Mycobacterium tuberculosis Beijing genotype and mycobacterial interspersed repetitive unit typing.

Rao et al. (3) described a method to identify Mycobacterium tuberculosis strains of the Beijing genotype. The authors suggested using simple PCR of mycobacterial interspersed repetitive unit–variable-number tandem repeat (MIRU-VNTR) locus 26 to identify these strains. This suggestion is based on the authors’ assumption that MIRU locus 26 has a specific seven-copy signature in these strains. Rao et al. (3) based their conclusions on 10 available Beijing strains (DNA samples) for which they indeed found sevencopy signatures. The authors justly mentioned that their test should be validated and confirmed with a large number of known Beijing strains. Prior to further experiments, it would be advisable, however, to compare their results against an already large number of available articles on MIRU typing of M. tuberculosis strains, published between 2001 and 2004, also containing information about Beijing strains (1–2, 4–6), including our article specially focused on the MIRU typing of the Beijing genotype (2). These articles clearly demonstrated for a large number of representative samples of strains from Russia (n 44), South Africa (n 38), Singapore (n 160), Bangladesh (n 15), and other locations that MIRU locus 26 is not monomorphic but instead moderately polymorphic in the Beijing genotype and may consist of two, three, four, five, six, seven, eight, or nine copies [1, 2, 4, 5, 6]). Furthermore, a brief check of the mentioned articles (1, 4–6) reveals that seven copies at MIRU locus 26 may occur not only in some Beijing strains but also in strains of other genotypes of M. tuberculosis. I am afraid that for these reasons, formally speaking, the proposed method dramatically lacks both sensitivity and specificity and cannot be used to identify M. tuberculosis Beijing genotype strains in any setting.