One of the most striking properties of members of the M. tuberculosis complex is the remarkable lack of genomic diversity observed at the nucleotide level and the rarity of synonymous or silent (third position) codon changes. This is important in the context of immunity and vaccine development as it means that the majority of the proteins should be exactly the same in all strains and, as a consequence, the potential for antigenic drift is restricted. Based on the systematic sequence analysis of 26 loci in a large number of independent isolates of the tubercle bacillus (Kapur et al 1994, Sreevatsan et al 1997), Musser and his colleagues have concluded that the genome of M. tuberculosis is either unusually inert for a bacterium or that the organism is relatively young in evolutionary terms, i.e. <15 000 years old. While this hypothesis is seductive, and consistent with both the period when the domestication of cattle occurred, (i.e. the possible conversion of the host range of the tubercle bacillus from bovines to humans) and the archaeological record (Salo et al 1994), other factors that might contribute to the remarkable lack of genetic diversity are the inefficiency of homologous recombination that has been observed (Kalpana et al 1991) and the long generation time. Alternatively, it is possible that, as a result of long exposure to the potentially mutagenic environment of the macrophage, the DNA repair system may be outstandingly efficient, although the fact that drug-resistant mutants emerge with the same frequency (10~6—10~7) in M. tuberculosis as in Escherichia coli argues against this (Sreevatsan et al 1997).
During the last five years, a large body of evidence has emerged indicating that the principal source of genomic polymorphism in M. tuberculosis is due to the transposition of an endogenous insertion sequence, IS6110, commonly used for typing purposes (Hermans et al 1990, McAdam et al 1990, Otal et al 1991, van Embden et al 1993), or, somewhat less frequently, to alterations of the PGRS or MPTR loci. It was known from mapping studies that 16 copies of IS6110 and six copies of the more stable element, IS1081, resided within the genome of H37Rv (Philipp et al 1996a), but scrutiny of the genomic sequences led to many more insertion sequence elements being discovered. In addition to these, >12 novel insertion sequences have been identified by sequence comparison, as well as a smaller number of repetitive sequences that bear some of the hallmarks of mobile genetic elements (Fig. 1). At least two prophages have been detected and more may be present. As one of the prophages and one of the new insertion sequences are missing from the genomes of recent clinical isolates of M. tuberculosis (Mahairas et al 1996, Philipp et al 1996b, S.T. Cole & B.G. Barrell, unpublished data 1997), this could indicate that horizontal transfer of genetic material occurs in nature and that, despite its intracellular niche, the tubercle bacillus might be able to acquire new genes if they were carried by mycobacteriophages or transposable elements. This is also consistent with the findings of a comparative mapping study (Philipp et al 1996b), that insertions/deletions are probably the principal source of genomic diversity within the M. tuberculosis complex. The distribution of these elements on the chromosome is particularly informative as there appears to have been a selection against insertions in the quadrant encompassing the origin of replication (Fig. 1). Strikingly, in a recent transposon mutagenesis study a similar bias was observed (Bardarov et al 1997), again suggesting that this chromosomal region may be less permissive to insertional events.
Was this article helpful?