hot springs and one from a coastal saline spring. Although all lacked fruiting structures, they were bacteriolytic and formed diffusive shallow crater-like colonies. Even after repeated enrichment procedures using Escherichia coli-prey agar media, the crater-like cultures were usually surrounded by thin, film-like, spreading colonies of other gliding bacteria that grew faster. Within the crater-like sunken area, rod-shaped bacterial cells were observed. 16S rRNA gene sequences from the myxobacterial cultures were amplified by nested PCR with myxobacteria-specific primers, subcloned, and phylogenetically analyzed. All of the myxobacterial clones obtained from the three cultures were assigned to the suborder Sorangineae. These clones were distantly positioned within this suborder, and mostly shown to represent new genera of myxobacteria. These results suggested that diverse species of moderately thermophilic myxobacteria, including many as yet undescribed, ubiquitously inhabit hot spring environments. Key words: moderately thermophilic myxobacteria, Sorangineae, gliding bacteria, hot springs, microbial mats Myxobacteria are known for the following unique characteristics. First, these bacteria move by gliding and form diffusive spreading colonies or so-called “swarms”. Second, they have a highly developed life cycle with complex cell to cell communication systems and show remarkable morphogenetic potential, typically, the ability to form fruiting bodies. Third, they are good producers of secondary metabolites with distinct chemical structures9,25). Myxobacteria have long been described as terrestrial and obligatory aerobic bacteria6,23). However, recent findings have revealed a far more diverse habitat and the actual metabolic activities of myxobacteria. For example, slightly halophilic marine myxobacteria have been isolated from coasts8,13–15). Further* Corresponding author; E-mail: takashi_iizuka@ajinomoto.com, Tel: +81–44–244–7181, Fax: +81–44–222–0129 more, anaerobic myxobacteria showing unique metabolic activity such as the dehalogenation of 2-chloro-phenol were recently discovered28). With respect to growth temperature, myxobacteria including the novel strains noted above have been regarded as mesophiles23,24). Our previous study showed that all of the authentic strains of myxobacteria in our collections failed to grow at temperatures over 40°C, except for a few strains able to grow at up to 45°C8). To our knowledge, myxobacteria that can grow at temperatures above 50°C have not been isolated to date. Our attempts to find “thermophilic” myxobacteria have focused on geothermal hot springs, because these environments harbor diverse microorganisms and are rich sources of thermophilic bacteria1,11,12,20,26). Another feature of myxobacteria is their predatory activity, lysing a variety of other microorganisms such as bacte- IIZUKA et al. 190 ria, yeast, and fungi. Microbial predators such as viruses, predatory bacteria, and protozoa are generally thought to profoundly influence the ecological and biochemical processes of microbial communities in various environments1–5). In recent years, the distribution of viruses and phages in hot springs has been reported2–4). A thermophilic Bdellovibrio has also been suggested to exist though it has not yet been isolated5). In addition, a thermophilic amoeba that can grow at up to 57°C was recently isolated1). However, the ecological niches of myxobacteria in hot spring environments are quite obscure20). By using both molecular and culture-dependent methods, we investigated the taxonomic characteristics of the myxobacterial cultures obtained from the hot springs, and concluded the existence of thermophilic myxobacteria growing at 50°C. Materials and Methods Sampling from hot springs Samples were collected during the period from December 1999 to March 2000 (Table 1). The sites included neutral to slightly alkaline springs, acidic springs, slightly saline springs, and coastal seawater springs. The temperature of sites ranged from 40 to 80°C, with a pH of 2.0 to 8.0 and NaCl concentration of 0.0 to 2.0% (w/v). Water samples were filtered through a G-V Durapore Membrane (500 ml content, pore size: 0.2 µm). Then, square pieces of filter (size: 5×5 mm) were cut out with a sterile dissecting knife and subjected to an isolation procedure. The samples were Table 1. Sampling sites and samples Sample site no. Location I II III-1 Physico-chemical properties Samples collected Temp (°C) pH % NaCl No. Type Hachinohe, bath, Aomori Nasu Shiobara, bath, Tochigi Atami, Hashiri-Yu, hot spring cave, Kanagawa 40–50 40–50 63–70 7.0–8.0 6.0–7.0 6.0–7.0 0.8–0.9 0 0.3–0.5 1 1 11 III-2 Atami, NaNa-Yu, hot spring wells (I–VII) 50–74 6.0–7.0 0.1–0.4 20 IV-1 Izu; Atagawa, hot spring well (I), Shizuoka 75–80 7.0–7.2 0.1–0.2 11 IV-2 Izu; Atagawa, Shima-mi-yu, hot spring well 45–50 7.0–7.1 0.2–0.3 8 IV-3 Izu; Atagawa, coastal warm water pond 45–60 7.0–7.1 0.1–0.2 14 IV-4 Izu; Atagawa, hot spring well (II) 50–60 7.0–8.0 0.1–0.2 12 V-1 V-2 V-3 Izu; Shimogamo, bath Izu; Shimogamo, hot spring well (I) Izu; Shimogamo, hot spring well (II–IV) 40–50 85–88 50–60 7.5–8.0 8.0–8.1 7.0–8.0 1 1 12 V-4 Izu; Shimogamo, hot spring well (V) 55–60 7.3–7.4 0.4–0.5 1.0–1.2 Not determined 1.0–1.1 V-5 Izu; Shimogamo, hot spring well (VI) 55–75 7.3–7.4 0.8–0.9 8 V-6 VI-1 VI-2 VI-3 VI-4 Izu; Shimogamo, hot spring effluent ditch Hakone; Yumoto, bath, Kanagawa Hakone; Gōra, hot steam outlet Hakone; Ōwaku-dani, hot spring (I), Hakone; Ōwaku-dani, hot spring (II) 50–70 40–50 45–50 50–55 45–48 6.0–7.0 8.3–8.4 7.0–8.0 2.8–3.0 2.0–3.0 0.0 0.1–0.2 0.0 0.0 0.0 5 3 6 1 9 VII VIII Takeo, hot spring well, Saga Tokachi; Fukiage, hot spring, Hokkaido 45–50 40–50 8.5–8.6 2.0–3.0 0.0–0.1 0.0 3 3 IX Ibusuki, Yuno-Hama, coastal beach hot spring, Kagoshima 40–50 8.0–8.2 2.5–3.0 4 Water (1) Water (1) Water (3), Sand (4), Gypsum precipitate (4) Water (7), Biomat (3), Mud (5), Fallen leaf (5) Water (1), Sand (5), Gypsum precipitate (5) Water (1), Biomat (2), Gypsum precipitate (3), Sand (2) Water (3), Fallen leaf (4), Mud (4), Sulfur precipitate (3) Water (2), Biomat (2), Mud (3), Fallen leaf (3), Wood piece (2) Water (1) Water (1) Biomat (3), Mud (5), Gypsum precipitate (4) Biomat (2), Mud (2), Gypsum precipitate (2) Water (1), Sand (2), Mud (2), Fallen leaf (3) Mud (2), Fallen leaf (3) Water (1), Biomat (2) Biomat (3), Mud (3) Water (1) Water (1), Mud (3), Fallen leaf (3), Sulfur precipitate (2) Water (3) Water (1), Sand (1), Sulfur precipitate (1) Water (1), Sand (3) 6 Myxobacteria from Hot Springs in Japan stored at 8°C, and used within 5 days after the sampling. Temperature was measured with an alcohol thermometer. The NaCl concentration at the sites was measured with a HORIBA model C-121 Compact Salt meter. The pH was measured with a HORIBA model F-22 pH meter. Culture media SWS-agar medium13), was used as a basal medium for the coastal hot spring samples. As a basal medium for the terrestrial hot spring samples; four different media were used to take account of the physico-chemical properties of each sampling site. WCX-agar25) was used for freshwater springs. HS-agar, which contained (per liter) 1.0 g of CaCl2·2H2O, 2.0 g of NaCl, 1.0 g of MgSO4·7H2O, 0.5 g of KCl, 1.0 ml of trace element solution25), and 15 g of Bacto Agar (pH 7.2) (Becton, Dickinson & Co., Sparks, MD, USA), was used for slightly saline springs. 1/2 SWS-agar, containing half strength components of SWS and Bacto Agar 15 g (pH 7.5), was used for the terrestrial saline springs. GG-medium, containing (per liter) 0.1 g of CaCl2·2H2O, 0.1 g of MgSO4·7H2O, 0.1 g of KCl, 0.1 ml of trace element solution25), and 6.0 g of Phytagel (gellan gum, Sigma-Aldrich, St. Louis, MO, USA), was used for the acidic springs. The gel and the other components were separately autoclaved, and the pH of the liquid medium was adjusted to 2.0 with 1 N H2SO4 prior to the sterilization. To prepare the media for the enrichment of myxobacteria, a live Escherichia coli cell paste was smeared linearly in parallel on the surface of the basal solidified media, which were then designated SWS-, WCX-, HS-, 1/2 SWS- and GG-Ec plate medium. For the maintenance of terrestrial cultures, VY2 (yeast)-agar25) was used. Associating gliding bacteria were purified and maintained on 1/3CY agar for terrestrial isolates and 1/3CY/SWS agar for marine isolates. 1/3CY agar contained (per liter) 1.0 g of Bacto Casitone (Becton, Dickinson & Co.), 0.3 g of Bacto Yeast Extract (Becton, Dickinson & Co.), 0.2 g of CaCl2·2H2O, 0.2 g of MgSO4·7H2O, and 15 g of Bacto Agar (pH 7.2). 1/3CY/SWS agar contained (per liter of SWS) 1.0 g of Bacto Casitone, 0.3 g of Bacto Yeast Extract, and 15 g of Bacto Agar (pH 7.5). To check cellulose-decomposing activity, modified ST-2125) agar was used; it contained (per liter) 1.0 g of KNO3, 0.2 g of CaCl2·2H2O, 0.2 g of MgSO4·7H2O, 50 mg of FeSO4·7H2O, 25 mg of MnSO4·5H2O, 0.2 g of K2HPO4, 2.0 g of HEPES, 0.2 ml of trace element solution, and 15 g of Bacto Agar (pH 7.2, KOH). To all media, 0.5 mg of cyanocobalamine (Wako Pure Chemicals, Osaka, Japan) and 25 mg of cycloheximide (Sigma-Aldrich) were added per liter from filter-sterilized stock solutions. 191 Enrichment and isolation Enrichment and isolation of the myxobacteria from hot springs were performed by the E. coli-baiting method25). Small pieces of sample were put on the terminal part of the streaks of the live E. coli cell paste on the solid media. The inoculated plates, contained in tightly closed plastic incubation boxes, were aerobically incubated in darkness at 50 or 60°C. The cultures of the myxobacteria were purified by several transfers on E. coli (Ec) plate medium25). Growth conditions The response of the hot spring cultures to temperatures was checked on Ec agar plates smeared with autoclaved cell paste. Agar pieces of the cultures, punctured from the edge of swarms on Ec agar plates using sterile plastic-straws (diameter: 3 mm), were picked out and inoculated onto the center of the smeared round area (diameter: 6 cm) of E. coli cell-paste. Terrestrial cultures were grown on WCX-Ec agar and coastal cultures, on SWS-Ec agar. Reference myxobacterial strains were cultured on VY2 agar for terrestrial isolates and VY2/SWS agar for marine isolates. The growth was checked at a temperature ranging from 8 to 55°C. Growth was determined by measuring the diameter of the cell-lysing swarm area after one month of incubation. Terrestrial and marine reference myxobacterial strains are shown in Table 3. Other physiological characteristics of the hot spring cultures To investigate salt requirements, the growth of the cultures was checked on the agar media with and without SWS. To check anaerobic growth, the cultures grown on SWS-Ec, WCX-Ec, VY2 or 1/3CY agar plates were incubated at 50°C in an Anaero-Pack system (Mitsubishi Gas Chemical, Tokyo, Japan). Cellulose-degrading activity was tested on sterile filter paper laid on the VY2 agar, 1/3CY agar, or modified ST-21 agar. Agar pieces of growing swarm cultures picked out from the VY2 agar and 1/3CY agar were put at the center of the filter paper. After one month of incubation at 50°C, the decomposition of filter papers was checked. 16S rRNA gene sequencing and phylogenetic analysis Cells were cut out from the swarm areas of co-cultures grown on WCX-Ec or SWS-Ec agar, and subjected to a DNA extraction procedure using the Blood & Cell culture DNA Midi Kit (Qiagen, Valencia, CA, USA) according to the manufacturer’s instructions. PCR primers Myxo1f: 5'- IIZUKA et al. 192 GCGKCSCATCAGCTAGTT-3' (E. coli position 252–270) and Myxo2f: 5'-AMGACGSGTAGCTGGTCT-5' (E. coli position 296–313) were designed on the basis of all myxobacterial 16S rRNA gene sequences available from the Ribosomal RNA Database Project II (RDP II)18). 16S rRNA genes were amplified by nested PCR using the myxobacteria-specific primer sets. The first PCR was performed with the eubacterial universal primer set 27f/1492r. Extracted bacterial genomic DNA and amplified PCR products were purified by Microcon-PCR (Millipore, Billerica, MA, USA). The second PCR was performed with the myxobacteria-specific primer set Myxo1f/1492r. The first PCR products and the purified second PCR products were used as the template of the third PCR with primers Myxo2f/1492r. All the PCRs were performed with an Extaq DNA polymerase kit according to the manufacturer’s instructions. Each 50-µl reaction mixture contained 1 µl of template DNA and 0.4 pmol/µl each of the primers. The reaction conditions for the first and second PCRs were 25 cycles of 94°C for 1 min, 50°C for 1 min, and 72°C for 1.5 min followed by postextension at 72°C for 5.5 min. The conditions for the third PCR were 94°C for 1 min, 68°C for 2.5 min, and 72°C for 5.5 min. The third PCR products were purified with a QIAEX II Gel Extraction Kit (Qiagen) and then cloned into a pGEM-T vector plasmid (Promega, Madison, WI, USA) according to the manufacturer’s instructions. The insertion of a DNA fragment of expected size (nearly 1,200 bp) was determined by PCR amplification with universal primers: 21M13f and M13RV. The amplified insert DNA was subjected to restriction fragment length polymorphism (RFLP) analysis using three restriction enzymes, Sau3AI, HaeIII, and HhaI (Takara, Otsu, Japan). Sequencing was performed using an ABI Prism BigDye Terminator cycle Sequencing Ready Reaction kit and an ABI model 310 automatic sequence analyzer (Applied Biosystems, Foster City, CA, USA). The sequencing primers used were described previously21). Sequence data used to infer phylogenetic trees were retrieved from the GenBank nucleotide sequence database and RDP II. The sequence data were aligned using the Clustal W package34) and checked manually. Distance matrices between sequences were calculated with Kimura’s two-parameter model in the program DNADIST of the PHYLIP ver. 3.5c package7) and a neighbor-joining tree27) was constructed from the distance matrices using NEIGHBOR in this package. A bootstrap with 100 replicates was generated using SEQBOOT and consensus trees were constructed using CONSENSE in the same package. Phylogenetic analyses of myxobacteria-associating gliding bacteria with a comparison of nearly 500 bases were performed by TechnoSuruga. Co. (Shizuoka, Japan). Results Emergence of swarm-forming cultures Various hot-spring samples yielded diffusive, E. coli celllysing colonies on Ec plates when incubated at 50°C for 2–4 weeks. These colonies grew in and around the E. coli smear on the surface of the agar media (Fig. 1). Swarms emerged from 27 out of the 140 samples, with the microbial mats (or biomats) having the highest frequency of occurrence among the tested samples. Thirteen swarms appeared from 17 samples of biomats, 4 from 29 samples of effluent water, 4 from 29 samples of mud, 4 from 21 samples of fallen leaves, 1 from 18 samples of gypsum precipitates, and 1 from 3 samples of coastal sand (Table 2). At 60°C, no distinct myxobacteria-like swarms could be observed. All of the hot spring swarm cultures thus obtained contained co-existing gliding bacteria that could not be separated from the myxobacterial colonies despite repeated transfers on agar plates. The associating gliding bacteria rapidly grew and always spread faster than the myxobacteria-like strains on the agar surfaces. Four cultures which reproducibly formed swarms, designated SIB-1, AT-1, AT-3 and YU-2 (Table 2), were obtained after one year of repeated transfers and used for further studies. The culture SIB-1 was obtained from a sand sample of a coastal hot spring, at Ibusuki. To date, our attempts to purify myxobacteria from SIB-1 by enrichment using antibiotics or SDS solution25) have been unsuccessful. AT-1 was from a dark-green biomat sample collected at a hot spring well, at Izu-Atagawa. AT-3 was obtained from a dark-green biomat growing on a gypsum stone, immersed in warm (around 45–50°C) spillage water of another hot spring well, at Izu-Atagawa. The remaining culture, YU-2, was obtained from a dark-green biomat growing on the concrete wall of an open-air hot spring bath, at HakoneYumoto. Colony morphology All the four cultures formed predatory, spreading, shallow, crater-like colonies or so-called swarms peculiar to myxobacteria when incubated on SWS-Ec or WCX-Ec agar at 50°C. None of the cultures formed the tough slime sheets which are observed in some species of myxobacteria. Culture SIB-1 formed shallow crater-like bacteriolytic colonies on SWS-Ec agar (Fig. 1A), but a distinct swarm-like formation was not observed on VY2/SWS or on 1/3CY/SWS agar, probably because of the vigorous growth of the associating bacteria on rather nutrient-rich media. After 2 months Myxobacteria from Hot Springs in Japan 193 in depth, on VY2 and 1/3CY agar after a month of incubation (Fig. 1C). The two cultures resembled each other in the morphology of their colonies. The swarms of AT-3 tended to form radiating ‘lobes’ around circular edges of the craters after prolonged incubation (Fig. 1C; indicated by arrows). The formation of depressions in an agar gel matrix was also observed in the swarms of marine myxobacteria of the genus Enhygromyxa15). Formation of Fruiting bodies Although the four cultures formed myxobacteria-like swarms at 50°C, no fruiting bodies, the most important morphological feature of myxobacteria in their classification19,25,29,30), were observed. In the case of SIB-1, however, large numbers of cell aggregates measuring 20–50 µm in diameter (Fig. 2C), could be observed within agar gel beneath the small holes of the swarm area (Fig. 1B; inside the dotted square). The aggregates may be incomplete fruiting bodies. Cellular morphology Fig. 1. Appearance of swarms of the hot spring myxobacteria cultured at 50°C. (A), culture SIB-1 grown on SWS-Ec agar for 3 weeks; (B), culture SIB-1 grown for 2 months on SWS-Ec agar; (C), culture AT-3 grown on 1/3CY agar for a month. of incubation at 50°C, many small holes appeared within the swarm area on the SWS-Ec agar (Fig. 1B). Similar small holes in the agar gel matrix were also reported for myxobacteria of the genus Nannocystis22). AT-1 formed shallow crater-like colonies on WCX-Ec, VY2, and 1/3CY agar. The cultures AT-3 and YU-2 formed shallow crater-like colonies on WCX-Ec agar and rather deeper craters, 0.5–1.0 mm The cellular morphology of SIB-1 and the terrestrial cultures was studied within the swarm area on SWS-Ec agar and 1/3CY agar, respectively. SIB-1 contained rod-shaped cells measuring 0.5–0.7×2.0–4.0 µm and very long filamentous cells measuring 0.3–0.5×40–100 µm (Fig. 2A), inside the dotted square area of Fig. 1A. Very long bacteria were also observed in the area outside of the swarm (Fig. 2B). AT-1 had blunt-ended larger rods (0.8–1.5×3.0–8.0 µm), smaller rods (0.5–0.8×2.0–4.0 µm), and thin filamentous cells (0.1–0.2×3.0–10 µm). AT-3 contained rod-shaped cells measuring 0.5–0.8×1.5–5.0 µm (Fig. 2D; indicated by arrows), inside the dotted square area of Fig. 1C. A few elongated cells were also observed among the rods (Fig. 2D; indicated by a dotted arrow). In a younger swarm culture of AT-3 that was incubated for 10 days, the rod-shaped cells were more distinctly observed (Fig. 2E). Cells of YU-2 were rods having a size of 0.5–0.8×1.0–3.0 µm. As for the cultures AT-3 and YU-2, we could not observe distinct rods in agar pieces taken from the outer area around the sunken swarms, and the cellular shapes of bacteria inside and outside of the swarm area were obviously different from each other (Fig. 2D and 2F). Irregularly-shaped filamentous cells, indicated by arrows in Fig. 2F, were observed outside of the swarm (Fig. 1C; indicated by a square). Growth temperatures As shown in Table 3, the hot spring cultures grew at higher temperatures than any of the myxobacterial strains IIZUKA et al. 194 Table 2. Myxobacterial cultures obtained from the hot springs Sampling site no. II III-1 III-2 IV-2 IV-3 IV-4 V-1 V-3 V-4 V-5 V-6 VI-1 VI-2 VII IX Samples positive for emergence of swarms No. Type of the sample(s) Enrichment culture medium used Myxobacterial culture obtained 1 1 2 2 2 2 1 4 2 1 1 3 3 1 1 Water Gypsum precipitate Dark-green biomats Dark-green biomats Fallen leaves 1, Dark-green biomat; 1, Mud Water 2, Dark-green biomats; 2, Mud 1, Dark-green biomat; 1, Mud Fallen leaf Fallen leaf 1, Water; 2, Dark-green biomats Green biomats Water Beach hot spring sand WCX-Ec HS-Ec WCX-Ec WCX-Ec WXC-Ec WXC-Ec 1/2SWS-Ec WXC-Ec 1/2SWS-Ec 1/2SWS-Ec WXC-Ec WXC-Ec WXC-Ec WXC-Ec SWS-Ec — — — AT-3 — AT-1 (from a biomat) — — — — — YU-2 (from a biomat) — used for comparison. The optimum temperature for growth of the thermophilic cultures ranged from 45 to 50°C (Table 4). However, even at the optimum, the four hot spring cultures grew much slower than any of the previously isolated strains of terrestrial and marine mesophilic myxobacteria. The growth declined sharply at above 50°C, and none of the cultures grew at 52°C (Table 4). After incubation at 55°C for a week, all the cultures seemed to die out. Below the temperature range of 27–30°C, the growth of the novel cultures was severely retarded or stopped, whereas most of the mesophilic myxobacteria grew well in the same range (Table 3, 4). At 25°C, none of the hot spring myxobacterial cultures showed growth after a month of incubation. Phylogenetic analysis The 16S rRNA genes from cultures SIB-1, AT-1 and AT3 were amplified using the myxobacterial specific PCR primers and subcloned. RFLP analyses of 15 clones from SIB-1, 13 from AT-1, and 15 from AT-3 resulted in groupings into 3, 3, and 5 phylotypes, respectively. Representative clones of each phylotype were partly sequenced. After sequences that were suspected of being chimeric or nonmyxobacterial were discarded, 6 clones related to myxobacteria were obtained (1 clone for SIB-1, 2 for AT-1, and 3 for AT-3). A neighbor-joining distance matrix tree showing the phylogenetic positions of the 6 clones and previously known myxobacteria based on partial 16S rRNA gene sequences are shown in Fig. 3. All of the 6 clones were related to the suborder Sorangineae myxobacteria. No SIB-1 clones related to the suborder of myxobacteria, Cystobacterineae22,24,25), were found. The representative clone of SIB-1 (SIBN-17) was distantly related to the genera Nannocystis and Enhygromyxa. On the other hand, both AT-1 and AT-3 were shown to be composed of phylogenetically different clones. For AT-1, two clone types (AT1-02 and AT1-01) were identified. Clone AT1-01 was related to the strains of the genera Chondromyces and Polyangium, especially to the unidentified myxobacterial strain NOSO1. The other clone, AT1-02, was positioned near the marine myxobacterial genus Enhygromyxa. For the culture AT-3, 3 clone types (AT3-01, -03, and -09) were scattered into the three different lineages. Clone AT3-01 was distantly related to the genera Chondromyces and Polyangium. The second clone AT3-03 was distantly related to Polyangium vitellinum and the marine myxobacteria of the genus Haliangium. The third clone AT3-09 was also related to another marine genus of myxobacteria, Enhygromyxa. Among the 6 clones, SIBN-17, AT3-01, and AT3-03 were remote from any known sequences. The homology values with closest relatives were less than 90%. These low values indicated that the clones might represent novel taxa above the generic level. Characterization of accompanying gliding bacteria As for the accompanying gliding strains, both terrestrial and coastal strains were isolated and were culturable axenically. The strains were designated as SIB-1C, AT-1C, AT3C, and YU-2C. They grew at higher temperatures than Myxobacteria from Hot Springs in Japan 195 Fig. 2. Phase-contrast micrographs of cells of the hot spring myxobacteria cultured at 50°C. (A), rods of SIB-1 in the swarm grown on SWS-Ec agar; (B), long cells of SIB-1C or Thermonema rossianum grown on 1/3CY/SWS agar; (C), cell aggregates of SIB-1 in the swarm on SWS-Ec agar; (D), rods of AT-3 in the swarm grown on 1/3CY agar for a month; (E), rods of AT-3 in the swarm grown on 1/3CY agar for 10 days; (F), irregularly-shaped filamentous cells around the swarm of AT-3 grown on 1/3CY agar for a month; Scale bar=10 µm. myxobacteria-like strains, up to 55–60°C. Based on the 16S rRNA gene sequence analysis, the coastal culture SIB-1C was assigned to Thermonema rossianum33) with 99.9% sim- ilarity. The long filamentous cellular morphology (Fig. 2B), and physiological characteristics of marine bacteria 17) coincided well with the description of the species33). As for the IIZUKA et al. 196 Table 3. Growth temperature ranges of the myxobacteria Organism tested Phylogenetic affiliation (suborder) Isolated from: New cultures SIB-1 AT-3 YU-2 Temperature (°C) for growth Range Optimum 37–51 30–50 28–50 45–50 45–50 45–49 Beach hot spring sand Land hot spring biomat Land hot spring biomat Sorangineae Sorangineae Coastal mud, Hokkaido Sorangineae 5–30 20–30 Coastal sand, Kanagawa Coastal soil, Kanagawa Coastal sand, Shizuoka Mangrove bark, Okinawa Sea grass, Okinawa Sorangineae Sorangineae Sorangineae Sorangineae Sorangineae 20–37 18–40 20–45 15–37 20–37 30–37 30–34 37–40 30–34 25–30 Coral beach sand, Okinawa Sea weed, Kanagawa Sorangineae Sorangineae 18–34 15–40 28–30 30–37 (Sorangineae)a Reference strains Enhygromyxa salina SHK-1T Enhygromyxa sp. SMH-97-3 Enhygromyxa sp. SMH-27-4 Enhygromyxa sp. SYM-1 Enhygromyxa sp. SIS-1 Plesiocystis pacifica SIR-1T Plesiocystis sp. SIS-2 Haliangium ochraceum SMP-2T Haliangium tepidum SMP-10T Sorangium cellulosum IS-1 Sorangium cellulosum YA-2 Sorangium cellulosum ATCC 25532 Sorangium cellulosum ATCC 25569 Sorangium cellulosum ATCC 29610 Nannocystis exedens DSM 71T Nannocystis exedens ATCC 35989 Chondromyces apiculatus DSM 436T Chondromyces apiculatus HT-1 Myxococcus xanthus ATCC 25232T Archangium gephyra ATCC 25201T Cystobacter fuscus ATCC 25194T Stigmatella aurantiaca ATCC 25190T Myxococcus sp. GT-7 b a b Sea grass, Kanagawa Sorangineae 25–45 40–45 Soil, Kanagawa Soil, Kanagawa Soil, Missouri, USA Soil, Costa Rica Soil Desert soil, Arizona, USA Sorangineae Sorangineae Sorangineae Sorangineae Sorangineae Sorangineae 15–40 20–40 15–40 20–44 15–37 20–42 28–30 37–40 28–30 37–40 28–30 30–40 Goat dung, Isreal Decayed wood, USA Sorangineae Sorangineae 15–37 15–30 25–30 25–30 Goat dung, Okinawa Soil Sorangineae Cystobacterineae 15–30 20–37 25–30 28–30 Soil, Canada Cystobacterineae 20–40 20–30 Soil, Canada Cystobacterineae 20–40 28–30 Bark Cystobacterineae 15–37 34–37 Soil Cystobacterineae 30–48 42–44 Although a phylogenetic analysis was not performed, the morphological features closely resembled those of culture AT-3. Data taken from Ref. (10). Table 4. Swarm diameter of hot spring myxobacterial cultures at each temperaturea Hot Spring Cultures SIB-1 AT-3 a Swarm diameters (mm) at: 25°C 30°C 40°C 45°C 50°C 52°C 0 0 0 1–2 8–9 4–5 10–11 6–7 12–13 6–7 0 0 The cultures were incubated for a month. The diameter of inoculation agar piece (3 mm), was subtracted from the swarm diameter. terrestrial cultures, all the associated gliding strains turned out to be related to Caldimonas manganoxidans JCM 10698T 31) with 99.9% similarity in the 16S rRNA gene sequence. The strains did not lyse live or autoclaved E. coli cells, and did not form a spreading sunken swarm. The filamentous bacterium accompanying culture AT-3 shown in Fig. 2F probably differs from C. manganoxidans whose cells are rods31). Myxobacteria from Hot Springs in Japan 197 Fig. 3. Distance matrix tree showing phylogenetic positions of the hot spring myxobacteria within the order Myxococcales. The tree was constructed based on 16S rRNA gene sequences by using the neighbor-joining method. The bar represents one nucleotide substitution per 10 nucleotides. Bootstrap values, greater than 70%, are indicated at the branch points. Other physiological characteristics The coastal culture SIB-1 showed the physiological properties of marine bacteria17) and required NaCl with an optimum concentration of around 2% (w/v) and other cationic components of seawater for its growth. Three cultures from microbial mats of terrestrial hot springs, AT-1, AT-3, and YU-2, possessed the physiological characteristics of terrestrial or freshwater bacteria requiring no NaCl for their growth. None of the four cultures degraded the filter paper, nor grew under anaerobic conditions. Discussion With regard to growth temperature characteristics, myxobacteria have long been considered mesophiles23). Recently, a thermotolerant strain GT-7, identified as a member of the genus Myxococcus within the Cystobacterineae, was isolated from a soil. The optimum and maximum temperatures for growth of this bacterium are 42–44°C and 48°C, respectively10). On the other hand, the myxobacterial cultures obtained from the hot springs in this study grow at temperatures up to 50°C and optimally at 45–49°C. Thus, the novel myxobacteria are regarded as moderately thermophilic bacteria. To our knowldge, this is the first report of thermophilic myxobacteria growing at 50°C. The novel myxobacteria seem to be adapted to temperature zones around 40–50°C, suggesting that they are indigenous to the hot springs. Our attempts to isolate the hot spring myxobacteria as axenic cultures have so far been unsuccessful. However, phylogenetic analyses basesd on 16S rRNA gene sequences have revealed that all clones from the novel myxobacterial cultures are confined to the suborder Sorangineae, though IIZUKA et al. 198 they are quite diverse within this suborder. In view of the phylogenetic positions of these clones, it seems evident that most of the moderately thermophilic myxobacteria represent new genera of Sorangineae. Further study with possible axenic cultures of the novel myxobacteria should offer formal taxonomic proposals for these bacteria. Although we could not find myxobacteria of the Cystobacterineae in hot spring samples using the culture-dependent isolation method, it seems probable that unknown strains of Cystobacterineae myxobacteria, which can grow at around 50°C, also inhabit the hot springs. The relatively high frequencies of the occurrence of the myxobacteria in hot spring microbial mats are probably due to stable microbial ecosystems held in the mats, which were sheltered from outer stress, such as nutrient depletion, temporal desiccation, or predation by the protozoa. The coexistence of diverse microorganisms; myxobacteria, the other groups of gliding bacteria, and the phototropic bacteria (or algae), might enhance the stability of microbial communities in hot spring biomats. Interestingly, all three strains of the accompanying gliding bacteria obtained from the terrestrial hot spring biomats have been identified as a single species, Caldimonas manganoxidans, which was originally described as a poly(3hydroxybutyrate)-degrading, manganese-oxidizing thermophile31). The type strain of the species was isolated from a hot spring with abundant cyanobacterial mats32). It could be speculated that microbial mats composed of green phototropic bacteria (or algae), C. manganoxidans, several species of myxobacteria, and other bacteria (such as filamentous bacteria), are probably widespread among terrestrial hot springs in Japan. Although the ecophysiology of such a complex microbial community is almost entirely unknown, it might offer a favorable niche for hot-spring myxobacteria in terms of nutritional acquisition. Since myxobacteria as predatory bacteria excrete a wealth of hydrolyzing enzymes such as proteases, glucanases and so on24,25), they might easily obtain nutrients by decomposing microbial cells or macromolecules produced by the other microbes in biomats. As for the mesophilic myxobacteria, mutualism between the myxobacterium Chondromyces crocatus and the Sphingobacterium-like companion bacterium in soil environments has been reported16). In that case, the relationship between the two different groups of bacteria might be defined as strict mutualism because the presence of C. crocatus was reported to be essential for the growth of the companion bacterium. Recently, a number of phylogenetic studies on microbial ecosystems of various hot spring environments have been conducted, revealing the genetic and physiological diversity of microbial communities11,12,26,32). However, as far as we know, only one report has referred to the existence of myxobacteria in a hot spring to date20). On the other hand, results of our study suggested that diverse myxobacteria inhabit hot springs, as shown in Table 2. The seeming discrepancy between the results of our study and those of previous ecological studies, is probably due to the different methods used for investigating the existence of myxobacteria. Assuming that myxobacterial populations in the hot spring environment are relatively small, we adopted the culturedependent enrichment procedure along with a phylogenetic analysis with myxobacteria-specific PCR. From the results of this study, the two-step (enrichment and analysis) method proved to be useful for the detection of hot spring myxobacteria. From an industrial point of view, myxobacteria have attracted much attention as producers of secondary metabolites9). This study and recent findings10,13,28) have broadened our understanding of the ecological and physiological aspects of the bacteria, and it is very likely that more novel myxobacteria will be isolated from unexploited environments, hopefully being good sources of natural products with extensive chemical diversity. Acknowledgements The authors wish to thank Masaru Ishihara of Ajinomoto Co., for collecting samples at Takeo hot spring in Saga prefecture. References 1) Baumgartner, M., A. Yapi, R. Gröbner-Ferreira and K.O. Stetter. 2003. Cultivation and properties of Echinamoeba thermarum n. sp., an extremely thermophilic amoeba thriving in hot springs. Extremophiles 7: 267–274. 2) Breitbart, M., L. Wegley, S. Leeds, T. Schoenfeld and F. Rohwer. 2004. Phage community dynamics in hot springs. Appl. Environ. Microbiol. 70: 1633–1640. 3) Chiura, H.X. 2002. Broad host range xenotrophic gene transfer by virus-like particles from a hot spring. Microbes Environ. 17: 53–58. 4) Chiura, H.X. and M. Umitsu. 2004. Isolation and characterisation of broad-host range gene transporter particles from geo-thermal vent of the Toyoha mine. Microbes Environ. 19: 20–30. 5) Davidov, Y. and E. Jurkevitch. 2004. Diversity and evolution of Bdellovibrio-and-like organisms (BALOs), reclassification of Bacteriovorax starrii as Peredibacter starrii gen. nov., comb. nov., and description of the Bacteriovorax-Peredibacter clade as Bacteriovoracaceae fam. nov. Int. J. Syst. Evol. Microbiol. 54: Myxobacteria from Hot Springs in Japan 1439–1452. 6) Dawid, W. 2000. Biology and global distribution of myxobacteria in soils. FEMS Microbiol. Rev. 24: 403–427. 7) Felsenstein, J. 1989. PHYLIP-phylogeny inference package (version 3.2). Cladistics 5: 164–166. 8) Fudou, R., Y. Jojima, T. Iizuka and S. Yamanaka. 2002. Haliangium ochraceum, gen. nov. sp. nov. and Haliangium tepidum sp. nov.: Novel moderately halophilic myxobacteria isolated from coastal saline environments. J. Gen. Appl. Microbiol. 48: 109–115. 9) Gerth, K., S. Pradella, O. Perlova, S. Beyer and R. Müller. 2003. Myxobacteria: proficient producers of novel natural products with various biological activities—past and future biotechnological aspects with the focus on the genus Sorangium. J. Biotechnol. 106: 233–253. 10) Gerth, K. and R. Müller. 2005. Moderately thermophilic Myxobacteria: novel potential for the production of natural products isolation and characterization. Environ. Microbiol. 7: 874–880. 11) Hanada, S. 2003. Filamentous anoxygenic phototrophs in hot springs. Microbes Environ. 18: 51–61. 12) Hugenholtz, P., C. Pitulle, K.L. Hershberger and N.R. Pace. 1998. Novel division level bacterial diversity in a Yellowstone hot spring. J. Bacteriol. 180: 366–376. 13) Iizuka, T., Y. Jojima, R. Fudou and S. Yamanaka. 1998. Isolation of myxobacteria from the marine environment. FEMS Microbiol. Lett. 169: 317–322. 14) Iizuka, T., Y. Jojima, R. Fudou, A. Hiraishi, J.-W. Ahn and S. Yamanaka. 2003. Plesiocystis pacifica gen. nov., sp. nov., a marine myxobacterium that contains dihydrogenated menaquinone, isolated from the Pacific coasts of Japan. Int. J. Syst. Evol. Microbiol. 53: 189–195. 15) Iizuka, T., Y. Jojima, R. Fudou, M. Tokura, A. Hiraishi and S. Yamanaka. 2003. Enhygromyxa salina gen. nov., sp. nov., a slightly halophilic myxobacterium isolated from the coastal areas of Japan. Syst. Appl. Microbiol. 26: 189–196. 16) Jacobi, C.A., B. Assmus, H. Reichenbach and E. Stackebrandt. 1997. Molecular evidence for association between the sphingobacterium-like organism “Candidatus comitans” and the myxobacterium Chondromyces crocatus. Appl. Environ. Microbiol. 63: 719–723. 17) Macleod, R.A. 1965. The question of the existence of specific marine bacteria. Bacteriol. Rev. 29: 9–23. 18) Maidak, B.L., J.R. Cole, C.T. Parker Jr., G.M. Garrity, N. Larsen, B. Li, T.G. Lilburn, M.J. McCaughey, G.J. Olsen, R. Overbeek, S. Pramanik, T.M. Schmidt, J.M. Tiedje and C.R. Woese. 1999. A new version of the RDP (ribosomal Database Project). Nucleic Acids Res. 27: 171–173. 19) McCurdy, H.D. 1989. Order Myxococcales TCHAN, POCHON and PRÉVOT 1948, 398AL (with contributions of E.R. Brockman, H. Reichenbach and D. White) p. 2139–2170. In J.T. Staley, M.P. Bryant, N. Pfennig and J.G. Holt (eds.), Bergey’s manual of systematic bacteriology, vol. 3. Williams and Wilkins, Baltimore, USA. 199 20) Moyer, C.L., F.C. Dobbs and D.M. Karl. 1995. Phylogenetic diversity of the bacterial community from a microbial mat at an active, hydrothermal vent system, Loihi Seamount, Hawaii. Appl. Environ. Microbiol. 61: 1555–1562. 21) Ohkuma, M., S. Noda and T. Kudo. 1999. Phylogenetic relationships of symbiotic methanogens in diverse termites. FEMS. Microbiol. Lett. 171: 147–153. 22) Reichenbach, H. 1989. Genus II. Nannocystis, p. 2162–2166. In J.T. Staley, M.P. Bryant, N. Pfennig and J.G. Holt (eds.), Bergey’s manual of systematic bacteriology, vol. 3. Williams and Wilkins, Baltimore, USA. 23) Reichenbach, H. 1999. The ecology of the myxobacteria. Environ Microbiol. 1: 15–21. 24) Reichenbach, H. 1993. Chapter 2. Biology of the myxobacteria: ecology and taxonomy, p. 22–23. In M. Dworkin and D. Kaiser (eds.), Myxobacteria II, American Society for Microbiology, Washington DC. 25) Reichenbach, H. and M. Dworkin. 1992. The myxobacteria, p. 3416–3487. In A. Balows, H.G. Trüper, M. Dworkin, W. Harder, K.-H. Schleifer (eds.), The Prokaryotes, 2nd ed., Springer-Verlag, Berlin, Germany. 26) Rothschild, L.J. and R.L. Mancinelli. 2001. Life in extreme environments. Nature 409: 1092–1101. 27) Saitou, N. and M. Nei. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 4: 406–425. 28) Sanford, R.A., J.R. Cole and J.M. Tiedje. 2002. Characterization and description of Anaeromyxobacter dehalogenans gen. nov., sp. nov., an aryl-halorespiring facultative anaerobic myxobacterium. Appl. Environ. Microbiol. 68: 893–900. 29) Shimkets, L. and C.R. Woese. 1992. A phylogenetic analysis of the myxobacteria: Basis for their classification. Proc. Natl. Acad. Sci. USA. 89: 9459–9463. 30) Spröer, C., H. Reichenbach and E. Stackebrandt. 1999. The correlation between morphological and phylogenetic classification of myxobacteria. Int. J. Syst. Bacteriol. 49: 1255–1262. 31) Takeda, M., Y. Kamagata, W.C. Ghiorse, S. Hanada and J. Koizumi. 2002. Caldimonas manganoxidans gen. nov., sp. nov., a poly(3-hydroxybutyrate)-degrading, manganese-oxidizing thermophile. Int. J. Syst. Evol. Microbiol. 52: 895–900. 32) Takeda, M., J. Koizumi, K. Yabe and K. Adachi. 1998. Thermostable poly (3-hydroxybutyrate) depolymerase of a thermophilic strain of Leptothrix sp. isolated from a hot spring. J. Ferment. Bioeng. 85: 375–380. 33) Tenreiro, S., M.F. Nobre, F.A. Rainey, C. Miguel and M.S. da Costa. 1997. Thermonema rossianum sp. nov., a new thermophilic and slightly halophilic species from saline hot springs in Naples, Italy. Int. J. Syst. Bacteriol. 47: 122–126. 34) Thompson, J.D., D.G. Higgins and T.J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22: 4673– 4680.