The genus Clostridium was proposed in 1880 by Prazmowski with the type species C. butryicum^[1]. Subsequently the genus Clostridium became a general depository for Gram-positive staining, sporeforming, anaerobic organisms. There are approximately 228 Clostridium species and subspecies with validly published names^[2] (http://www.bacterio.net). The species of the genus Clostridium are extremely heterogeneous, with many species displaying a wide range of phenotypes including psychrophiles, thermophiles and acidophiles, organisms that synthesize cytochromes and quinones. In addition, the G+C content of chromosomal DNA ranges from approximately 21% to 54% which is considered to be too extensive a range for a single genus^[3-5]. Johnson and Francis^[6] began to pioneer molecular insights into the considerable diversity within the genus using DNA-rRNA pairing, this was later expanded using 16S rRNA oligonucleotide cataloguing studies^[7-9]. It was not until the advent of the complete (or almost complete) sequencing of the 16S rRNA gene that the true extent of the diversity of the genus and the interrelationships with other taxa was realized. Sequencing studies demonstrated that many Clostridium species were phylogenetically distantly related to the type species of the genus, Clostridium butyricum and in addition that species of a number of other genera were closely related to this type species^[10-12]. Collectively, these studies established that the genus Clostridium was in need of revision and the first comprehensive effort was attempted by Collins and co-workers^[13] who proposed a possible hierarchical framework for the classification of clostridia. A total of 19 clusters or groups, (cluster I being recognized as comprising the “true” representatives of the genus Clostridium sensu stric to based on it containing the type species of the genus Clostridium, C. butryicum) were identified leading to the description of five new genera and proposal of eleven new species combinations^[13]. Successive studies supported and expanded this framework^{[5, 14-15]} (Figure 1) with many organisms formally recognized as Clostridium species being transferred to new or established genera^{[3, 16-23]}. However, the reclassification of organisms that fall outside the Clostridium cluster I continues and the reader should consult List of Prokaryotic Names with Standing in Nomenclature^[2] (http://www.bacterio.cict.fr/), which is updated regularly on publication of the International Journal of Systematic and Evolutionary Microbiology (IJSEM).

Figure 1　Neighbor-joining phylogenetic tree showing the relationship of species with Clostridium sensu stricto (rRNA clostridial cluster I) and some other members of the family Clostridiaceae The tree was constructed using the neighbor-joining method based on the pairwise comparison of approximately 1 340 nucleotides. Atopobium parvulum was used as the out-group. Bootstrap values (>90%), expressed as a percentage of 1 000 replications. The scale bar indicates 1% sequence divergence.

图选项

In addition to phylogenetic studies based on comparisons of 16S rRNA gene sequences, Gupta and Gao^[24] identified three conserved indels in three highly conserved proteins (a 4 amino acid insert in DNA gyrase A, a 1 amino acid deletion in ATP synthase beta subunit and a 1 amino acid insert in ribosomal protein S2) that were unique to the species of Clostridium cluster I; these conserved indels were not found in any other known bacterial species for which sequences are available^[24].

Yutin and Galperin^[25] provided a potentially important addition to the restructuring of the clostridia using genomic analyses to investigate Gram-negative staining spore formers and other misplaced clostridia. This study, based on a concatenated set of 50 widespread ribosomal proteins with the trees for beta subunits of the RNA polymerase (RpoB) and DNA gyrase (GyrB) and with the 16S rRNA-based phylogeny, resulted in the proposal of a number of novel genera^[25]. These descriptions do not comply with the Bacteriological Code and therefore the names cannot be validly published in IJSEM

Although it is now generally accepted that Clostridium cluster I represents the true genus Clostridium and the species included in it should be recognized as Clostridium sensu stricto, an anomaly was revealed when a number of species assigned to the genera Anaerobacter, Eubacterium and Sarcina were found to fall within the radiation of Clostridium cluster I^{[3, 5]}. Based on these observations, in his chapter on the Clostridum in the Bergey’s Manual of Systematic Bacteriology Rainey (2009) transferred Anaerobacter polyendosporus to the genusClostridium^[3]. The species Eubacterium moniliforme and Eubacterium tarantellae are phylogenetically distant to the type species of the genus Eubacterium namely, E. limosum, and these two species should be transferred to Clostridium. The situation with the genus Sarcina is more complex as this genus was first proposed in 1842 by Goodsir some 38 years prior to the genus Clostridium^[1] giving the genus Sarcina nomenclatural priority over the genus Clostridium^{[1, 26]}. Consequently, one could interpret this as meaning all Clostridium should be transferred to the genus Sarcina. Clearly, with numerous organisms being of clinical significance (C. botulinum, C. perfringens, C. novyi, C. septicum) it would cause consternation and confusion amongst the scientific, and especially, the medical community. However, the Bacteriological Code comes to our aid in Rule 24 which states “Priority of publicationdates from 1 January 1980. On that date all names published prior to 1 January 1980 and included in the Approved Lists of Bacterial Names of the ICSB are treated for all nomenclatural purposes as though they had been validly published for the first time on that date, the existing types being retained”. Therefore, both genera Clostridium and Sarcina (the latter with just two species) have equal standing in the nomenclature, and the date of priority of valid publication of both genus names is January 1, 1980. Furthermore, support for retaining the names of medically important species comes from the Code with Rule 56 (5) “perilous names - (nomina periculosa). i.e., names whose application is likely to lead to accidents endangering health or life or both or of serious economic consequences”. The transfer of all Clostridium species to the genus Sarcina would clearly pose a problem and fall into this category.

The taxonomy of the genus Clostridium is in a continued state of flux and several important issues remain with medically important species both within (C. botulinum complex) and outside rRNA cluster I (Clostridium difficile. The C. botulinum complex in particular causes a problem with the application of 16S rRNA sequencing reveling at four phylogenetically separate clusters that should represent distinct species or even genera. This highlighted the problems of describing taxa on a single trait such as toxin production especially when said traits are transmissible leading to scientific inconsistencies that are no longer tenable in updated taxonomies^[27]. Another very important issue centers on the important human pathogen C. difficile and the condition known as Clostridium difficile associated diarrhea (CDAD) one of the major scourges of modern health care. A consequence of the publication of Lawson and Rainey^[28] and the proposal to restrict the genus Clostridium to C. butyricum and close relatives is that C. difficile located in cluster XI and should not be considered as a bonafide member of the genus and should be reclassified. This is likely to cause some consternation among clinical colleagues but change is inevitable, an earlier attempt proposing the name Peptoclostridium difficile^[25] was unsuccessful. This issue must be treated in a sensitive manner to accommodate the medical community and will be the subject of a future paper.

Although Clostridium cluster I represents the true genus Clostridium and should be recognized as Clostridium sensu stricto, unfortunately authors continue to assign organisms to the genus Clostridium even when it is obvious using 16S rRNA gene sequence data that these novel organisms fall outside of Clostridium cluster I. Therefore, it should be formally proposed that members of the genus Clostridium (Prazmowski) be restricted to the type species of the genus, Clostridium butyricum and relatives in Clostridium cluster I. It is proposed that Eubacterium moniliforme, Eubacterium tarantellae, Sarcina maxima, and Sarcina ventriculi be transferred to this genus as Clostridium moniliforme comb. nov., Clostridium tarantellae comb. nov., Clostridium maximum comb. nov., and Clostridium ventriculi comb. nov. In addition, it is also recommended that new species that are unrelated to C. butryicum and fall outside of Clostridium cluster I should not be added to the genus Clostridium. A novel genus is also proposed to encompass proteolytic species designated as rRNA cluster II; Hathewaya gen. nov. and that Clostridium histolyticum, Clostridium limosum and Clostridium proteolyticum be transferred to this new genus as Hathewaya histolytica gen. nov. com. nov., Hathewaya limosa com. nov. and Hathewaya proteolytica comb. nov. The type species of the genus Hathewaya is Hathewaya histolytica^[28].

Even with these proposed changes, it is acknowledged that future revisions may be necessary due to the internal structure and phylogenetic depth present within clostridia cluster I. However, it is another important step in resolving the taxonomy of the genus Clostridium and prevent future assignment of organisms to the genus that are phylogenetically removed from C. butyricum and close relatives.