Multilocus Sequence Typing for Clostridium difficile

Multilocus sequence typing (MLST), a nucleotide sequence-based characterization of allelic polymorphism of housekeeping genes, has been proposed as a new approach for population and evolutionary genetics and global epidemiology of bacterial pathogens. MLST provides unambiguous sequence data that can be generated from various laboratories and should be shared in a common web database. Here are presented most of materials, methods, and programs or software necessary to perform MLST on Clostridium difficile . We also describe an example of an MLST scheme for C. difficile based on sequence analysis of six housekeeping gene loci and use a set of 74 C. difficile isolates from various hosts, geographic sources, and PCR-toxigenic types (A+B+, A–B+, and A–B–). Thirty-two “sequence types” (ST) are defined from the combination of allelic data, which correlate well with toxigenic types. The estimation of linkage disequilibrium between loci reveals a clonal population structure. Mutational evolution of C. difficile is characterized, with point mutation generating new alleles at a frequency eightfold higher than recombinational exchange. Phylogenetic analysis shows that human and animal isolates do not cluster in distinct lineages, and that no hypervirulent lineage can be characterized within the population of toxigenic human isolates studied (strains from pseudomembranous colitis and antibiotic-associated diarrhea do not cluster in distinct lineages). However, all A–B+ variant isolates belong to a divergent but very homogeneous lineage in the population studied. An MLST database specific for this species is now hosted at the web site of the Institut Pasteur Paris. Since MLST data reflect evolutionary genetics of the species, they could be used as typing markers, possibly in combination with virulence genes data, for long-term global epidemiology of C. difficile .