International Journal of Systematic and Evolutionary Microbiology (2002), 52, 423–428 DOI: 10.1099/ijs.0.01874-0

Reclassification of Eubacterium NOTE formicigenerans Holdeman and Moore 1974 as gen. nov., comb. nov., and description of sp. nov., isolated from human faeces

1 Department of David Taras,1 Rainer Simmering,1 Matthew D. Collins,2 Paul A. Lawson2 Gastrointestinal 1 Microbiology, German and Michael Blaut Institute of Human Nutrition, Arthur- Scheunert-Allee 114-116, Author for correspondence: Michael Blaut. Tel: j49 33200 88470. Fax: j49 33200 88407. 14558 Bergholz- e-mail: blaut!www.dife.de Rehbru$ cke, Germany 2 School of Food Biosciences, University of Reading, Two strains of a Gram-positively staining, obligately anaerobic, non-spore- Reading RG6 6AP, UK forming, rod-shaped bacterium, designated strains 111-13A and 111-35T, were isolated from human faeces. Analysis of the 16S rRNA gene sequences indicated that these strains were members of the Clostridium coccoides rRNA group of organisms. The nearest relatives of the unknown bacterium were Eubacterium formicigenerans (having a sequence similarity of 94%) and an uncultured bacterium (similarity " 99%). Characterization studies indicated that the unidentified faecal bacterium was biochemically distinct from Eubacterium formicigenerans, members of the Clostridium coccoides group and all other described Eubacterium . On the basis of the data from these studies, it is proposed that the hitherto unknown rod-shaped bacterium be designated a species of a novel genus, namely Dorea longicatena gen. nov., sp. nov., and that Eubacterium formicigenerans be transferred to this genus as Dorea formicigenerans gen. nov., comb. nov.

Keywords: Dorea longicatena, Dorea formicigenerans, 16S rRNA, , phylogeny

The intestinal microbiota of humans consists of more human gut (Suau et al., 1999). During an investigation than 400 bacterial species (Finegold et al., 1974; Moore of the human faecal flora, we isolated two strains & Holdeman, 1974). Despite extensive efforts to of a strictly anaerobic, non-spore-forming, Gram- characterize the intestinal microflora, it is now recog- positively staining, rod-shaped organism that dis- nized that a significant proportion of the dominant played " 99% 16S rRNA gene sequence similarity to flora has so far eluded scientific description (Lan- one of the uncultured species reported by Suau et al. gendijk et al., 1995; Zoetendal et al., 1998). Recently, (1999). On the basis of the results of a polyphasic taxo- Suau et al. (1999) conducted a molecular genetic nomic investigation, we propose that these strains be analysis of rDNA amplicons generated directly from a classified, alongside their nearest phylogenetic rela- human faecal sample and showed that more than 90% tive, Eubacterium formicigenerans, in a novel genus, of the flora could be assigned to three major phylo- Dorea gen. nov. genetic lineages (the Bacteroides, Clostridium coccoides T and Clostridium leptum groups). It was evident from Strains 111-13A and 111-35 were isolated from a this molecular taxonomic inventory that the vast screening programme (for hydrogen-producing micro- majority of rDNA sequences generated (76%) did not organisms) using faecal samples from a healthy vol- correspond to known organisms and were clearly unteer who had not undergone antibiotic therapy for derived from hitherto unknown species within the the preceding 6 months. For this purpose, fresh faecal samples were transferred into an anaerobic work- station (MK3; DW Scientific) and diluted serially 10- ...... "# The GenBank/EMBL/DDBJ accession number for the 16S rRNA gene fold up to 10 in Wilkins–Chalgren anaerobic broth sequence of strain DSM 13814T is AJ132842. (WCA; Oxoid), a peptone- and yeast extract-con-

01874 # 2002 IUMS Printed in Great Britain 423 D. Taras and others

(a)

(b)

...... Fig. 1. Scanning electron micrographs of isolate 111-35T grown in HA medium with 10 mM glucose and with 0n5g − Proteose-Peptone l 1 for 12 h at 37 mC. Bars, 5 (a) and 1 (b) µm.

taining complex medium, supplemented with 20 mM pH of the medium. Cells were tested for catalase and 2-bromoethanesulfonate and 20 mM Na#MoO% (to oxidase as described previously (Smibert & Krieg, give medium mod-WCA). An aliquot (0n5 ml) of an 1994) after cultivation on WCA agar. Other bio- enrichment culture that developed from the highest chemical features were determined with the API 50 dilution and showed H# production and morphology CHL system (bioMe! rieux). Experiments with resting of long cell chains was plated onto mod-WCA agar cells were performed as described by Kamlage et al. (1n5% agar, w\v) plates. Single colonies were re- (1997). Hydrogen (Hartmann et al., 2000) and acetate, peatedly picked and streaked until pure cultures were butyrate, propionate, valerate and isovalerate (Kam- obtained. Unless indicated otherwise, all incubations lage et al., 1997) were determined by GC as described. were performed at 37 mC under a N#\CO# (80:20, v\v) Succinate, ethanol, formate, acetate, -lactate, - gas phase, using strictly anaerobic conditions (Hun- lactate and glucose were determined enzymically gate, 1969; Bryant, 1972). For morphological and (Bergmeyer & Graßl, 1984). DNA extraction was done physiological studies, both strains were grown on as described by Schwiertz et al. (2000). The GjC Columbia blood agar (bioMe! rieux), in ST medium content of the DNA was determined by HPLC (Schwiertz et al., 2000) or in a medium used for according to Mesbah et al. (1989) except that the culturing acetogenic (Kamlage et al., 1997) methanol content of the chromatographic buffer was but modified by the addition of 0n5 g Proteose-Peptone reduced to 8% and the temperature was raised to −" no. 2 (Difco) l (HA medium). Agar plates were 37 mC. Lambda DNA (Sigma) served as a standard. prepared by adding 1n5% agar to the appropriate DNA–DNA hybridization was carried out according liquid media, and the plates were poured inside an to Johnson (1994). The DNA–DNA hybrids were anaerobic workstation (N#\CO#\H#; 80:10:10, by detected with the DIG luminescent detection kit vol.). The morphology of the isolates was examined by (Roche) and quantified with an image analyser (Fuji- phase-contrast microscopy (Axioplan 2; Zeiss) and film LAS-1000). For the phylogenetic analysis, the 16S by scanning electron microscopy (model DSM 950; rRNA genes of the two isolates were amplified by PCR Zeiss) after growth on WCA or Columbia blood agar. and sequenced directly using a Taq Dye-Deoxy ter- The methods of Grund et al. (1995) were used for the minator cycle sequencing kit (Applied Biosystems) and scanning electron microscopy. Growth was monitored an automatic DNA sequencer (model 373A, Applied using changes in optical density (at 600 nm) and in the Biosystems). A phylogenetic tree was constructed

424 International Journal of Systematic and Evolutionary Microbiology 52 Dorea gen. nov., containing two species according to the neighbour-joining method, and the showed that the two strains had identical 16S rRNA confidence values of the groupings were estimated by gene sequences, and searches of the GenBank and bootstrap analysis (Felsenstein, 1989). Ribosomal Database Project databases revealed that the isolates were closely related to members of the The two faecal isolates (111-13A and 111-35T) were Clostridium coccoides rRNA group of organisms non-spore-forming, non-motile, strictly anaerobic, (rRNA cluster XIVa; Collins et al., 1994). The highest rod-shaped organisms. The cells were 0 5–0 6 µm wide n n level of sequence relatedness was shown with respect to by 2 0–4 3 µm long and occurred in chains of approxi- n n an rDNA clone derived from an uncultured faecal mately 4–200 cells (Fig. 1). In the exponential growth bacterium (Suau et al., 1999). The described species phase, cells stained Gram-positive, whereas stationary- nearest to the unknown isolates was Eubacterium phase cells exhibited Gram-negative staining behav- formicigenerans (94% sequence similarity). A tree iour. Cultures in ST medium and in HA medium had constructed by neighbour-joining and depicting the a ropy sediment with little or no turbidity. When ST phylogenetic position of the unknown bacterium medium was supplemented with 0 15% agar, the n within the Clostridium coccoides group is shown in cultures exhibited turbidity with dense areas of a fluffy, Fig. 2. ‘woolly’ appearance. Both strains formed white– opaque, non-haemolytic colonies on Columbia blood From these results, it is evident that the two faecal agar and on WCA agar; the colonies were approxi- isolates belong to a hitherto undescribed species, mately 1–3 mm in diameter, circular, convex, smooth, probably related to an uncultured species detected shiny and mucoid. The strains behaved the same with previously in human faeces by direct PCR rDNA respect to all of the biochemical tests. Both isolates community analysis (Suau et al., 1999). The novel rod- produced acid from -glucose, -lactose, -maltose, shaped bacterium forms a distinct subline within the -galactose, amygdalin, -arabinose, aesculin, -fruc- Clostridium coccoides group of organisms. The cul- tose, -salicin, -sorbitol, sucrose, inositol, arbutin, β- tured relative nearest to the unknown bacterium, gentiobiose, lactulose, -glucosamine, -xylose, inulin Eubacterium formicigenerans, displays a sequence di- and fructo-oligosaccharides. The reactions with α-- vergence value of 6%, which shows clearly that they raffinose, xylitol, -lyxose and -arabitol were very represent closely related, albeit different, species (Fig. weak or doubtful. The organisms failed to produce 2). The generic assignment of the novel bacterium acid from -arabitol, -arabinose, -cellobiose, gly- is somewhat problematic, given the phylogenetic cogen, glycerol, erythritol, adonitol, methyl β-xyloside, inter-mixing of species from several genera within -sorbose, dulcitol, -fucose, -fucose, gluconate, 2- the Clostridium coccoides rRNA grouping. Prior to the ketogluconate, 5-ketogluconate, methyl α--manno- widespread adoption of 16S rRNA sequencing, the side, methyl α--glucoside, -turanose, -mannose, unknown rod-shaped bacterium would taxonomically -mannitol, α--melibiose, α--melezitose, N-acetyl- have been considered a member of the genus Eubac- glucosamine, pyruvate, -rhamnose, -ribose, starch, terium. It is, however, now universally acknowledged -trehalose, -tagatose or -xylose. Nitrate was not that eubacteria represent a phylogenetically very het- reduced. No liquefaction of gelatin or peptonization of erogeneous group of organisms, and there is a growing milk occurred. To ascertain further information about consensus that the genus Eubacterium sensu stricto the catabolic potential of the two isolates, resting cells should be restricted to the type species, Eubacterium in 50 mM anoxic potassium phosphate buffer (pH 7 0, " n limosum, and its close phylogenetic relatives (Willems 1 mg resazurin l− , 5 mM dithioerythritol) were incu- & Collins, 1996; Kageyama et al., 1999). The novel bated with glucose under N and CO and the # # rod-shaped bacterium described here is phylogene- degradation products were analysed. Resting cells tically far removed from Eubacterium limosum and converted glucose to ethanol, formate, acetate, hy- related species (approx. 20% 16S rRNA sequence drogen and CO . The fermentation balance was as # divergence; data not shown). Furthermore, the proper- follows (CO was calculated from the redox balance): # ties of the unknown bacterium are incompatible with 1 glucose ! 1 01 ethanol 0 71 acetate 1 70 for- n j n j n the definition of Eubacterium sensu stricto, because it mate 0 95 H 0 64 CO . j n #j n # does not produce butyrate and lactate as major The DNA base composition of the faecal isolates fermentation products. Consistent with 16S rRNA T 111-13A and 111-35 was respectively 43n8 and phylogenetic inferences, the unknown bacterium 45n6mol%GjC. The genotypic relatedness of the phenotypically closely resembles its nearest neighbour, isolates was further investigated by chromosomal Eubacterium formicigenerans. The unidentified species DNA–DNA pairing. The two strains displayed 96n7p and Eubacterium formicigenerans form a robust and 7n4% DNA–DNA relatedness, demonstrating clearly statistically significant subcluster (bootstrap value that they represent a single genomic species. To 100%) within the Clostridium coccoides group of determine the phylogenetic position of the unknown organisms. Furthermore, both species can be dis- organism, comparative 16S rRNA gene sequence tinguished phenotypically readily from all other mem- analysis was performed. The almost complete se- bers of the Clostridium coccoides group of organisms. quences (corresponding to positions 37–1448 of the For example, they can be distinguished from most Escherichia coli 16S rRNA gene) of strains 111-13A clostridial species in not producing endospores, from and 111-35T were determined. Pairwise analysis Lachnospira by the absence of curved cellular shapes, http://ijs.sgmjournals.org 425 D. Taras and others

...... Fig. 2. Unrooted tree showing the phylogenetic relationships of Dorea formicigenerans gen. nov., comb. nov., and Dorea longicatena sp. nov. with some other members of the Clostridium coccoides group. The tree, constructed using the neighbour-joining method, was based on a comparison of 1330 nucleotides. Bootstrap values, each expressed as a percentage of 500 replications, are given at branching points.

from Coprococcus by cellular morphology and end- produced by D. longicatena may be utilized by aceto- products of glucose fermentation and from Roseburia genic bacteria such as Clostridium coccoides or metha- and Butyrivibrio by the end-products of glucose meta- nogenic members of the Archaea. bolism (i.e. by not producing butyric acid) and by being non-motile. Therefore, on the basis of tree Description of Dorea gen. nov. topology and sequence divergence evidence, combined with their close phenotypic resemblance to each other Dorea (Do.reha. N.L. gen. n. Dorea of Joel Dore! ,in and the separateness of other members of the Clostri- honour of the French microbiologist, in recognition dium coccoides group, we believe that the novel faecal of his many contributions to gut microbiology). bacterium and Eubacterium formicigenerans merit classification in a novel genus, Dorea gen. nov., as Gram-positive-staining, non-spore-forming, non- Dorea longicatena sp. nov. and Dorea formicigenerans motile, rod-shaped cells. Obligately anaerobic and comb. nov. Tests that are useful in distinguishing catalase- and oxidase-negative. Glucose and some D. longicatena and D. formicigenerans from each other other sugars are fermented. The major end-products and from non-butyrate-producing Eubacterium species of glucose metabolism are ethanol, formate, acetate, are given in Table 1. H# and CO#; lactate may or may not be formed but butyrate is not produced. Gelatin and starch are not The novel rod-shaped bacterium D. longicatena de- hydrolysed. Nitrate is not reduced to nitrite. The scribed here represents a hitherto unknown member of GjC content of the DNA is 40–45n6 mol%. The the dominant human gut microflora. Experiments with type species of the genus is Dorea formicigenerans specific 16S rRNA-targeted oligonucleotides indicate (Holdeman and Moore 1974). that D. longicatena is present in all human volunteers * −" studied so far, at cell counts of approximately 10 g Description of Dorea formicigenerans comb. nov. (dry weight) faeces (D. Taras and M. Blaut, manuscript in preparation). This conforms with the isolation of Dorea formicigenerans (for.mi.ci.gehne.rans. N.L. adj. the two strains of D. longicatena in faecal dilutions of formicigenerans formic acid-producing; referring to its ) * 10− and 10− , respectively. Like the group of aceto- production of large amounts of formic acid from genic bacteria, D. longicatena probably contributes to carbohydrate fermentation). the daily production of approximately 10–30 g acetate Basonym: Eubacterium formicigenerans Holdeman in the human intestine (Roediger, 1980). Hydrogen, and Moore 1974. acetate and formate formed by D. longicatena as end- products can be used by other microbial populations Gram-positive rods that are obligately anaerobic, non- of the human gut ecosystem. For instance, formate motile and non-spore-forming. Cells are 0n6–1n4 µm

426 International Journal of Systematic and Evolutionary Microbiology 52 Dorea gen. nov., containing two species

Table 1. Characteristics useful in differentiating D. longicatena and D. formicigenerans from some other related species ...... Species are given as: 1, D. longicatena;2,D. formicigenerans;3,Eubacterium fissicatena;4,Eubacterium contortum; 5, Eubacterium hadrum;6,Collinsella aerofaciens;7,Eggerthella lenta;8,Eubacterium elligens. The characteristics of D. longicatena were determined in this study. Culturing was performed in ST medium as well as in HA medium supplemented with substrates. Data for other non-butyrate-producing species of the genus Eubacterium were taken from Moore & Holdeman Moore (1986). Reactions were scored as follows: k, negative; j, positive; , variable; k, mostly negative; , weak; k, mostly weak; , not reported.

Characteristic 1 2 3 4 5 6* 7† 8

Motility kkk kkkkj Utilization of: Amygdalin jk kk kkkk Arabinose j  k jk kkk Aesculin jk kk kk kk Inulin  kkk kk Mannose kk  jjkk Rhamnose kk j  kkkk Sorbitol jk k kk kkk Trehalose kk kk kk kk Raffinose  kkwkk kkk Melibiose kkkjkkkk Inositol jk j   kkk Maltose jj j jk jkk Glucose jjjjjjkk

* Kageyama et al. (1999) proposed the transfer of Eubacterium aerofaciens to the genus Collinsella gen. nov. as Collinsella aerofaciens gen. nov., comb. nov. † Wade et al. (1999) proposed the transfer of Eubacterium lentum to the genus Eggerthella gen. nov. as Eggerthella lenta gen. nov., comb. nov.

wide by 0n8–4n7 µm long and occur in pairs or chains. bose or trehalose. Hippurate hydrolysis is variable. On rumen fluid\glucose\cellobiose agar, colonies are Starch and gelatin are not hydrolysed. Nitrate is not 0n5–1n0 mm in diameter, white to tan, circular to reduced. Indole is not produced. Isolated from human lenticular and often have fuzzy edges or a woolly-ball faeces. The GjC content of DNA is within the range T appearance. Colonies on blood-agar plates are 0n5– 40–44 mol%. The type strain is ATCC 27755 . 3n0 mm in diameter, circular to slightly irregular, entire to slightly erose, convex to umbonate, opaque, white to tan, shiny and smooth. PYG broth cultures have Description of Dorea longicatena sp. nov. little or no turbidity, have a stringy or flocculent (oc- Dorea longicatena (lon.gi.ca.tehna. L. adj. longus long; casionally smooth) sediment and reach a pH of 4n7– L. fem. n. catena chain; N.L. fem. n. longicatena long 5n0 in 5 days. Growth in PYG is generally not affected chain, referring to the long chains that this organism by the addition of 0n02% Tween 80 or 10% (v\v) develops in culture medium). rumen fluid. Growth is inhibited by 6n5% NaCl. The optimum growth temperature is 37 mC; most strains Gram-positive rods that are obligately anaerobic, non- grow moderately well at 30 and 45 mC, but (usually) motile and non-spore-forming. Individual cells are not at 25 mC. Moderate to abundant gas is produced 0n5–0n6 µm wide by 2n0–4n3 µm long and occur in in glucose-agar deep cultures. Acetic, formic and chains of 4–200 cells. Sometimes cells from old cultures lactic acids and ethanol are the major products of stain Gram-negative. On Columbia blood and WCA glucose metabolism. Pyruvate is converted to acetate, agar, cells form white–opaque colonies that are formate and ethanol, usually with a trace of lactate 1–3 mm in diameter, circular, convex, smooth, shiny and sometimes succinate. Lactate and gluconate are and sticky. Non-haemolytic. Catalase- and oxidase- not utilized. Threonine is not converted to propi- negative. Ethanol, formate and acetate are the major onate. Acid is produced from fructose, galactose, glu- products of glucose metabolism. Acid is produced cose, lactose and maltose. Acid is not produced from from -glucose, -lactose, -maltose, -galactose, aesculin, adonitol, amygdalin, cellulose, dextrin, dulci- amygdalin, -arabinose, aesculin, -fructose, -salicin, tol, glycerol, glycogen, inulin, mannitol, mannose, -sorbitol, sucrose, inositol, arbutin, β-gentiobiose, melibiose, melezitose, rhamnose, salicin, sucrose, sor- lactulose, -glucosamine, -xylose, inulin, fructo- http://ijs.sgmjournals.org 427 D. Taras and others oligosaccharides, α--raffinose (weak), xylitol (weak), Hungate, R. E. (1969). A roll tube method for cultivation of strict -lyxose (weak) and -arabitol (weak). Acid is not anaerobes. Methods Microbiol 3B, 117–132. produced from -arginine, -arabitol, -arabinose, Johnson, J. L. (1994). Similarity analysis of DNAs. In Methods for General and Molecular Bacteriology, pp. 655–682. Edited by P. -cellobiose, glycogen, glycerol, erythritol, adonitol, Gerhardt, R. G. E. Murray, W. A. Wood & N. R. Krieg. Washington, methyl β-xyloside, -sorbose, dulcitol, -fucose, glu- DC: American Society for Microbiology. conate, 2-ketogluconate, 5-ketogluconate, methyl α- Kageyama, A., Benno, Y. & Nakase, T. (1999). Phylogenetic and -mannoside, methyl α--glucoside, -turanose, - phenotypic evidence for the transfer of Eubacterium aerofaciens to the mannose, -mannitol, α--melibiose, α--melezitose, genus Collinsella as Collinsella aerofaciens gen. nov., comb. nov. Int J N-acetylglucosamine, pyruvate, -rhamnose, -ribose, Syst Bacteriol 49, 557–565. starch, -trehalose, -tagatose or -xylose. Nitrate is Kamlage, B., Gruhl, B. & Blaut, M. (1997). Isolation and characteriz- not reduced. Gelatin and hippurate are not hydrolysed. ation of two new homoacetogenic hydrogen-utilizing bacteria from the human intestinal tract that are closely related to Clostridium coccoides. Isolated from human faeces. The GjC content of the T Appl Environ Microbiol 63, 1732–1738. DNA is 43n8–45n6 mol%. The type strain is 111-35 T T T Langendijk, P. S., Schut, F., Jansen, G. J., Raangs, G. C., Kamphuis, (l DSM 13814 l CCUG 45247 l JCM 11232 ). G., Wilkinson, M. H. F. & Welling, G. W. (1995). Quantitative fluorescence in situ hybridization of Bifidobacterium spp. with genus- specific 16S rRNA-targeted probe and its application in fecal samples. Appl Environ Microbiol 61, 3069–3075. Acknowledgements Mesbah, M., Premachandran, U. & Whitman, W. B. (1989). Precise measurement of the GjC content of deoxyribonucleic acid by high- We thank Ba$ rbel Gruhl and Verena Eckert-Funke for performance liquid chromatography. Int J Syst Bacteriol 39, 159–167. technical assistance. We are grateful to Lynne Rogers-Blaut Moore, W. E. C. & Holdeman, L. V. (1974). Human fecal flora: the for critical reading of the manuscript. This study was normal flora of 20 Japanese-Hawaiians. Appl Microbiol 27, 961–979. supported by grants from the Deutsche Forschungsgemein- Moore, W. E. C. & Holdeman Moore, L. V. (1986). Genus Eubac- schaft (BL 257\6-1, 6-2) and the European Union (FAIR terium Pre! vot 1938. In Bergey’s Manual of Systematic Bacteriology, vol. CT97-3035). 2, pp. 1353–1373. Edited by P. H. A. Sneath, N. S. Mair, M. E. Sharpe & J. G. Holt. Baltimore: Williams & Wilkins. Roediger, W. E. W. (1980). Role of anaerobic bacteria in the metabolic welfare of the colonic mucosa in man. Gut 21, 793–798. References Schwiertz, A., Le Blay, G. & Blaut, M. (2000). Quantification of different Eubacterium spp. in human fecal samples with species-specific Bergmeyer, J. & Graßl, M. (1984). Methods of Enzymatic Analysis, 16S rRNA-targeted oligonucleotide probes. Appl Environ Microbiol 66, 3rd edn, vol. 6. Weinheim: Verlag Chemie. 375–382. Bryant, M. P. (1972). Commentary on the Hungate technique for Smibert, R. M. & Krieg, N. R. (1994). Phenotypic characterization. In culture of anaerobic bacteria. Am J Clin Nutr 25, 1324–1328. Methods for General and Molecular Bacteriology, pp. 611–651. Edited by P. Gerhardt, R. G. E. Murray, W. A. Wood & N. R. Krieg. Collins, M. D., Lawson, P. A., Willems, A., Cordoba, J. J., Fernan- Washington, DC: American Society for Microbiology. dez-Garayzabal, J., Garcia, P., Cai, J., Hippe, H. & Farrow, J. A. E. (1994). The phylogeny of the genus Clostridium: proposal of five new Suau, A., Bonnet, R.,! Sutren, M., Godon, J.-J., Gibson, G. R., Collins, M. D. & Dore, J. (1999). Direct analysis of genes encoding 16S genera and eleven new species combinations. Int J Syst Bacteriol 44, rRNA from complex communities reveals many novel molecular species 812–826. within the human gut. Appl Environ Microbiol 65, 4799–4807. Felsenstein, J. (1989).  – phylogeny inference package (version Wade, W. G., Downes, J., Dymock, D., Hiom, S. J., Weightman, 3.2). Cladistics 5, 164–166. A. J., Dewhirst, F. E., Paster, B. J., Tzellas, N. & Coleman, B. (1999). Finegold, S. M., Attebery, H. R. & Sutter, V. L. (1974). Effect of diet The family Coriobacteriaceae: reclassification of Eubacterium exiguum on human fecal flora: comparison of Japanese and American diets. Am (Poco et al. 1996) and Peptostreptococcus heliotrinreducens (Lanigan J Clin Nutr 27, 1456–1469. 1976) as Slackia exigua gen. nov., comb. nov. and Slackia heliotrini- reducens gen. nov., comb. nov., and Eubacterium lentum (Prevot 1938) Grund, S., Nattermann, H. & Horsch, F. (1995). Zum elektronen- as Eggerthella lenta gen. nov., comb. nov. Int J Syst Bacteriol 49, mikroskipischen Spirochaten-Nachwies bei der Dermatitis digitalis des 595–600. Rindes. J Vet Med B 42, 533–542. Willems, A. & Collins, M. D. (1996). Phylogenetic relationships of the Hartmann, L., Taras, D., Kamlage, B. & Blaut, M. (2000). A new genera Acetobacterium and Eubacterium sensu stricto and reclassifi- technique to determine hydrogen excreted by gnotobiotic rats. Lab cation of Eubacterium alactolyticum as Pseudoramibacter alactolyticus Anim 34, 162–170. gen. nov., comb. nov. Int J Syst Bacteriol 46, 1083–1087. Holdeman, L. V. & Moore, W. E. C. (1974). New genus, Coprococcus, Zoetendal, E. G., Akkermans, A. D. L. & De Vos, W. M. (1998). twelve new species, and emended descriptions of four previously Temperature gradient gel electrophoresis analysis of 16S rRNA from described species of bacteria from human feces. Int J Syst Bacteriol 24, human fecal samples reveals stable and host-specific communities of 260–277. active bacteria. Appl Environ Microbiol 64, 3854–3859.

428 International Journal of Systematic and Evolutionary Microbiology 52