Detection of Staphylococcus aureus and enterotoxin genotype ...

Letters in Applied Microbiology ISSN 0266-8254

ORIGINAL ARTICLE

Detection of Staphylococcus aureus and enterotoxin genotype diversity in Monte Veronese, a Protected Designation of Origin Italian cheese A. Poli1, E. Guglielmini2, S. Sembeni1, M. Spiazzi2, F. Dellaglio2, F. Rossi2 and S. Torriani2 1 Dipartimento di Medicina e Sanita` Pubblica, Universita` di Verona, Verona, Italy 2 Dipartimento Scientifico e Tecnologico, Universita` di Verona, Verona, Italy

Keywords food safety, multiplex PCR, PDO cheese, realtime PCR detection, SE genes diffusion, Staphylococcus aureus. Correspondence Sandra Torriani, Dipartimento Scientifico e Tecnologico, Universita` di Verona, Strada Le Grazie 15, 37134 Verona, Italy. E-mail: [email protected]

2006 ⁄ 1796: received 20 December 2006, revised and accepted 28 June 2007 doi:10.1111/j.1472-765X.2007.02224.x

Abstract Aims: To evaluate the risk associated with the load and enterotoxigenicity of Staphylococcus aureus in Monte Veronese, a PDO (Protected Designation of Origin) cheese of the Lessinia area (Verona, Italy). Methods and Results: Staphylococcus aureus was quantified by a conventional culture method and by a nucA targeted real-time PCR assay developed in this study. Staphylococcus aureus numbers in cheese were higher than the limit tolerated by the Italian food legislation in 78% instances, according to both detection methods. Multiplex PCR tests for 17 Staph. aureus enterotoxin (SE) genes were applied to nucleic acids extracted from curds, cheeses and Staph. aureus isolates. The SE gene diversity appeared reduced after ripening. The gene encoding SED was found most frequently in dairy samples and the enterotoxin genes ser, sed, seg and sem predominated in the isolates. Conclusions: The occurrence of enterotoxigenic Staph. aureus strains with complex SE genotypes in this PDO cheese at numbers often exceeding the Italian tolerance threshold represents an important risk factor. Significance and Impact of the Study: The high frequency of contamination of Monte Veronese PDO cheese and, expectedly, similar typical productions from raw milk, by enterotoxigenic Staph. aureus imposes a tighter hygienic control in the earlier manufacturing phases.

Introduction Typical dairy productions of the European countries survive thanks to protective policies that defend their quality and genuineness. The assignment of the ‘Protected Designation of Origin’ (PDO) status (Commission Regulation 1107 ⁄ 96 ⁄ EC 1996) fixes the requisites of authenticity for products from defined geographical regions, that comprise the observance of specific manufacturing procedures. The majority of these products is made with raw milk. Much research has been published for the definition of their microbiological typicality, but information on the microbiological safety is nearly absent in the recent literature with exception of a report regarding food-borne disease outbreaks involving dairy products (De Buyser et al. 2001). More investigations on the risks

related to the possible microbial hazards in dairy products are required to check and improve the healthiness of such products. Strains of Staphylococcus aureus are frequent cause of diseases in humans and animals and alimentary toxinfection due to the ingestion of staphylococcal enterotoxins (SEs) is frequent. Mildly or not heated foods are often implicated in poisoning by SEs (Le Loir et al. 2003). To date 18 SE genes, sea–see, seg–ser and seu, have been identified on the basis of sequence similarities. The encoded products act as superantigens, by stimulating T-cells proliferation with release of a massive amount of cytokines that cause the illness symptoms (Omoe et al. 2005). The aim of this study was to evaluate the risk associated with the load and enterotoxigenicity of Staph.

ª 2007 The Authors Journal compilation ª 2007 The Society for Applied Microbiology, Letters in Applied Microbiology 45 (2007) 529–534

529

Staph. aureus SEs in a PDO cheese

A. Poli et al.

aureus in Monte Veronese, a PDO cheese of semi hard or hard consistence, varying with the ripening time of at least 25 days, produced in the Lessinia area (Verona, Italy) from raw cow milk. The production volume per year is 60 000 wheels, from the daily transformation of about 15 t of milk. A real-time PCR assay, sensitive enough to quantify Staph. aureus at levels lower than the tolerance threshold of 103 CFU g)1 fixed by the Italian food protection legislation, was developed to analyse dairy samples collected during cheese manufacturing and ripening in winter, spring and summer. Moreover, the multiplex PCR assays designed by Omoe et al. (2005) for the detection of SE genes were applied to analyse their presence in samples and in some Staph. aureus isolates. This improved the estimation of the enterotoxins related risk compared to the commercial SET-RPLA kits, as almost all known SEs were analysed. Materials and methods Reference staphylococcal strains and cultural methods Staphylococcus aureus LMG 8224, ATCC 8095 and ATCC 8096, Staphylococcus epidermidis LMG 10474T, Staphylococcus carnosus DSM 11676T, Staphylococcus simulans DSM 20322T, Staphylococcus xylosus DSM 20266T were used to assess the specificity of the thermonuclease gene (nucA) targeted PCR assay. The strains were propagated in Brain Heart Infusion broth (BHI; Oxoid, Basingstoke, UK) by 24–48 h incubation in aerobiosis at 37C. Counts of presumptive Staph. aureus in curd and cheese were carried out by plating serial dilutions on Baird Parker agar medium supplemented with egg yolk and tellurite (Merck, Darmstadt, Germany). Coagulase positive colonies were recognized by addition of 500 ll rabbit plasma (Oxoid) to 500 ll of a 24-h culture in BHI and incubation at 37C for 4 h. Sample treatment Monte Veronese PDO cheese from small dairies of the Lessinia area was sampled in years 2005–2006 during November–February, March–May and July–September, at the mature curd stage (21 samples) and after ripening for one (16 samples) and 3 months (nine samples). A 25 g sample was homogenized in 225 ml of physiological solution (1 g l)1 peptone, 9 g l)1 NaCl) in a Stomacher Laboratory Blender (Stomacher 400, Seward; PBI international, Milan, Italy) for plate counts. For direct DNA extraction from samples, 25 g of curd or cheese were suspended in 100 ml of 20 g l)1 trisodium citrate solution, and homogenized in filter bags. Sample 530

suspensions were centrifuged at 8000 g for 10 min at 4C. The cell pellet and sample fine particulate were resuspended in 20 ml TE buffer (10 mmol l)1 Tris ⁄ HCl, 1 mmol l)1 EDTA, pH 8Æ0). To this suspension, 2 ml ammonium hydroxide, 2 ml absolute ethanol, 4 ml petroleum ether, 2 ml 6 mol l)1 urea and 130 ll sodium acetate were added according to the cell purification procedure developed by Drake et al. (1996). Such mixture was centrifuged for 10 min at 8000 g and the pellet resuspended in 6 ml TE buffer. DNA extraction DNA extraction from pure cultures of 20 ml was carried out on cell pellets harvested by centrifugation for 10 min at 8000 g and washed in an equal volume of TE buffer. DNA extraction from cheese samples was carried out on the cell suspensions obtained as above after centrifugation for 5 min at 8000 g. For both sample types, the cell pellets were resuspended in 500 ll of TE buffer containing 10 mg ml)1 lysozyme (Sigma-Aldrich, Milan, Italy) and incubated at 37C for 1 h. After 30-min incubation at 37C with 50 lg ml)1 ribonuclease A (SigmaAldrich) and 5 g l)1 SDS, and a second incubation at 55C for 30 min with 10 lg ml)1 proteinase K (SigmaAldrich), the DNA was purified by chloroform extraction and precipitated with two volumes of ethanol plus 20 mg ml)1 glycogen (MBI Fermentas, Milan, Italy). After centrifugation at maximum speed for 10 min at 4C, the DNA pellet was dried and resuspended in 50 ll Tris ⁄ HCl 10 mmol l)1, pH 8Æ0. Quantitative real-time PCR The primers s.au-F (CAAGGCTTGGCTAAAGTTGC) and s.au-R (CTGAATCAGCGTTGTCTTCG) (Sigma-Genosys, Haverhill, UK), flanking a 127-bp DNA region of the Staph. aureus nucA gene, were designed by the Primer Express software (Applied Biosystems, Foster City, CA, USA) verifying specificity by BLASTN alignment in GenBank. Amplifications were carried out in a 25 ll volume, using 1x Platinum SYBR Green I SupermixUDG reaction mixture, 1x Rox (Invitrogen, Milan, Italy), 0Æ2 lmol l)1 each primer and 2 ll of DNA suspension. The thermal cycling programme was 50C for 2 min, 95C for 5 min and 45 cycles of 95C for 20 s, 58C for 40 s, 72C 1 min. PCR and data analysis were performed by the ABI Prism 7000 sequence detection system (Applied Biosystems). The correlation coefficient between the quantitative real-time PCR data series, expressed in log CFU g)1, and conventional counts was calculated by the ‘correl’ function in Microsoft Excel.


A. Poli et al.


Multiplex PCR assays

40

Multiplex PCR tests with the primers designed by Omoe et al. (2005) were performed in 25 ll reaction mixtures using 0Æ3 lmol l)1 of each primer and 2 ll of DNA in 2x QIAGEN Multiplex PCR Master Mix (QIAGEN Spa, Milan, Italy). For PCR sets 1 and 4, the PCR mix contained also 5 ll of 5x Q Solution (QIAGEN). Thermal cycling conditions were: 95C for 15 min, 35 cycles of 95C for 30 s, 57C (for sets 1 and 2) or 58C (for sets 3 and 4) for 90 s and 72C for 90 s plus final extension at 72C for 10 min.

35 30 Ct

25 20 15 10 5 0

0

1

2

3

4

5 6 log CFU

7

8

9

10

Figure 2 Standard curves log CFU per PCR reaction vs Ct obtained with DNA from cell culture dilutions (4) and from experimentally inoculated cheese (h).

Immunological assay Cheese samples positive for classical SE genes sea–sed were homogenized and tested for SEA–SED by the Staphylococcus enterotoxin reversed passive latex agglutination test (SET-RPLA Kit; Oxoid). Results Detection of Staph. aureus in dairy samples by quantitative real-time PCR The primers designed on the nucA gene were tested on staphylococcal collection strains and showed the formation of a unique amplicon dissociating at 74–75C only for the three Staph. aureus reference strains. Figure 1 shows the melting curves of PCR reactions from different cell numbers of Staph. aureus LMG 8224. Triplicate trials on DNA samples from known cell numbers resulted in a reproducible detection limit of approx. 102 Staph. aureus CFU per reaction from pure cultures. The standard curve log CFU against threshold cycle (Ct) indicated a high amplification efficiency, with a slope value of )3Æ58, close to the theoretical optimum of )3Æ30, linearity range

7–3 log CFU and correlation coefficient of 0Æ99 (Fig. 2). A standard curve log CFU against Ct was constructed also for amplifications from artificially inoculated cheeses. Reproducible and accurate quantification, with linearity in the range 102–106 CFU per PCR reaction, slope value of )3Æ06, correlation coefficient of 0Æ99 and detection limit of 102 CFU, was accomplished (Fig. 2). The quantitative real-time PCR assay was applied to 46 dairy samples (curds and cheeses) analysed also by standard cultural methods. Staphylococcus aureus negative samples were concordantly recognized, however, culturing often showed slightly higher contamination levels than PCR (Table 1). Nevertheless, the correlation coefficient for the two series of data was high (0Æ98). Staphylococcus aureus population in cheese exceeded the threshold of 103 CFU g)1 for 78% cheese samples and was higher of at least one order of magnitude in 40% and 52% cases, based on the real-time PCR and count analyses, respectively (Table 1).

dF/dT

Detection of SE genes in dairy samples by multiplex PCR 0·5 0·45 0·4 0·35 0·3 0·25 0·2 0·15 0·1 0·05 0

70

75 80 Temperature (°C)

85

Figure 1 Melting curves of quantitative real-time PCR with primers s.au-F ⁄ s.au-R for reaction mixtures containing DNA from 6 · 102 to 6 · 105 cells of Staphylococcus aureus LMG 8224.

The presence of 17 SE genes was examined in DNAs from all the Staph. aureus positive samples and all were found to contain one or more target gene. These, listed in the order of decreasing detection frequencies, were sed, ser, seb, sem, sea, seg and sek, sei, sen and seo, while genes see, seh, sel and seq were never detected (Table 1). The Staph. aureus loads in curd and cheese were comparable, but the frequency and diversity of SE genes was higher in curd; seg and sep were found only in curd samples or in isolates from curd (Table 1). Seasonal variations were observed: seb and sei diminished in summer, while sec occurred mostly in summer, sek increased in spring and summer, sen was found only in winter and seo was not detected in spring (Table 1).


531


A. Poli et al.

Table 1 Quantification of Staphylococcus aureus and SE gene sets found in curd (C) and in Monte Veronese cheese ripened for 1 month (MF) and 3 months (MA) in winter (I), spring (P) and summer (E). SE genotypes of single isolates from each sample. The dairies of provenance are indicated by numerals. Staph. aureus (CFU g)1)*

quantification

Sample

pH

Real-time PCR

Cultural method

SE genes detected

SE genotype of one isolate

1CI 2CI 3CI 4CI 5CI 6CI 1MFI 2MFI 3MFI 4MFI 5MFI 1MAI 2MAI 1CP 2CP 3CP 4CP 5CP 6CP 7CP 8CP 1MFP 2MFP 3MFP 4MFP 5MFP 1MAP 2MAP 3MAP 4MAP 5MAP 1CE 2CE 3CE 4CE 5CE 6CE 7CE 1MFE 2MFE 3MFE 4MFE 5MFE 6MFE 1MAE 2 MAE

5Æ31 5Æ28 5Æ33 5Æ41 5Æ36 5Æ28 5Æ66 5Æ62 5Æ74 5Æ67 5Æ63 5Æ72 5Æ78 5Æ29 5Æ24 5Æ29 5Æ32 534 5Æ29 5Æ22 5Æ25 5Æ71 5Æ78 5Æ63 5Æ74 5Æ80 5Æ79 5Æ76 5Æ79 5Æ8 5Æ81 5Æ24 5Æ31 5Æ40 5Æ32 5Æ28 5Æ28 5Æ33 5Æ68 5Æ65 5Æ69 5Æ71 5Æ69 5Æ77 5Æ72 5Æ75

3 · 104 4 · 103