A new phenotype of resistance to lincosamide and streptogramin A-type antibiotics in Streptococcus agalactiae in New Zealand

Transcription

1 Journal of Antimicrobial Chemotherapy (2004) 54, DOI: /jac/dkh493 Advance Access publication 10 November 2004 A new phenotype of resistance to lincosamide and streptogramin A-type antibiotics in Streptococcus agalactiae in New Zealand Brigitte Malbruny 1, Anja M. Werno 2, Trevor P. Anderson 2, David R. Murdoch 2,3 and Roland Leclercq 1 * JAC 1 Service de Microbiologie, CHU Côte de Nacre, Caen, France; 2 Microbiology Unit, Canterbury Health Laboratories, Christchurch; 3 Department of Pathology, Christchurch School of Medicine and Health Sciences, Christchurch, New Zealand Received 29 June 2004; returned 10 August 2004; revised 1 October 2004; accepted 5 October 2004 Objectives: To characterize a new type of resistance to clindamycin in Streptococcus agalactiae. Methods: Nineteen erythromycin-susceptible, clindamycin-resistant S. agalactiae isolates from New Zealand were studied. MICs of macrolide, lincosamide and streptogramin antibiotics were determined. Clindamycin and streptogramin resistance genes were searched for by PCR. Isolates were compared by serotyping and by DNA macrorestriction patterns determined by PFGE. Conjugative transfer of resistance traits to recipient strains of S. agalactiae and Enterococcus faecium was assayed. Results: The 19 S. agalactiae isolates were intermediate or resistant to clindamycin (MIC range: mg/l) and lincomycin (MIC range: 1 8 mg/l) and had high MICs of dalfopristin (4 32 mg/l), a streptogramin A-type antibiotic, compared with controls. By contrast, the strains were susceptible to macrolides and quinupristin, a streptogramin B-type antibiotic. This new phenotype was called LSA (lincosamide streptogramin A). Clindamycin resistance could not be transferred to recipient strains. Thirteen isolates belonged to serotype III and to a single PFGE genotype A, and five isolates belonged to serotype I and to genotype B. One isolate was non-typeable and belonged to a distinct genotype C. Conclusions: We have characterized a new LSA phenotype in S. agalactiae. Analysis of restriction patterns of S. agalactiae chromosomal DNA showed that the resistance was spread in a minimum of three bacterial clones. The genetic and biochemical basis for the resistance remains unknown. Keywords: streptococci, resistance mechanism, inhibition of protein synthesis, drug resistance, group B streptococcus Introduction Streptococcus agalactiae (group B streptococcus) is part of the commensal flora of the genital and digestive tracts. This bacterial species is also a common cause of severe infections in neonates and can cause various infections in adults, including bacteraemia and endocarditis. Penicillin G and ampicillin are the drugs of choice for prevention or treatment of S. agalactiae infections, whereas erythromycin or clindamycin are the recommended alternatives for patients who are intolerant to b-lactams. Whereas clinical isolates of S. agalactiae usually remain susceptible to penicillins, resistance to erythromycin has increased during the last decade in several countries. 1 5 The presence of mef(a) genes or of erm(b) or erm(tr) [subset of the erm(a) gene class] genes generally accounts for erythromycin resistance in clinical isolates. 3,4,6 The mef(a) gene confers resistance by drug efflux to 14- and 15-membered ring macrolides only, whereas the erm genes confer cross-resistance to erythromycin and clindamycin by ribosomal modification (macrolide lincosamide streptogramin B phenotype). Therefore, resistance to macrolides and related antibiotics in S. agalactiae usually presents as resistance to erythromycin alone or to both erythromycin and clindamycin. The observation that in New Zealand, among 117 S. agalactiae isolates, the most common resistance phenotype (11% of the strains) combined a low-level resistance to clindamycin (MIC range: 1 to 4 mg/l) with susceptibility to erythromycin (MIC < 0.25 mg/l) prompted us to study further this unusual type of resistance *Corresponding author. Tel: ; Fax: ; leclercq-r@chu-caen.fr JAC vol.54 no.6 q The British Society for Antimicrobial Chemotherapy 2004; all rights reserved.

2 Clindamycin resistance in Streptococcus agalactiae Materials and methods Bacterial strains Nineteen S. agalactiae susceptible to erythromycin but resistant to clindamycin were obtained from vaginal swabs (n = 9), urine (n =2) and various other sites (n = 8). Eighteen were isolated from women (age: years) and one from a newborn. The isolates were identified by conventional methods and by latex agglutination assay (Murex Diagnostics, Dartford, UK). Serotyping of isolates was determined by using a latex agglutination assay (Bio-Rad, Marnesla-Coquette, France). Antibiotic susceptibility The MICs of ampicillin, clindamycin, dalfopristin, erythromycin, lincomycin, quinupristin, quinupristin dalfopristin and doxycycline were determined by the microbroth dilution method with Mueller Hinton broth (biomérieux) supplemented with 5% defibrinated sheep blood, as recommended by the NCCLS (tested range: mg/l). 8 All antibiotics were provided by their manufacturers. The plates were incubated overnight at 358C in ambient air. MICs of clindamycin were also determined by the Etest technique, as recommended by the manufacturer (AB Biodisk, Uppsala, Sweden). MICs were also determined in the presence and absence of 10 mg/l of reserpine (Sigma Chemicals, St Louis, MO, USA), as described previously. 9 An efflux mechanism could be suspected when there was at least a four-fold lower MIC in the presence of reserpine. A double disc-diffusion test with discs of erythromycin and clindamycin was used to test inducibility of resistance. 10 Blunting of the clindamycin zone of inhibition proximal to the erythromycin disc was taken to indicate an inducible type of resistance. Inactivation of antimicrobials Inactivation of clindamycin, lincomycin or dalfopristin was checked using a previously described technique. 11 Briefly, strains were streaked on brain heart infusion agar plates containing a heavy inoculum of Micrococcus luteus ATCC 9341 as an indicator organism, and 0.5 mg/l of dalfopristin or 0.2 mg/l of clindamycin, a concentration slightly higher than the MIC for M. luteus. Inactivation of the antibiotic in the culture medium would allow growth of the indicator in the surrounding medium. Staphylococcus aureus RN4220/pVMM26 inactivating lincomycin and clindamycin and S. aureus BM3002 inactivating dalfopristin were used as controls. 11,12 Molecular techniques PFGE was used to compare the isolates, as previously described. 13 Genomic DNA was isolated from an overnight-grown culture and the agarose-embedded DNA was digested with the enzyme SmaI (New England BioLabs Inc., Beverly, MA, USA). PFGE was performed using the CHEF-DR-III system (Bio-Rad). Gels were run at 6 V/cm, 148C, on a 1.2% agarose gel with pulse times of 5 35 s for 24 h. The bacteriophage l DNA ladder (New England BioLabs Inc.) was used as a size standard. The PFGE banding patterns were compared visually and interpretation of gels was performed using the criteria of Tenover et al. 14 Strains were considered genetically distinct if their restriction patterns differed by three or more bands. The lincomycin-resistant isolates were screened for lincosamide and streptogramin A-type resistance genes. The erm(a), erm(tr), erm(b) and erm(c) genes encoding ribosomal methylases and the lnu(a) and lnu(b) genes (previously called lina and linb) encoding lincosamide nucleotidyltransferases were detected by PCR amplification, as described previously. 12,15 The primers used to detect dalfopristin resistance genes were those described previously: vat(a), 16 vat(b), 17 vat(c), 18 vat(e), 19 vga(a), 20 vga(b) 21 and vga(av). 22 S. aureus strain BM3002 for vga(a) and vat(a), 11 BM3318 for vga(av), 22 HM290 for erm(a), HM1054R for erm(c), BM4611 for lnu(a ), 23 RN4220/pVMM26 for lnu(b), 12 Staphylococcus haemolyticus BM4610 for lnu(a), 23 E. faecium HM1032 for erm(b) and vat(d), and E. faecium UW 1965 for vat(e) 19 were used as controls. To detect mutations in the ribosomal target of lincosamides, we amplified a portion of the rrl gene for domain II from nt (Escherichia coli numbering), four fragments of domain V of 23S rrna (nt 1601 to 2900) and the entire rplv gene (for L22 ribosomal protein) using primers previously described for Streptococcus pyogenes and Streptococcus pneumoniae. 24,25 For amplification of the entire rpld gene (L4 ribosomal protein), primers 5 0 -ggtaacgtaccaggtgctaag (direct) and 5 0 -gatttcaacgcaaggcgacg (reverse), and primers 5 0 -ggtggtggtgttgtctttgg (direct) and 5 0 -gcacgtgtgtcaagttcaaatg (reverse) were used. The amplicons were analysed by single-strand conformation polymorphism (SSCP) analysis, as described previously. 25 Plasmids were extracted from streptococcal strains, as described previously by Ehrenfeld & Clewell. 26 Enterococcus faecalis JH2-2 containing plasmid pad1 was used as a control. 26 Filter mating Transfer of lincomycin and dalfopristin resistance from strains of S. agalactiae to the recipient strains of E. faecium HM1070 and S. agalactiae 132 (both susceptible to lincosamides and streptogramins A-type and resistant to rifampicin and fusidic acid) was attempted by filter mating, as previously described. 12 Transconjugants were selected on brain heart infusion containing rifampicin (20 mg/l), fusidic acid (10 mg/l) and lincomycin (2 mg/l) or dalfopristin (10 mg/l). Experiments were repeated independently three times. Results Characterization of clindamycin and dalfopristin resistant strains MICs of macrolide, lincosamide and streptogramin antibiotics for the 19 strains and for controls are shown in Table 1 and Figures 1 and 2. All strains were susceptible to ampicillin and resistant to doxycycline (not shown). Compared with susceptible controls, MICs of clindamycin, lincomycin and dalfopristin were increased by a two- to four-fold dilution factor. This shift in MICs of lincosamides and dalfopristin contrasted with maintained susceptibility to 14-, 15- and 16-membered macrolides, telithromycin and quinupristin. The synergism between quinupristin (a streptogramin B) and dalfopristin (a streptogramin A) was retained with no significant increase in MICs. The doubledisc diffusion test did not demonstrate inducible resistance to erythromycin. The MICs of lincosamides and dalfopristin were unchanged in the presence of the efflux pump inhibitor, reserpine. To the best of our knowledge, this phenotype is new in S. agalactiae and is called LSA (lincosamide streptogramin A). Thirteen LSA isolates belonged to serotype III, five to serotype I and one was non-typeable. The SmaI macrorestriction of genomic DNA generated about fragments of different sizes. Analysis of macrorestriction patterns led to the identification of three different genotypes. All serotype III isolates had DNA profiles that were indistinguishable (five isolates) or differing by 1041

3 B. Malbruny et al. Table 1. MICs of antibiotics for clinical isolates of S. agalactiae susceptible to macrolide, lincosamide and streptogramin antibiotics, with the LSA phenotype and for the susceptible control S. agalactiae 132 MIC range (mg/l) S. agalactiae (no. of strains) AMP ERY AZM SPI TEL LIN CLI CLI (Etest) QUI DAL Q/D Susceptible (10) ND S. agalactiae With LSA phenotype (19) AMP, ampicillin; AZM, azithromycin; CLI, clindamycin; DAL, dalfopristin; ERY, erythromycin; LIN, lincomycin; QUI, quinupristin; Q/D, quinupristin/dalfopristin; SPI, spiramycin; TEL, telithromycin; ND, not determined. Figure 1. Distribution of MICs of clindamycin for S. agalactiae isolates susceptible to clindamycin (white bars) and with the LSA phenotype (black bars). Figure 2. Distribution of MICs of dalfopristin for S. agalactiae isolates susceptible to dalfopristin (white bars) and with the LSA phenotype (black bars). Figure 3. PFGE patterns for some of the S. agalactiae isolates with the LSA phenotype. Total DNA from 13 S. agalactiae isolates (1 13) was digested with the SmaI restriction enzyme and submitted to PFGE. The characterized genotypes are indicated in parentheses. one or two bands (eight isolates) and were therefore clustered in a single genotype A (Figure 3). The five serotype I isolates formed a single genotypic cluster with indistinguishable profiles, whereas the non-typeable isolate displayed a distinct genotype C. Therefore, the spread of resistance appeared to be multiclonal. 1042

4 Clindamycin resistance in Streptococcus agalactiae Mechanism of clindamycin resistance To determine the resistance genotype of strains, we used primers specific for genes encoding resistance to lincosamides or to streptogramin A-type antibiotics in clinical isolates. Although these genes confer resistance to lincosamides or streptogramins A alone, and have not been previously reported to confer an LSA phenotype, a combination was possible. All PCR controls gave the expected results. However, no PCR product was obtained for the study isolates. Genes that encode ribosomal structures composing the target of lincosamides, rrl, rpld and rplv genes, from two strains with serotype I and III were analysed by PCR-SSCP. Compared with the sequences of S. agalactiae 2603 V/R and S. agalactiae NEM316 obtained from The Institute for Genomic Research website at and from The Institut Pasteur de Paris at recherche/unites/gmp/, respectively, 27,28 no mutation was found. Four strains of S. agalactiae of serotype I and III were tested for the transfer of clindamycin- or dalfopristin-resistance trait. No transfer to S. agalactiae 132 or to E. faecium HM1070 could be detected. No plasmid could be visualized after DNA extraction from the cells. Discussion S. agalactiae isolates intermediate or resistant to clindamycin but susceptible to erythromycin are already widespread in New Zealand. 7,29 Other reports from Taiwan show frequencies of clindamycin resistance greater than for erythromycin, suggesting that isolates similarly resistant may also be present in Asia. 1,2 In Canada, a single clindamycin-resistant and erythromycin-susceptible strain has been reported, which contained a lnu(b) [lin(b)] gene responsible for lincosamide nucleotidylation. 3 Although dalfopristin was not tested in the Canadian study, the lnu(b) gene has been shown in E. faecium to confer resistance to lincosamides only. 12 In contrast, the strains from New Zealand were co-resistant to lincosamide and streptogramin A-type antibiotics. The biochemical and genetic basis for this new LSA phenotype of resistance remains obscure. No known genes of resistance to lincosamide or streptogramin A-type antibiotics were found by PCR; no inactivation of antibiotics was detected by the microbiological screening test that we used and MICs of clindamycin and dalfopristin were unaffected by the pump inhibitor, reserpine. However, this latter result does not exclude an efflux mechanism. 30,31 The presence of the LSA phenotype in three S. agalactiae clones suggested a horizontal transfer of the resistance, although no plasmid could be visualized from the isolates and no conjugative transfer of resistance traits could be detected. The LSA phenotype of S. agalactiae is reminiscent of similar LSA phenotypes reported as intrinsic in E. faecalis 32,33 and acquired in S. aureus 34,35 In E. faecalis, the LSA intrinsic resistance is related to the expression of an Lsa protein that is similar to members of a superfamily of transport-related proteins known as ABC transporters capable of transporting molecules in response to ATP hydrolysis. 32,33 However, the efflux mechanism has not been proven. An outbreak with 27 isolates of methicillin-resistant S. aureus, displaying a similar LSA phenotype, has been reported in a French hospital. 34 However, the mechanism of resistance has not been further studied. Clindamycin resistance may be misidentified in strains with the LSA phenotype if only erythromycin is tested. This may be of clinical importance in countries where clindamycin is recommended as an ampicillin substitute for the treatment or prevention of S. agalactiae infections in case of allergy. The emergence of LSA resistance justifies the in vitro test of clindamycin, in particular in countries where this phenotype is prevalent. Acknowledgements We thank Patrick Trieu-Cuot for the gift of S. agalactiae 132, Wolfgang Witte for the gift of E. faecium UW 1965 and Nevine El Sohl for the gift of S. aureus BM3318. References 1. Wu, J.-J., Lin, K.-Y., Hsueh, P.-R. et al. (1997). High incidence of erythromycin-resistant streptococci in Taiwan. Antimicrobial Agents and Chemotherapy 41, Ko, W. C., Lee, H. C., Wang, L. R. et al. (2001). Serotyping and antimicrobial susceptibility of group B Streptococcus over an eight-year period in southern Taiwan. European Journal of Clinical Microbiology and Infectious Diseases 20, de Azavedo, J. C., McGavin, M., Duncan, C. et al. (2001). Prevalence and mechanisms of macrolide resistance in invasive and noninvasive group B Streptococcus isolates from Ontario, Canada. Antimicrobial Agents and Chemotherapy 45, Betriu, C., Culebras, E., Gómez, M. et al. (2003). Erythromycin and clindamycin resistance and telithromycin susceptibility in Streptococcus agalactiae. Antimicrobial Agents and Chemotherapy 47, Biedenbach, D. J., Stephen, J. M. & Jones, R. N. (2003). Antimicrobial susceptibility profile among b-haemolytic Streptococcus spp. collected in the SENTRY Antimicrobial Surveillance Program- North America, Diagnostic Microbiology and Infectious Disease 46, Poyart, C., Jardy, L., Quesne, G. et al. (2003). Genetic basis of antibiotic resistance in Streptococcus agalactiae strains isolated in a French hospital. Antimicrobial Agents and Chemotherapy 47, Werno, A. M., Anderson, T. P. & Murdoch, D. R. (2003). Antimicrobial susceptibilities of group B streptococci in New Zealand. Antimicrobial Agents and Chemotherapy 47, National Committee for Clinical Laboratory Standards. (2000). Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow Aerobically Fifth Edition: Approved Standard M7-A5. NCCLS, Wayne, PA, USA. 9. Brenwald, N. P., Gill, M. J. & Wise, R. (1998). Prevalence of a putative efflux mechanism among fluoroquinolone-resistant clinical isolates of Streptococcus pneumoniae. Antimicrobial Agents and Chemotherapy 42, Seppälä, H., Nissinen, A., Yu, Q. et al. (1993). Three different phenotypes of erythromycin-resistant Streptococcus pyogenes in Finland. Journal of Antimicrobial Chemotherapy 32, Bozdogan, B. & Leclercq, R. (1999). Effects of genes encoding resistance to streptogramins A and B on the activity of quinupristindalfopristin against Enterococcus faecium. Antimicrobial Agents and Chemotherapy 43, Bozdogan, B., Berrezouga, L., Kuo, M. S. et al. (1999). A new resistance gene, linb, conferring resistance to lincosamides by nucleotidylation in Enterococcus faecium HM1025. Antimicrobial Agents and Chemotherapy 43, Gordillo, M. E., Singh, K. V., Baker, C. J. et al. (1993). Typing of group B streptococci: comparison of pulsed-field gel electrophoresis and conventional electrophoresis. Journal of Clinical Microbiology 31, Tenover, F. C., Arbeit, R. D., Goering, R. V. et al. (1995). Interpreting chromosomal DNA restriction patterns produced by 1043

5 B. Malbruny et al. pulsed-field gel electrophoresis: criteria for bacterial strain typing. Journal of Clinical Microbiology 33, Angot, P., Vergnaud, M., Auzou, M. et al. (2000). Macrolide resistance phenotypes and genotypes in French clinical isolates of Streptococcus pneumoniae. European Journal of Clinical Microbiology and Infectious Diseases 19, Allignet, J., Loncle, V., Simenel, C. et al. (1993). Sequence of a staphylococcal gene, vat, encoding an acetyltransferase inactivating the A-type compounds of virginiamycin-like antibiotics. Gene 130, Allignet, J. & El Solh, N. (1995). Diversity among the grampositive acetyltransferases inactivating streptogramin A and structurally related compounds and characterization of a new staphylococcal determinant, vatb. Antimicrobial Agents and Chemotherapy 39, Allignet, J., Liassine, N. & El Solh, N. (1998). Characterization of a staphylococcal plasmid related to pub110 and carrying two novel genes, vatc and vgbb, encoding resistance to streptogramins A and B and similar antibiotics. Antimicrobial Agents and Chemotherapy 42, Werner, G. & Witte, W. (1999). Characterization of a new enterococcal gene, satg, encoding a putative acetyltransferase conferring resistance to streptogramin A compounds. Antimicrobial Agents and Chemotherapy 43, Allignet, J., Loncle, V. & El Solh, N. (1992). Sequence of a staphylococcal plasmid gene, vga, encoding a putative ATP-binding protein involved in resistance to virginiamycin A-like antibiotics. Gene 117, Allignet, J. & El Solh, N. (1997). Characterization of a new staphylococcal gene, vgab, encoding a putative ABC transporter conferring resistance to streptogramin A and related compounds. Gene 202, Haroche, J., Allignet, J., Buchrieser, C. et al. (2000). Characterization of a variant of vga(a) conferring resistance to streptogramin A and related compounds. Antimicrobial Agents and Chemotherapy 44, Leclercq, R., Brisson-Noël, A., Duval, J. et al. (1987). Phenotypic expression and genetic heterogeneity of lincosamide inactivation in Staphylococcus spp. Antimicrobial Agents and Chemotherapy 31, Bingen, E., Leclercq, R., Fitoussi, F. et al. (2002). Emergence of group A streptococcus strains with different mechanisms of macrolide resistance. Antimicrobial Agents and Chemotherapy 46, Canu, A., Malbruny, B., Coquemont, M. et al. (2002). Diversity of ribosomal mutations conferring resistance to macrolides, clindamycin, streptogramin, and telithromycin in Streptococcus pneumoniae. Antimicrobial Agents and Chemotherapy 46, Ehrenfeld, E. E. & Clewell, D. B. (1987). Transfer functions of the Streptococcus faecalis plasmid pad1: organization of plasmid DNA encoding response to sex pheromone. Journal of Bacteriology 169, Tettelin, H., Masignani, V., Cieslewicz, M. J. et al. (2002). Complete genome sequence and comparative genomic analysis of an emerging human pathogen, serotype V Streptococcus agalactiae. Proceedings of the National Academy of Sciences, USA 17, Glaser, P., Rusniok, C., Buchrieser, C. et al. (2002). Genome sequence of Streptococcus agalactiae, a pathogen causing invasive neonatal disease. Molecular Microbiology 45, Grimwood, K., Stone, P. R., Gosling, I. A. et al. (2002). Late antenatal carriage of group B Streptococcus by New Zealand women. Australian & New Zealand Journal of Obstetrics & Gynaecology 42, Jones, H. E., Brenwald, N. P., Owen, K. A. et al. (2003). A multidrug efflux phenotype mutant of Streptococcus pyogenes. Jounal of Antimicrobial Chemotherapy 51, Ahmed, M., Borsch, C. M., Neyfakh, A. A. et al. (1993). Mutants of the Bacillus subtilis multidrug transporter Bmr with altered sensitivity to the antihypertensive alkaloid reserpine. Journal of Biological Chemistry 268, Singh, K. V., Weinstock, G. M. & Murray, B. E. (2002). An Enterococcus faecalis ABC homologue (Lsa) is required for the resistance of this species to clindamycin and quinupristin-dalfopristin. Antimicrobial Agents and Chemotherapy 46, Dina, J., Malbruny, B. & Leclercq, R. (2003). Nonsense mutations in the lsa-like gene in Enterococcus faecalis isolates susceptible to lincosamides and streptogramins A. Antimicrobial Agents and Chemotherapy 47, Arpin, C., Lagrange, I., Gachie, J. P. et al. (1996). Epidemiological study of an outbreak of infection with Staphylococcus aureus resistant to lincosamides and streptogramin A in a French hospital. Journal of Medical Microbiology 44, Zarrouk, V., Bozdogan, B., Leclercq, R. et al. (2000). Influence of resistance to streptogramin A type antibiotics on the activity of quinupristin-dalfopristin in vitro and in experimental endocarditis due to Staphylococcus aureus. Antimicrobial Agents and Chemotherapy 44,