New insights into molecular typing methods for Staphylococcus aureus

Staphylococcus aureus (SA) remains a significant problem causing infections in both hospital and community settings. Methicillin-resistant SA (MRSA) continues to evolve and pose a great challenge through outbreaks and pandemic spread. Humans are no longer the only and the most important reservoir of SA. The nature of especially MRSA has started to change and MRSA emerged in animal reservoirs particularly in bovine and pigs. These MRSA can be transmitted to humans. Rapid typing of pathogens is important for infection control. Several typing methods have been developed to characterize SA. These methods have different resolutions for the detection of outbreaks or for population studies. However, the currently available methods are not satisfactory in particular for outbreak detection. We developed and evaluated a variable number tandem repeat (VNTR) analysis (MLVA) typing scheme for outbreak detection and population study purposes of SA in humans (HSA) and SA from bovine mastitis (BSA) and pigs-related MRSA (PSA). This typing scheme employed a simple DNA amplification of six VNTRs of SA termed staphylococcal interspersed repeat units (SIRUs), SIRU01, -05, -07, -13, -15, and -21. Evaluation of the MLVA was based on typeability, reproducibility, stability, discriminatory power, and epidemiologic concordance. The MLVA was validated and compared with currently available typing methods. Furthermore, we investigated a DNA microarray with a limited number of probes for typing of HSA on both outbreak and population levels. Using these methods we correlated virulence factors which might underlie the mastitis, in Dutch BSA that were genotyped by MLVA. In addition, we investigated transmission of MRSA between pig farms in The Netherlands. Our study demonstrated that the MLVA performed very well for HSA and was suitable for typing of BSA and PSA. The method is fast, cheap, highly discriminatory, reproducible, stable and portable, which can be widely implemented in the institutions without access to more advanced techniques such as DNA sequencing. However, because the amplification of some SIRUs in particular SIRU05 in BSA and PSA was poor, the replacement of SIRU05 by another locus or the addition of another locus should be investigated. Ideally, this new locus should also be used for HSA to keep the MLVA scheme as universal as possible. Moreover, the usefulness of the scheme would be greatly enhanced when a library with a MLVA web-based database for international comparison is set up. Microarrays with a limited number of probes can be a useful typing method for SA on outbreak and population level by using different cut-offs. Quality and utility of the microarray should be improved and enhanced by a well-defined genetic distance cut-off, removal of unstable probes, and inclusion of probes for resistance genes or specific virulence factors as shown by BSA. Certain genes appear to be required for the pathogenesis of bovine mastitis, although an alternative explanation is that different gene profiles lead to mastitis. The study of PSA from different farms which showed that transmission within the production chain occurs, would have profited from the MLVA scheme we developed if it was available at the time of the study.