Novel insights into the enterotoxigenic potential of Staphylococcus aureus isolated from raw milk.

Enterotoxigenic Staphylococcus (S.) aureus continue to be a great concern for raw milk and raw milk cheeses safety. Apart from the classical staphylococcal enterotoxins (SEs) (SEA-SEE), new SEs and staphylococcal enterotoxin-like toxins (SEls) have been detected and for most of them the emetic activity has been demonstrated or their genes have been found in S. aureus strains involved in staphylococcal food poisoning (SFP) cases. These findings suggest that the new SE/SEls are a potential cause of SFP outbreaks and call for studies aimed at investigating the actual and up-to-date enterotoxigenic potential of S. aureus harboring these toxins. Whole genome sequencing is being widely used to characterize S. aureus strains, but incorrect annotations and the use of pipelines targeting only a limited number of se/sel genes provide only a partial and often misleading description of the actual and overall enterotoxigenic potential of these pathogens. Confusion has further been created by the incorrect nomenclature given to the latest SEls discovered. In this study 53 S. aureus isolated from raw milk were typed by enterotoxin gene- (Seg) and 16S-23S rDNA intergenic spacer region- (ISR) typing. Six strains, one per each Seg- type and ISR-type obtained, were genome sequenced and the ability to produce classical and new SEs in milk was investigated. Seg- and ISR-typing showed the same discriminatory power, which was higher than that of multi locus sequence typing. Manual genomic analyses allowed to confirm, correct, or widen the results of the common available pipelines. In particular, novel allelic variants of se and sel genes and pseudogenes were detected. The actual location of se and sel genes was found, even for genes such as selY, whose location (in the core genome) was so far unknown. A novel type of enterotoxin gene cluster (egc), i.e. the egc-type 5 in S. aureus 356P and 364P, was detected while S. argenteus MSHR1132 harbored the novel egc-type 6. The correct annotation of se and sel genes and pseudogenes will contribute to improve and standardize the comparative genomics of S. aureus and clarify the confused scenario of the se and sel nomenclature. Four of the six S. aureus strains produced in milk sufficient amounts of SEA, SEC, SED and SEH to cause SFP, with S. aureus 372P as the highest producer of SED known to date. By contrast, S. aureus 372P released a low amount of SER in milk, most likely because the seR gene was present as a pseudogene encoding 51 amino acids. These findings confirm that not only the classical SEs but also the new ones can represent a hazard for the consumers' health and that the detection of SEs in raw milk and other matrices, especially if involved in SFP cases, should focus not only on classical, but also on all the new SEs and SEls known to date. Where reference methods are unavailable for the detection of these toxins, the presence of the relevant genes should be investigated using the (real time) PCR protocols refined by our group.