Abstract
Staphylococcus lugdunensis is an emergent virulent coagulase-negative Staphylococcus that is increasingly responsible for severe infections. In an attempt to generate informative sequence data for subtyping S. lugdunensis, we selected and sequenced seven polymorphic variable number of tandem repeats (VNTRs) to develop two new methods: a classic length-based multiple-locus VNTR analysis (MLVA) method and a tandem repeat sequence typing (TRST) method. We assessed their performances compared to two existing methods, multilocus sequence typing (MLST) and multivirulence-locus sequence typing (MVLST) for 128 isolates from diverse clinical settings and geographical origins. The clustering achieved by the four methods was highly congruent, with MLVA discriminating within clonal complexes as defined by MLST. Indeed, MLVA was highly discriminant compared to MLST and MVLST in terms of number of genotypes as well as diversity indexes. Sequencing of the seven VNTRs showed that they were stable, and analysis of sequence polymorphisms provided superior discriminatory power. The typeability, reproducibility, and epidemiological concordance of these new methods were excellent. Of note, no link between clustering and clinical settings was identified. This study demonstrates that MLVA and TRST provide valuable information for molecular epidemiological study of S. lugdunensis, and represent promising tools to distinguish between strains of homogenous lineages in this clonal species.