Abstract
Recently, Binnewies and colleagues described the proteome comparisons of eight Mycoplasma genome sequences in Microbiology Comment (Binnewies et al., 2005). The presence of a species from a different genus, Ureaplasma urealyticum, between the different Mycoplasma species is merely indicative of the problems taxonomy may experience in the genomic era.
The diversity of the different Mycoplasma species has been attributed to the rapid (degenerative) evolution of the species (Rogers et al., 1985). The diversity within the genus is illustrated by the large differences in genome size (ranging from 0·58 to 1·36 Mbp) and percentage G+C (from 24 to 40 mol%) between the eight sequenced Mycoplasma species.
With the current availability of whole-genome sequences, some efforts have been made to initiate genome-based taxonomy (Coenye et al., 2005; Konstantinidis & Tiedje, 2005). Previously, Coenye & Vandamme (2003) described various phylogenetic information sources, and compared the sequences of 16S RNA genes and several housekeeping genes with the fraction of shared orthologous genes between related genomes, the conservation of gene order, the dinucleotide relative abundance values and the codon usage. They concluded, using the lactic acid bacteria as a test case, that the different kinds of phylogenetic information agree and that whole-genome sequence comparisons may be of value in taxonomy, e.g. to enhance the phylogenetic signal derived by more traditional methods. Would this approach also show that the different Mycoplasma species are closely related?
The parameters used by Coenye & Vandamme show that for Mycoplasma, the sequence identity values for 16S RNA genes are generally low (between 71·5 and 98·1 %) and that dinucleotide relative abundance differences between their genomes, also known as the genome signature dissimilarity, are very high (Table 1), much higher than between representatives of the Enterobacteriaceae (Table 2). For three Mycoplasma species, this had already been found (Sandberg et al., 2003), but this trend continues with the eight known Mycoplasma genome sequences. Recently, Gophna et al. (2005) devised several weighted whole-genome trees in which it is clearly visible that the Mycoplasma genus branches at a great depth, indicating a high level of diversity within this genus. Together, these data confirm the limits of current taxonomy for the genus Mycoplasma. Results with Buchnera and Wolbachia indicate that this may be true for all genera which are thought to undergo rapid genome evolution, such as endosymbionts (data not shown).