B. Heym, M. Le Moal, L. Armand-Lefevre and M.-H. Nicolas-Chanoine*

Transcription

1 Journal of Antimicrobial Chemotherapy (2002) 50, DOI: /jac/dkf132 Multilocus sequence typing (MLST) shows that the Iberian clone of methicillin-resistant Staphylococcus aureus has spread to France and acquired reduced susceptibility to teicoplanin B. Heym, M. Le Moal, L. Armand-Lefevre and M.-H. Nicolas-Chanoine* Microbiology Department, Hôpital Ambroise Paré, AP-HP and Faculté de Médecine Paris-Ouest, Université Paris V, 9 avenue Charles de Gaulle, Boulogne-Billancourt, France Received 2 January 2002; returned 25 March 2002; revised 9 May 2002; accepted 31 May 2002 Infection by methicillin-resistant Staphylococcus aureus (MRSA) presents a serious problem, as these organisms are resistant to a wide range of antibiotics. Moreover, MRSA with reduced susceptibility to glycopeptides (GISA) have emerged in recent years. In our hospital we were faced with an outbreak of GISA between January and March 2000, involving six patients who were free of MRSA when the GISA isolate emerged, and five of whom had not been given glycopeptides. The initial GISA isolate from the six patients was compared with six historical multiple-resistant MRSA isolates ( ), which had not been found to be GISA by the routinely used method. Antibiotic susceptibility was studied through the disc diffusion method, and MICs of glycopeptides were determined by Etest, agar and broth dilution techniques. Molecular strain typing was done by PFGE and multilocus sequence typing (MLST). All 12 isolates that belonged to the gentamicin-resistant group of MRSA were susceptible to vancomycin, but showed reduced susceptibility to teicoplanin through at least one MIC method. PFGE typing yielded five different but closely related profiles, and eight of the 12 clinical isolates were classified into a single PFGE type. MLST yielded an identical allelic profile for all 12 isolates, corresponding to the profile ( ) of the Iberian clone HPV107 of MRSA. MLST allowed us to confirm the presence of the Iberian MRSA clone HPV107 with the same allelic profile in our hospital for several years. We can now note that strains of this clone have acquired reduced susceptibility to teicoplanin. Introduction Infection by methicillin-resistant Staphylococcus aureus (MRSA) presents a serious problem in hospital practice, as these organisms are resistant to a wide range of otherwise effective antibiotic agents. 1 In recent times, these already multidrug-resistant organisms have been shown to be able to develop resistance to the last antibiotic family to which they were still susceptible, the glycopeptides. The history of S. aureus with reduced susceptibility to glycopeptides (GISA) began with the discovery of clinical isolates with intermediate susceptibility to teicoplanin. The first case, described in 1990, concerned an isolate susceptible to methicillin (MSSA), whereas the second case, in 1995, concerned clonally related strains of MRSA isolated in inpatients of a French hospital. 2,3 The publication of sporadic MRSA isolates less susceptible to both vancomycin and teicoplanin, first in Japan, 4 then in the USA 5,6 and finally in France, 7 triggered deep concern in the international medical community, because glycopeptides have been the drug of choice and commonly the sole antimicrobial agent available for the treatment of multiple-resistant S. aureus isolates for the past 15 years. 8,9 The concern in the medical community was enhanced by the microbiologists uncertainty about the tests routinely used to detect GISA isolates, which did not seem to be reliable. 10 The microbiological analysis performed by Sieradzki et al. 6 on three MRSA isolates, obtained consecutively over a period of 3 months from a haemo-dialysed patient repeatedly treated with vancomycin, showed that a subpopulation which required... *Corresponding author. Tel: ; Fax: ; marie-helene.nicolas-chanoine@apr.ap-hop-paris.fr The British Society for Antimicrobial Chemotherapy

2 B. Heym et al. a higher inhibitory concentration of vancomycin (MIC 16 mg/l) could be selected in vitro from the last isolated MRSA strain, which was less susceptible to vancomycin (MIC 8 mg/l). Moreover, they demonstrated that MRSA isolates collected in their hospital and clonally related to the MRSA isolates of the haemo-dialysed patient also contained subpopulations of bacteria for which the MIC of vancomycin was greater than that required by the majority of bacterial cells. 6 A systematic search for such hetero-visa subpopulations in MRSA isolates was recently performed by the Central Public Health Laboratory for epidemiological typing in the UK, and among S. aureus in general by the French National Reference Centre in France. 11,12 Approximately three-quarters of the MRSA strains studied in the UK from May 1998 to April 1999 belonged to one or other of the two principal UK epidemic strains (EMRSA-15 and EMRSA-16), whereas those studied in France were representative of strains prevalent in France between 1987 and In the UK, no hetero-visa were found in the above-mentioned study, whereas in France hetero-visa were found in six of 11 gentamicin-resistant MRSA strains, but not in gentamicin-susceptible MRSA strains. 11,12 In our hospital, we were faced with an outbreak of MRSA with reduced susceptibility to teicoplanin (GISA) between January and March 2000, involving six patients who were hospitalized in three different wards and who were free of MRSA infection and colonization when the GISA isolate emerged. These GISA isolates belonged to the gentamicinresistant MRSA group, which are still isolated nowadays in our hospital, although their frequency has decreased continuously for the past 5 years. 13 Table 1. The initial GISA isolate of six patients involved in an outbreak and MRSA isolates of six unlinked control patients according to hospitalization ward, and type and year of sample Patient Ward Sample Year 1 ICU1 nasal swab surgery urine orthopaedic hip prosthesis surgery wound ICU1 nasal swab ICU1 nasal swab 2000 PAG rheumatology scar 1999 VIL internal medicine urine 1999 BEC rheumatology wound 1998 SOP ICU2 PBS 1997 PEY surgery urine 1997 LEV surgery wound 1996 We decided to investigate this outbreak, which was suspected, on the basis of the clinical data, to be due to the spread of the same strain, by using two typing methods: pulsed-field gel electrophoresis (PFGE) and multilocus sequence typing (MLST). The PFGE typing method, which is the technique usually applied to type multiple-resistant MRSA isolates, generally classifies such isolates into one or few clones Consequently, by this technique it is difficult to prove that an outbreak results from the transmission of the same strain. For this reason, we also used the MLST method, which has been shown to be able to characterize a clone in an MRSA population. 17,18 Indeed, by sequencing an internal fragment of seven unlinked housekeeping genes, allelic profiles can be determined, and a clone is defined as a group of isolates having a strictly identical sequence of all seven genes. 18 This method has the advantage of not succumbing to interpretative ambiguities, as is the case for the PFGE technique Materials and methods Outbreak description Six patients were involved in an outbreak of GISA infection or colonization (Table 1). The epidemiological graph (Figure 1) shows overlapping stays of patients 1 and 2, then patients 2 and 4 in the digestive surgery ward, and patients 1, 4 and 5, then patients 1, 5 and 6 in the surgery intensive care unit (ICU) of the Ambroise Paré hospital. Patient 3, who was being hospitalized in the orthopaedic ward when the GISA isolate was obtained from hip prosthesis haematoma, had spent 2 days in the digestive surgery ward before being transferred to the orthopaedic ward. This stay was concomitant to that of patient 2, who was hospitalized in the same unit of the digestive surgery ward as patient 3, and who was found, 5 days after the transfer of patient 3 to the orthopaedic ward, to have a GISA urinary tract infection. None of the six patients had been concerned by MRSA nasal carriage or MRSA infections before the isolation of the GISA isolate. None of them had been given glycopeptides, except patient 5 who had been treated with vancomycin (1.5 g per day in continuous infusion) for 8 days because of a Staphylococcus hominis bacteraemia when the GISA isolate was found for the first time in his nose. The isolation barriers that have been set up for the last 5 years in all wards of the hospital for MRSA colonized or infected patients were reinforced at the end of February in the surgery ICU when the second case appeared in this unit. The last GISA isolation took place in this unit at the beginning of March. Bacterial strains ICU1, surgery intensive care unit; ICU2, medical intensive care unit; PBS, protected brush specimen. The first one of several GISA clinical isolates of each of the six patients was studied (Figure 1; Table 1). Six multiple- 324

3 Multilocus sequence typing of MRSA Figure 1. Epidemiological graph showing the hospital stays of six patients involved in an outbreak of GISA infection or colonization. 1, patient 1, hospitalized from 17th of December 1999 to 29th of March 2000; 2, patient 2, hospitalized from 30th of November 1999 to 1st of February 2000; 3, patient 3 hospitalized from 12th of January 2000 to 13th of April 2000; 4, patient 4 hospitalized from 14th to 20th of December 1999, then from 11th of January 2000 to 23rd of February 2000; 5, patient 5 hospitalized from 11th of January 2000 to 26th of February 2000; 6, patient 6 hospitalized from 5th of February 2000 to 8th of March White triangles, negative surveillance sample; grey triangles, GISA positive surveillance sample; white ovals, GISA negative clinical sample; grey ovals, GISA positive clinical sample; grey bars, surgery ICU; striped bars, digestive surgery ward; spotted bar, internal medicine ward; white bar, orthopaedics ward. resistant MRSA isolates that were apparently susceptible to glycopeptides, and that were isolated up to 5 years before this outbreak ( historical isolates) in our hospital (isolate LEV, 1996; isolates PEY and SOP, 1997; isolate BEC, 1998; and isolates VIL and PAG, 1999), and the methicillin-resistant S. aureus reference strain COL 22 were also studied by the same means with regard to their antibiotic susceptibility and their molecular typing profile (Table 1). Antibiotic susceptibility Antimicrobial susceptibility was determined by the agar disc diffusion method on Mueller Hinton agar (biomérieux, Marcy l Étoile, France) according to the recommendations of the French Antibiogram Committee. 23 The antibiotics (Bio- Rad, Marne la Coquette, France) tested were: penicillin G, oxacillin, erythromycin, lincomycin, tetracycline, chloramphenicol, gentamicin, trobramycin, amikacin, rifampicin, pefloxacin, fusidic acid, sulfamethoxazole, trimethoprim, cotrimoxazole, fosfomycin, vancomycin and teicoplanin. The breakpoint for glycopeptide susceptibility was an inhibition zone diameter 17 mm. For diameters <17 mm, the French Antibiogram Committee recommends that the MIC is determined. 23 The MICs of vancomycin and teicoplanin were determined by the Etest system (AB Biodisk, Solna, Sweden), and by agar (French Antibiogram Committee) and broth (NCCLS) dilution methods. 23,24 Etest was carried out in three independent assays on Mueller Hinton agar with an inoculum of 10 5 cfu/ml (0.5 McFarland). The MICs of vancomycin and teicoplanin were read after 24 and 48 h of incubation at 37 C according to the recommendations of the manufacturer. A triplicate determination of teicoplanin (Hoechst Marion Roussel, Romainville, France) and vancomycin (Lilly, Saint Cloud, France) MICs was performed on Mueller Hinton agar using a Steers replicator with an inoculum of 10 4 cfu/spot. The determination of the MICs according to NCCLS standards was carried out in liquid medium (Mueller Hinton broth; biomérieux). All MIC methods were interpreted according to the recommendations of the French Antibiogram Committee. 23 Molecular typing systems PFGE was carried out using the GenePath system with a CHEF DRII apparatus (Bio-Rad Laboratories), as described previously. 19 Briefly, SmaI-digested DNA was separated for 20 h with a current range of ma and an angle of 325

4 B. Heym et al. Table 2. Clinical data concerning the six patients involved in a GISA outbreak Patient Underlying disease GISA infection site Antibiotic treatment Cure of infection Outcome 1 colon adenocarcinoma UTI, KT, bacteraemia vancomycin a + fusidic acid + chloramphenicol yes recovered 2 colon adenocarcinoma UTI vancomycin a + fusidic acid + chloramphenicol yes recovered 3 hip fracture hip prosthesis vancomycin a + fusidic acid + chloramphenicol yes deceased 4 colon adenocarcinoma parietal wound vancomycin a yes recovered 5 acute pancreatitis pancreatitis vancomycin a + fusidic acid no deceased 6 acute pancreatitis nasal carrier none na recovered UTI, urinary tract infection; KT, catheter infection; na, not applicable. a Correct vancomycin serum concentrations The profile interpretation was performed according to Tenover et al. 19 MLST analysis was carried out as described previously, including internal fragments of the following seven housekeeping genes: arcc (carbamate kinase), aroe (shikimate dehydrogenase), glpf (glycerol kinase), gmk (guanylate kinase), pta (phosphate acetyltransferase), tpi (triosephosphate isomerase) and yqil (acetyl coenzyme A acetyltransferase). 17 For specific DNA amplification of an 450 bp fragment of each of these genes, the primer sequences given at the MLST website were used ( Sequencing was performed by the big dye technique with an ABI Prism sequencer (Applied Biosystems, Courtaboeuf, France). The profile analysis was carried out with the S. aureus MLST website scheme (multiple locus query and allelic profile query) ( Results Patients, treatment and outcome Salient clinical data concerning the six patients involved in the GISA outbreak are presented in Table 2. Three patients had been hospitalized for surgery of colon adenocarcinoma, and two patients for acute pancreatitis. Patient 3 had been admitted following an accidental hip fracture. All patients except patient 3 had received non-glycopeptide antibiotics prior to the isolation of the GISA. Patients 1 5 were infected by the GISA isolate, whereas patient 6 became a chronic nasal carrier of the GISA isolate. This patient did not receive antistaphylococcal treatment and recovered without further septic complications. Patients 1 3 had been treated with vancomycin by continuous infusion in combination with fusidic acid and chloramphenicol. Patients 4 and 5 had undergone surgery and had been treated by vancomycin alone (patient 4) or vancomycin plus fusidic acid (patient 5). The infection of patients 1 and 2 was cured and both patients recovered and left the hospital, one of them having become a chronic carrier of the GISA isolate. In patient 3 the infection was cured, but the patient died of heart failure. Patient 4 was cured of his infection and recovered, whereas patient 5 died of septic shock (Table 2). Antibiotic susceptibility The six isolates obtained from the six patients involved in the outbreak, and the six historical MRSA isolates, which all belonged to the gentamicin-resistant MRSA group, were susceptible only to chloramphenicol, fusidic acid, trimethoprim and sulfamethoxazole, with one exception among the historical isolates (isolate PEY 1997), which was additionally susceptible to erythromycin, lincomycin and fosfomycin. The size of the inhibition zone diameter for teicoplanin was <17 mm for all outbreak isolates, but 17 mm for the historical MRSA isolates, whereas that for vancomycin was 17 mm in all cases (Table 3). The MRSA reference strain COL was susceptible to teicoplanin and vancomycin according to the critical inhibition zone diameters (Table 3). The MICs of vancomycin for all outbreak and historical isolates were 4 mg/l, i.e. all isolates were susceptible to vancomycin by Etest, agar dilution and broth dilution testing. The MICs of teicoplanin for the six outbreak isolates varied between 6 and 16 mg/l by the Etest system, whereas they varied between 8 and 16 mg/l by the agar dilution method. The broth dilution method yielded similar results (6 8 mg/l) for four of the six outbreak isolates (1 4), but for isolates 5 and 6 MICs were equal to 4 mg/l, thus categorizing these isolates as susceptible to teicoplanin by the broth dilution method. One of the historical MRSA isolates (SOP 1997) was found to be completely susceptible to teicoplanin and vancomycin by Etest and agar dilution (MICs 2 and 4 mg/l, respectively), but intermediately susceptible to teicoplanin by the broth dilution method (MIC 6 mg/l). The MIC of teicoplanin 326

5 Multilocus sequence typing of MRSA Table 3. Vancomycin (VAN) and teicoplanin (TEC) susceptibility of outbreak isolates, historical isolates and MRSA reference strain COL MIC (mg/l) Inhibition zone diameter (mm) Etest agar dilution broth dilution Year Isolate TEC VAN TEC VAN TEC VAN TEC VAN PAG VIL BEC SOP PEY LEV COL for isolates PAG (1999), BEC (1998), PEY (1997) and LEV (1996) was found to be in the intermediately susceptible range by the Etest and agar dilution methods (MICs 8 16 mg/l), but two of them [PAG (1999) and LEV (1996)] were found to be susceptible (MIC 4 mg/l) by the broth dilution method. For the last historical isolate [VIL (1999)], the MIC of teicoplanin was 32 mg/l by Etest, 8 mg/l by agar dilution and 14 mg/l by broth dilution (Table 3). Molecular typing PFGE. As shown in Figure 2, five of the six outbreak isolates (lanes 1 4 and 6) displayed an identical SmaI-restricted PFGE profile (profile A). The profile of the isolate of patient 4 (lane 5) differed from profile A by two bands (profile A 1 ) (Figure 2; Table 4). Among the six MRSA historical isolates, three displayed a profile strictly identical to profile A (isolates SOP 1997, BEC 1998 and PAG 1999; Figure 2, lanes 8, 11 and 12, respectively). The profiles displayed by the three remaining MRSA historical isolates differed from profile A by either one additional band (isolate LEV, Figure 2, lane 9; profile A 4 ) or three additional bands (isolate PEY, Figure 2, lane 10; profile A 3 ), or a deletion of one band and an addition of one band (isolate VIL, Figure 2, lane 13; profile A 2 ). MRSA reference isolate COL yielded a profile close to that of our clinical isolates, with a difference of three bands compared with profile A (Figure 2, lane 7; profile A 5 ). MLST. All six outbreak isolates displayed an identical MLST profile for the seven housekeeping genes, namely for the genes arcc (allele 3), aroe (allele 3), glpf (allele 1), gmk (allele 12), pta (allele 4), tpi (allele 4) and yqil (allele 16), respectively. This MLST allelic profile was also found for all six historical MRSA isolates. Concerning MRSA strain COL, the nucleotide sequences of the seven housekeeping genes were obtained by BLAST of the MLST reference sequence from the Institute for genomic research (TIGR) website ( mdbinprogress.html), where the sequencing project of strain COL is currently underway. The sequence obtained for its seven genes arcc, aroe, glpf, gmk, pta, tpi and yqil was identified with 100% identity to the sequence of alleles , respectively. Neither allelic profile is included in the MLST website database. Discussion Faced with an outbreak of MRSA with reduced susceptibility to teicoplanin detected by the disc diffusion method, we studied bacteriologically and molecularly the six outbreak isolates in comparison with six teicoplanin-susceptible historical MRSA isolates, because both groups of isolates were multiply antibiotic resistant and belonged to the gentamicinresistant group of MRSA. 15 This comparative analysis emphasized the difficulty in determining reduced susceptibility to glycopeptides by the disc diffusion method. 10 In fact, 327

6 B. Heym et al. Figure 2. PFGE profiles of the six outbreak GISA and the six historical MRSA isolates and the MRSA reference strain COL. M, molecular weight markers; lane 1, patient 6; lane 2, patient 5; lane 3, patient 3; lane 4, patient 1; lane 5, patient 4; lane 6, patient 2; lane 7, strain COL; lane 8, patient SOP; lane 9, patient LEV; lane 10, patient PEY; lane 11, patient BEC; lane 12, patient PAG; lane 13, patient VIL. reduced susceptibility to teicoplanin was never detected by the disc diffusion method in the historical isolates, whereas it was found in three of six of these isolates by the three MIC methods, and in three isolates by at least one MIC method. Regarding the concordance of the teicoplanin MICs, we observed that it was the best between the Etest and agar dilution methods for all 12 isolates. To overcome the problem of misinterpreted teicoplanin susceptibility by the disc diffusion method, we now determine MICs of teicoplanin and vancomycin for all MRSA belonging to the gentamicin-resistant group by Etest. The clinical and epidemiological data strongly suggested that the isolation of the GISA isolates from the six patients resulted from the transmission of the same strain. Classically, molecular typing systems are used to confirm the clonal relatedness of outbreak isolates. For S. aureus the reference typing system is PFGE. However, for MRSA, it has been shown that PFGE profiles of epidemiologically unlinked isolates yielded remarkable uniformity due to the geographic dissemination of MRSA. 14,15,25 27 Our study confirmed this finding, as the six historical isolates, which were unlinked to each other and to the six outbreak isolates, displayed identical or similar PFGE profiles. Overall, according to the PFGE clone definition established by Tenover et al., 19 the six outbreak isolates, the six historical isolates and MRSA reference strain COL are the same clone. By using the MLST technique, we definitively confirmed that the outbreak and historical isolates belonged to the same clone, whose allelic profile corresponds to that of the Iberian MRSA clone HPV107 (MLST ), which was recently identified in several MRSA isolates belonging to this clone. 26,28 This clone, which has previously been found among several historical strains isolated in Denmark between 1964 and 1969, has spread from Portugal and Spain to Belgium and France since the beginning of the 1990s. 27,29 Regarding the MLST profile of the MRSA reference strain COL ( ), it represents a single locus variant of the MLST profile of the Iberian clone, which was also identified in several Iberian MRSA isolates such as the Spanish isolate PER34 and several MRSA strains isolated in the UK during the years ,28,30 However, the allelic profile of strain COL has also been identified in several S. aureus strains susceptible to methicillin (MSSA), whereas the allelic profile of our clinical isolates has only been found in MRSA. 26 The MLST method allowed us to confirm unambiguously what has been suspected on the basis of PFGE results before, 29 namely the spread of the Iberian clone HPV107 in France, and our study demonstrates that strains belonging to this clone became less susceptible to teicoplanin. Table 4. PFGE and MLST types of outbreak isolates, historical isolates and MRSA reference strain COL Year Isolate PFGE type MLST type A A A A A A PAG A VIL A BEC A SOP A PEY A LEV A COL A References 1. Waldvogel, F. A. (1999). New resistance in Staphylococcus aureus. New England Journal of Medicine 340, Kaatz, G. W., Seo, S. M., Dorman, N. J. & Lerner, S. A. (1990). Emergence of teicoplanin resistance during therapy of Staphylococcus aureus endocarditis. Journal of Infectious Diseases 162, Mainardi, J. L., Shlaes, D. M., Goering, R. V., Shlaes, J. H., Acar, J. D. & Goldstein, F. W. (1995). Decreased teicoplanin susceptibility of methicillin-resistant strains of Staphylococcus aureus. Journal of Infectious Diseases 171, Hiramatsu, K., Hanaki, H., Ino, T., Yabuta, K., Ogury, T. & Tenover, F. C. (1997). Methicillin-resistant Staphylococcus aureus clinical strain with reduced vancomycin susceptibility. Journal of Antimicrobial Chemotherapy 40, Centers for Disease Control and Prevention. (1997). Staphyloccus aureus with reduced susceptibility to vancomycin United States, Morbidity and Mortality Weekly Report 46,

7 Multilocus sequence typing of MRSA 6. Sieradzki, K., Roberts, R. B., Haber, S. W. & Tomasz, A. (1999). The development of vancomycin resistance in a patient with methicillin-resistant Staphylococcus aureus infection. New England Journal of Medicine 340, Ploy, M. C., Grélaud, C., Martin, C., De Lumley, L. & Denis, F. (1998). First clinical isolate of vancomycin-intermediate Staphylococcus aureus in a French hospital. Lancet 351, Tabaqchali, S. (1997). Vancomycin-resistant Staphylococcus aureus: apocalypse now? Lancet 350, Centers for Disease Control and Prevention. (1997). Interim guidelines for prevention and control of staphylococcal infection associated with reduced susceptibility to vancomycin. Morbidity and Mortality Weekly Report 46, Tenover, F. C., Lancaster, M. V., Hill, B. C., Steward, C. D., Stocker, S. A., Hancock, G. A. et al. (1998). Characterization of staphylococci with reduced susceptibilities to vancomycin and other glycopeptides. Journal of Clinical Microbiology 36, Aucken, H. M., Warner, M., Ganner, M., Johnson, A. P., Richardson, J. F., Cookson, B. D. et al. (2000). Twenty months of screening for glycopeptide-intermediate Staphylococcus aureus. Journal of Antimicrobial Chemotherapy 46, Chesneau, O., Morvan, A. & El Solh, N. (2000). Retrospective screening for heterogeneous vancomycin resistance in diverse Staphylococcus aureus clones disseminated in French hospitals. Journal of Antimicrobial Chemotherapy 45, Lelievre, H., Lina, G., Jones, M. E., Olive, C., Forey, F., Roussel- Delvallez, M. et al. (1999). Emergence and spread in French hospitals of methicillin-resistant Staphylococcus aureus with increasing susceptibility to gentamicin and other antibiotics. Journal of Clinical Microbiology 37, Aires de Sousa, M., Miragaia, M., Santos Sanches, I., Avila, S., Adamson, I., Casagrande, S. T. et al. (2001). Three-year assessment of methicillin-resistant Staphylococcus aureus clones in Latin America from Journal of Clinical Microbiology 39, Blanc, D. S., Francioli, P., Le Coustumier, A., Gazagne, L., Lecaillon, E., Gueudet, P. et al. (2001). Reemergence of gentamicinsusceptible strains of methicillin-resistant Staphylococcus aureus in France: a phylogenetic approach. Journal of Clinical Microbiology 39, Salmenlinna, S. & Vuopio-Varkila, J. (2001). Recognition of two groups of methicillin-resistant Staphylococcus aureus strains based on epidemiology, antimicrobial susceptibility, hypervariable region type, and ribotype in Finland. Journal of Clinical Microbiology 39, Enright, M. C., Day, N. P. J., Davies, C. E., Peacock, S. J. & Spratt, B. G. (2000). Multilocus sequence typing for characterization of methicillin-resistant and methicillin-susceptible clones of Staphylococcus aureus. Journal of Clinical Microbiology 38, Day, N. P. J., Moore, C. E., Enright, M. C., Berendt, A. R., Maynard Smith, J., Murphy, M. F. et al. (2001). A link between virulence and ecological abundance in natural populations of Staphylococcus aureus. Science 292, Tenover, F. C., Arbeit, R. D., Goering, R. V., Mickelsen, P. A., Murray, B. E., Persing, D. H. et al. (1995). Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. Journal of Clinical Microbiology 33, Guerin, F., Buu-Hoï, A., Mainardi, J. L., Kac, G., Colardelle, N., Vaupré, S. et al. (2000). Outbreak of methicillin-resistant Staphylococcus aureus with reduced susceptibility to glycopeptides in a Parisian hospital. Journal of Clinical Microbiology 38, Chung, M., de Lencastre, H., Matthews, P., Tomasz, A., Adamsson, I., Aries de Sousa, M. et al. (2000). Molecular typing of methicillin-resistant Staphylococcus aureus by pulsed-field gel electrophoresis: comparison of results obtained in a multilaboratory effort using identical protocols and MRSA strains. Microbial Drug Resistance 6, de Lencastre, H. & Tomasz, A. (1994). Reassessment of the number of auxiliary genes essential for expression of high-level methicillin resistance in Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 38, Soussy, C. J., Carret, G., Cavallo, J. D., Chardon, H., Chidiac, C., Choutet, P. et al. (2000). Antibiogram Committee of the French Microbiology Society. Report Pathologie Biologie 48, 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, Villanova, PA, USA. 25. van Leeuwen, W., van Belkum, A., Kreiswirth, B. N. & Verbrugh, H. (1998). Genetic diversification of methicillin-resistant Staphylococcus aureus as a function of prolonged geographic dissemination and as measured by binary typing and other genotyping methods. Research in Microbiology 149, Crisostomo, M. I., Westh, H., Tomasz, A., Chung, M., Oliveira, D. C. & de Lencastre, H. (2001). The evolution of methicillin resistance in Staphylococcus aureus: similarity of genetic backgrounds in historically early methicillin-susceptible and -resistant isolates and contemporary epidemic clones. Proceedings of the National Academy of Sciences, USA 98, Santos Sanches, I., Ramirez, M., Troni, H., Abecassis, M., Padua, M., Tomasz, A. et al. (1995). Evidence for the geographic spread of a methicillin-resistant Staphylococcus aureus clone between Portugal and Spain. Journal of Clinical Microbiology 33, Oliveira, D. C., Tomasz, A. & De Lencastre, H. (2001). The evolution of pandemic clones of methicillin-resistant Staphylococcus aureus: identification of two ancestral genetic backgrounds and the associated mec elements. Microbial Drug Resistance 7, Deplano, A., Witte, W., van Leeuwen, W. J., Brun, Y. & Struelens, M. J. (2000). Clonal dissemination of epidemic methicillinresistant Staphylococcus aureus in Belgium and neighboring countries. Clinical Microbiology and Infection 6, Dominguez, M. A., de Lencastre, H., Linares, J. & Tomasz, A. (1994). Spread and maintenance of a dominant methicillin-resistant Staphylococcus aureus (MRSA) clone during an outbreak of MRSA disease in a spanish hospital. Journal of Clinical Microbiology 32,

8