[Molecular epidemiology of Mycobacterium tuberculosis: its accomplishment and future perspective]

In the traditional study of tuberculosis epidemiology, information about social contact of persons and patient's illness history used to be an only relevant basis for elucidating transmission of tuberculosis infection. Therefore, it was very difficult to give a clear conclusion of whether isolates from different patients derived from a common source of infection or not. Recently, the subspecies typing of M. tuberculosis strains has become possible, based on the visualization of multiple loci of an insertion sequence (IS6110) that is a relatively stable gene fragment existing in a specific region of the genome. The variability of the number of copies and locations of this IS6110 in a genome is the basis that enables this technique to be used for the above purpose, which is a unique tool applicable to the analysis of M. tuberculosis. Generally, this technique, i.e., restriction fragment length polymorphism (RFLP) analysis, depends on the diversity of pattern of any polymorphic marker found in a genome of a strain. Among various markers so far developed and examined, IS6110 has been proved most appropriate for the purpose of typing strains of M. tuberculosis complex, especially in such circumstances as in Japan where isolated strains' RFLP patterns are similar each with others so that finer subtyping is needed. In this lecture, I would like to review the following topics based on the world literature of molecular epidemiology and the findings of our own that we have achieved during 1992 through 2001 in our Institute; 1) typing of the isolates for the identification of the infection source, 2) pathogenesis of tuberculosis under low incidence situation, 3) predominance of certain genotypes endemic in an area, 4) cross-contamination of isolates in the laboratory, 5) the stability of IS6110 patterns, 6) phylogeny of M. tuberculosis complex, and 7) differentiation between M. tuberculosis and M. bovis BCG.