Abstract
A case of staphylococcal food poisoning was observed in two individuals of the same family after consumption of primosale, a semiripened sheep cheese produced in Sicily. Staphylococcus aureus isolated from the cheese produced enterotoxin C (SEC) and carried both the enterotoxin C (sec) and the toxic shock syndrome toxin (tsst-1) gene. Following this case, an extensive survey was conducted on 971 food samples (raw milk, cheese, meat, and food preparations). S. aureus was detected in 102 of 971 food samples, from all types of food with the exception of ricotta cheese. The tsst-1 gene was present in 42% of the strains, either alone or in combination with other toxin genes. The enterotoxin C gene was the most represented enterotoxin, but it was only found in dairy products. Six S. aureus isolates carried the sea gene alone, two isolates carried both sea and seb, and one isolate carried both sea and sec. A significant percentage (46%) of all isolates carried a toxin gene, creating significant concern that virulent S. aureus can be transmitted through food in Sicily.
Introduction
S
A screening on food samples in several Italian regions reported the presence of coagulase-positive staphylococci (Normanno et al., 2005). In a study conducted in Sicily, S. aureus strains were isolated from milk of cows with clinical mastitis (Ferguson et al., 2007). However, until the present time, no screening for toxigenic S. aureus strains has been performed on food samples from the Sicilian islands.
A case of food poisoning was registered in Sicily in 2009; the subjects (a married couple) reported vomiting and abdominal pain twice in a week, 1 h after eating a typical Sicilian sheep cheese (primosale). S. aureus poisoning was confirmed by bacterial isolation and detection of staphylococcal enterotoxin C (SEC) in the cheese sample.
The aim of this study was the analysis of food samples for the presence of S. aureus strains and related toxins, to determine the prevalence of virulent isolates and to evaluate potential risks for the consumer in Sicily.
Materials and Methods
Food sampling
Raw milk was collected from apparently healthy individual animals in sterile 50-mL tubes, in farms of different Sicilian provinces.
Other food products, such as cheese (231), fresh meat (95 steaks and 83 mincemeat/hamburgers), and preparations (meatballs, sausages, baked pasta, eggplant-based “Parmigiana di Melanzane,” cheese and mushroom omelettes, fish balls), were purchased from 24 large retail stores and supermarkets in the island's 9 main cities in all Sicilian provinces. At least two large supermarkets were randomly sampled from each town. All food samples were collected and sent refrigerated to the laboratory within 24 h.
Isolation of S. aureus from food samples
Microbiological analysis of food and S. aureus microbial characterization were performed as described in Normanno et al. (2005). The staphylococcal strains were identified using the API Staph system (BioMérieux, Marcy l'Etoile, France).
Enterotoxin production analysis in the S. aureus strains
The strains were grown in 10 mL of Brain Heart Infusion Broth (Oxoid, Basingstoke, Hampshire, UK) for 24 h at 37°C and the broth was filtered with a 0.45-μm membrane (Nalgene Labware, Rochester, NY). The Staphylococcal Enterotoxin Reversed Passive Latex Agglutination (SET-RPLA) kit (Oxoid) was used to detect staphylococcal enterotoxin A–D (SEA, SEB, SEC, and SED) in the culture fluid of each S. aureus strain, following the manufacturer's instructions.
DNA isolation and PCR detection of toxin genes
Total DNA was extracted from each isolate by boiling the samples for 5 min in 1 mL of Tris-EDTA (10 mM Tris chloride, 1 mM EDTA, pH 8.0) and it was directly used in multiplex polymerase chain reactions (PCRs) (MPX-A and MPX-B), as described in Mehrotra et al. (2000). Taq Gold DNA polymerase was used for the PCR in a 9700 thermo cycler (Applied Biosystem) and the amplicons were detected using a 2.5% agarose gel containing ethidium bromide and visualized under ultraviolet light.
Results and Discussion
A case of food poisoning due to S. aureus was registered in Sicily in 2009. Molecular analysis of the toxin genes from this strain revealed the presence of both sec and toxic shock syndrome toxin (tsst-1) genes. In a screening on 971 different food samples, a total of 102 S. aureus strains were isolated, mostly from dairy products (89.2%), and less frequently from food preparations (5.9%) and meat (4.9%). No S. aureus was isolated from the 72 samples of ricotta cheese, which is a fresh cheese made by heating whey at 80–85°C (Table 1). In a previous screening, it was found that 2.4% of all ricotta samples in Italy were contaminated by S. aureus (Normanno et al., 2005). The absence of S. aureus contamination in the ricotta samples of the present study probably hints at higher hygiene standards in the Sicilian production. Ricotta is an important ingredient in Sicilian cuisine and it is mostly consumed raw (e.g., in the filling of cakes). It is likely that special attention is paid on the island, to ensure good hygiene levels and a well-controlled cold chain in handling and processing.
One hundred two Staphylococcus aureus isolates were detected from 711 milk and dairy samples and 260 samples from meat and food preparations (10 meatballs, 11 sausages, 26 baked pasta, 21 “parmigiana eggplants,” 12 cheese and mushrooms omelettes, 12 fish balls; subtotals in bold). A high prevalence of tsst-1 gene was detected in S. aureus isolates. All strains carrying and expressing an enterotoxin showed the simultaneous presence of the tsst1 gene. Three strains carried two enterotoxin genes at the same time, one sea and sec and two sea and seb. Data from food preparations showed a higher presence of sea compared to dairy products, which, in contrast, contained mostly sec-producing strains. No S. aureus was isolated from ricotta cheese samples.
Mincemeat was from single species (cow or pork).
tsst-1, toxic shock syndrome toxin; eta and etb (genes for exfoliative toxins A and B); se a–e (staphylococcus enterotoxins from A to E); SE, staphylococcal enterotoxins.
The molecular analysis of the toxin genes in the 102 isolates showed that >42% (43/102) harbored the tsst-1 gene. Almost 80% of the strains that carried the tsst-1 gene (35/43) were isolated from dairy products (Table 1). These data are in contrast with previous studies, which reported a lower occurrence of TSST-1 in S. aureus from food samples (e.g., only 13.5% of the samples in Korea [Oh et al., 2007] and 12% in Iran [Dallal et al., 2010]). TSTT-1 is related to diseases, such as mastitis in cows, and it is often found in combination with enterotoxin C (Takeuchi et al., 1998; Srinivasan et al., 2006). In our study, milk collection was performed in farms with apparently healthy animals to avoid contamination. However, ruminant mastitis is a common problem in Sicily, and the presence of pathogenic staphylococci in apparently healthy herds cannot be excluded. Exotoxin genes for exfoliative toxins (eta and etb) were only present in 7 and 5 of 102 isolates, respectively, from raw milk samples (Table 1).
Thirty-nine isolates carrying the tsst-1 gene contained one or two enterotoxin genes. SEC was the most commonly identified enterotoxin (26/39), and it was almost exclusively present in dairy product isolates. Few isolates harbored seb, sed, and see genes. Six S. aureus isolates carried sea alone, two bacterial isolates carried both sea and seb, and one isolate carried both sea and sec. The sea gene was detected more frequently in isolates from meat and food preparations (4/11) than from dairy products (5/91). SEA was previously reported to be involved in outbreaks of food poisoning in Italy from seafood salads (Gallina et al., 2013) and in Japan from skimmed milk and yogurt (Asao et al., 2003).
Conclusions
The present study of toxigenic S. aureus strains is the first to be performed on food samples, following a case of food poisoning in Sicily, and revealed that many isolates contain at least one exotoxin gene. Since pathogenic bacteria can circulate in healthy farms, controls “from farm to fork” are important to enhance food safety.
Footnotes
Acknowledgments
The authors wish to thank Drs. Anna Carrozzo, Maria Anna La Giglia, and Franco Sciurba for scientific and technical support and Luca Dolce for reading the manuscript. This work was financially supported by research grants provided by the Italian Ministry of Health RC IZS Si 5/ 2011 to C.P. and RC IZS Si 15/ 2011 to M.V.
Disclosure Statement
No competing financial interests exist.