Matched-cohort DNA microarray diversity analysis of methicillin sensitive and methicillin resistant Staphylococcus aureus isolates from hospital admission patients.

As genotyping of S. aureus is important for epidemiologic research and for hygiene management, methods are required for standardized fast and easily applicable evaluation of closely related epidemic strains with high prevalence in hospitals. In this single centre matched control study we compared a new commercially available DNA microarray (IdentiBAC) with standard spa-typing for S. aureus genotyping. Included in the study was a subgroup of 46 MRSA and matched 46 MSSA nasal isolates of the Saarland University Medical Center collected during a state-wide admission prevalence screening. Microarray (MA) and also spa-typing could easily differentiate the genetically diverse MSSA group. However, due to the predominance of CC5/t003 in the MRSA group a sufficient subtyping required analysis of more complex genetic profiles as was shown here by the MA comprising a total number of 334 different hybridization probes. The genetic repertoire of the MRSA group was characterized by more virulence genes as compared to the MSSA group. The standard evaluation of MA results by the original software into CCs, agr-, SCCmec- and capsule-types was substituted in the present study by implementation of multivariate subtyping of closely related CC5 isolates using three different bioinformatic methods (splits graph, cluster dendrogram, and principal component analysis). Each method used was applicable for standardized and highly discriminative subtyping with high concordance. We propose that the identified S. aureus subtypes with characteristic virulence gene profiles are presumably associated also with virulence and pathogenicity in vivo; however, this remains to be analyzed in future studies. MA was superior to spa-typing for epidemiologic and presumably also provide functional respectively virulence associated characterization of S. aureus isolates. This is of specific importance for the hospital setting. In future, MA could become a new standard test for S. aureus typing in combination with multivariate bioinformatic analysis.