Pheno-and Genotyping of Staphylococcus aureus I solates of Sheep Origin

The aim of the study was to determine the prevalence of genes encoding virulence factors in Staphylococcus aureus strains isolated from raw sheep milk, sheep cheese and Bryndza cheese. Genes encoding staphylococcal enterotoxin (sea, seb, sec, sed and see), toxic shock syndrome toxin-1 (tst), exfoliative toxins (eta and etb) and collagen-binding protein (cna) were detected. In a total of 79 S. aureus isolates all assessed toxins encoding genes were found, except for see, eta and etb. Overall, 75.9% of S. aureus isolates were found to be positive for one or more toxin genes. The sec gene was found most frequently (24.1%), followed by tst (22.8%), seb (13.9%), sed (10.1%) and sea (5.1%). The cna gene was detected in 55.7% of S. aureus isolates. Based on tandem repeats in coa gene, five coa types were observed, further divided into 16 subtypes based on their RFLP pattern. Similarly tandem repeats in spa gene divided S. aureus isolates into 7 types. In the parallel antibiotic resistance study, 69.6% isolates were resistant to at least one of the 11 tested antibiotics. The phenoand genotyping of S. aureus isolates of sheep origin presented in this work update the epidemiological data in Slovakia. Milk, cheese, sheep-bacteria, toxins, coa and spa genotyping, antimicrobial resistance Staphylococcus aureus, the most important aetiological agent of contagious bovine and ovine mastitis, has attracted attention by its presence in the infected udder and by environmental contamination of milk during handling and processing (Scherrer et al. 2004; Jorgensen et al. 2005). Epidemiological survey is therefore important to prevent the spread of S. aureus. The virulence of S. aureus has been postulated to depend on the expression of a wide range of cell wall-associated and secreted molecules that are believed to promote colonization of host tissues and evasion of the host immune response (Foster and Hook 1998; Dinges et al. 2000). Adherence to host tissue is the first critical step required to initiate infection and is mediated by collagen-binding protein (Cna) and adhesins of the microbial surface components, recognizing adhesive matrix molecules (MSCRAMM) (Foster and Hook 1998). S. aureus also produces a wide variety of extracellular toxins of which the most important are staphylococcal enterotoxins (SEs), toxic shock syndrome toxin-1 (TSST-1), and exfoliative toxins (ET). Each of these toxins has a potent effect on cells of the immune system (Dinges et al. 2000). Some SEs are potent emetics (Boerema et al. 2006). TSST-1 is unique in its ability to cross mucosal surfaces and to develop life-threatening toxic shock syndrome (Llewelyn and Cohen 2002; Proft and Fraser 2003). Because S. aureus is a common pathogen for humans and animals, many studies were orientated towards staphylococcal antibiotic resistance (Lyon and Skurray 1987; Werckenthin et al. 2001; Strommenger et al. 2003; Normanno et al. 2007; Wang et al. 2008). In human medicine, antimicrobial multiresistance is frequently encountered and methicillin-resistant S. aureus strains (MRSA) belong to the most life threatening nosocomial bacteria. In veterinary medicine, however, MRSA and multiresistant S. aureus strains are only reported occasionally (Werckenthin et al. 2001; Normanno et al. 2007). Little is known about the occurrence of virulence factors in ovine S. aureus strains. The ACTA VET. BRNO 2009, 78: 345–352; doi:10.2754/avb200978020345 Address for correspondence: Doc. MVDr. Ludmila Tkáčiková, Ph.D. Department of Microbiology and Immunology University of Veterinary Medicine Komenského 73, 941 81 Košice Slovakia Phone: +421 915 984 603 E-mail: tkacikova@uvm.sk http://www.vfu.cz/acta-vet/actavet.htm objective of this study was phenoand genotyping of ovine S. aureus strains isolated from sheep milk, sheep cheese and Bryndza cheese (typical Slovak sheep cheese) samples from Slovakia. Materials and Methods Bacterial strains and culture media During the years 2004 2007, raw sheep milk and cheese samples were screened for staphylococcal contamination by classical microbiological examination (ISO 6888-1). Based on colony morphology (growth on Baird-Parker agar) and coagulase production 79 possible S. aureus isolates (49 raw sheep milk, 24 sheep cheese and 6 Bryndza cheese) were selected for further genospecies confirmation and geno-phenotyping. Reference strains used in PCR based assays were: S. aureus CCM 2353 (cna gene), S. aureus CCM 7056 (eta and etb genes), S. aureus CCM 5756 (sea gene), S. aureus CCM 5984 (sec gene) and S. aureus CCM 5972 (see gene) (Czech Collection of Microorganisms, Brno, Czech Republic). Working cultures of all strains were prepared by inoculation from frozen glycerol stocks into brain heart infusion (BHI) broth (Oxoid, Basingstoke, Hampshire, UK) followed by incubation at 37 °C for 16 18 h. Antibiotic susceptibility testing disc diffusion assays Discs (Oxoid) with oxacillin (OXA), chloramphenicol (CMP), tetracycline (TET), erythromycin (ERY), gentamycin (GEN), kanamycin (KAN), streptomycin (STR), vankomycin (VAN), linkomycin (LCM), clindamycin (CLI) and spiramycin (SPI) were added onto S. aureus inoculated Mueller-Hinton agar plates (Oxoid) (Table 6). The zone of inhibition was determined after 24 h of incubation at 37 °C. The zone size was interpreted according to the recommendations of the CLSI criteria (CLSI, 2004). Nucleic acid amplification PCR: Nucleic acid amplification was performed on S. aureus genomic DNA isolated according to (Hein et al. 2005). PCRs were performed either in multiplex (set A, set B and set C) (Sharma et al. 2000) or as a single reaction (mecA, cna, spa and coa) (Table 1). Cycling conditions for each type of PCR were according to the authors depicted in Table 1. Amplified products were detected in 1.5% agarose gel. DNA restriction endonuclease analysis of the PCR-amplified coagulase gene For coagulase genotyping, a PCR-RFLP analysis of isolates was performed according to the method of Hookey et al. (1998). Ten μl of PCR product was digested with 2 U of HaeIII (Fermentas, Vilnius, Lithuania) at 37 °C overnight. After restriction the samples were analysed by electrophoresis in 2% agarose gel. Results and Discussion Specific gene amplification of 16S rRNA and S. aureus (Martineau et al. 1998) confirmed 79 S. aureus isolates already selected based on the microbiological criteria like colony characterization and coagulase production. In 79 S. aureus isolates all the assessed toxin genes were found, except for see, eta and etb. Overall, 75.9% S. aureus isolates were found to be positive for one or more toxin genes. The sec gene was found most frequently (24.1%), followed by tst (22.8%), seb (13.9%), sed (10.1%) and finally, sea (5.1%). The tst gene was the only gene present in all types of isolates (Table 2). The presence of sea, seb, sed, see and tst genes in S. aureus strains associated with bovine mastitis has been described earlier (Akineden et al. 2001; Stephan et al. 2001; Scherrer et al. 2004; Zschock et al. 2004; Morandi et al. 2007). Of the staphylococcal enterotoxin encoding genes, the sec gene has been reported to have the highest frequency in bovine as well as ovine isolates (Scherrer et al. 2004). We obtained similar results, where out of 42 enterotoxin positive strains, 19 (45.2%) were sec positive. The presence of more than one toxin genes was found in 15.2% S. aureus isolates. In our study, from all 18 tst positive isolates, 6 (33.6%) were also positive for the sec gene. The strains positive for sec-tst were isolated mostly from sheep cheese; only one with the sebsec-tst gene combination was found in sheep milk isolate (Table 3). A high co-occurrence between sec and tst has also been reported for bovine S. aureus strains (Akineden et al. 2001; Stephan et al. 2001; Zschock et al. 2004). Smyth et al. (2005) published a high prevalence (19.2%, 46.2% and 60.9%) of the tst gene in bovine, goat and sheep isolates, respectively. In the present study the tst gene was the only gene found in isolates from Bryndza cheese. These isolates did not show any double presence of toxin genes. Whether TSST-1 plays a role in the pathogenesis of mastitis is still unknown. 346

Staphylococcus aureus, the most important aetiological agent of contagious bovine and ovine mastitis, has attracted attention by its presence in the infected udder and by environmental contamination of milk during handling and processing (Scherrer et al. 2004;Jorgensen et al. 2005).Epidemiological survey is therefore important to prevent the spread of S. aureus.
The virulence of S. aureus has been postulated to depend on the expression of a wide range of cell wall-associated and secreted molecules that are believed to promote colonization of host tissues and evasion of the host immune response (Foster and Hook 1998;Dinges et al. 2000).Adherence to host tissue is the first critical step required to initiate infection and is mediated by collagen-binding protein (Cna) and adhesins of the microbial surface components, recognizing adhesive matrix molecules (MSCRAMM) (Foster and Hook 1998).S. aureus also produces a wide variety of extracellular toxins of which the most important are staphylococcal enterotoxins (SEs), toxic shock syndrome toxin-1 (TSST-1), and exfoliative toxins (ET).Each of these toxins has a potent effect on cells of the immune system (Dinges et al. 2000).Some SEs are potent emetics (Boerema et al. 2006).TSST-1 is unique in its ability to cross mucosal surfaces and to develop life-threatening toxic shock syndrome (Llewelyn and Cohen 2002;Proft and Fraser 2003).
Because S. aureus is a common pathogen for humans and animals, many studies were orientated towards staphylococcal antibiotic resistance (Lyon and Skurray 1987;Werckenthin et al. 2001;Strommenger et al. 2003;Normanno et al. 2007;Wang et al. 2008).In human medicine, antimicrobial multiresistance is frequently encountered and methicillin-resistant S. aureus strains (MRSA) belong to the most life threatening nosocomial bacteria.In veterinary medicine, however, MRSA and multiresistant S. aureus strains are only reported occasionally (Werckenthin et al. 2001;Normanno et al. 2007).
Little is known about the occurrence of virulence factors in ovine S. aureus strains.The objective of this study was pheno-and genotyping of ovine S. aureus strains isolated from sheep milk, sheep cheese and Bryndza cheese (typical Slovak sheep cheese) samples from Slovakia.

Materials and Methods
Bacterial strains and culture media During the years 2004 -2007, raw sheep milk and cheese samples were screened for staphylococcal contamination by classical microbiological examination (ISO 6888-1).Based on colony morphology (growth on Baird-Parker agar) and coagulase production 79 possible S. aureus isolates (49 raw sheep milk, 24 sheep cheese and 6 Bryndza cheese) were selected for further genospecies confirmation and geno-phenotyping.Reference strains used in PCR based assays were: S. aureus CCM 2353 (cna gene), S. aureus CCM 7056 (eta and etb genes), S. aureus CCM 5756 (sea gene), S. aureus CCM 5984 (sec gene) and S. aureus CCM 5972 (see gene) (Czech Collection of Microorganisms, Brno, Czech Republic).Working cultures of all strains were prepared by inoculation from frozen glycerol stocks into brain heart infusion (BHI) broth (Oxoid, Basingstoke, Hampshire, UK) followed by incubation at 37 °C for 16 -18 h.
Nucleic acid amplification -PCR: Nucleic acid amplification was performed on S. aureus genomic DNA isolated according to (Hein et al. 2005).PCRs were performed either in multiplex (set A, set B and set C) (Sharma et al. 2000) or as a single reaction (mecA, cna, spa and coa) (Table 1).Cycling conditions for each type of PCR were according to the authors depicted in Table 1.Amplified products were detected in 1.5% agarose gel.

DNA restriction endonuclease analysis of the PCR-amplified coagulase gene
For coagulase genotyping, a PCR-RFLP analysis of isolates was performed according to the method of Hookey et al. (1998).Ten μl of PCR product was digested with 2 U of HaeIII (Fermentas, Vilnius, Lithuania) at 37 °C overnight.After restriction the samples were analysed by electrophoresis in 2% agarose gel.

Results and Discussion
Specific gene amplification of 16S rRNA and S. aureus (Martineau et al. 1998) confirmed 79 S. aureus isolates already selected based on the microbiological criteria like colony characterization and coagulase production.
In 79 S. aureus isolates all the assessed toxin genes were found, except for see, eta and etb.Overall, 75.9% S. aureus isolates were found to be positive for one or more toxin genes.The sec gene was found most frequently (24.1%), followed by tst (22.8%), seb (13.9%), sed (10.1%) and finally, sea (5.1%).The tst gene was the only gene present in all types of isolates (Table 2).
The presence of sea, seb, sed, see and tst genes in S. aureus strains associated with bovine mastitis has been described earlier (Akineden et al. 2001;Stephan et al. 2001;Scherrer et al. 2004;Zschock et al. 2004;Morandi et al. 2007).Of the staphylococcal enterotoxin encoding genes, the sec gene has been reported to have the highest frequency in bovine as well as ovine isolates (Scherrer et al. 2004).We obtained similar results, where out of 42 enterotoxin positive strains, 19 (45.2%) were sec positive.
The presence of more than one toxin genes was found in 15.2% S. aureus isolates.In our study, from all 18 tst positive isolates, 6 (33.6%) were also positive for the sec gene.The strains positive for sec-tst were isolated mostly from sheep cheese; only one with the sebsec-tst gene combination was found in sheep milk isolate (Table 3).
A high co-occurrence between sec and tst has also been reported for bovine S. aureus strains (Akineden et al. 2001;Stephan et al. 2001;Zschock et al. 2004).Smyth et al. (2005) published a high prevalence (19.2%, 46.2% and 60.9%) of the tst gene in bovine, goat and sheep isolates, respectively.In the present study the tst gene was the only gene found in isolates from Bryndza cheese.These isolates did not show any double presence of toxin genes.Whether TSST-1 plays a role in the pathogenesis of mastitis is still unknown.The common "forward" primer; 2 Variable product sizes depending on the number of repeat units TSST-1 and staphylococcal enterotoxins may act as superantigens for cells of the bovine immune system and may potentially contribute to the pathological mechanisms of bovine mastitis.Some other authors also described high co-occurrence of sec and tst genes in sheep (Orden et al. 1992;Hayakawa et al. 2000;Scherrer et al. 2004).This data suggest that the sec gene may be present on a pathogenicity island together with the tst gene (Fitzgerald et al. 2001).
Many authors studied the production of exfoliative toxins among S. aureus isolates from bovine mastitis and have reported a rare prevalence of these toxins (Hayakawa et al. 2000;Akineden et al. 2001;Larsen et al. 2002;Endo et al. 2003).The present study confirms that these genes are rare in S. aureus isolates from sheep; eta and etb genes were not found in any of our isolates.
In this study we found a high prevalence of the collagen binding protein (cna) gene (55.7%).The gene was found most frequently in isolates from Bryndza cheese (83.3%), followed by sheep milk (55.1%) and sheep cheese (50%).The collagen binding protein mediates bacterial adherence to collagen substrates and collagenous tissue and can therefore be a virulence factor of S. aureus (Peacock et al. 2002;Zong et al. 2005).The cna gene is usually not present in S. aureus strains (Elasri et al. 2002).
Production of coagulase encoding by the coa gene is an important phenotypic feature used worldwide to identify S. aureus (Sutra and Poutrel 1994;Panizzi et al. 2004).In the present study five different amplicons of the coa gene with sizes ranging from 500-820 bp were found: type A (500 bp with 4 tandem repeats; 4TR), type B (580 bp, 5TR), type C (660 bp, 6TR), type D (740 bp, 7TR) and type E (820 bp, 8TR).Only type D (7TRs) was present in all isolates.6TR and 7TR together account for the majority of the isolates (63%) (Table 4).
The use of PCR-RFLP for analysis of the coa gene allowed a more detailed characterization of the S. aureus isolates.In this study the highest polymorphism was recorded in type C, with 4 or 5 subtypes in sheep milk and sheep cheese isolates, respectively, and in type D with 4, 2 or 1 subtype in sheep milk, sheep cheese and Bryndza cheese isolates, respectively.Some coa types or subtypes were recorded only in S. aureus isolates from sheep cheese (B, E and C5-C7) or sheep milk (C3, C4, D3 and D4) (Table 4).Katsuda et al. (2005) and other authors (Lange et al. 1999;Schlegelová et al. 2003) described a number of tandem repeats in the coa gene ranging from 3 to 9 and bovine 348 S. aureus strains mostly showing five tandem repeats.Scherrer et al. (2004) studied polymorphisms in the coa gene of S. aureus isolates from sheep and goat milk.They report an 80.6% prevalence of eight tandem repeats in sheep milk samples.In sheep isolates we only found one with 5TR.This could indicate a connection between mastitis and 5TR in the coa gene of S. aureus isolates, as our isolates were not specifically from animals with mastitis.
The detection of variation in the X region of the spa gene is also used for the epidemiological study of S. aureus.Amplicons of the spa gene with different sizes ranging from 224 -392 bp were observed in 82.4% S. aureus isolates (Table 5).Based on the size of the corresponding amplicons, 4 to 11 highly polymorphic tandem repeats (TR, 24 bp long) (Frenay et al. 1994;Shopsin et al. 1999) were supposed to be present in this X region.In the present study only two spa types (5TR and 6TR) were found in all isolates.Some spa  types were observed only in isolates from sheep milk (4TR, 7TR and 11TR) or only from sheep cheese (9TR) (Table 5).
Several authors studied polymorphisms in the X region of the spa gene of bovine S. aureus.Lange et al. (1999) recorded the presence of 5TR to 12TR (except 10TR) with a predominance of 6TR.Stephan et al. (2001) recorded the presence of 2TR, 6TR, 10TR and 11TR with a predominance of 6TR and Kalorey et al. (2007) recorded the presence of 4TR, 5TR, and 8TR with a predominance for 4TR.To the best of our knowledge, this publication describes the variation of the X-region of the spa gene in isolates from sheep milk and sheep cheese for the first time.
Over the last few decades, there has been an enormous increase and emergence of S. aureus strains resistant to the antibiotic methicillin (MRSA strains).MRSA is known to be one of the most prevalent nosocomial pathogens throughout the world and to be capable of causing a wide range of hospital-linked infections (Mehrotra et al. 2000).In the present study, methicillin resistance was not present in any of the tested S. aureus isolates.Pengov and Ceru (2003) also recorded a low prevalence of resistance to penicillin and ampicillin in ovine isolates.This can be explained by the reduction of the use of b-lactam antibiotics for treatment of sheep; the presence of a different population of S. aureus genotypes in the ovine mammary gland is likely to be partly responsible for this as well.
In the present study all staphylococcal isolates were susceptible to oxacillin, tetracycline and gentamycin.A low level of resistance to chloramphenicol, kanamycin and streptomycin was recorded in isolates from sheep milk only.Resistance to vankomycin was recorded only in one isolate from sheep cheese.A high prevalence of resistance to linkomycin, spiramycin and to clindamycin was recorded (Table 6).The prevalence of resistance was higher in isolates from sheep milk than in isolates from sheep cheese and from Bryndza cheese.We also found multiresistant S. aureus isolates.Multiresistance to CMP-ERY-KAN-STR-LCM-SPI, ERY-VAN-LCM-CLI-SPI or LCM-CLI-SPI was recorded in one staphylococcal isolate from sheep milk, sheep cheese and Bryndza cheese, respectively.Resistance to LCM-SPI was recorded in 10 isolates; resistance to LCM-CLI was recorded in 4 isolates.
The reason for a high prevalence of resistance to linkomycin, spiramycin and clindamycin in our study is probably the use of these antibiotics for treatment of sheep mastitis.Wang at al. (2008) also reported a considerably higher resistance to lincosamides (linkomycin and clindamycin) or macrolides (erythromycin and spiramycin) in S. aureus strains isolated from bovine mastitis.
To our knowledge, this is the first study providing comprehensive characterisation data of S. aureus strains originating from sheep milk and sheep cheese samples in Slovakia.
CGG TAC ACG ATA TTC TTC ACG 107 sau2 CGT AAT GAG ATT TCA GTA GAT AAT ACA ACA Sa-U TGT ATG TAT GGA GGT GTA AC

Table 1 .
Characteristics of the primers used in this study

Table 2 .
Distribution of toxin encoding genes among S. aureus isolates

Table 4 .
Polymorphism in coa gene of S. aureus isolates

Table 5 .
Polymorphism in numbers of tandem repeats in X region of spa gene of S. aureus isolates

Table 6 .
Distribution of antibiotic resistance among of S. aureus isolates