A coagulase-negative variant of Staphylococcus aureus from bovinemastitis milk

Ömer Akineden1*, Abdulwahed Ahmed Hassan2, Elisabeth Schneider3 and Ewald Usleber3

1Food Science, Institute of Veterinary Food Science, Justus-Liebig-University Giessen, Frankfurter Strasse 92, D-35392 Giessen, Germany2GD Animal Health Service, Department of Bacteriology and Parasitology, Postbus 9, 7400 AA Deventer, The Netherlands3Dairy Sciences, Institute of Veterinary Food Science, Justus-Liebig-University Giessen, Ludwigstrasse 21, D-35390 Giessen, Germany

Received 23 June 2010; accepted for publication 3 August 2010; first published online 1 December 2010

Bacteriological analysis of milk samples from quarters of a dairy cow suffering from subclinicalmastitis yielded two isolates of Staphylococcus aureuswhich gave a negative reaction in the standardcoagulase test. Both isolates were also clumping factor and thermonuclease negative, and gavea negative reaction in the Staphaurex® test. The isolates were identified by using commercialbiochemical systems, and by PCR analysis of different staphylococcal cell surface protein andexoprotein genes. Further molecular identification of the isolates, which included sequencing of the16S rRNA gene and RT-PCR of coagulase (coa), clumping-factor (clfA) and thermonuclease (nuc)genes, was consistent with the diagnosis phenotypically ‘coagulase-negative variant of Staph.aureus’. The fact that coagulase-negative Staph. aureus variants can occur in the context ofintramammary infections in cattle may result in the incorrect diagnosis ‘coagulase-negativestaphylococci (CNS)’ in routine mastitis diagnostic, at least in rare cases. To fully ensure correctspecies diagnosis, sequencing of the 16S rRNA gene and amplification of specific genes such as coais necessary in these cases.

Keywords: Staphylococcus aureus, coagulase negative, bovine mastitis, beta-haemolysin.

The differentiation between coagulase-positive and coagu-lase-negative staphylococci (CNS) is of major relevance inroutine mastitis diagnostics. Among the three coagulase-positive Staphylococcus species (Staph. aureus, Staph.intermedius and Staph. hyicus), the most important causeof bovinemastitis is Staph. aureus, which species differs fromthe other two by frequently having beta-haemolytic activityon blood agar plates (Kloos & Schleifer, 1975). The diagnosiscoagulase-positive staphylococci has a major impact ontreatment of a cow or a dairy herd, while CNS are stillfrequently regarded as ‘minor pathogens’.

The decisive criterion for differentiation of mastitisstaphylococci is the standard coagulase tube method, inwhich a coagulase-positive Staphylococcus isolate gelati-nizes with plasma via the activity of extracellular freecoagulase. Therefore, presence of beta-haemolysin on bloodagar in combination with a positive coagulase activity seemsto represent the optimal criterion for the identification ofStaph. aureus from mastitis milk samples (Varshney et al.1993; Lam et al. 1995; Boerlin et al. 2003). Additionally,testing for the presence of surface proteins, clumping factor

(CF) and/or protein-A, are helpful in routine analysis for therapid identification of Staph. aureus. Finally, the presence ofa DNase activity is often used as a surrogate marker for theidentification of coagulase-positive staphylococci, particu-larly of Staph. aureus, in milk samples (Menzies, 1977;Boerlin et al. 2003). Most of these test parameters have beenextensively validated for use in identifying Staph. aureusfrom human clinical specimens.The known specificity of all these tests has previously

been used to identify atypical variants of Staph. aureusstrains isolated from clinical specimens and mastitis milksamples (Kloos & Schleifer, 1975; Heltberg & Bruun, 1984;Smyth et al. 1988; Bodén et al. 1989; Fox et al. 1993;Mackay et al. 1993; Laevens et al. 1996; Matthews et al.1997; Personne et al. 1997; Wilkerson et al. 1997;Mlynarczyk et al. 1998; Smole et al. 1998; Al Obaid et al.1999; van Griethuysen et al. 2001; Garbacz et al. 2002;Malinowski et al. 2009). In fact, Staph. aureus strainsshowing non-expression of these major characteristics arerare but not absolutely uncommon in routine analysis.A certain percentage of Staph. aureus isolates is thereforemisidentified as CNS (Ruane et al. 1986; Neville et al. 1991;Wanger et al. 1992; Vandenesch et al. 1993; Vandeneschet al. 1994a, b).*For correspondence: Oemer.Akineden@vetmed.uni-giessen.de

Journal of Dairy Research (2011) 78 38–42. © Proprietors of Journal of Dairy Research 2010doi:10.1017/S0022029910000774

38

Other diagnostic methods based on DNA sequence havebeen significantly improved for identification of mastitisstaphylococci during recent years. Such methods, forexample sequencing of 16S rRNA genes, and PCR ofspecies-specific fragments of ribosomal RNA, thermonu-clease, coagulase, clumping factor and protein A genes,were used in numerous studies (Brakstad et al. 1992;Forsman et al. 1997; Martineau et al. 1998; Straub et al.1999; Stephan et al. 2000; Akineden et al. 2001; Woo et al.2001; Boerlin et al. 2003).

To reduce the impact of errors in routine diagnostics, or atleast to evaluate the dimensions of the margin of error inroutine analysis due to atypical strains, control analyses areadvisable using molecular characterization of a certainpercentage of isolates, and in particular of suspect isolates,seems to be advisable. In the present study, two atypicalvariants of Staph. aureus strains isolated from milk samplesof a dairy cow with subclinical mastitis were investigatedusing standard methods for identification of mastitis samplesand subsequently confirmed with genotypic methods.

Materials and Methods

Isolation and characterization of the bacteria

During routine mastitis diagnostics in our laboratory, milksamples from two quarters of a dairy cow suffering fromsubclinical mastitis were subjected to the standard analyticalscheme (Anonymous, 2009). From each milk sample 0·1 mlinitially was spread on sheep blood agar and incubatedaerobically for 24–48 h at 37 °C as described previouslyafter incubation colonies were weakly positive for beta-haemolysin. Two isolates (k54-2 and k54-3) were furthertested for catalase, clumping factor reactions with EDTArabbit plasma (Merck, Darmstadt, Germany), coagulase tubetest, aerobic production of acid from mannitol (mannitol saltagar, Oxoid), Staphaurex® test (Murex, Wiesbaden,Germany) for the detection of surface protein ofStaph. aureus and cultured on Baird Parker agar (Merck)for lipase activity.

DNase and thermonuclease activity was performed onToluidinblue O-DNase agar (Merck). Production of haemo-lysins was determined by the interaction of haemolysins withthe β-toxin of a Staph. aureus strain as described by Skalkaet al. (1979). Both isolates were characterized using twocommercial systems, the BD BBL Crystal system (BectonDickinson, Darmstadt, Germany) and the ID 32 Staphsystem (bioMerieux, Wiesbaden, Germany).

The antibiotic sensitivity test was performed as follows:four to five identical colonies were incubated in 3 ml ofTodd-Hewitt broth (THB, Oxoid) for 2 h at 37 °C. Bacterialsuspension (0·1 ml) was plated on Mueller-Hinton agar(Merck), followed by addition of antibiotic disks [ampicillin,10 μg; bacitracin, 10 U; cefquinome sulphate, 10 μg; clox-acillin, 25 μg; enrofloxacin, 5 μg; kanamycin, 30 μg; tri-methoprim-sulphamethoxazole, 25 μg; tetracycline, 30 μg;

gentamicin, 10 μg; amoxicillin-clavulanic acid, 20/10 μg;lincomycin at 60 μg with neomycin at 15 μg; colistinsulphate, 10 μg; penicillin G, 10 IU (Oxoid), danofloxacin,5 μg; and cefoperazone, 10 μg (Pfizer, Karlsruhe, Germany)].

Testing for enterotoxin production

Both strains were tested for their ability to produceenterotoxins A, B, C, D and E in brain heart infusion culture(BHI (Merck), 18–24 h at 37 °C) by using a commercialenzyme immunoassay kit (Ridascreen SET, R-Biopharm,Darmstadt, Germany). Tests were performed from BHIculture material after filtration (0·2 μm, FP30/0·2CA-S,Schleicher and Schuell, Dassel, Germany).

Molecular identification of the bacteria

DNA extraction. Bacterial DNAwas isolated after cultivationof the bacteria in 5·0 ml BHI. From this culture, 0·1 ml wascentrifuged at 10000 g for 10min, and the pellet resus-pended in 180 μl TE buffer (10 mmol/l of Tris–HCl, 1 mmol/lof EDTA; pH 8·0) containing 5 μl of lysostaphin (1·8 U/μl;Sigma, Deisenhofen, Germany) and incubated for 1 h at37 °C. Finally, 10 μl proteinase K was added to thesuspension and incubated for 2 h at 56 °C. DNA wassubsequently isolated with the DNeasy tissue kit (Qiagen,Hilden, Germany) according to the manufacturer’s rec-ommendations.

Sequencing of the 16S rRNA gene. The two isolates wereidentified by sequencing of the 16S rRNA gene using theprimers 16SUNI-L and 16SUNI-R as described previously(Kuhnert et al. 1996). The PCR reaction was carried out in a30-μl total volume in a 0·2-ml reaction tube containing 3 μlGeneAmp 10x PCR Gold Buffer (150mmol/l Tris–HCl,500mmol/l KCl; pH 8·0) (Applied Biosystem, Darmstadt,Germany), 1·8 μl MgCl2 (25 mmol/l) (Applied Biosystem),1·0 μl from each primer (10 pmol/μl), 0·6 μl dNTP-mix(10 mmol/l) (MBI Fermentas, St Leon-Rot, Germany), 0·2 μlAmpliTaq Gold® polymerase (5 U/μl, Applied Biosystem),19·9 μl sterile Aqua dist. and 2·5 μl DNA template. PCRconditions for the iCycler (BioRad, Munich, Germany) wereas follows: 1 cycle at 95 °C for 10min followed by 30 cyclesat 95 °C for 30 s, 54 °C for 30 s and 72 °C 60 s and a finalextension 1 cycle at 72 °C for 7 min. The purified DNA(30 μl volume) was sequenced by SEQLAB SequenceLaboratories (Göttingen, Germany) in ABI DNA SequencerSystem (Applied Biosystems, Darmstadt, Germany). The16S rRNA sequences of the two isolates were aligned againstsequences of Staph. aureus (accession no. Y15856), Staph.intermedius (D83369), Staph. hyicus (D83368) and Staph.epidermidis (AM157417) and analysed by using theMegAlign program (DNASTAR, Inc., MadisonWI, USA).

PCR-amplification of genes encoding staphylococcal pro-teins and toxins. The two coagulase-negative Staph. aureus

Coagulase-negative variant of Staphylococcus aureus 39

isolates were additionally characterized by PCR amplifica-tion of species-specific parts of the 23S rRNA, 16S-23S rRNAintergenic spacer region, coagulase (coa), clumping-factor(clfA), alpha haemolysin (hla), beta haemolysin (hlb),thermonuclease (nuc), and IgG binding region and x-regionof protein A (spa) genes, as well as the staphylococcal entero-toxins genes sea to selo, the toxic shock syndrome toxin genetst, and exfoliative toxins A (eta) and B (etb). The sequencesof the oligonucleotide primers, the thermocycler programs,and the corresponding literature references were describedby Akineden et al. (2008).

PCR reaction was carried out in a 30-μl volume containingthe same composition of reagents as described above. PCRproducts were determined by electrophoresis of 12 μl of thereaction product in a 2% agarose gel (Biozym, Hessisch-Oldendorf, Germany) at 120 Volt in 1x Tris–acetate-electro-phoresis buffer (TAE) [(0·04 mol/l Tris, 0·001mol/l EDTA; pH7·8)]. The molecular marker GeneRuler 50 and a 100-bpDNA ladder (MBI Fermentas) were used.

Reverse transcriptase-PCR. Total RNA (approximately8 μg) was extracted from isolates after 18 h of culture usingthe RNeasy-Mini kit (Qiagen) as described by Akineden et al.(2008). DNA digestion was performed by incubating 10 μlextracted RNA with 3 μl of RQ1 RNase-free DNase(Promega, Mannheim, Germany) (1 U/μl) in a 500-μlRNase-free Eppendorf cap at 37 °C for 30min. DNase wasinactivated by heating the solution at 75 °C for 5 min. Then

the solution was cooled on ice and either immediately usedfor RT-PCR or stored frozen at �80 °C.cDNA was synthesized with Omniscript RT kit (Qiagen)

by incubating at 42 °C for 1 h. The reaction mixture (20 μl)contained 2 μl of 5×Qiagen RT-PCR buffer (including12·5 mmol/l MgCl2), 2 μl of dNTP mix (containing 5mmol/l each dNTP), 1 μl of Rnasin inhibitor (40 U/μl; N2111,Promega), 1 μl of a random hexamer primer (Promega), 2 μlof Qiagen RT polymerase (4 U/μl), template RNA (2 μl of theRNA extract) and 9 μl of Rnase-free water. As a control forgenomic DNA contamination, total RNAwas also subjectedto reverse transcription (RT)-PCR but without the RT step.The coa, clfA, nuc and sec cDNAs were then detected usingthe same primers and PCR protocol described by Akinedenet al. (2008).

Results

Both isolates showed identical properties with regard to alltested biochemical reactions, and were also identical inaspects of the tested genetic traits (Table 1).

Phenotypic properties

Both isolates showed morphological characteristics consist-ent with Staph. aureus on blood agar, yellow pigmentationand a very weak beta-haemolysin reaction. They formedtypical colonies on Baird Parker agar but were egg yolk-negative. They were catalase-positive, clumping factor-negative, coagulase-negative, DNase-negative, thermonu-clease-negative and also negative in the Staphaurex® testsystem. The BD BBL Cyristall® system and ID 32 Staph®

identified both isolates as Staph. aureus with ‘confident0·9275’ and ‘identity 99·8%’ (ID number 36733261),respectively. Both isolates were positive for enterotoxin Cin the RIDASCREEN SET EIA. They were sensitive towardsampicillin, penicillin G, cefoperazone, bacitracin, amoxi-cillin-clavulanic acid, lincomycin-neomycin, trimethoprim-sulphamethoxazole, enrofloxacin, gentamicin, tetracycline,cefqiunom sulphate, cloxacillin, danofloxacin, and resistanttowards colistin sulphate and kanamycin.

16S rRNA sequence and other genotypic characteristics

PCR of 16S rRNA gene of both isolates yielded an ampliconsize of 1360 bp. These amplicons were sequenced and thesequence submitted to the NCBI GenBank [accession no.GU459255 (strain k54-2) and GU459256 (strain k54-3)].Comparison of both 16S rRNA sequences with the corre-sponding sequences of Staph. aureus (accession no.Y15856), Staph. intermedius (D83369), Staph. hyicus(D83368) and Staph. epidermidis (AM157417) showed99·7%, 96·0%, 96·7% and 98·4% similarity, respectively.Additionally, the species identity of both isolates as Staph.

aureus could be confirmed by PCR amplification of thegenes encoding for 23 rRNA (1·270 bp), coa (600 bp), clfA

Table 1. Summarized phenotypic and genotypic characteristics ofthe two Staphylococcus aureus isolates obtained from mastitis milk

Parameter†Results for isolates K54-2 andK54-3

Phenotypic characterizationHaemolysin weak βCatalase +Egg yolk �Coagulase �Clumping factor �DNase �TNase �Staphaurex® test �Enterotoxin (ELISA) SECGenotypic characterization23S rDNA (bp) 1270ISR (bp) 420 and 520coa (bp) 780clfA (bp) 1000nuc (bp) 280IgG-binding region spa (bp) 920hla (bp) 535hlb (bp) 833Enterotoxin/Enterotoxin-likegene(s)

sec, seg, sei, selm, seln, selo, tst

†See text for explanation of abbreviations

40 O Akineden and others

(1000 bp), nuc (280 bp) and IgG binding region of protein Aspa (920 bp), hla (535 bp), hlb (833 bp). Amplification of the16S-23S rRNA intergenic spacer region yielded two distinctamplicons (420 bp and 520 bp) in the same PCR. Furthergenotypic properties of the Staph. aureus isolates revealedthat certain genetic traits such as the enterotoxin genes sec,seg, sei, selm, seln, selo and tst could be detected in bothisolates. None of the isolates harboured the genes encodingsea, seb, sed, see and sej.

Reverse transcriptase-PCR

Reverse transcription-PCR analysis proved that the mRNAsof coagulase, clumping factor and thermonuclease weretranscribed in both Staph. aureus isolates. In comparison,both isolates were also positive for the genes coa, clfAand nuc, but were negative in the biochemical tests forcoagulase, clumping factor or thermonuclease. In thiscontext, the staphylococcal enterotoxin C gene was alsoused as a positive control in RT-PCR.

Discussion

Coagulase-negative variants of Staph. aureus strains werefirst isolated from milk samples derived from subclinicalmastitis cases in dairy cattle (Fox et al. 1996; Laevens et al.1996; Matthews et al. 1997; Malinowski et al. 2009). Insome other clinical studies, detection of coagulase-negativeStaph. aureus was described but the reported results fromphenotypic tests, such as beta-haemolysin, strong DNaseactivity, production of phosphates and slide clumping factor,nitrate reduction, and utilization of certain sugars made theidentification of Staph. aureus questionable (Golledge &Gordon, 1989, Mlynarczyk et al. 1998, Aarestrup et al.1999). Biochemical and other phenotypic tests per definitioncannot consistently identify bacterial species having anambiguous biochemical profile (Notarnicola et al. 1985).Amplification of the coagulase gene of staphylococci canbe performed by PCR used for identification of atypicalcoagulase negative Staph. aureus (Luijendijk et al. 1996).However, economic reasons prohibit a broader use of suchmethods in routine mastitis diagnostic. Therefore, reportson the occurrence of coagulase-negative Staph. aureus inbovine milk are still extremely rare, considering the numberof analyses performed worldwide. In this study, two tentativestaphylococcal isolates frommastitis milk appeared negativefor the clumping factor, coagulase, thermonuclease, andStaphaurex® test. In a normal situation, this would have beenmore than enough to ‘identify’ the isolates as coagulase-negative staphylococci. However, a very weak beta-haemolysin activity of these isolates motivated us to performfurther work on identification.

PCR-based systems for identification of Staph. aureusisolates were used as described earlier for genes encodingthe 16S rRNA, 16S–23S rRNA intergenic spacer region,23S rRNA, as well as the genes encoding staphylococcal

thermonuclease (nuc), clumping-factor (clfA), IgG bindingregion encoding part of protein A and coagulase (coa)(Brakstad et al. 1992; Lange et al. 1999; Stephan et al. 2000;Akineden et al. 2001). Although all these target genes allowa rapid identification of this species with high sensitivity andspecificity, it is of some importance that we could detect notonly the DNA sequences but also the corresponding mRNAby RT-PCR. However, only expression of enterotoxin Cwas detected by protein-based tests. This again shows thateven positive results for gene transcription (RT-PCR) doesnot necessarily mean that it is translated in quantities abovethe detection level. The sequencing of the 16S rRNA gene ofboth isolates was used as a complementary method tothe phenotypic tests. Although this is still far from being aroutine method, sequence analysis seems to be a promisingtechnique for identification and differentiation of staphylo-cocci from cases of bovine mastitis in the future.Routine mastitis laboratories should keep in mind that

identification of Staph. aureus based on the coagulase test asthe decisive criterion may give misleading results, althoughsuch cases are probably very rare. As in our case, a veryweak beta-haemolysin activity – in spite of negative clump-ing factor and coagulase reactions –may be a reason totrigger further detailed investigation. At the moment,however, it is impossible to estimate the frequency ofcoagulase-negative Staph. aureus in the context of bovinemastitis. However, since the consequences of a Staph.aureus diagnosis are much more serious than that ofcoagulase-negative staphylococci for dairy cows, a betterknowledge of the epidemiology of coagulase-negativeStaph. aureus strains in dairy production would be helpful.

References

Aarestrup FM, Larsen HD, Eriksen NH, Elsberg CS & Jensen NE 1999Frequency of alpha- and beta-haemolysin in Staphylococcus aureus ofbovine and human origin. A comparison between pheno- and genotypeand variation in phenotypic expression. Acta Pathologica, Micro-biologica et Immunologica Scandinavica 107 425–430

Akineden Ö, Annemüller C, Hassan AA, Lämmler C, Wolter W & ZschöckM 2001 Toxin genes and other characteristics of Staphylococcus aureusisolates from milk of cows with mastitis. Clinical and DiagnosticLaboratory Immunology 8 959–964

Akineden Ö, Hassan AA, Schneider E & Usleber E 2008 Enterotoxigenicproperties of Staphylococcus aureus isolated from goats’ milk cheese.International Journal of Food Microbiology 124 211–216

Al Obaid AI, Udo EE, Jacob LE & Johny M 1999 Isolation and charac-terization of cogulase-negative methicillin-resistant Staphylococcusaureus from patients in an intensive care unit. Medical Principles andPractice 8 230–236

Anonymous 2009 [Guidelines for the collection of milk samples underantiseptic conditions and for identification of mastitis pathogens. Expertcommittee on subclinical mastitis] Deutsche VeterinärmedizinischeGesellschaft (DVG), Gießen

Brakstad OG, Aasbakk K & Maeland JA 1992 Detection of Staphylococcusaureus by polymerase chain reaction amplification of the nuc gene.Journal of Clinical Microbiology 30 1654–1660

Bodén MK & Flock JI 1989 Fibrinogen-binding protein/clumping factor fromStaphylococcus aureus. Infection and Immunity 57 2358–2363

Coagulase-negative variant of Staphylococcus aureus 41

Boerlin P, Kuhnert P, Hussy D & Schaellibaum M 2003 Methods foridentification of Staphylococcus aureus isolates in cases of bovinemastitis. Journal of Clinical Microbiology 41 767–771

Forsman P, Tilsala-Timisjärvi A & Alatossava T 1997 Identificationof staphylococcal and streptococcal causes of bovine mastitis using16S-23S rRNA spacer regions. Microbiology 143 3491–3500

Fox LK & Gay JM 1993 Contagious mastitis. Veterinary Clinics of NorthAmerica Food Animal Practice 9 475–487

Garbacz K, Piechowicz L,Wisniewska K &Galinski J 2002 Characteristics ofdefective strains of methicillin resistance Staphylococcus aureus that donot produce coagulase or CF.Medycyna Doswiadczalna i Mikrobiologia54 1–8

Golledge C & Gordon A 1989 Slide coagulase positive, tube coagulasenegative Staphylococcus aureus. Journal of Clinical Pathology 42 443

Heltberg O & Bruun B 1984 Recognition of coagulase-negativeStaphylococcus aureus strains by primary polymyxin susceptibilitytesting. Acta Pathologica et Microbiologica Scandinavica 92 115–118

Kloos WE & Schleifer KH 1975 Simplified scheme for routine identificationof human Staphylococcus species. Journal of Clinical Microbiology 182–88

Kuhnert P, Capaul SE, Nicolet J & Frey J 1996 Phylogenetic positions ofClostridium chauvoei andClostridium septicum based on 16S rRNA genesequences. International Journal of Systematic Bacteriology 46 1174–1176

Laevens H, Devriese LA, Deluyker H, Hommez J & de Kruif A 1996 Anatypical Staphylococcus aureus intramammary infection in a dairy herd.Veterinary Microbiology 52 271–275

Lam TJ, Pengov A, Schukken A, Smit YH & Brand A 1995 The differentiationof Staphylococcus aureus from other Micrococcaceae isolated frombovine mammary glands. Journal of Applied Bacteriology 79 69–72

Lange C, Cardoso M, Senczek D & Schwarz S 1999 Molecular subtyping ofStaphylococcus aureus isolates from cases of bovine mastitis in Brazil.Veterinary Microbiology 67 127–141

Luijendijk A, van Belkum A, VerbrughH & Kluytmans J 1996 Comparison offive tests for identification of Staphylococcus aureus from clinicalsamples. Journal of Clinical Microbiology 34 2267–2269

Mackay AD, Quick A, Gillespie SH & Kibbler CC 1993 Coagulase-negativemethicillin-resistant Staphylococcus aureus infection. Lancet 342 492

Malinowski E, Lassa H, Klossowska A, Smulski S & Kaczmarowski M 2009Atypical Staphylococcus aureus as an aetiological agent of mastitis incows. Bulletin of the Veterinary Institute in Pulawy 53 383–387

Martineau F, Picard FJ, Roy PH, Ouellette M & BergeronMG 1998 Species-specific and ubiquitous-DNA based assays for rapid identification ofStaphylococcus aureus. Journal of Clinical Microbiology 36 618–623

Matthews KR, Roberson J, Gillespie BE, Luther DA, Oliver SP 1997Identification and differentiation of coagulase-negative Staphylococcusaureus by polymerase chain reaction. Jounal of Food Protection 60686–688

Menzies ER 1977 Comparison of coagulase, deoxyribonuclease (DNase),and heat-stable nuclease tests for identification of Staphylococcusaureus. Journal of Clinical Pathology 30 606–608

Mlynarczyk G, Kochman M, Lawrynowicz M, Fordymacki P, Mlynarczyk A& Jeljaszewicz J 1998 Coagulase-negative variants of methicillin-resistant Staphylococcus aureus subsp. aureus strains isolated fromhospital specimens. Zentralblatt für Bakteriologie 288 373–381

Neville LO, Billington OJ, Kibbler CC & Gillespie SH 1991 Methicillinresistant Staphylococcus aureus without clumping factor, protein A, andDNAse. Lancet 338 518

Notarnicola SM, Zamarchi GR & Onderdonk AB 1985 Misidentification ofmucoid variants of Staphylococcus aureus by standard laboratorytechniques. Journal of Clinical Microbiology 22 459–461

Ruane PJ, Morgan MA, Citron DM & Mulligan ME 1986 Failure of rapidagglutination methods to detect oxacillin-resistant Staphylococcusaureus. Journal of Clinical Microbiology 24 490–492

Personne P, Bes M, Lina G, Vandenesch F, Brun Y & Etienne J 1997Comparative performances of six agglutination kits assessed by usingtypical and atypical strains of Staphylococcus aureus. Journal of ClinicalMicrobiology 35 1138–1140

Smole SC, Aronson E, Durbin A, Brecher SM & Arbeit RD 1998 Sensitivityand specificity of an improved rapid latex agglutination test for identi-fication of methicillin-sensitive and -resistant Staphylococcus aureusisolates. Journal of Clinical Microbiology 36 1109–1112

Smyth EG, Wright ED & Marples RR 1988 New type of staphylococcalenteriditis. Journal of Clinical Pathology 41 809–810

Stephan R, Annemüller C, Hassan AA & Lämmler C 2000 Characterizationof enterotoxigenic Staphylococcus aureus strains isolated from bovinemastitis in north-east Switzerland. Veterinary Microbiology 78 373–382

Straub JA, Hertel C & Hammes WP 1999 A 23S rRNA-targeted polymerasechain reaction-based system for detection of Staphylococcus aureus inmeat starter cultures and dairy products. Journal of Food Protection 621150–1156

Van Griethuysen A, Bes M, Etienne J, Zbinden R & Kluytmans J 2001International multicenter evaluation of latex agglutination tests foridentification of Staphylococcus aureus. Journal of Clinical Microbiology39 86–89

Vandenesch F, Lebeau C, Bes M, McDevitt D, Greenland T, Novick RP &Etienne J 1994a Coagulase deficiency in clinical isolates ofStaphylococcus aureus involves both transcriptional and post-transcrip-tional defects. Journal of Medical Microbiology 40 344–349

Vandenesch F, Lebeau C, BesM, Lina G, Lina B, Greenland T, Benito Y, BrunY, Fleurette J & Etienne J 1994b Clotting activity in Staphylococcusschleiferi subspecies from human patients. Journal of Clinical Micro-biology 32 388–392

Vandenesch F, Bes M, Lebeau C, Greenland T, Brun Y & Etienne J 1993Coagulase-negative Staphylococcus aureus. Lancet 342 994–995

Varshney JP, Kapur MP & Sharma A 1993 Studies on some biochemicalcharacteristics of Staphylococcus aureus of buffalo mammary origin.Comparative Immunology Microbiology and Infectious Diseases 16317–321

Wanger AR, Morris SL, Ericsson C, Singh KV & LaRocco MT 1992 Latexagglutination-negative methicillin-resistant Staphylococcus aureus re-covered from neonates: epidemiologic features and comparison of typingmethods. Journal of Clinical Microbiology 30 2583–2588

Wilkerson M, McAllister S, Miller JM, Heiter BJ & Bourbeau PP 1997Comparison of five agglutination tests for identification of Staphylo-coccus aureus. Journal of Clinical Microbiology 35 148–151

Woo PC, Leung AS, Leung KW & Yuen KY 2001 Identification of slidecoagulase positive, tube coagulase negative Staphylococcus aureus by16S ribosomal RNA gene sequencing. Molecular Pathology 54 244–2447

42 O Akineden and others