INTERNATIONAL JOURNAL OF SYSTEMATIC BACTERIOLOGY, Apr. 1992, p. 312-314 0020-7713/92/020312-O3$02.00/0 Copyright 0 1992, International Union of Microbiological Societies
Vol. 42, No. 2
Phylogenetic Analysis of the Pathogenic Anaerobe Clostridium perjhngens Using the 16s rRNA Nucleotide Sequence BRUNO CANARD,^ THIERRY GARNIER,~BENEDICTELA FAY,^ RICHARD CHRISTEN,’ AND STEWART T. COLE1* Laboratoire de Gtnttique Molkculaire Bactkrienne, Institut Pasteur, 28 rue du Docteur R o u , 75724 Paris Cedex 15,’ and Biologie Cellulaire Marine URA 671 CNRS, Universitk Pierre & Marie Curie, Station Zoologique, 06230 Kllefranche sur Mer, France Clostridium perpingens, the first pathogenic clostridium examined, was placed in the nonmycoplasma subgroup of the low-dG+dC-content gram-positive cluster on the basis of the results of a phylogenetic analysis in which we used 16s rRNA comparisons. The closest relative that has been identified to date is Clostridium pasteurianum
.
rRNA sequence of C. perfrzngens with all previously published rRNA eubacterial sequences (about 370 sequences aligned on the basis of conserved domains and secondary structure). Molecular cloning and sequencing of the complete 16s rRNA gene, rrs, from the rmB operon have been described recently (1, 5) (GenBank accession number M69264). In a first approach, conserved domains (positions 1 to 67, 101 to 150,171 to 180,220 to 436,500 to 577,752 to 820,880 to 995, 1044 to 1117, 1156 to 1238, and 1299 to 1429 of the Escherichia coli sequence) were aligned, and a phylogenetic analysis was performed; in this analysis representatives of all of the eubacterial phyla (and their subdivisions) as defined by Woese (15) were included. The results of this study, in which the neighbor-joining distance matrix method of Saitou and Nei (9) was used, led to the affiliation of C. perfnngens with the low-dG+dC-content subgroup of the gram-positive phylum as defined by Woese and coworkers. Additional phylogenetic studies that were restricted to sequences of bacteria belonging to this group showed that C. perfnngens is part of a distinct cluster that contains several other Clostridium species (Fig. 1). The alignment of the rRNA sequence of C. perfnngens was then refined in the most variable domains for bacterial sequences belonging to this cluster. The apparently monophyletic Clostridium cluster containing C. pasteurianum (Fig. 1)was consistently located as an early radiation at the base of the low-dG+dC-content cluster. Both this early radiation and the fact that the dG+dC contents of these clostridia have been reported to vary from 24 mol% (C. perfrzngens) to 58 mol% (C. barkeri [ll]) suggest that these clostridia could be placed in an independent group that is closely related to the low-dG+dC-content bacterial cluster. The grouping of these species in a single monophyletic unit can also be inferred from a striking molecular signature in the form of a loop deletion at the positions equivalent to positions 455 to 479 of the E. coli sequence (a region not included in the phylogenetic analysis), whereas none of the previously published sequences for bacteria belonging to the low-dG+dC-content cluster had such a deletion. This was true even for C. barkeri, although this organism does not belong to any of the homology groups of Johnson and Francis (6) and shares two characteristics (acetate synthesis and murein structure) with Eubacterium sp. (12, 13). In contrast, this deletion is present in bacteria belonging to other subgroups (for example, Heliobacterium
Clostridia are obligately anaerobic, gram-positive, sporeforming, rod-shaped bacteria that belong to a large and somewhat heterogeneous taxonomic group. These longstudied eubacteria have interesting properties, ranging from organic solvent production (concomitant with intriguing metabolic developmental stages) to devastating clinical conditions caused by the synthesis of some of the deadliest toxins known. Within the genus Clostridium, this heterogeneity is reflected by various physiological and biochemical characteristics, including ambiguous Gram stain reactions and genomic dG+dC contents that vary from 24 to 58 mol% (2). Recent work based on 16srRNA analyses has indicated that the clostridia make up a major subgroup within the grampositive eubacterial phylum (4, 15). Clostridium perjhngens is a commensal of humans and animals. The natural habitat of this organism is intestinal tracts, and under appropriate circumstances the bacterium can produce a wide variety of illnesses, ranging from enterotoxemia to the more serious condition invasive gas gangrene (8). Pathogenicity has been attributed mainly to the release of highly destructive exotoxins, which constitute the basis for dividing the species into serotypes (3, 8). Recently, a genome map has been constructed for the type A reference strain, C. perfnngens CPN5O (1). Among the genes and functional loci identified on the 3.6-megabase genome were the operons that encode rRNA, and these operons were subsequently characterized at the molecular level (5). On the basis of 23s rRNA homologies and the results of 16s rRNA cataloging analyses (4, 6, 10, 15), the genus Clostridium has been divided into four clusters (clusters I through IV) (4,6,10,15), which lack obvious cluster-specific physiological or biochemical traits. However, on the basis of the results of 165 rRNA sequence analyses, none of the three major pathogenic clostridia, C. botulinum (cluster I), C. p e m n g e n s (cluster I), and C. tetani (cluster 11), has been placed in one of these subgroups; this is also true for the nonhistotoxic pathogens (e.g., C. sporogenes [cluster I], C. dijqicile, and C. tertium [cluster 111) which occasionally occur in patients with similar necrotic diseases. To begin to address the question of evolutionary relatedness among these important pathogens, we analyzed and compared the 16s
* Corresponding author. 312
NOTES
VOL. 42, 1992
313
H . chlorum S . ruminantium
c.l ( U I K I S u m I
C . inocuum i
.
I
C . sticklandii I
C . amimvalericum
chlorum).The sequence alignments in this region are shown in Fig. 2 (equivalent to E. coli positions 455 to 479). Two sequences of Clostn'dium species belonging to another cluster were included as evidence of a specific deletion. Table 1 shows the evolutionary distances calculated for each sequence pair, and these distances were used to construct the tree in Fig. 1by using the neighbor-joining procedure. Both types of data, as well as direct visualization of the aligned sequences, show clearly that C. pasterianum is the closest relative of C. perfhngens that has been identified to date. Similar conclusions were drawn from the results of compe-
tition hybridization experiments in which 23s rRNA was used as a probe by Johnson and Francis (6). Our findings provide the first phylogenetic classification based on an analysis of the 1 6 s rRNA sequence of an important human pathogen belonging to the genus Clostridium. Our results are in agreement with those of previous workers (6, 12, 13) and, taken together with the data of Tanner et al. (12,13), exclude the C. barkeri branch from the clostridial tree. All four homology groups (clusters I through IV) that are based on both 23s rRNA reassociation data and phenotypic traits have representatives in the true clostridial
E.coli -
GGAGGAAGGG AGUAAAGUUA AUACCUUUGC UCAUUGACGU UACCCGCA -__---___----*****G *****AAG C.perfringens **+A**UAA- - - _ - - _ _ _ - _
-
C .pasteur.
*+*C**UAA- - - - - - - - - - - - - - - - - - - - - ----*****G ****AAAG
C . barkeri
***A**f*AA GA-------- - - - - - - - - - - ----*****G *****AAU
C.stickla. -~
***A**UAAU - - _ - - - - - - - - _ _ - - - - - - ------****G ****AUAG
C.aminova. -~
***C**UAA- - - - - - _ - - - - - - - - - - - _ - - - - - - * * f * * G
****U*AC
C.mayombe.
***A**UAA- _ _ _ - - _ - - _ -
* * *UU*AG
C ._ innocu. __
U**A***C** CUC*U*GAGG *A*UGC*AUG GG*G*****G **G*UUAC ***A***C** C*GCU*CAGG *A*UGG*A** CG*G*****G ***UUUAU
C . ramosum -~
---__-_---
----*****G
FIG. 2. Sequence alignment around the region from position 455 to position 479 (E. coli numbering). The C. innocuum and C. ramosum sequences were aligned with the E. coli region in order to fit secondary-structure requirements; the complementary domains of the additional helix (boldface type) are aligned within these three sequences. Deletions are indicated by dashes, and identical nucleotides are indicated by asterisks. C. pasteur., C. pasteurianum; C. stickla., C. sticklandii; C. aminova., C. aminovalencum; C. mayombe, C. mayombeii; C. innocu., C. innocuum.
314
INT. J. S Y S T . BACTERIOL.
NOTES TABLE 1. Evolutionary distances between gram-positive bacterial speciesa
~~
Evolutionary distance from:
Species
Bacillus subtilis Clostridium aminovalericum Clostridium barkeri Clostridium innocuum Clostridium mayombeii Clostridium pasteunanum Clostridium perjhngens Clostridium ramosum Clostndium sticklandii Heliobacter chlorum Lactococcus gawieae Leuconostoc cremoris Mycoplasma capricolum Selenomonas ruminantium
13.3 14.9 15.4 12.0 16.7 16.4 16.1 10.5 15.6 16.9 14.7 16.7 15.3 17.5
13.4 12.4 11.7 12.5 11.5 11.1 10.0 12.0 12.5 10.9 12.3 14.5 11.6
13.1 13.9 10.4 11.8 10.5 13.0 10.5 14.3 16.2 13.7 16.8 12.6
13.7 11.6 12.1 11.6 13.6 11.9 14.0 14.3 16.5 19.1 12.5
14.6 14.2 13.9 9.9 15.0 15.5 12.8 14.4 13.1 13.2
9.6 8.1 15.1 6.5 14.9 14.7 16.3 15.7 13.5
5.9 13.2 9.0 14.1 15.2 15.3 15.2 13.2
11.7 7.8 15.0 13.6 13.1 13.9 13.2
14.2 13.9 12.1 13.9 13.8 13.3
14.0 15.7 15.4 15.5 13.3
14.0 15.8 18.6 11.1
10.7 15.9 13.2
15.4 13.8
17.5
Evolutionary distances were calculated by taking into account the positions that were aligned with certainty and showed no signs of multiple substitutions (i.e., one base accounted for 50% of the positions considered) and by using the Jukes-Cantor formula (7). a
branch of our tree. However, C. rumosum (cluster 111) and C. innocuum (cluster IV) are not included, a finding apparent in previous studies (6, 12-14), because of their lack of 23s rRNA homology with the other bacteria in their respective groups. Financial support was provided by the Institut Pasteur and by grant CRE 883003 from the Institut National de la SantC et de la Recherche MCdicale to S.T.C., as well as by funds from the Association Recherche et Partage, Association pour la Recherche contre le Cancer, Centre National de la Recherche Scientifique to R.C. REFERENCES 1. Canard, B., and S. T. Cole. 1989. Genome organization of the anaerobic pathogen Clostridium perjhngens. Proc. Natl. Acad. Sci. USA 86:6676-6680. 2. Cato, E. P., W. L. George, and S. M. Finegold. 1986. Clostridium, p. 1141-1200. In P. H. A. Sneath, N. S. Mair, M. E. Sharpe, and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 2. The Williams & Wilkins Co., Baltimore. 3. Finegold, S. 1977. Anaerobic bacteria in human disease. Academic Press, New York. 4. Fox, G. E., E. Stackebrandt, R. B. Hespell, J. Gibson, J. Maniloff, T. A. Dyer, R. S. Wolfe, W. E. Balch, R. S. Tanner, L. J. Magrum, L. B. Zablen, R. Blakemore, R. Gupta, L. Bonen, B. J. Lewis, D. A. Stahl, K. R. Luehrsen, K. N. Chen, and C. R. Woese. 1980. The phylogeny of prokaryotes. Science 209:457463. 5. Garnier, T., B. Canard, and S. T. Cole. 1991. Cloning, mapping, and molecular characterization of the ribosomal RNA operons of Clostndium perjhngens. J. Bacteriol. 1735434-5438.
6. Johnson, J. L., and B. S. Francis. 1975. Taxonomy of the clostridia: ribosomal ribonucleic acid homologies among species. J. Gen. Microbiol. 88:229-244. 7. Jukes, T. H., and C. R. Cantor. 1969. Evolution of protein molecules, p. 21-132. In H. N. Munro (ed.), Mammalian protein metabolism. Academic Press, Inc., New York. 8. Rood, J. I., and S. T. Cole. 1991. Molecular genetics and pathogenesis of Clostridium perjhngens. Microbiol. Rev. 55: 621-648. 9. Saitou, N., and M. Nei. 1987. A neighbour joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 44406425. 10. Stackebrandt, E., and M. Teuber. 1988. Molecular taxonomy and phylogenetic position of lactic acid bacteria. Biochimie 70:3 17-324. 11. Stadtman, E. R., T. C. Stadtman, I. Pastan, and L. D. S. Smith. 1972. Clostridium barkeri sp. nov. J. Bacteriol. 110:758-760. 12. Tanner, R. S., E. Stackebrandt, G. E. Fox, R. Gupta, L. J. Magrum, and C. R. Woese. 1982. A phylogenetic analysis of anaerobic eubacteria capable of synthesizing acetate from carbon dioxide. Curr. Microbiol. 7:127-132. 13. Tanner, R. S., E. Stackebrandt, G. E. Fox, and C. R. Woese. 1981. A phylogenetic analysis of Acetobacter woodii, Clostridium barkeri, Clostridium butyricum , Clostndium lituseburense, Eubacterium limosum, and Eubacterium tenue. Curr. Microbiol. 935-38. 14. Weisburg, W. G., J. G. Tully, D. L. Rose, J. P. Petzel, H. Oyaizu, D. Yang, L. Mandelco, J. Sechrest, T. G. Lawrence, J. van Etten, and C. R. Woese. 1989. A phylogenetic analysis of the mycoplasmas: basis for their classification. J. Bacteriol. 171: 6455-6467. 15. Woese, C. R. 1987. Bacterial evolution. Microbiol. Rev. 51:221271.
Vol. 42, No. 2
Phylogenetic Analysis of the Pathogenic Anaerobe Clostridium perjhngens Using the 16s rRNA Nucleotide Sequence BRUNO CANARD,^ THIERRY GARNIER,~BENEDICTELA FAY,^ RICHARD CHRISTEN,’ AND STEWART T. COLE1* Laboratoire de Gtnttique Molkculaire Bactkrienne, Institut Pasteur, 28 rue du Docteur R o u , 75724 Paris Cedex 15,’ and Biologie Cellulaire Marine URA 671 CNRS, Universitk Pierre & Marie Curie, Station Zoologique, 06230 Kllefranche sur Mer, France Clostridium perpingens, the first pathogenic clostridium examined, was placed in the nonmycoplasma subgroup of the low-dG+dC-content gram-positive cluster on the basis of the results of a phylogenetic analysis in which we used 16s rRNA comparisons. The closest relative that has been identified to date is Clostridium pasteurianum
.
rRNA sequence of C. perfrzngens with all previously published rRNA eubacterial sequences (about 370 sequences aligned on the basis of conserved domains and secondary structure). Molecular cloning and sequencing of the complete 16s rRNA gene, rrs, from the rmB operon have been described recently (1, 5) (GenBank accession number M69264). In a first approach, conserved domains (positions 1 to 67, 101 to 150,171 to 180,220 to 436,500 to 577,752 to 820,880 to 995, 1044 to 1117, 1156 to 1238, and 1299 to 1429 of the Escherichia coli sequence) were aligned, and a phylogenetic analysis was performed; in this analysis representatives of all of the eubacterial phyla (and their subdivisions) as defined by Woese (15) were included. The results of this study, in which the neighbor-joining distance matrix method of Saitou and Nei (9) was used, led to the affiliation of C. perfnngens with the low-dG+dC-content subgroup of the gram-positive phylum as defined by Woese and coworkers. Additional phylogenetic studies that were restricted to sequences of bacteria belonging to this group showed that C. perfnngens is part of a distinct cluster that contains several other Clostridium species (Fig. 1). The alignment of the rRNA sequence of C. perfnngens was then refined in the most variable domains for bacterial sequences belonging to this cluster. The apparently monophyletic Clostridium cluster containing C. pasteurianum (Fig. 1)was consistently located as an early radiation at the base of the low-dG+dC-content cluster. Both this early radiation and the fact that the dG+dC contents of these clostridia have been reported to vary from 24 mol% (C. perfrzngens) to 58 mol% (C. barkeri [ll]) suggest that these clostridia could be placed in an independent group that is closely related to the low-dG+dC-content bacterial cluster. The grouping of these species in a single monophyletic unit can also be inferred from a striking molecular signature in the form of a loop deletion at the positions equivalent to positions 455 to 479 of the E. coli sequence (a region not included in the phylogenetic analysis), whereas none of the previously published sequences for bacteria belonging to the low-dG+dC-content cluster had such a deletion. This was true even for C. barkeri, although this organism does not belong to any of the homology groups of Johnson and Francis (6) and shares two characteristics (acetate synthesis and murein structure) with Eubacterium sp. (12, 13). In contrast, this deletion is present in bacteria belonging to other subgroups (for example, Heliobacterium
Clostridia are obligately anaerobic, gram-positive, sporeforming, rod-shaped bacteria that belong to a large and somewhat heterogeneous taxonomic group. These longstudied eubacteria have interesting properties, ranging from organic solvent production (concomitant with intriguing metabolic developmental stages) to devastating clinical conditions caused by the synthesis of some of the deadliest toxins known. Within the genus Clostridium, this heterogeneity is reflected by various physiological and biochemical characteristics, including ambiguous Gram stain reactions and genomic dG+dC contents that vary from 24 to 58 mol% (2). Recent work based on 16srRNA analyses has indicated that the clostridia make up a major subgroup within the grampositive eubacterial phylum (4, 15). Clostridium perjhngens is a commensal of humans and animals. The natural habitat of this organism is intestinal tracts, and under appropriate circumstances the bacterium can produce a wide variety of illnesses, ranging from enterotoxemia to the more serious condition invasive gas gangrene (8). Pathogenicity has been attributed mainly to the release of highly destructive exotoxins, which constitute the basis for dividing the species into serotypes (3, 8). Recently, a genome map has been constructed for the type A reference strain, C. perfnngens CPN5O (1). Among the genes and functional loci identified on the 3.6-megabase genome were the operons that encode rRNA, and these operons were subsequently characterized at the molecular level (5). On the basis of 23s rRNA homologies and the results of 16s rRNA cataloging analyses (4, 6, 10, 15), the genus Clostridium has been divided into four clusters (clusters I through IV) (4,6,10,15), which lack obvious cluster-specific physiological or biochemical traits. However, on the basis of the results of 165 rRNA sequence analyses, none of the three major pathogenic clostridia, C. botulinum (cluster I), C. p e m n g e n s (cluster I), and C. tetani (cluster 11), has been placed in one of these subgroups; this is also true for the nonhistotoxic pathogens (e.g., C. sporogenes [cluster I], C. dijqicile, and C. tertium [cluster 111) which occasionally occur in patients with similar necrotic diseases. To begin to address the question of evolutionary relatedness among these important pathogens, we analyzed and compared the 16s
* Corresponding author. 312
NOTES
VOL. 42, 1992
313
H . chlorum S . ruminantium
c.l ( U I K I S u m I
C . inocuum i
.
I
C . sticklandii I
C . amimvalericum
chlorum).The sequence alignments in this region are shown in Fig. 2 (equivalent to E. coli positions 455 to 479). Two sequences of Clostn'dium species belonging to another cluster were included as evidence of a specific deletion. Table 1 shows the evolutionary distances calculated for each sequence pair, and these distances were used to construct the tree in Fig. 1by using the neighbor-joining procedure. Both types of data, as well as direct visualization of the aligned sequences, show clearly that C. pasterianum is the closest relative of C. perfhngens that has been identified to date. Similar conclusions were drawn from the results of compe-
tition hybridization experiments in which 23s rRNA was used as a probe by Johnson and Francis (6). Our findings provide the first phylogenetic classification based on an analysis of the 1 6 s rRNA sequence of an important human pathogen belonging to the genus Clostridium. Our results are in agreement with those of previous workers (6, 12, 13) and, taken together with the data of Tanner et al. (12,13), exclude the C. barkeri branch from the clostridial tree. All four homology groups (clusters I through IV) that are based on both 23s rRNA reassociation data and phenotypic traits have representatives in the true clostridial
E.coli -
GGAGGAAGGG AGUAAAGUUA AUACCUUUGC UCAUUGACGU UACCCGCA -__---___----*****G *****AAG C.perfringens **+A**UAA- - - _ - - _ _ _ - _
-
C .pasteur.
*+*C**UAA- - - - - - - - - - - - - - - - - - - - - ----*****G ****AAAG
C . barkeri
***A**f*AA GA-------- - - - - - - - - - - ----*****G *****AAU
C.stickla. -~
***A**UAAU - - _ - - - - - - - - _ _ - - - - - - ------****G ****AUAG
C.aminova. -~
***C**UAA- - - - - - _ - - - - - - - - - - - _ - - - - - - * * f * * G
****U*AC
C.mayombe.
***A**UAA- _ _ _ - - _ - - _ -
* * *UU*AG
C ._ innocu. __
U**A***C** CUC*U*GAGG *A*UGC*AUG GG*G*****G **G*UUAC ***A***C** C*GCU*CAGG *A*UGG*A** CG*G*****G ***UUUAU
C . ramosum -~
---__-_---
----*****G
FIG. 2. Sequence alignment around the region from position 455 to position 479 (E. coli numbering). The C. innocuum and C. ramosum sequences were aligned with the E. coli region in order to fit secondary-structure requirements; the complementary domains of the additional helix (boldface type) are aligned within these three sequences. Deletions are indicated by dashes, and identical nucleotides are indicated by asterisks. C. pasteur., C. pasteurianum; C. stickla., C. sticklandii; C. aminova., C. aminovalencum; C. mayombe, C. mayombeii; C. innocu., C. innocuum.
314
INT. J. S Y S T . BACTERIOL.
NOTES TABLE 1. Evolutionary distances between gram-positive bacterial speciesa
~~
Evolutionary distance from:
Species
Bacillus subtilis Clostridium aminovalericum Clostridium barkeri Clostridium innocuum Clostridium mayombeii Clostridium pasteunanum Clostridium perjhngens Clostridium ramosum Clostndium sticklandii Heliobacter chlorum Lactococcus gawieae Leuconostoc cremoris Mycoplasma capricolum Selenomonas ruminantium
13.3 14.9 15.4 12.0 16.7 16.4 16.1 10.5 15.6 16.9 14.7 16.7 15.3 17.5
13.4 12.4 11.7 12.5 11.5 11.1 10.0 12.0 12.5 10.9 12.3 14.5 11.6
13.1 13.9 10.4 11.8 10.5 13.0 10.5 14.3 16.2 13.7 16.8 12.6
13.7 11.6 12.1 11.6 13.6 11.9 14.0 14.3 16.5 19.1 12.5
14.6 14.2 13.9 9.9 15.0 15.5 12.8 14.4 13.1 13.2
9.6 8.1 15.1 6.5 14.9 14.7 16.3 15.7 13.5
5.9 13.2 9.0 14.1 15.2 15.3 15.2 13.2
11.7 7.8 15.0 13.6 13.1 13.9 13.2
14.2 13.9 12.1 13.9 13.8 13.3
14.0 15.7 15.4 15.5 13.3
14.0 15.8 18.6 11.1
10.7 15.9 13.2
15.4 13.8
17.5
Evolutionary distances were calculated by taking into account the positions that were aligned with certainty and showed no signs of multiple substitutions (i.e., one base accounted for 50% of the positions considered) and by using the Jukes-Cantor formula (7). a
branch of our tree. However, C. rumosum (cluster 111) and C. innocuum (cluster IV) are not included, a finding apparent in previous studies (6, 12-14), because of their lack of 23s rRNA homology with the other bacteria in their respective groups. Financial support was provided by the Institut Pasteur and by grant CRE 883003 from the Institut National de la SantC et de la Recherche MCdicale to S.T.C., as well as by funds from the Association Recherche et Partage, Association pour la Recherche contre le Cancer, Centre National de la Recherche Scientifique to R.C. REFERENCES 1. Canard, B., and S. T. Cole. 1989. Genome organization of the anaerobic pathogen Clostridium perjhngens. Proc. Natl. Acad. Sci. USA 86:6676-6680. 2. Cato, E. P., W. L. George, and S. M. Finegold. 1986. Clostridium, p. 1141-1200. In P. H. A. Sneath, N. S. Mair, M. E. Sharpe, and J. G. Holt (ed.), Bergey's manual of systematic bacteriology, vol. 2. The Williams & Wilkins Co., Baltimore. 3. Finegold, S. 1977. Anaerobic bacteria in human disease. Academic Press, New York. 4. Fox, G. E., E. Stackebrandt, R. B. Hespell, J. Gibson, J. Maniloff, T. A. Dyer, R. S. Wolfe, W. E. Balch, R. S. Tanner, L. J. Magrum, L. B. Zablen, R. Blakemore, R. Gupta, L. Bonen, B. J. Lewis, D. A. Stahl, K. R. Luehrsen, K. N. Chen, and C. R. Woese. 1980. The phylogeny of prokaryotes. Science 209:457463. 5. Garnier, T., B. Canard, and S. T. Cole. 1991. Cloning, mapping, and molecular characterization of the ribosomal RNA operons of Clostndium perjhngens. J. Bacteriol. 1735434-5438.
6. Johnson, J. L., and B. S. Francis. 1975. Taxonomy of the clostridia: ribosomal ribonucleic acid homologies among species. J. Gen. Microbiol. 88:229-244. 7. Jukes, T. H., and C. R. Cantor. 1969. Evolution of protein molecules, p. 21-132. In H. N. Munro (ed.), Mammalian protein metabolism. Academic Press, Inc., New York. 8. Rood, J. I., and S. T. Cole. 1991. Molecular genetics and pathogenesis of Clostridium perjhngens. Microbiol. Rev. 55: 621-648. 9. Saitou, N., and M. Nei. 1987. A neighbour joining method: a new method for reconstructing phylogenetic trees. Mol. Biol. Evol. 44406425. 10. Stackebrandt, E., and M. Teuber. 1988. Molecular taxonomy and phylogenetic position of lactic acid bacteria. Biochimie 70:3 17-324. 11. Stadtman, E. R., T. C. Stadtman, I. Pastan, and L. D. S. Smith. 1972. Clostridium barkeri sp. nov. J. Bacteriol. 110:758-760. 12. Tanner, R. S., E. Stackebrandt, G. E. Fox, R. Gupta, L. J. Magrum, and C. R. Woese. 1982. A phylogenetic analysis of anaerobic eubacteria capable of synthesizing acetate from carbon dioxide. Curr. Microbiol. 7:127-132. 13. Tanner, R. S., E. Stackebrandt, G. E. Fox, and C. R. Woese. 1981. A phylogenetic analysis of Acetobacter woodii, Clostridium barkeri, Clostridium butyricum , Clostndium lituseburense, Eubacterium limosum, and Eubacterium tenue. Curr. Microbiol. 935-38. 14. Weisburg, W. G., J. G. Tully, D. L. Rose, J. P. Petzel, H. Oyaizu, D. Yang, L. Mandelco, J. Sechrest, T. G. Lawrence, J. van Etten, and C. R. Woese. 1989. A phylogenetic analysis of the mycoplasmas: basis for their classification. J. Bacteriol. 171: 6455-6467. 15. Woese, C. R. 1987. Bacterial evolution. Microbiol. Rev. 51:221271.
