Original articles. Robert L. Skov*, Lars V. Pallesen, Rikke L. Poulsen and Frank Espersen

Transcription

1 Journal of Antimicrobial Chemotherapy (1999) 43, Original articles JAC Evaluation of a new 3-h hybridization method for detecting the meca gene in Staphylococcus aureus and comparison with existing genotypic and phenotypic susceptibilty testing methods Robert L. Skov*, Lars V. Pallesen, Rikke L. Poulsen and Frank Espersen Department of Research and Development, Statens Serum Institut, Artillerivej 5, Copenhagen, 2300S, Denmark A new 3-h hybridization assay for detection of the staphylococcal meca gene and the Staphylococcus aureus nuclease gene was evaluated by comparing the assay with existing genotypic and phenotypic methods. A total of 275 S. aureus strains were tested, including 257 epidemiologically unrelated strains (135 meca-positive and 122 meca-negative; collection I), and 18 strains with known borderline resistance to methicillin (collection II). Complete agreement was obtained for both collections when comparing the new assay with genotypic methods. We further evaluated a range of phenotypic susceptibility methods recommended in Europe and/or USA using the presence of the meca gene as the defining standard. For collection I a high degree of agreement was found for both Etests (256 strains) and the oxacillin screen plate test (255 strains); the degree of agreement was lower for agar dilution methicillin (250 strains) and oxacillin 1 g discs (239 strains). For the borderline strains a high degree of agreement was only obtained by the oxacillin screen plate test (17 of 18 strains). The other tests were less accurate, in the following order: agar dilution methicillin, Etest methicillin, Etest oxacillin and oxacillin discs with disagreement for four, five, nine and 13 strains, respectively. In conclusion, the new hybridization assay is a rapid and exact method for detecting the meca gene and the S. aureus nuclease gene. This study confirms that phenotypic tests for methicillin resistance in S. aureus strains creates both false-susceptible and false-resistant results, especially for borderline resistant strains. Introduction Rapid and reliable detection of methicillin-resistant Staphylococcus aureus (MRSA) is essential in order to secure the optimal treatment of patients with staphylococcal infections, as well as for infection control procedures. Phenotypic susceptibility testing often takes 1 2 days and sometimes even longer before a definitive result is obtained. Methicillin resistance is almost exclusively caused by production of an additional penicillin binding protein (PBP 2a) encoded by the meca gene, 1 although other mechanisms have been described. 2,3 There is, however, a great variation in the phenotypic expression of methicillin resistance of individual MRSA strains due to the heterogeneous nature of the resistance mechanism. 4,5 Furthermore, the meca-negative strains with borderline susceptibility (i.e. MICs of methicillin of 4 16 mg/l) can be difficult to classify by phenotypic methods. 6 Altogether this makes phenotypic susceptibility testing less accurate. 4,6 9 The meca gene is highly conserved among staphylococcal species, and there is no meca homologue in susceptible strains, so it is possible to detect methicillin resistance reliably by detection of this gene. 5,10 This is now considered the gold standard for detection of methicillin resistance. 5 Several reliable methods for DNA-based detection such as probe detection of the meca gene (both radioactive and non-radioactive) and different PCR-based assays have been published. These methods, however, are at present either too laborious or too technically demanding for routine use in most laboratories. In this paper we present an evaluation of a newly developed hybridization assay for detection of the meca gene 18 and the S. aureus *Corresponding author. Tel: ; Fax: ; rsk@ssi.dk 1999 The British Society for Antimicrobial Chemotherapy 467

2 R. L. Skov et al. nuclease gene (nuc) 19 which can be performed in a routine microbiology laboratory. The assay is compared with results obtained by Southern blot and PCR, both established methods for detection of the meca gene. Methicillin resistance is still determined predominantly by phenotypic methods; we therefore compared a range of phenotypic susceptibility testing methods recommended in Europe and/or in the USA as well as the Etest (AB Biodisk, Solna, Sweden), in order to describe the sensitivity and specificity of the tested methods using the presence of the meca gene as the defining standard. Materials and methods Bacterial strains A total of 275 S. aureus strains were investigated. Two hundred and fifty-seven of the strains (collection I) were selected from the strain collection at the Staphylococcus Laboratory, Statens Serum Institut, Copenhagen, Denmark, to include 137 methicillin-resistant S. aureus (MRSA) according to earlier phenotypic testing (107 isolated in Denmark from 1988 to 1994, and 30 MRSA isolated from different parts of the world e.g. Europe, Russia, Pakistan and Brazil) and 120 consecutive Danish blood culture isolates from The MRSA strains investigated wereepidemiologically unrelated, i.e. only one isolate from a given institution or a given outbreak was included. Eighteen S. aureus strains with borderline susceptibility to methicillin were also tested (collection II): Three mecanegative reference strains were kindly provided by B. Olsson-Liljequist, Smittskydsinstituttet, Stockholm, Sweden and 15 clinical strains, 11 meca-negative and four meca-positive isolates (two heteroresistant class I and two heteroresistant class II MRSA strains), were kindly provided by A. Österlund, Uppsala University Hospital, Sweden. The nature of the resistance mechanism of the latter 15 strains was not disclosed to us until after the investigation. In the rest of this paper the meca-negative borderline susceptible S. aureus will be referred to as BORSA strains. All strains were stored at 80 C in Oxbroth (Statens Serum Institute) with 10% glycerine before and during the investigation. Purification of DNA The cells from an overnight culture (6 ml) were harvested and resuspended in 500 L buffer A (50 mm Tris HCl, 50 mm EDTA (ph 8.0)). Twenty microlitres of a lysozyme/ RNase mixture (500 L of lysozyme 100 mg/ml and 25 L of RNase 10 mg/ml) was added and the mixture was incubated for 30 min at 37 C. Fifty microlitres of proteinase K 10 mg/ml and 20 L of 10% sodium dodecyl sulphate (SDS) was added, and the mixture was incubated at 56 C until the lysate cleared. The lysate was extracted twice with phenol:chloroform:isoamylalcohol (25:24:1 by volume) using Phase Lock Gel I Heavy tubes (5 Prime 3 Prime Inc., Boulder, CO, USA) to separate the phases before the DNA was precipitated by adding isopropanol (400 L). The precipitated DNA was collected by centrifugation at 5000g for 5 min and redissolved in 300 L TE buffer (10 mm Tris HCl, 1 mm EDTA (ph 8.0)). The DNA was reprecipitated for 30 min at 4 C by adding 100 L of 7.5 M ammonium acetate and 800 L of 99% ethanol. The precipitated DNA was collected by centrifugation at 5000g for 5 min, dried and dissolved in TE buffer. Southern blot HinfI digests of DNA were separated in a 0.8% agarose gel and transferred to Hybond N membrane (Amersham plc, Amersham, UK) in a Model 785 Vacuum Blotter (Bio- Rad, Hercules, CA, USA) using the Bio-Rad Rapid Transfer Procedure. The membrane was prehybridized with 80 ml of hybridization solution (5 SSC (0.75 M NaCl, M sodium citrate; ph 7.0), 0.5% blocking reagent (Boehringer-Mannheim, Mannheim, Germany), 0.1% sodium laurylsarcosine, 0.02% SDS) in a hybridization oven (Hybaid, Middlesex, UK) at 5 rpm for 1 h at 50 C. The prehybridization solution was replaced with hybridization solution (12 ml) containing 60 ng of biotinylated probe (see Primers and Probes ). The membrane was hybridized overnight at 50 C in the Hybaid hybridization oven at 5 rpm, and then washed twice with 2 SSC (100 ml) for 5 min at 50 C and twice with 0.1 SSC (100 ml) for 15 min at 50 C. The membrane was washed in 10 mm Tris HCl, 15 mm NaCl (ph 7.5) and incubated in 100 ml of 100 mm maleic acid, 150 mm NaCl, 1% blocking reagent for 30 min at room temperature. The membrane was incubated in 100 ml of 100 mm maleic acid, 150 mm NaCl, 1% blocking reagent containing alkaline phosphatase conjugated streptavidin (DAKO, Glostrup, Denmark) in a dilution of 2000:1 for 30 min at room temperature. The membrane was washed twice in 10 mm Tris HCl, 15 mm NaCl (ph 7.5) before being equilibrated in 20 ml of freshly made 100 mm Tris HCl, 100 mm NaCl, 50 mm MgCl 2 (ph 9.5) for 2 min. The membrane was placed in a plastic bag and 50 ml of 100 mm Tris HCl, 100 mm NaCl, 50 mm MgCl 2 (ph 9.5) mixed with 225 L of nitroblue tetrazolium (NBT; Boehringer-Mannheim), and 180 L of 5-bromo-4-chloro- 3-indolyl phosphate (Boehringer-Mannheim) was added. The bag was sealed and incubated in the dark. When the colour development was satisfactory, the membrane was removed from the bag, washed with water and dried at room temperature. meca PCR Two overnight colonies were suspended in autoclaved Milli-Q water (300 L) in a tube and heated at 95 C for 10 min. The tube was centrifuged at 20,000g for 5 min and 10 L of the supernatant was used as template. Each PCR was carried out in a total volume of 100 L, with 10 mm Tris HCl (ph 8.3), 50 mm KCl, 1.5 M MgCl 2, 0.9 M of 468

3 meca gene detection in Staphylococcus aureus each of the primers mecapcr4 and mecapcr5, 0.1 M of each of the primers unip1sa and unip2sa, 100 M each of datp, dttp, dctp and dgtp (Pharmacia Biotech, Uppsala, Sweden), 2.5 units of AmpliTaq (Roche Molecular Systems Inc., Branchburg, CA, USA) and 10 L of template DNA, overlaid with 60 L of mineral oil. The PCR was carried out in an OmniGene thermocycler (Hybaid) using the following program: (i) 5 min at 94 C; (ii) 30 cycles of 1 min at 94 C, 1 min at 45 C, 1 min at 72 C; (iii) 5 min at 72 C; and (iv) hold at 25 C. PCR products were analysed by electrophoresis in 2% agarose gels followed by staining with ethidium bromide. The primer pair unip1sa unip2sa produces a 150 bp fragment of the 16S rrna gene and functions as an internal control of the quality of the template DNA. The primer pair mecapcr4 mecapcr5 produces a 413 bp fragment of the meca gene. Hybridization assay Using a 10 L disposable plastic inoculating loop, 10 L of bacteria were transferred from a 5% blood agar plate to a 2 ml SafeLock microcentrifuge tube (Eppendorf, Hamburg, Germany) containing 350 L of Reagent A. The bacteria were completely suspended by vortexing and the tube was incubated in a heating block at 37 C for 20 min. Reagent B (175 L) was added to the tube and the sample was mixed by vortexing. The tube was incubated in a heating block at 100 C for 5 min, vortexed thoroughly and incubated at 100 C for another 5 min. Reagent C (125 L) containing the detection probes (see Primers and Probes ) was added to the tube and the sample was mixed by vortexing. Immediately, 125 L of the sample was transferred to each of the four wells in a test. The wells were sealed with tape and incubated at 400 rpm and 50 C for 1 h in an ELISA incubator/shaker (Thermo Mixer, National Labnet Company, Woodridge, NJ, USA). The wells were washed four times with 200 L of Wash Solution. Reagent D (100 L) was added to each well, and the wells were sealed with tape and incubated at 37 C for 30 min. The wells were washed four t i m e s with Wash Solution (200 L) and Reagent E (100 L) was added to each well followed by incubation in the dark at 37 C for 30 min. Finally Reagent F (50 L) was added to each well and the result was read by the naked eye. The reagents A F and wash solution were made at the Department of Research and Development, Statens Serum Institute. Primers and probes The following sequences from GenBank were used for designing primers and probes: meca, accession number X52593; nuc, accession number J01785; 16S rrna gene, accession number All primers and probes were made at the Department of Clinical Biochemistry, Statens Serum Institut. The detection probes mecares1, mecares2, nucf2, nucf3 and PSA16S-2 were labelled with seven biotin molecules at the 3 end. The mecares1 probe was used in the meca Southern blots and the nucf3 probe was used in the nuc Southern blots. The capture probes were as follows: mecarcap, 5 - C C A A A C C C G A C A A C T A C A A C T A T T A A A A T A A GTGGAACAA-3 ; nucf1, 5 -AAGGTGTAGAGAAAT A T G G T C C T G A A G C A A G T G C A T T T A C - 3; PSA16S- 3, 5 - G G G C G G A A A C C C C C T A A C A C T T A G C A C T C A - 3. Detection probes were as follows: mecares1, 5 - T T T A T C G G A C G T T C A G T C A T T T C T A C T T C A C C ATTATCGC-3 ; mecares2, 5 -TGAACGTTGCGATC AATGTTACCGTAGTTTGTTTTAATTT-3 ; nucf2, 5 - C A A A G G T C A A A G A A C T G A T A A A T A T G G A C G T G GCTTAGCG-3 ; nucf3, 5 -ATGTTTACAAACCTAAC AATACACATGAACAACATTTAAG-3 ; PSA16S-2, 5 -TTCGCCACTGGTGTTCCTCCATATCTCTGC-3. PCR primers were as follows: mecapcr4, 5 -TCATA TGTGTTCCTGTATTG-3 ; mecapcr5, 5 -TACTATT GATGCAATTGAAG-3 ; unip1sa, 5 -GCTTGCTTCTC TGATGTT-3 ; unip2sa, 5 -TAAGTGACAGCAAGAC CG-3. Antimicrobial susceptibility testing Strains were grown on 5% blood agar plates (Danish blood agar, Statens Serum Institut), incubated overnight at 35 C in ambient air and checked for purity. The inocula were prepared by suspending a few isolated colonies in 0.15 M NaCl giving a density of cfu/ml. Colony counts of the reference strains in each experiment ensured the expected density of the suspensions. The suspensions were then further diluted in 0.15 M NaCl to the desired inoculum for each test. As many MRSA strains grow with different colony morphology after exposure to antibiotics, purity was ensured by plating aliquots from the final dilution of each strain. S. aureus ATCC 29213, S. aureus ATCC 25923, S. aureus 6893 ( meca positive) and Enterococcus faecalis ATCC were used as control strains. The following MICs, as published by the NCCLS, 20,21 were used to determine susceptibility (S) or resistance (R) for agar dilution and Etests: methicillin: S, 8 mg/l; R, 16 mg/l; oxacillin: S, 2 mg/l; R, 4 mg/l. For disc diffusion, the zone diameters of oxacillin disc 1 g were interpreted according to the recommendation of the Swedish Reference Group for Antimicrobials (SRGA-M): S, 12 mm; R, 9 mm. 22 Agar dilution MICs of methicillin (Lucopenin, Durascan, Odense, Denmark) were determined by applying 1 L of a cfu/ml suspension, using a multipoint inoculator (Denley Instruments Ltd, Billingshurst, UK),on tomueller Hinton II agar (Becton Dickinson, Oxford, UK) supplemented with 2% (w/v) NaCl containing methicillin in two-fold dilutions from 1 to 256 mg/l. The plates were incubated for a 469

4 R. L. Skov et al. full 24 h in ambient air at 35 C. The MIC was read as the first plate with >99.9% inhibition of growth (in practice read as fewer than three colonies); a faint haze was disregarded. Etest MICs of methicillin and oxacillin were determined by inoculating Mueller Hinton II agar plates supplemented with 2% (w/v) NaCl by flooding the plates with a 10 7 cfu/ml suspension followed by immediate removal of excess fluid as described by the manufacturer. The surface of the plates was allowed to dry before Etest strips (AB Biodisk) were applied. The plates were incubated for 24 h in ambient air at 35 C and carefully inspected using transmitted light for faint growth or microcolonies. If there were one or more colonies in the ellipse, the MIC was read as the intersection of the Etest strip for the colony with the highest MIC. Etest MICs are reported using the same dilution steps as used for agar dilution, except that 256 mg/l and >256 mg/l are reported in one group since differentiation between these values by the Etest isdifficult. Disc diffusion test Disc diffusion tests were performed using PDM II plates (7 cm) (AB Biodisk), which were inoculated with 50 L of a 10 8 cfu/ml suspension and spread using an L -shaped glass stick resulting in a dense confluent growth. Discs containing 1 g oxacillin (AB Biodisk) (both collection I and II) were applied. The plates were incubated at 30 C for 24 h in ambient air as recommended by the SRGA-M. 22 The zone diameters of inhibition were measured after careful inspection including using transmitted light for faint growth or microcolonies within the zone of inhibition in which case the smaller zone was reported. Oxacillin screening plate test Fifty microlitres of a suspension of 10 8 cfu/ml was applied as a spot 23 on to Mueller Hinton II agar plates supplemented with 4% (w/v) NaCl containing 6 mg/l oxacillin (Sigma, St Louis, MO, USA); the plates were incubated for 24 h at 35 C in ambient air and inspected for growth. The test was read as positive if there was any growth on the plate. Results Comparison of the new hybridization method with Southern blot and PCR All the strains were tested for the presence of the meca gene by the new 3-h hybridization method as well as by Southern blot and PCR. The 3-h hybridization method identified 139 of the 275 S. aureus isolates as being mecapositive and 136 isolates as being meca-negative; these results were in total agreement with the results obtained both by Southern blot and PCR. Two strains of collection I previously classified as MRSA strains were meca-negative by all three DNA-based tests. One of these strains was found to be phenotypically sensitive, whereas the other was found to be highly resistant by all methods (see Phenotypic results ). Four of the strains from collection II were correctly identified as meca-positive. All of the S. aureus strains were positive for the nuc gene both in Southern blot using the nucf3 probe and in the hybridization assay. Phenotypic resistance of S. aureus strains of collection I The bimodal distribution of MICs and zone diameters by different phenotypic methods for the 135 meca-positive and 122 meca-negative strains of collection I is shown in Figure 1. The two Etests gave a good separation between meca-positive and meca-negative strains, while for agar dilution and for the 1 g oxacillin disc there was some overlap. For meca-negative strains no differences in methicillin MICs between those obtained by Etest and agar dilution were found (i.e. variations within a range of 1 log 2 ). However, for meca-positive strains Etest MICs were higher than MICs obtained by agar dilution. Using the breakpoints recommended by NCCLS, 20,21 the correlation between the phenotypic results and presence of the meca gene is shown in Table I. The oxacillin screen plate method Table I. Correlation between phenotypic test results (number of strains found to be resistant and sensitive) and presence of the meca gene for the 257 strains of collection I Susceptibility based on MIC from Oxacillin 1 g Phenotypic screening oxacillin methicillin methicillin oxacillin classification plate disc agar dilution Etest Etest meca positive resistant (n 135) sensitive meca negative resistant (n 122) sensitive

5 meca gene detection in Staphylococcus aureus Figure 1. Distribution of MICs obtained by various methods for meca-positive ( ) and meca-negative ( ) S. aureus strains from collection I. (a) Methicillin, agar dilution method; (b) oxacillin, disc diffusion (1 g oxacillin); (c) methicillin, Etest; (d) oxacillin, Etest. The breakpoints for resistance for oxacillin ( 4 mg/ml), methicillin ( 16 mg/ml) 27 and for oxacillin disc diffusion ( 9 mm) 22 are shown by dotted lines. and Etest MIC of both oxacillin and methicillin gave good agreement (>99%) for both meca-positive and mecanegative strains. For agar dilution using methicillin, seven discrepant results were found, five of 135 (3.7%) mecapositive strains were classified as sensitive, and two of 122 (1.7%) meca-negative strains were classified as resistant. The 1 g oxacillin disc fully agreed for the meca-positive strains, but 18 out of 122 (14.8%) meca-negative strains were classified as resistant (i.e. zone diameter 9 mm). Phenotypic resistance of S. aureus strains of collection II The results for these strains are shown in Table II. One of the four meca-positive strains was found to be sensitive to methicillin by both agar dilution and Etest, while the rest of the results were in agreement with the DNA methods. For the 14 meca-negative BORSA strains a high number of discrepant results were found. Only one strain was sensitive in all phenotypic tests. The oxacillin screen plate test was the most accurate with corresponding results in 13/14 (93%), whereas the other tests had a higher number of discrepant results, highest for 1 g oxacillin (13/14) and oxacillin Etest (9/14). Detailed results of strains giving discrepant genotypic and phenotypic test results Table III shows the actual MICs and zone diameters for the strains where the genotypic and the phenotypic resistance testing gave discrepant results. In collection I, five mecapositive strains were found to be susceptible by agar dilution. Repeated testing by agar dilution of these five strains resulted in MICs of 8 mg/l (one strain), 16 mg/l (one strain), 32 mg/l (two strains) and 64 mg/l (one strain). One of these strains (ID 25), was also negative by the oxacillin screen plate test. Repeated testing gave a negative result on two occasions and only a faintly positive result on a third occasion. Nineteen meca-negative strains from collection I were found to be resistant in one or more of the methods. One strain (ID 38) was highly resistant in all methods. One strain (ID 155) was resistant to methicillin according to the agar dilution method but not by Etest. Repeated agar dilution testing gave MICs of 8 and 16 mg/l. The rest of the strains with discrepant results were only found to be resistant by the disc diffusion method. Several of these strains had higher MICs than isolates that were sensitive to oxacillin in the disc diffusion assay. One of the heteroresistant class I MRSA strains (ID 307, collection II) was found to be susceptible to methicillin 471

6 R. L. Skov et al. Table II. Correlation between phenotypic test results (number of strains found to be resistant and sensitive) and presence of the meca gene for 14 BORSA strains and four MRSA (collection II) Susceptibility based on MIC from Oxacillin 1 g Phenotypic screening oxacillin methicillin methicillin oxacillin classification plate disc agar dilution Etest Etest meca positive resistant (n 4) sensitive meca negative resistant (n 14) sensitive both by agar dilution and Etest, but resistant by the rest of the phenotypic methods. All except one of the BORSA strains were found to be resistant by the 1 g oxacillin disc testing and nine of these had MICs indicating resistance to either oxacillin and/or methicillin. Discussion The spread of MRSA strains during the last decade in various parts of the world, especially the spread outside hospitals (e.g. in nursing homes), has made it both difficult and expensive to control or prevent the dissemination of these strains. 24 Glycopeptides, especially vancomycin, are the basis for treatment of infections due to methicillinresistant staphylococci. With the emergence of infections caused by coagulase-negative staphylococci this has resulted in increased use of vancomycin, encouraging the spread of vancomycin-resistant enterococci and, recently, S. aureus strains with reduced susceptibility to vancomycin. 25,26 Rapid and exact detection of methicillin resistance will decrease unnecessary use of vancomycin and also provide a tool to control the spread of MRSA strains. The presence of the meca gene is considered the hallmark for identification of MRSA strains. 1,5 The presented method was designed to be a fast, ready-to-use kit for routine microbiology laboratories and as presented here showed total agreement with established DNA-methods. Besides the reagents in the kit, the method requires facilities for incubation at 37 C and 100 C, and for hybridization at 50 C while shaking. For the latter we have used a heated microplate shaker. The assay time for performing the test manually on one strain is 3 h,with hands-on time of 40 min. For 24 strains the corresponding times are 4 h 30 min and 2 h 10 min (i.e. 5 6 min per sample) respectively. Hands-on time is defined as the total assay time minus all incubation steps that last for at least 20 min. The presented kit is an alternative to other DNA-based methods. It will probably result in increased expense for the laboratory and is more technically demanding and laborious than phenotypic methods. However, it will, especially for borderline strains, result in earlier and more accurate determination, leading to lower use of vancomycin and thereby reducing both cost of care and the risk of resistance development. The strains comprising collection II, with known borderline methicillin resistance, were included because such strains can be either meca-positive heterogeneous MRSA strains (class I and II) or meca-negative (BORSA) strains; these two categories can be very difficult to distinguish using phenotypic methods. 5,27 The differentiation is very important, as the former are fully capable MRSAs while the BORSA strains can be treated as methicillin-susceptible strains, both in regards of choice of antibiotics and infection control measures. 5 In this study, as described in previous publications, 8,9 the agar screen is a highly reliable method for classifying S. aureus strains as MRSA or MSSA for borderline resistant strains. The Etest methods had a high degree of accuracy for both the MRSA and MSSA strains from collection I. For BORSA strains, however, 64% and 29% were falsely classified as resistant to oxacillin and methicillin, respectively. Etest MICs were higher than those obtained by agar dilution for MRSA strains. This difference, resulting in the Etest being slightly more accurate than agar dilution (Table II), resulted in several cases from the Etest MIC being generated by one or a few colonies within the ellipse while the majority of colonies had a much lower MIC. That a greater proportion of BORSA strains were misclassified by the oxacillin Etest than the methicillin Etest probably reflects the lower -lactamase stability of the former. 28 The oxacillin disc diffusion test has previously been found to be less reliable, with high numbers of both false-susceptible and false-resistant results In order to avoid falsesusceptible results, the inoculum used in this study was slightly higher than what is obtained by using the cotton swab method. 22 As a result, all meca-positive strains were found to be resistant. However, using this inoculum 13.9% of the meca-negative collection I strains (strain ID 38 excluded) and 93% of the BORSA strains were falsely found to be resistant. As seen in Figure 1, the zone diameters of meca-negative strains showed considerable 472

7 meca gene detection in Staphylococcus aureus Table III. Detailed results of isolates showing disagreement between susceptibility test results and the presence of the meca gene MIC (mg/l) Isolate meca methicillin methicillin oxacillin Oxacillin Disc diffusion no. Collection gene agar dilution Etest Etest screen plate (mm) 25 I I I I I II I 64 >256 > I I I I I I I I I I I I I I I I I I II II II II II II II II II II II II II Divergent phenotypic results are marked in grey. variability, probably reflecting differences in the amount of -lactamase produced. This study confirms that this method is less accurate, and that an increase in inoculum in order to avoid false-susceptible results yields a high number of false-resistant strains. Mechanisms other than meca-mediated methicillin resistance have been described, including modified penicillin binding proteins (PBPs) or overexpression of existing PBPs. The frequency of such strains is unknown but is believed to be low and without epidemiological significance. 5 One such unusually highly resistant meca-negative, -lactamase-positive strain (ID 38, Table III) was found in this study. The strain was also found to be resistant to co-amoxiclav (data not shown). The resistance mechanism 473

8 R. L. Skov et al. has not been further explored, but is probably not conferred by -lactamase production alone as the strain is coamoxiclav-resistant and -lactamase hyperproducing strains only have marginally elevated MICs of methicillin. 32,33 In summary, this study has shown that the newly developed 3-h hybridization assay is both exact and reliable. The study further confirms that phenotypic tests of methicillin resistance in S. aureus strains creates both false-susceptible and false-resistant results. These problems are best avoided using a meca gene-based detection system. False classification would then only be reported for strains which are highly methicillin-resistant due to mechanisms not based on the meca gene. Such strains are presumed to be very rare. Acknowledgement Part of this work was presented at the 37th ICAAC, Toronto, 1997 (abstract D-24). References 1. Hiramatsu, K. (1995). Molecular evolution of MRSA. Microbiology and Immunology 39, Tomasz, A., Drugeon, H. B., de Lencastre, H. M., Jabes, D., McDougall, L. & Bille, J. (1989). New mechanism for methicillin resistance in Staphylococcus aureus: clinical isolates that lack the PBP 2a gene and contain normal penicillin-binding proteins with modified penicillin-binding capacity. Antimicrobial Agents and Chemotherapy 33, Henze, U. U. & Berger-Bachi, B. (1996). Penicillin-binding protein 4 overproduction increases -lactam resistance in Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 40, Matthews, P. R. & Stewart, P. R. (1984). Resistance heterogeneity in methicillin-resistant Staphylococcus aureus. FEMS Microbiology Letters 22, Chambers, H. F. (1997). Methicillin resistance in staphylococci: molecular and biochemical basis and clinical implications. Clinical Microbiology Reviews 10, Bignardi, G. E., Woodford, N., Chapman, A., Johnson, A. P. & Speller, D. C. E. (1996). Detection of the mec-a gene and phenotypic detection of resistance in Staphylococcus aureus isolates with borderline or low-level methicillin resistance. Journal of Antimicrobial Chemotherapy 37, Hartman, B. J. & Tomasz, A. (1986). Expression of methicillin resistance in heterogeneous strains of Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 29, Skulnick, M., Simor, A. E., Gregson, D., Patel, M., Small, G. W., Kreiswirth, B. et al. (1992). Evaluation of commercial and standard methodology for determination of oxacillin susceptibility in Staphylococcus aureus. Journal of Clinical Microbiology 30, Mackenzie, A. M., Richardson, H., Lannigan, R. & Wood, D. (1995). Evidence that the National Committee for Clinical Laboratory Standards disk test is less sensitive than the screen plate for detection of low-expression-class methicillin-resistant Staphylococcus aureus. Journal of Clinical Microbiology 33, Archer, G. L. & Pennell, E. (1990). Detection of methicillin resistance in staphylococci by using a DNA probe. Antimicrobial Agents and Chemotherapy 34, Youmans, G. R., Davis, T. E. & Fuller, D. D. (1993). Use of chemiluminescent DNA probes in the rapid detection of oxacillin resistance in clinically isolated strains of Staphylococcus aureus. Diagnostic Microbiology and Infectious Disease 16, Kolbert, C. P., Connolly, J. E., Lee, M. J. & Persing, D. H. (1995). Detection of the staphylococcal meca gene by chemiluminescent DNA hybridization. Journal of Clinical Microbiology 33, Vannuffel, P., Gigi, J., Ezzedine, H., Vandercam, B., Delmee, M., Wauters, G. et al. (1995). Specific detection of methicillinresistant Staphylococcus species by multiplex PCR. Journal of Clinical Microbiology 33, Geha, D. J., Uhl, J. R., Gustaferro, C. A. & Persing, D. H. (1994). Multiplex PCR for identification of methicillin-resistant staphylococci in the clinical laboratory. Journal of Clinical Microbiology 32, Hiramatsu, K., Kihara, H. & Yokota, T. (1992). Analysis of borderline-resistant strains of methicillin-resistant Staphylococcus aureus using polymerase chain reaction. Microbiology and Immunology 36, Ubukata, K., Nakagami, S., Nitta, A., Yamane, A., Kawakami, S., Sugiura, M. et al. (1992). Rapid detection of the meca gene in methicillin-resistant staphylococci by enzymatic detection of polymerase chain reaction products. Journal of Clinical Microbiology 30, Unal, S., Hoskins, J., Flokowitsch, J. E., Wu, C. Y. E., Preston, D. A. & Skatrud, P. L. (1992). Detection of methicillin-resistant staphylococci by using the polymerase chain reaction. Journal of Clinical Microbiology 30, Ryffel, C., Tesch, W., Birch-Machin, I., Reynolds, P. E., Barberis-Maino, L., Kayser, F. H. et al. (1990). Sequence comparison of meca genes isolated from methicillin-resistant Staphylococcus aureus and Staphylococcus epidermidis. Gene 94, Shortle, D. (1983). A genetic system for analysis of staphylococcal nuclease. Gene 22, National Committee for Clinical Laboratory Standards. (1997). Performance Standards for Antimicrobial Disk Susceptibility Tests Sixth Edition: Approved Standard M2-A6. NCCLS, Wayne, PA. 21. National Committee for Clinical Laboratory Standards. (1997). Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically Fourth Edition: Approved Standard M7-A4. NCCLS, Wayne, PA. 22. Olsson-Liljequist, B., Larsson, P., Walder, M. & Miörner, H. (1997). Antimicrobial susceptibility testing in Sweden. III. Methodology for susceptibility testing. Scandinavian Journal of Infectious DiseasesSuppl. 105, Alfa, M. J., Ongsansoy, E., Hoban, S., Mathew, T. & Greenaway, C. (1994). Methicillin-resistant Staphylococcus aureus. Optimization of the agar screen plate method. Diagnostic Microbiology and Infectious Disease 18, Voss, A. & Doebbeling, B. N. (1995). The worldwide prevalence of methicillin-resistant Staphylococcus aureus. International Journal of Antimicrobial Agents 5, Jones, R. N., Marshall, S. A., Pfaller, M. A., Wilke, W. W., Hollis, R. J., Erwin, M. E. et al. (1997). Nosocomial enterococcal 474

9 meca gene detection in Staphylococcus aureus blood stream infections in the SCOPE program: antimicrobial resistance, species occurrence, molecular testing results, and laboratory testing accuracy. Diagnostic Microbiology and Infectious Disease 29, Hiramatsu, K., Hanaki, H., Ino, T., Yabuta, K., Oguri, T. & Tenover, F. C. (1997). Methicillin-resistant Staphylococcus aureus clinical strain with reduced vancomycin susceptibility. Journal of Antimicrobial Chemotherapy 40, Gerberding, J. L., Miick, C., Liu, H. H. & Chambers, H. F. (1991). Comparison of conventional susceptibility tests with direct detection of penicillin-binding protein 2a in borderline oxacillin-resistant strains of Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 35, Frimodt-Moller, N., Rosdahl, V. T., Sorensen, G., Hartzen, S. H. & Bentzon, M. W. (1986). Relationship between penicillinase production and the in-vitro activity of methicillin, oxacillin, cloxacillin, dicloxacillin, flucloxacillin, and cephalothin against strains of Staphylococcus aureus of different phage patterns and penicillinase activity. Journal of Antimicrobial Chemotherapy 18, Kotilainen, P., Hyvarinen, J., Jarvinen, H., Linko, L., Eerola, E., Lehtonen, O. P. et al. (1995). Testing of methicillin resistance by in vitro susceptibility and the presence of the meca gene in clinical Staphylococcus aureus isolates in Finland. Scandinavian Journal of Infectious Diseases 27, Huang, M. B., Gay, T. E., Baker, C. N., Banerjee, S. N. & Tenover, F. C. (1993). Two per cent sodium chloride is required for susceptibility testing of staphylococci with oxacillin when using agarbased dilution methods. Journal of Clinical Microbiology 31, de Lencastre, H., Sá Figueiredo, A. M., Urban, C., Rahal, J. & Tomasz, A. (1991). Multiple mechanisms of methicillin resistance and improved methods for detection in clinical isolates of Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 35, McMurray, L. W., Kernodle, D. S. & Barg, N. L. (1990). Characterization of a widespread strain of methicillin-susceptible Staphylococcus aureus associated with nosocomial infections. Journal of Infectious Diseases 162, Woods, G. L. & Yam, P. (1988). Bactericidal activity of oxacillin against -lactamase hyperproducing Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 32, Received 15 July 1998; returned 9 September 1998; revised 19 October 1998; accepted 20 November

10