High resolution, on-line identification of strains from the Mycobacteriumtuberculosis complex based on tandem repeat typing

Abstract

Background: Currently available reference methods for the molecular epidemiology of the Mycobacterium tuberculosis complex either lack sensitivity or are still too tedious and slow for routine application. Recently, tandem repeat typing has emerged as a potential alternative. This report contributes to the development of tandem repeat typing for M. tuberculosis by summarising the existing data, developing additional markers, and setting up a freely accessible, fast, and easy to use, internet-based service for strain identification. Results: A collection of 21 VNTRs incorporating 13 previously described loci and 8 newly evaluated markers was used to genotype 90 strains from the M. tuberculosis complex (M. tuberculosis (64 strains), M. bovis (9 strains including 4 BCG representatives), M. africanum (17 strains)). Eighty-four different genotypes are defined. Clustering analysis shows that the M. africanum strains fall into three main groups, one of which is closer to the M. tuberculosis strains, and an other one is closer to the M. bovis strains. The resulting data has been made freely accessible over the internet [http://bacterial-genotyping.igmors.u-psud.fr/bnserver] to allow direct strain identification queries. Conclusions: Tandem-repeat typing is a PCR-based assay which may prove to be a powerful complement to the existing epidemiological tools for the M. tuberculosis complex. The number of markers to type depends on the identification precision which is required, so that identification can be achieved quickly at low cost in terms of consumables, technical expertise and equipment.