Building a better bacillus: the emergence of Mycobacterium tuberculosis

Abstract

The genus Mycobacterium is comprised of more than 150 species that reside in a wide variety of habitats. Most mycobacteria are environmental organisms that are either not associated with disease or are opportunistic pathogens that cause non-transmissible disease in immunocompromised individuals. In contrast, a small number of species, such as the tubercle bacillus, Mycobacterium tuberculosis, are host-adapted pathogens for which there is no known environmental reservoir. In recent years, gene disruption studies using the host-adapted pathogen have uncovered a number of "virulence factors," yet genomic data indicate that many of these elements are present in non-pathogenic mycobacteria. This suggests that much of the genetic make-up that enables virulence in the host-adapted pathogen is already present in environmental members of the genus. In addition to these generic factors, we hypothesize that molecules elaborated exclusively by professional pathogens may be particularly implicated in the ability of M. tuberculosis to infect, persist, and cause transmissible pathology in its host species, Homo sapiens. One approach to identify these molecules is to employ comparative analysis of mycobacterial genomes, to define evolutionary events such as horizontal gene transfer (HGT) that contributed M. tuberculosis-specific genetic elements. Independent studies have now revealed the presence of HGT genes in the M. tuberculosis genome and their role in the pathogenesis of disease is the subject of ongoing investigations. Here we review these studies, focusing on the hypothesized role played by HGT loci in the emergence of M. tuberculosis from a related environmental species into a highly specialized human-adapted pathogen.

Keywords: M. tuberculosis-specific genes; Mycobacterium kansasii; Mycobacterium tuberculosis; comparative genomics; horizontal gene transfer.

FIGURE 1

Phylogeny of M. tuberculosis and closely related Mycobacterium species. The un-rooted phylogenetic tree was generated by MEGA6.0 using 20 randomly selected genes conserved across eight Mycobacterium species (Schwab et al., 2009). The blue arrows schematically represent where putative HGT events may have occurred, resulting in M. tuberculosis-specific genomic islands. The scale bar indicates 0.02 substitutions per nucleotide position, and the bootstrap values calculated using the neighbor-joining method (expressed as a percentage of 1000 replicates) are shown at the branch points. The fast growing species M. smegmatis is used as the out-group. Genes used are listed below (represented as M. tuberculosis genes): Rv0001-dnaA, Rv0041-leuS, Rv0236A–Rv0236A, Rv0248c–Rv0248c, Rv0285-PE5, Rv0287-esxG, Rv0288-esxH, Rv1085c–Rv1085c, Rv0197–Rv0197, Rv1304-atpB, Rv1305-atpE, Rv1894c–Rv1894c, Rv2172c–Rv2172c, Rv2392-cysH, Rv2440c-obg, Rv2477c–Rv2477c, Rv3019c-esxR, Rv3045-adhC, Rv3392c-cmaA1, Rv3502c- hsd4A.

FIGURE 2

Genomic organization of operons in M. tuberculosis, M. kansasii, and M. marinum. (A) Rv0986-0988; (B) Rv3376-3378c; (C) Rv3108-3126c; (D) Rv2954c-2961. Protein-coding genes are represented by arrows and orthologous genes are indicated by arrows of the same color. Yellow and blue arrows mark the “boundary” of each M. tuberculosis-specific locus. Red arrows indicate genes discussed in this review. Dark green arrows indicate M. tuberculosis genes with no orthologs within the corresponding M. kansasii and M. marinum genomic regions. White arrows in M. kansasii and M. marinum genomes indicate genes present but not orthologs to M. tuberculosis genes. Black arrows indicate transposases; orange arrows indicate toxin-antitoxin genes. Genome organizations for M. tuberculosis, M. kansasii, M. marinum and gene clusters were obtained from the Kyoto Encyclopedia of Genes and Genomes (http://www.kegg.jp/) based on databases available at the Sanger Institute, Tuberculist, and McGill University, respectively. Orthologs were verified by comparing each predicted protein against the H37Rv genome using the BLAST program. Only proteins with >50% coverage, 60% identity and E-value <e^-20 were used.