Analysis of the allelic diversity of the mycobacterial interspersed repetitive units in Mycobacterium tuberculosis strains of the Beijing family: practical implications and evolutionary considerations.

Abstract

A study set comprised 44 Mycobacterium tuberculosis strains of the Beijing family selected for their representativeness among those previously characterized by IS6110-RFLP and spoligotyping (Northwest Russia, 1997 to 2003). In the present study, these strains were subjected to mycobacterial interspersed repetitive units (MIRU) typing to assess a discriminatory power of the 12-MIRU-loci scheme (P. Supply et al., J. Clin. Microbiol. 39:3563-3571, 2001). The 44 Russian Beijing strains were subdivided into 12 MIRU types with identical profiles: 10 unique strains and two major types shared by 10 and 24 strains. Thus, basically, two distinct sublineages appear to shape the evolution of the Beijing strains in Russia. Most of the MIRU loci were found to be (almost) monomorphic in the Russian Beijing strains; the Hunter-Gaston discriminatory index (HGDI) for all 12 loci taken together was 0.65, whereas MIRU26 (the most variable in our study) showed a moderate level of discrimination (0.49). The results were compared against all available published MIRU profiles of Beijing strains from Russia (3 strains) and other geographic areas (51 strains in total), including South Africa (38 strains), East Asia (7 strains), and the United States (4 strains). A UPGMA (unweighted pair-group method with arithmetic averages)-based tree was constructed. Interestingly, no MIRU types were shared by Russian and South African strains (the two largest samples in this analysis), whereas both major Russian types included also isolates from other locations (United States and/or East Asia). This implies the evolution of the Beijing genotype to be generally strictly clonal, although a possibility of a convergent evolution of the MIRU loci cannot be excluded. We propose a dissemination of the prevailing local Beijing clones to have started earlier in South Africa rather than in Russia since more monomorphic loci were identified in Russian samples than in South African samples (mean HGDI scores, 0.08 versus 0.17). To conclude, we suggest to use a limited number of MIRUs for preliminary subdivision of Beijing strains in Russian (loci 26 + 31), South African (10 + 26 + 39), and global settings (10 + 26 + 39).