Staphylococcal food poisoning represents the most prevalent foodborne intoxication worldwide. It is caused by oral intake of enterotoxins preformed by Staphylococcus aureus in food. The relevance of newly described enterotoxins in outbreaks of staphylococcal food poisoning is controversially discussed. Although the staphylococcal enterotoxins SEG, SEI, SEM, SEN, and SEO elicit emesis in a monkey feeding assay, there has been no conclusive proof of their emetic activity in humans. In this study, we provide further evidence suggesting that one of these enterotoxins or a combination of SEG, SEI, SEM, SEN, and SEO cause staphylococcal food poisoning. We investigated two outbreaks registered with the Swiss Federal Office of Public Health, in which only Staphylococcus aureus strains harboring the egc cluster, including seg, sei, sem, sen, and seo linked to typical signs of staphylococcal food poisoning were isolated. The outbreaks were caused by consumption of raw goat cheese and semi-hard goat cheese, and were linked to strains assigned to CC45 (agr type I) and CC9 (agr type II), respectively. These outbreaks provide further evidence that newly-described staphylococcal enterotoxins are likely to cause staphylococcal food poisoning in humans.

The objective of this study was to characterize Staphylococcus aureus isolates from Swiss raw milk cheeses that had been found to be contaminated with coagulase-positive staphylococci and to estimate the frequency of the various genotypes, in particular the mastitis-associated Staph. aureus genotype B (GTB). The isolates were also tested for staphylococcal enterotoxin (SE) genes and other virulence factors. From 623 coagulase-positive staphylococci isolated from 78 contaminated raw milk cheeses, 609 were found to be Staphylococcus aureus. Genotyping of all Staph. aureus isolates was performed by PCR amplification of the 16S-23S rRNA intergenic spacer region, as this method was used previously to differentiate between mastitis subtypes associated with their clinical outcome. In total, 20 different genotypes were obtained and the 5 most frequently occurring genotypes were distributed in 6.4% or more of the samples. The enterotoxin-producing Staph. aureus GTB, known for its high contagiousness and increased pathogenicity in Swiss mastitis herds, was found to be the most abundant subtype at the sample level (71.8%) as well as among the isolates (62.0%). A subset of 107 isolates of the different genotypes were analyzed for the presence of SE genes and revealed 9 different SE gene patterns, with sed being most frequently detected and 26% being PCR-negative for SE genes. Almost all isolates of the major contaminant GTB contained the SE gene pattern sed, sej, ser, with half of them additionally carrying sea. Production of SE in vitro was consistent with the SE genes detected in most of the cases; however, some isolated GTB did not produce SEA. Staphylococcus aureus Protein A (spa) typing revealed 30 different subtypes and most GTB isolates belonged to the bovine spa type t2953; GTB/t2953 was linked among other subtypes to SE production in cheese and staphylococcal intoxication cases. Furthermore, 1 of the 623 isolates was a methicillin-resistant Staph. aureus, which was an seh-carrying Staph. aureus spa type tbl 0635 (non-GTB). We conclude that control and reduction of enterotoxigenic Staph. aureus GTB in dairy herds in Switzerland will not only prevent economic losses at the farm level but also improve the safety of raw milk cheeses; distribution of methicillin-resistant Staph. aureus via raw milk cheese is of less concern.

The objective of this study was to compare the biofilm-forming capabilities of different genotypes of Staphylococcus aureus dairy isolates from Switzerland and northern Italy, including Staph. aureus genotype B (GTB) and methicillin-resistant Staph.aureus (MRSA). We hypothesized that biofilm formation might be more pronounced in the contagious GTB isolates compared with other genotypes affecting individual animals. Twenty-four dairy isolates, including 9 MRSA, were further characterized by genotyping by using ribosomal spacer PCR, spa typing, biofilm formation under static and dynamic conditions, and scanning electron microscopy. The GTB isolates (n = 6) were more able to form biofilms than other genotypes at 37°C and at 20°C after 48 and 72 h of incubation in the static assay using polystyrene microtiter plates. This result was supported by scanning electron micrographs showing a GTB isolate producing strong biofilm with extracellular matrix in contrast to a genotype C isolate. Furthermore, none of the MRSA isolates formed strong biofilms in the static assay. However, some MRSA produced low or moderate amounts of biofilm depending on the applied conditions. Under dynamic conditions, a much more diverse situation was observed. The ability of GTB isolates to be strong biofilm formers was not observed in all cases, emphasizing the importance of growth conditions for the expression of biofilm-related genes. No specific genotype, spa type, or MRSA isolate could be categorized significantly into one level of biofilm formation. Nineteen percent of isolates behaved similarly under static and dynamic conditions. The results of this study expand our knowledge of different dairy-related Staph. aureus subtypes and indicate the benefit of genotyping when biofilms are studied.

Here we report the isolation of heat-resistant Escherichia coli from raw milk cheeses. Detection of the heat-resistance markers clpK and orfI by PCR was followed by phenotypical confirmation of increased heat-resistance. These strains were Shiga toxin-negative and, although several were found to be multidrug resistant, no plasmids encoding extended-spectrum β-lactamases (ESBL) were found in any of the isolates. The aim of this study was to assess the potential of these strains to acquire ESBL plasmids and a modified Shiga toxin-encoding phage. Only 4 ESBL-encoding, heat-sensitive E. coli strains were isolated from 1,251 dairy samples (2/455 raw milk and 2/796 raw milk cheese samples). One incompatibility group FII plasmid (CTX-M-14, 79.0 kb) and 3 incompatibility group I1 plasmids (CTX-M-15, 95.2, 96.1, and 97.8 kb) were fully sequenced and de novo assembled. All 4 plasmids are readily transferred to heat-resistant E. coli isolates in plate matings (9.7×10-5 to 3.7×10-1 exconjugants per recipient) and, to a lesser extent, in milk (up to 7.4×10-5 exconjugants per recipient). Importantly, the plasmids are stably maintained during passaging in liquid media without antimicrobial pressure. The heat-resistant isolate FAM21805 was also shown to be capable of acting as donor of all 4 ESBL plasmids. In addition, 3 of 11 tested ESBL exconjugants of heat-resistant strains were lysogenized by the modified Shiga toxin-encoding phage 933W ∆stx::gfp::cat. The higher fraction of heat-resistant E. coli (93 of 256 isolates) compared with the estimated 2% previously predicted based on genomic prevalence of heat resistance genes seems to indicate a selection advantage in the raw milk cheese production environment. The combination of 2 factors may lead to said advantage: increased survival during thermization of raw milk (heating to subpasteurization temperatures) and increased survival rates during cheese ripening. Should these strains acquire ESBL-encoding plasmids, Shiga toxin-encoding phages, or both, these genetic elements would profit from the selection advantage of their host and become more abundant in this particular environment, which in turn could lead to an increased threat to consumers of raw milk products.

We tested the biofilm formation potential of 30 heat-resistant and 6 heat-sensitiveEscherichia colidairy isolates. Production of curli and cellulose, static biofilm formation on polystyrene (PS) and stainless steel surfaces, biofilm formation under dynamic conditions (Bioflux), and initial adhesion rates (IAR) were evaluated. Biofilm formation varied greatly between strains, media, and assays. Our results highlight the importance of the experimental setup in determining biofilm formation under conditions of interest, as correlation between different assays was often not a given. The heat-resistant, multidrug-resistant (MDR) strain FAM21845 showed the strongest biofilm formation on PS and the highest IAR and was the only strain that formed significant biofilms on stainless steel under conditions relevant to the dairy industry, and it was therefore fully sequenced. Its chromosome is 4.9 Mb long, and it harbors a total of five plasmids (147.2, 54.2, 5.8, 2.5, and 1.9 kb). The strain carries a broad range of genes relevant to antimicrobial resistance and biofilm formation, including some on its two large conjugative plasmids, as demonstrated in plate mating assays.IMPORTANCEIn biofilms, cells are embedded in an extracellular matrix that protects them from stresses, such as UV radiation, osmotic shock, desiccation, antibiotics, and predation. Biofilm formation is a major bacterial persistence factor of great concern in the clinic and the food industry. Many tested strains formed strong biofilms, and especially strains such as the heat-resistant, MDR strain FAM21845 may pose a serious issue for food production. Strong biofilm formation combined with diverse resistances (some encoded on conjugative plasmids) may allow for increased persistence, coselection, and possible transfer of these resistance factors. Horizontal gene transfer may conceivably occur in the food production setting or the gastrointestinal tract after consumption.

The number of environmental antibiotic-resistant bacteria (ARB) has increased dramatically since the start of antibiotic mass production for broad bacterial infection treatment in 1944. Nowadays, ARB and their resistance-determining genes (ARGs) are readily detected in all environments, including the human food chain. A highly relevant food group in this context is fresh produce, frequent raw consumption of which facilitates direct transfer of ARB and ARGs to the consumer. Here, we investigate the persistence of an extended-spectrum β-lactamase (ESBL)-producing Escherichia coli (E. coli) pEK499 and its clinically most important ARG (bla_CTX-M-15), after introduction via irrigation water or manure into a lettuce-growing system. Culturable ESBL-producing E. coli persisted longest in soil and when introduced via manure (until 9 weeks after introduction), while being undetectable on lettuce beyond day 7. In contrast, qPCR detection of bla_CTX-M-15 was much more frequent: introduction via water significantly increased bla_CTX-M-15 on lettuce until week 4, as opposed to manure, which affected the soil in the long-term (9 weeks) while leading to bla_CTX-M-15 detection on lettuce until day 7 only. Our findings demonstrate long-term persistence of undesired ARB and ARG after their introduction via both irrigation and amendment. Such an understanding of the persistence kinetics of an ESBL-producing E. coli and plasmid-encoded bla_CTX-M-15 aids the determination of critical actions in order to mitigate their transfer to the consumer.