Characterization of enterotoxigenic Staphylococcus aureus strains isolated from bovine mastitis in north-east Switzerland

Veterinary Microbiology 78 (2001) 373±382

Characterization of enterotoxigenic Staphylococcus aureus strains isolated from bovine mastitis in north-east Switzerland R. Stephana, C. AnnemuÈllerb, A.A. Hassanb, Ch. LaÈmmlerb,* a

Institut fuÈr Lebensmittelsicherheit und -hygiene der UniversitaÈt ZuÈrich, Winterthurerstr. 270, CH-8057 ZuÈrich, Switzerland b Institut fuÈr TieraÈrztliche Nahrungsmittelkunde, Bakteriologie und Hygiene der Milch, der Justus-Liebig-UniversitaÈt Giessen, Frankfurter Str. 107, 35392 Giessen, Germany

Received 21 April 2000; received in revised form 15 August 2000; accepted 28 August 2000

Abstract Thirty-four strains of enterotoxin-producing Staphylococcus aureus obtained from milk samples of 34 dairy cows suffering from mastitis from 34 different farms in north-east Switzerland were identi®ed and further characterized by pheno- and genotypic methods. This included the identi®cation of staphylococcal enterotoxin (SE) types, an antibiotic resistance testing, the appraisal of hemolysis, the egg yolk reaction, the detection of the clumping factor and protein A by means of a latex agglutination, the PCR ampli®cation of a S. aureus speci®c part of the gene encoding the 16S±23S rRNA ``intergenic spacer'' region and a species speci®c part of the 23S rRNA-gene, the PCR ampli®cation of the clumping factor (clfA) gene, the X region and the IgGbinding region of the protein A (spa) gene, the coagulase (coa) gene and additionally a macrorestriction analysis of the chromosomal DNA. Within the 26 cultures which formed a single SE, there were 23 SEC- and three SED-formers. Eight cultures were SEAD formers. It was remarkable that 22 SEC formers were also positive for TSST-1. Eighteen of the 23 SEC-formers could be classi®ed as being of the same phenotype. Most of the cultures of one enterotoxin type also showed a great uniformity in the size and number of repeats of the X region as well as in the size of the IgG-binding region of protein A gene and in the size of the coagulase gene. Macrorestriction analysis revealed 11 PFGE patterns. These were in part only different from each other in a few fragments and thus displayed close clonal relations. The results of the present investigation show that a broad distribution of identical or closely related enterotoxin-producing S. aureus clones seem to contribute to the bovine mastitis problem in north-east Switzerland. # 2001 Elsevier Science B.V. All rights reserved. Keywords: Cattle-bacteria; Staphylococcus aureus; Toxins; Mastitis; Epidemiology; Typing; Switzerland

0378-1135/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 1 3 5 ( 0 0 ) 0 0 3 4 1 - 2

374

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

1. Introduction Clinical and subclinical staphylococcal mastitis is also of considerable importance in Switzerland. According to SchaÈllibaum (1999), staphylococcal isolates were recovered from approximately 40% of mastitic milk samples from cows. Some authors have shown that in different countries only a few S. aureus clones are responsible for most of the cases of bovine mastitis and that these clones have a broad geographic distribution (Fitzgerald et al., 1997; Lange et al., 1999; AnnemuÈller et al., 1999). Some of the S. aureus cultures have the ability to form different enterotoxins and also the toxic shock syndrome toxin 1 (TSST-1) (ZschoÈck et al., 1998). This might be of signi®cance for food hygiene especially in cases of subclinical mastitis due to S. aureus and might also contribute to an increased udder pathogenicity of the organisms. The objective of this study was to characterize the pheno- and genotype of enterotoxinforming S. aureus strains which were derived from mastitic milk specimens from cows originating from a closely de®ned region of north-east Switzerland. This should provide information on the spreading of individual clones and on epidemiological correlations in this region. 2. Materials and methods 2.1. Bacterial isolates, identi®cation and further biochemical characterization Between February and June 1998, 34 enterotoxin-forming (SE) S. aureus strains were isolated from mastitic milk samples from 34 cows which came from 34 different herds in north-east Switzerland. The isolates were investigated for enterotoxin production with a sandwich-ELISA-kit (Bommeli, Bern, Switzerland). The 34 cultures were selected from a total number of 63 S. aureus isolated in this region. The other isolates were not enterotoxin producers (Stephan et al., 1999). The samples were taken and the pathogens were isolated by conventional methods (Anon., 1981). The cultures were identi®ed by means of API ID 32 Staph according to the recommendations of the manufacturer (bioMeÂrieux, GeneÁve, Switzerland) as well as by characterization of a S. aureus-speci®c section of the 16S±23S rRNA intergenic spacer region (Forsman et al., 1997; AnnemuÈller et al., 1999). According to Straub et al. (1999), the species speci®c primers Staur4 50 ACG GAG TTA CAA AGG ACG AC-30 and Staur6 50 -AGC TCA GCC TTA ACG AGT AC-30 were used for the detection of the 23S rRNA gene. The reaction mixture (30 ml) contained 1 ml primer 1 (10 pmol), 1 ml primer 2 (10 pmol), 0.6 ml dNTP (10 mM; MBI Fermentas, St. Leon Rot, Germany), 3:0 ml 10  thermophilic buffer (Promega/Boehringer, Ingelheim, Germany), 1.8 ml MgCl2 (25 mM; Promega/Boehringer), 0.2 ml Taq DNA polymerase (5 U/ml; Promega/ Boehringer) and 21.4 ml aquadest. Finally, 1 ml DNA preparation was added to each 0.2 ml reaction tube. For DNA preparation 5±10 colonies of the bacteria were incubated in 100 ml TE buffer [pH 7.6] containing 5 ml lysostaphin (1.8 U/ml; Sigma, Deisenhofen, Germany) for 45 min at 378C and with proteinase K (14.8 mg/ml, Boehringer) for 120 min at 568C. The suspension was heated 10 min at 958C to inactivate proteinase K, centrifuged (10,000g, 15 s) and cooled before use.

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

375

The production of hemolysins by the isolates were determined by the interaction of the hemolysins with the b-toxin of a S. aureus strain as described by Skalka et al. (1979). For this the b-toxin producing S. aureus was streaked down the centre of a sheep blood agar plate. The isolates were cultivated at an angle of 908 but not touching the b-toxin producing strain. After 24 h incubation at 378C the plates were examined for hemolysin enhancement (d) or inhibition (a). The cultures were phenotyped further by appraising the egg yolk reaction on Baird±Parker agar (Oxoid Wersel, Germany), and by detection of the clumping factor and protein A by means of latex agglutination (Staphaurex-Plus, Murex Diagnostika, Dantford, UK). 2.2. Toxin production Enterotoxin formation (SEA±SED) was again investigated for all strains using the SET RPLA toxin detection kit (Oxoid) in accordance with the manufacturer's instructions. The enterotoxins together with the TSST-1 (tst) and the exfoliative toxin (eta, etb) were determined by means of PCR as described by Johnson et al. (1991) and Tsen and Chen (1992) using the oligonucleotide primers for enterotoxin A sea1 50 -AAA GTC CCG ATC AAT TTA TGG CTA-30 and sea2 50 -GTA ATT AAC CGA AGG TTC TGT AGA-30 , for enterotoxin B seb1 50 -TCG CAT CAA ACT GAC AAA CG-30 and seb2 50 -GCA GGT ACT CTA TAA GTG CC-30 , for enterotoxin C-1 sec1 50 -GAC ATA AAA GCT AGG AAT TT-30 and sec2 50 -AAA TCG GAT TAA CAT TAT CC-30 , for enterotoxin D sed1 50 -CTA GTT TGG TAA TAT CTC CT-30 and sed2 50 -TAA TGC TAT ATC TTA TAG GG-30 , for enterotoxin E see1 50 -TAG ATA AAG TTA AAA CAA GC-30 and see2 50 -TAA CTT ACC GTG GAC CCT TC-30 , for TSST-1 tst1 50 -ATG GCA GCA TCA GCT TGA TA-30 and tst2 50 -TTT CCA ATA ACC ACC CGT TT-30 , for exfoliative toxin A eta1 50 -CTA GTG CAT TTG TTA TTC AA-30 and eta2 50 -TGC ATT GAC ACC ATA GTA CT-30 and for exfoliative toxin B etb1 50 -ACG GCT ATA TAC ATT CAA TT-30 and etb2 50 -TCC ATC GAT AAT ATA CCT AA-30 . The ampli®cation was carried out using the thermal cycler Techne Progene (thermo-DUX, Wertheim, Germany) with the following programme: 1  4 min precycle at 948C, 30  2 min at 948C, 2 min at 558C, 1 min at 728C, followed by a ®nal extension incubation at 728C for 5 min. Reference cultures for the various toxins were kindly provided by W. Witte, Robert Koch-Institut, Wernigerode, Germany. 2.3. Antibiotic resistance testing Sensitivity to antibiotics was determined by the standardized agar diffusion test (Bauer et al., 1966) on MuÈller±Hinton agar (Oxoid) using the following antibiotic impregnated disks: ampicillin (10 mg), cefoperazone (30 mg), cephalothin (30 mg), chloramphenicol (30 mg), erythromycin (15 mg), gentamicin (10 mg), lincomycin (15 mg), neomycin (30 mg), oxacillin (1 mg), penicillin G (10 mg), polymyxin B (300 U/IE), sulfamethoxazole/trimethoprim (23.75 mg/1.25 mg) and tetracycline (30 mg) (bioMeÂrieux). S. aureus ATCC 25923 was used as a reference strain. Zones of growth inhibition were evaluated according to the NCCLS (Anon., 1998).

376

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

2.4. PCR ampli®cation of the clumping factor (clfA) gene, the X region and the IgGbinding region of the protein A (spa) gene and of the 30 end of the coagulase (coa) gene The strains were additionally identi®ed and genotyped by PCR ampli®cation of the clumping factor (clfA) gene. The clfA speci®c oligonucleotide primers with the sequence 50 -GGC TTC AGT GCT TGT AGG-30 and 50 -TTT TCA GGG TCA ATA TAA GC-30 were designed according to the clfA sequence obtained from EMBL gene databank accession number Q 53653 with the computer programme OLIGO (primer analysis software, version 4.0). The ampli®cation was carried out using the following programme: 1  4 min precycle at 948C, 35  1 min at 948C, 1 min at 578C, 1 min at 728C, followed by a ®nal extension incubation at 728C for 5 min. A further genotyping was performed by ampli®cation of the X region and the IgGbinding region of the protein A (spa) gene and by PCR-ampli®cation of the coagulase (coa) gene. The determination of the repeat number of the X region of the protein A (spa) gene was performed with the oligonucleotide primers 50 -CAA GCA CCA AAA GAG GAA-30 and 50 -CAC CAG GTT TAA CGA CAT-30 and the following programme: 1  4 min precycle at 948C, 35  1 min at 948C, 1 min at 608C, 1 min at 728C, followed by a ®nal extension incubation at 728C for 5 min (FreÂnay et al., 1996). The number of repeats in the IgG-binding region of the protein A (spa) gene was determined according to Seki et al. (1998) by using the oligonucleotide primers 50 -CAC CTG CTG CAA ATG CTG CG-30 and 50 -GGC TTG TTG TTG TCT TCC TC-30 and the following programme: 1  2 min precycle at 948C, 30  1 min at 948C, 1 min at 588C, 1 min at 728C, followed by a ®nal extension incubation at 728C for 5 min. The PCR ampli®cation of the coagulase (coa) gene was carried out by using the oligonucleotide primers 50 -ATA GAG ATG CTG GTA CAG G-30 and 50 -GCT TCC GAT TGT TCG ATG C-30 and the programme: 1  4 min precycle at 948C, 30  1 min at 948C, 1 min at 588C, 1 min at 728C, followed by a ®nal extension incubation at 728C for 5 min (Hookey et al., 1998). 2.5. Macrorestriction analysis Finally, a macrorestriction analysis of the chromosomal DNA of the cultures was performed with the restriction enzyme SmaI and subsequent pulsed-®eld gel electrophoresis. The preparation of the chromosomal DNA and the programme for pulsed-®eld gel electrophoresis was performed as described by Toshkova et al. (1997). 3. Results and discussion According to the results of API ID 32 Staph system as well as the ampli®cation of a S. aureus speci®c section of the 23S rRNA gene, all 34 isolates used in the present investigation were identi®ed as S. aureus. An ampli®cation of a S. aureus speci®c section of the 16S±23S rRNA ``intergenic spacer'' region could be determined for 32 of the 34 isolates. The results on further characterization of the 34 enterotoxin-forming S. aureus cultures are summarized in Table 1.

Table 1 Pheno- and genotypic characteristics of the 34 enterotoxin-producing S. aureus strains Enterotoxinb

1 2 1 2 1 1

TSST-1

Phenotype

Protein A (spa) gene (bp)

Hemolysis

Egg yolk

Staphaurex

Resistancesc

IgG-binding region

X region

Repeats X region

Size of amplicons of coa gene (bp)

PFGE pattern

AD AD AD AD AD AD

ÿ ÿ ÿ ÿ ÿ ÿ

a d d a d d

ÿ ÿ ‡ ‡ ‡ ‡

‡ ÿ ‡ ‡ ‡ ‡

ÿ ÿ ÿ ÿ ÿ ÿ

920 920 920 920 750 750

300 300 200 300 300 300

11 11 6 11 11 11

580 580 580 580 580 580

Ia Ia Ia Ib If Ig

16 1 1 1 1 2 1

C C C C C C C

‡ ‡ ‡ ‡ ‡ ‡ ÿ

ab ab ab b b ab d

ÿ ÿ ÿ ÿ ÿ ÿ ‡

ÿ ÿ ‡ ÿ ÿ ÿ ‡

ÿ Poly ÿ ÿ ÿ ÿ Pen G/Amp

920 920 920 920 920 920 750

100 100 100 100 100 100 300

2 2 2 2 2 2 11

580 580 580 580 580 580 660

IIa IIa IIa IIa IIb IIc III

1 1 1

D D D

ÿ ÿ ÿ

ab ab ab

‡ ‡ ‡

‡ ‡ ‡

ÿ ÿ Pen G/Amp

920 920 920

300 280 300

11 10 11

580 580 580

Ic Id Ie

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

na

a

Number of strains. Results obtained with SET RPLA and PCR. c Amp: ampicillin, Pen G: penicillin, Poly: polymyxin B. b

377

378

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

Within the 34 cultures, 23 cultures showed ab-, six cultures d-, three cultures a- and two cultures b-hemolysis. Only 11 of the investigated cultures were positive for clumping factor and/or protein A determined with the Staphaurex-Plus test system. Only nine cultures were egg yolk-positive, disclosing that the Staphaurex and the egg yolk reaction are not suitable for diagnosis of S. aureus from milk samples. Antibiotic resistance testing revealed that all 34 S. aureus strains were susceptible to cefoperazone, oxacillin, cephalothin, gentamicin, neomycin, tetracycline, chloramphenicol, lincomycin and erythromycin. A total of 31 isolates (91.2%) were susceptible to all antibiotics tested. Two of the remaining three isolates were resistant to penicillin G/ ampicillin, one to polymyxin B. Within the 26 cultures which formed a single SE, there were 23 SEC- and three SEDformers. Eight cultures were SEAD formers. It was of interest that apart from one strain all other 22 SEC formers were TSST-1 positive and egg yolk-negative. A signi®cant correlation between the detection of SEC and TSST-1 in S. aureus from mastitis milk has already been reported in the literature (Ho et al., 1989; Kenny et al., 1993; ZschoÈck et al., 1998). All three SED formers were both egg yolk- and Staphaurex-Plus positive. None of the strains harboured the seb, see, eta or etb gene. A production of enterotoxins and TSST-1 by S. aureus strains associated with bovine mastitis has also been described by Matsunaga et al. (1993), Takeuchi et al. (1996), Lee et al. (1998) and Cardoso et al. (1999). According to Hayakawa et al. (1998) the production of exfoliative toxins among S. aureus isolated from mastitic cow's milk or from bulk milk seems to be rare. The importance of toxin formation of S. aureus for udder pathogenicity remains unclear. However the superantigenic toxins might overstimulate and reduce the hosts immune response. Amongst the SEC formers of the present investigation, four phenotypes could be distinguished. Nineteen of the 23 cultures could be classi®ed as being of the same phenotype (ab-hemolysis, Staphaurex-Plus negative, egg yolk-negative, no antibiotic resistancies). The three SED cultures showed two phenotypes and the eight SEAD formers showed four different phenotypes. Ampli®cation of the clumping factor (clfA) gene resulted in a single amplicon with a size of approximately 980 bp for all 34 S. aureus indicating no size polymorphisms of this gene (Fig. 1). Ampli®cation of the X region of the spa gene yielded a single amplicon for each isolate. Four different sized amplicons of 300, 280, 200, 100 bp and the calculated number of 11, 10, 6 and 2 repeats, respectively, were observed (Fig. 2). Corresponding to AnnemuÈller et al. (1999) the difference in the amplicon size of the X region of the protein A gene of one SEAD strain and one SED strain could be explained by a deletion of repeats and might give evidence for evolutionary changes. The PCR ampli®cation of the gene encoding the IgG-binding region of protein A revealed a size of 920 bp for most of the isolates investigated. However, the protein A gene of three cultures revealed an amplicon size of 750 bp (Fig. 2). Because in these three strains, the PCR products were 170 bp smaller in size, and because 170 bp is the fragment size that is required to encode one IgG-binding domain, a lack of one domain is assumed for these strains. It was of interest that the clfA- and the spa gene could also be ampli®ed for the S. aureus strains showing a negative clumping factor and/or protein A latex agglutination reaction. This could possibly be explained by a reduced expression of both surface proteins or by

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

379

Fig. 1. Typical amplicons of the clumping factor (clfA) gene of S. aureus. M ˆ 100 bp ladder (Gibco BRL, Eggenstein, Germany) served as marker.

capsular material covering the cell wall bound receptors. PCR ampli®cation of the 30 end of the coa gene and subsequent agarose gel analysis of the ampli®ed products showed a single band for each of the 34 S. aureus isolates. Two different sized PCR products of 580 and 660 bp were distinguished (Fig. 3). PCR products of 580 bp were found in 33 of the 34 strains. In the vast majority of the cultures of one enterotoxin type, a great uniformity

Fig. 2. Different amplicons of the repeats of the X region (1: 100 bp; 2: 200 bp; 3: 280 bp; 4: 300 bp) and the IgG-binding region (5: 750 bp; 6: 920 bp) of the protein A (spa) gene of some S. aureus cultures (for explanation of M see Fig. 1).

380

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

Fig. 3. Typical amplicons of the coagulase (coa) gene: (1, 2, 4, 5, 6: 580 bp; 3: 660 bp) of some S. aureus-strains (for explanation of M see Fig. 1).

was shown in the size and number of repeats of the X region as well as in the size of the IgG-binding region of protein A gene, and also in the coagulase gene. Eleven PFGE patterns (Ia±Ig; IIa±IIc, III) could be shown by means of SmaI macrorestriction analysis. However, some of the PFGE patterns differed from each other only in few fragments and thus displayed clonal relations. In Figs. 4 and 5, the PFGE

Fig. 4. Pulsed-®eld electrophoretic restriction pattern of chromosomal DNA of SEAD-positive S. aureus cultures with DNA restriction pattern I (type Ib: lane 1, 2; type Ia: lane 3, 4, 6; type If: lane 5) and the different PFGE pattern II of a SEC- and TSST-1-positive S. aureus culture (lane 7) after digestion with endonuclease SmaI. M ˆ lambda DNA/Hind III fragments at 0.1±200 kb and lambda DNA concatemeres at 50±1000 kb (Sigma, Deisenhofen, Germany) served as standards.

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

381

Fig. 5. Pulsed-®eld electrophoretic restriction pattern of chromosomal DNA of SEC- and TSST-1-positive S. aureus cultures with DNA restriction pattern II (type IIb: lane 1; type IIa: lane 2; type IIc: lane 3) and pattern III (lane 4) after digestion with endonuclease SmaI (for explanation of M see Fig. 4).

pattern I of SEAD-positive S. aureus cultures and the PFGE pattern II of SEC- and TSST1-positive S. aureus cultures are shown as examples. The genotypic uniformity within the SEC formers was particularly striking. Of the 23 SEC-forming cultures, 15 cultures showed the same genotype. Seven further cultures showed close relatedness and only one strain could be distinguished unequivocally on the basis of both the phenotype and genotype results. The results of the present investigation show that a broad distribution of identical or closely related S. aureus clones are responsible for the mastitis situation in north-east Switzerland. This supports the ®ndings of AnnemuÈller et al. (1999) for a region in Germany. Further studies will aim to obtain more data from a greater number of strains from different regions in Switzerland. References AnnemuÈller, C., LaÈmmler, Ch., ZschoÈck, M., 1999. Genotyping of S. aureus isolated from bovine mastitis. Vet. Microbiol. 1757, 1±8. Anon., 1981. Laboratory methods for use in mastitis work. Bulletin 132. International Dairy Federation, Brussels, Belgium. Anon., 1998. National Committee of Clinical Laboratory Standards (NCCLS): performance Standards for Antimicrobial Disc Susceptibility Tests. Approved Standard ASM-2, Vol. 14, No. 16. Villanova, PA. Bauer, A.W., Kirby, W.M.M., Sherris, J.C., Turck, M., 1966. Antibiotic susceptibility. Testing by a standardized single disc method. Am. J. Clin. Pathol. 45, 493±496. Cardoso, H.F.T., Silva, N., Sena, M.J., Carmo, L.S., 1999. Production of enterotoxins and toxic shock syndrome toxin by Staphylococcus aureus isolated from bovine mastitis in Brazil. Lett. Appl. Microbiol. 29, 345±349.

382

R. Stephan et al. / Veterinary Microbiology 78 (2001) 373±382

Fitzgerald, J.R., Meaney, W.J., Hartigan, P.J., Smyth, C.J., Kapur, V., 1997. Fine structure molecular epidemiological analysis of Staphylococcus aureus recovered from cows. Epidemiol. Infect. 119, 261±269. Forsman, P., Tilsala-TimisjaÈrvi, A., Altossava, T., 1997. Identi®cation of staphylococcal and streptococcal causes of bovine mastitis using 16S±23S rRNA spacer regions. Microbio. 143, 3491±3500. FreÂnay, H.M., Bunschoten, A.E., Schouls, L.M., Van Leeuwen, W.J., Vandenbroucke-Grauls, C.M.J.E., Verhoef, J., Mooi, F.R., 1996. Molecular typing of methicillin-resistant Staphylococcus aureus strains on the basis of protein A gene polymorphism. Eur. J. Clin. Micobiol. Infect. Dis. 15, 60±64. Hayakawa, Y., Hayashi, M., Shimano, T., Komae, H., Takeuchi, K., Endou, M., Igarashi, H., Hashimoto, N., Takeuchi, S., 1998. Production of exfoliative toxin A by Staphylococcus aureus isolated from mastitic cow's milk and farm bulk milk. J. Vet. Med. Sci. 60, 1281±1283. Ho, G., Campell, W.H., Bergdoll, M.S., Carlson, E., 1989. Production of a toxic shock syndrome toxin variant by S. aureus strains associated with sheep, goats, and cows. J. Clin. Microbiol. 27, 1946±1948. Hookey, J.V., Richardson, J.F., Cookson, B.D., 1998. Molecular typing of Staphylococcus aureus based on PCR restriction fragment length polymorphism and DNA sequence analysis of the coagulase gene. J. Clin. Microbiol. 36, 1083±1086. Johnson, W.M., Tyler, S.D., Ewan, E.P., Ashton, F.E., Pollard, E.R., Rozee, K.R., 1991. Detection of genes for enterotoxins, exfoliative toxins and toxic shock syndrome toxin 1 in Staphylococcus aureus by the polymerase chain reaction. J. Clin. Microbiol. 29, 426±430. Kenny, K., Reiser, R.F., Bastida-Corcuera, F.D., Norcross, N.L., 1993. Production of enterotoxins and toxic shock syndrome toxin by bovine mammary isolates of S. aureus. J. Clin. Microbiol. 31, 706±707. Lange, C., Cardoso, M., Sencek, D., Schwarz, S., 1999. Molecular subtyping of Staphylococcus aureus from cases of bovine mastitis in Brazil. Vet. Microbiol. 67, 127±141. Lee, S.U., Quesnell, M., Fox, L.K., Yoon, J.W., Park, Y.H., Davis, W.C., Falk, D., Deobald, C.F., Bohach, G.A., 1998. Characterization of staphylococcal bovine mastitis isolates using the polymerase chain reaction. J. Food Prot. 61, 1384±1386. Matsunaga, T., Kamata, S., Kakiichi, N., Uchida, K., 1993. Characteristics of Staphylococcus aureus isolated from peracute, acute and chronic bovine mastitis. J. Vet. Med. Sci. 55, 297±300. SchaÈllibaum, M., 1999. Mastitis pathogens isolated in Switzerland (1987±1996). Mastitis Newsletter 23. Bulletin 338. International Dairy Federation, Brussels, Belgium. Seki, K., Sakurada, J., Kyong Seong, H., Murai, M., Tachi, H., Ishii, H., Masuda, S., 1998. Occurrence of coagulase serotype among Staphylococcus aureus strains isolated from healthy individuals Ð special reference to correlation with size of protein-A gene. Microbiol. Immunol. 42, 407±409. Skalka, B., Smola, J., Pillich, J., 1979. A simple method of detecting staphylococcal hemolysins. Zbl. Bakt. Hyg. I. Abt. Orig. A 245, 283±286. Stephan, R., Dura, U., Untermann, F., 1999. Resistenzsituation und EnterotoxinbildungsfaÈhigkeit von Staphylococcus aureus StaÈmmen aus bovinen Mastitismilchproben. Schweiz. Arch. Tierheilk. 141, 287±290. Straub, J.A., Hertel, C., Hammes, W.P., 1999. A 23S rDNA-targeted polymerase chain reaction-based system for detection of Staphylococcus aureus in meat starter cultures and dairy products. J. Food Prot. 62, 1150±1156. Takeuchi, S., Ishiguro, K., Ikegami, M., Kaidoh, T., Hayakawa, Y., 1996. Detection of toxic shock syndrome toxin-1 gene in Staphylococcus aureus bovine isolates and bulk milk by polymerase chain reaction. J. Vet. Med. Sci. 58, 1133±1135. Toshkova, K., Savov, E., Soedarmanto, I., LaÈmmler, Ch., Chankova, D., van Belkum, A., Verbrugh, H.A., van Leeuwen, W., 1997. Typing of S. aureus isolated from nasal carriers. Zbl. Bakt. 286, 547±559. Tsen, H.Y., Chen, T.R., 1992. Use of the polymerase chain reaction for speci®c detection of type A, D and E enterotoxigenic Staphylococcus aureus in foods. Appl. Microbiol. Biotechnol. 37, 685±690. ZschoÈck, M., Botzler, D., BloÈcher, S., SommerhaÈuser, J., 1998. Nachweis von Genen fuÈr Enterotoxine (ent) und Toxic Shock Syndrome Toxin 1 (tst) in Staphylococcus aureus-Isolaten aus subklinischer Mastitis des Rindes mittels Polymerase-Kettenreaktion 39. Arbeitstagung des Arbeitsgebietes Lebensmittelhygiene der Deutschen VeterinaÈrmedizinischen Gesellschaft, 22±25.09.1998. Garmisch-Partenkirchen, pp. 251±256.