Genetic Characterization of Multidrug-Resistant Mycobacterium bovis ...

JOURNAL OF CLINICAL MICROBIOLOGY, June 1997, p. 1390–1393 0095-1137/97/$04.0010 Copyright © 1997, American Society for Microbiology

Vol. 35, No. 6

Genetic Characterization of Multidrug-Resistant Mycobacterium bovis Strains from a Hospital Outbreak Involving Human Immunodeficiency Virus-Positive Patients ´ S BLA ´ ZQUEZ,1* LUZ ELENA ESPINOSA DE LOS MONTEROS,2 SOFIA SAMPER,3 CARLOS MARTIŃ,3 JESU ANTONIO GUERRERO,1 JAVIER COBO,1 JAN VAN EMBDEN,4 FERNANDO BAQUERO,1 ´ MEZ-MAMPASO1 AND ENRIQUE GO Servicio de Microbiologıá, Hospital Ramo ń y Cajal, 28034 Madrid,1 and Departamento de Microbiologıá y Medicina Preventiva, Facultad de Medicina, Universidad de Zaragoza, Zaragoza,3 Spain; Servicio de Microbiologıá, Hospital Infantil de Me´xico Federico Go ´mez, Me´xico 06720 D.F., Me´xico2; and Department of Bacteriology, Research Laboratory for Infectious Diseases, National Institute of Public Health and Environmental Protection, 3720 BA Bilthoven, The Netherlands4 Received 6 November 1996/Accepted 28 February 1997

Nineteen multidrug-resistant (MDR) Mycobacterium complex strains isolated in a nosocomial outbreak were characterized at the molecular level. The strains were microbiologically characterized as Mycobacterium bovis. The mpt40 sequence was not present in chromosomal DNA from these strains, supporting the fact that they were M. bovis. All of the isolates were resistant to isoniazid, rifampin, pyrazinamide, ethambutol, streptomycin, para-aminosalicylic acid, clarithromycin, cycloserine, ethionamide, ofloxacin, capreomycin, and amikacin. By performing the standardized IS6110 fingerprinting by restriction fragment length polymorphism (RFLP) analysis, we were able to differentiate two groups (groups A and B) containing two (16 isolates) and three (3 isolates) IS6110 copies, respectively. These strains were typed by spoligotyping, developed to distinguish M. bovis strains and also to distinguish them from M. tuberculosis strains (J. Kamerbeek et al., J. Clin. Microbiol. 35:907–914, 1997). All the strains were confirmed to be M. bovis. In addition, spoligotyping showed a difference in only 1 of 43 spacers between RFLP groups A and B. The rpob region of several strains representative of each identified group was cloned and sequenced, and identical mutations (Ser-531 to Leu) responsible for the rifampin resistance phenotype were found. To our knowledge, this is the first characterization at the molecular level of an MDR M. bovis strain responsible for a nosocomial outbreak. to type M. tuberculosis complex strains, most molecular methods are based on the detection of repetitive elements. The repetitive elements described so far in M. tuberculosis are the insertion sequences IS6110 and IS1081, the major polymorphic tandem repeats, the polymorphic GC-rich repetitive sequences, and the direct repeat sequence (9, 18). At present the most extensively used method for differentiating M. tuberculosis is the internationally standardized Southern blotting-based technique consisting of restriction fragment length polymorphism (RFLP) analysis with insertion sequence IS6110 as the probe (17). Several epidemics caused by multidrug-resistant (MDR) M. tuberculosis strains have recently been detected; however, only some epidemics due to MDR strains have been properly documented from a genetic point of view (1). Mycobacterium bovis is the main organism responsible for tuberculosis in cattle, but its involvement as a causative agent of tuberculosis in humans was recognized a long time ago. Nevertheless, primary human disease due to M. bovis is very rare in developed countries, although a reactivated form of the disease is still encountered. Immunological factors preventing progression of infection caused by M. bovis, whether of animal or human origin, to overt human disease could well be suppressed by coinfection with HIV. Cases of HIV-related human tuberculosis due to M. bovis have been reported in England, France, and the United States (21). Human-to-human transmission of disease due to M. bovis in HIV-positive persons has recently been confirmed. A strain of M. bovis resistant to many antituberculosis drugs was the source of infection in five patients in a Paris hospital (2), but this nosocomial outbreak of MDR M. bovis was not characterized at the molecular level.

The line of progressive decrease in the cases of human tuberculosis in developed countries unexpectedly stopped and then moved upward. During the last few years we are witnessing an impressive increase in the number of cases of human tuberculosis, particularly among human immunodeficiency virus (HIV)-positive individuals (14). The situation may be dramatic in those countries such as Spain, where high rates of HIV infection are superimposed on a high prevalence of tuberculosis. As expected, the resulting high number cases of tuberculosis in HIV-positive patients frequently occurs in a disadvantaged population in which the compliance with the scheduled standard antituberculosis treatments is very low. The consequence is the progressive selection of multiresistant Mycobacterium strains by uncontrolled treatments. Worth consideration is whether multiresistant strains produce a lower rate of infection compared with that produced by wild-type (not multiresistant) Mycobacterium tuberculosis strains when they are acquired by immunocompetent people. Nevertheless, organisms that acquire high transmissibility may evolve to higher pathogenicity. Studies on the epidemiology of tuberculosis are therefore crucial in a social landscape where a high number of immunocompromised patients may come into contact with a high number of potentially susceptible hosts. Epidemiological studies of tuberculosis are greatly facilitated by the application of strain-specific markers (13). In order * Corresponding author. Mailing address: Servicio de Microbiologıá, Hospital Ramo ń y Cajal, Ctra. Colmenar Km 9.100, 28034 Madrid, Spain. Phone: 34 1 336 83 30. Fax: 34 1 336 88 09. E-mail: jesus [email protected]. 1390

GENETIC CHARACTERIZATION OF MDR M. BOVIS OUTBREAK

VOL. 35, 1997

Techniques other than RFLP for differentiating M. bovis strains with low IS6110 copy numbers have been based on polymorphic GC-rich repetitive sequence or direct repeat polymorphism (6). Recently, we have communicated the occurrence in our hospital of a presumptive M. tuberculosis complex outbreak. From December 1993 through February 1995, 19 cases of primary MDR tuberculosis in HIV-infected patients (16 males and 3 females; mean age, 31 6 6.5 years; age range, 24 to 54 years) were detected. In all patients the strains were resistant to all the drugs tested (5). The index patient and two other patients had previously attended another hospital in Madrid that had an M. tuberculosis MDR epidemic. The other 16 patients were exposed in our infectious diseases ward to one of these three patients. All patients died, despite treatment with multiple first- and second-line antituberculosis drugs. Standard microbiological identification techniques demonstrated that all strains were M. bovis. The epidemic was controlled after implementation of a control policy. In this report we present data resulting from the application of different molecular biology-based methods for characterization of the strains isolated in the cited outbreak. Strains were typed by the standardized method for M. tuberculosis strains, RFLP analysis with IS6110, and by spoligotyping, a more suitable procedure for typing and identifying M. bovis strains (4, 8). We cloned and sequenced the rpob region responsible for the rifampin resistance phenotype for several representative strains of each identified MDR M. bovis group. The spoligotype patterns were almost identical between the groups, suggesting that the strains that we studied and that caused the nosocomial outbreak are members of an MDR M. bovis family of clones. (This work was presented in part at the 36th Interscience Conference on Antimicrobial Agents and Chemotherapy, New Orleans, La., 15 to 18 September 1996.) MATERIALS AND METHODS Microbiological data on the Mycobacterium strains. The Mycobacterium isolates were identified as M. tuberculosis complex strains with nucleic acid probes (GenProbe, San Diego, Calif.). All isolates had the same microbiological characteristics: they were slowly growing mycobacteria, nitrate negative, niacin negative, sensitive to thiopin-2-carboxylic acid, and resistant to pyrazinamide. Drug susceptibility testing was performed by Canetti’s proportions method on Lowenstein-Jensen and/or Middlebrook 7H10 medium with isoniazid (0.2 and 1 mg/ml), rifampin (1 mg/ml), pyrazinamide (50 mg/ml), ethambutol (5 and 10 mg/ml), streptomycin (2 and 10 mg/ml), para-aminosalicylic acid (2 and 10 mg/ml), clarithromycin (10 mg/ml), cycloserine (20 an 30 mg/ml), ethionamide (5 mg/ml), ofloxacin (2.5 mg/ml), capreomycin (10 mg/ml), and amikacin (0.5 and 1 mg/ml) (5a). Escherichia coli strains and plasmids. The strain used as the recipient of plasmids was E. coli TG1. Plasmid pGEM-T (Promega Corp., Madison, Wis.) was used for direct cloning of the PCR-amplified DNA from the M. bovis rpob gene. PCR amplifications and primers. PCR amplifications were developed under conditions previously described for each primer set (see below). (i) For detection of the IS6110 sequence, the primers used were T4 (59-CCTGCGAGCGTAGG CGTCGG-39) and T5 (59-CTCGTCCAGCGCCGCTTCGG-39) (3). (ii) For detection of the pab sequence, the primers used were MT1 (59-ACCACCGAGC GGTTCGCCTGA-39) and MT2 (59-GATCTGCGGGTCGTCCCAGGT-39) (12). (iii) For amplification of a 159-bp fragment from the Rifr region of the rpob gene, primers TBrpo8 (59-TGCACGTCGCGGACCTCCA-39) and Tbrpo9 (59TCGCCGCGATCAAGGAGT-39) were used (7). Cloning and DNA sequence analysis. Direct cloning of amplified DNA fragments into the pGEM-T vector was carried out according to the manufacturer’s instructions (Promega Corp., Madison). Hybrid plasmid DNAs prepared with the Wizard Minipreps kit (Promega Corp.) were used as templates for nucleotide sequencing. Sequencing was performed by using the pUC forward and reverse sequencing primers that initiated the dideoxy-nucleotide chain termination reaction. Molecular typing of the strains. (i) IS6110-RFLP. Extraction of mycobacterial DNA was performed as described previously (10, 17). DNA (1 to 4 mg) from each sample was digested for 1 h with PvuII for analysis with the IS6110 probe. Digests

1391

were electrophoresed in 20-cm gels of 0.8% agarose. Southern blotting and hybridization studies were performed as described previously (11). (ii) Spoligotyping. The oligonucleotides DRa (59 biotinylated) and DRb were used as primers for PCR amplification (8). Approximately 10 ng of purified mycobacterial DNA was used as the target. The DNA was resuspended in 50 ml of a reaction mixture containing Taq buffer, 200 mM (each) deoxynucleoside triphosphate, 20 pmol (each) of primers DRa and DRb, and 0.5 U of Taq polymerase (Perkin-Elmer). Thermal cycling was performed as described before (8). Twenty microliters of the reaction mixture was hybridized with 43 different spacer oligonucleotides, which were covalently linked to a filter (8).

RESULTS Characterization of M. tuberculosis complex isolates. The strains belonged to the M. tuberculosis complex, as determined in clinical samples by positive PCR amplification with both IS6110-specific (3) and pab-specific (12) primers. From the end of 1993 to 1995, 19 Mycobacterium complexpositive cultures were obtained from 19 HIV-positive patients. All isolates were identified as M. bovis by standard procedures, as described in Materials and Methods. By PCR amplification with specific primers, we showed that the mpt40 sequence was present in chromosomal DNAs from several M. tuberculosis strains but not in DNAs from the strains from the HIV-positive patients, supporting the fact that they were M. bovis. All strains were resistant to isoniazid, rifampin, pyrazinamide, ethambutol, streptomycin, para-aminosalicylic acid, clarithromycin, cycloserine, ethionamide, ofloxacin, capreomycin, and amikacin. RFLP analysis of the strains. IS6110 fingerprinting by RFLP analysis resulted in two different patterns (designated groups A and B). Two copies of IS6110 of identical size (2.45 and 1.95 kb) were found in 16 isolates, and three copies (2.45, 1.8, and 1.5 kb) were found in 3 isolates. Figure 1 presents the results of RFLP analysis for two strains representative of each pattern. Spoligotyping. The strains were studied by spoligotyping, a technique developed to distinguish among M. bovis strains and also to distinguish them from M. tuberculosis strains (8). In all cases, the results confirmed the previous identification as M. bovis. In addition, spoligotyping showed only 1 of 43 spacer differences between RFLP groups A and B (Fig. 2). Characterization of rpoB mutations responsible for rifampin resistance. Substitution of a limited number of highly conserved amino acids of the b subunit of the RNA polymerase results in high-level resistance to rifampin (15, 16). By using the oligonucleotides described previously (7), we amplified a fragment of DNA of 159 bp that contained the cited region of rpoB from two of each of the two different IS6110 clusters and the susceptible strain H37rv. The amplified fragments were ligated to plasmid pGEM-T and were introduced by transformation into E. coli TG1. Two transformant clones from each RFLP type and one from H37rv were sequenced. Analysis of the sequence showed that the same change appeared in the two clusters, Leu-531 (codon TTG), with respect to the wild type, Ser-531 (codon TCG). DISCUSSION AIDS is changing dramatically the epidemiology of tuberculosis (14). M. bovis was recognized as a relatively common cause of tuberculosis, being implicated in a number of cases, ranging from 0.1% in France to 5% in the United Kingdom (2). The classical observations supported the conclusion that transmission of M. bovis between humans was a very infrequent event (20). The application of strain-specific markers for differentiating M. tuberculosis strains is a useful tool for epidemiological studies of tuberculosis. At present the most extensively used method for differentiating among M. tuberculosis strains is the stan-

1392

´ ZQUEZ ET AL. BLA

J. CLIN. MICROBIOL.

FIG. 1. RFLPs of M. bovis DNA digested with PvuII and probed with the right arm of IS6110. Results for two strains representative of each group are shown. The letters A and B indicate the respective patterns. RFLP of M. tuberculosis 14323 is shown at the right. Arrows indicate IS6110-containing PvuII fragments of approximately 14.4, 7.2, 7.0, 4.4, 3.6, 3.0, 2.3, 2.0, 1.8, and 1.5 kb.

dardized RFLP analysis (17), but the use of alternative techniques to distinguish M. tuberculosis complex strains within the group having low IS6110 copy numbers remains necessary (19). Most M. bovis isolates carry only a single IS6110 element and are difficult to differentiate by IS6110 RFLP analysis. Spoligotyping has been a useful tool for the detection of an MDR M. bovis outbreak in our hospital. The results presented strongly suggest that the strains characterized here are members of an MDR M. bovis clone family. They share almost identical spoligotypes (only 1 difference in more than 40 positions) both for the two-band and for the three-band IS6110 RFLP clusters. Since the clones containing two bands were isolated before those harboring three bands, the transposition of IS6110 during the course of the outbreak could be responsible for the new three-band RFLP. New experiments are being developed in order to determine if both clones belong to a unique outbreak. The detection of a particular rpob mutation (Ser-531 to Leu), although identical to the mutation most frequently found in rifampin-resistant M. tuberculosis isolates (7), again supports the clonality of the isolates. The present work confirms at the molecular level that M. bovis was responsible for a hospital MDR outbreak involving transmission between AIDS patients. To our knowledge this is the first characterization at the molecular level of an MDR-M. bovis outbreak in a hospital. Spoligotyping, a PCR amplification technique, makes possible the accurate typing of mycobacteria directly from pathological material in only 2 days. In addition, this technique is able to discriminate M. bovis from M. tuberculosis. A comparison of the different spoligotypes in a database

FIG. 2. Spoligotyping of MDR M. bovis DNA. The spoligotype of one strain representative of each pattern (A and B) is shown. The spoligotypes of M. bovis BCG (Pasteur) and M. tuberculosis H37rv are also shown.

containing MDR strains from different hospitals and countries could facilitate the early detection of interhospital or intercountry outbreaks. ACKNOWLEDGMENTS We thank Annelies van Bunschoten for perfect technical assistance. The group from Zaragoza, Spain, was supported in part by grant FIS/0051 from Spanish Fondo de Investigaciones Sanitarias de la Seguridad Social and by the European Commission CA on the Epidemiology on Tuberculosis (BIOMED1, CT93-1614). The group from Bilthoven, The Netherlands, was supported in part by the Dutch Foundation of Technical Sciences. Sofia Samper was the recipient of an FIS fellowship. L. E. Espinosa de Los Monteros was the recipient of a fellowship from the Patronato del Hospital Infantil de Me´xico “Federico Go ´mez.” REFERENCES 1. Bifani, P. J., B. B. Plikaytis, V. Kapur, K. Stockbauer, X. Pan, M. L. Luftey, S. L. Moghazeh, W. Wisner, T. M. Daniel, M. H. Kaplan, J. T. Crawford, J. M. Musser, and B. N. Kreiswith. 1996. Origin and interstate spread of a New York City multidrug-resistant Mycobacterium tuberculosis clone family. JAMA 275:452–457. 2. Bouvet, E., E. Casalino, G. Mendoza-Sassi, S. Lariven, E. Valleé, M. Pernet, S. Gottot, and F. Vachon. 1993. A nosocomial outbreak of multidrug-resistant Mycobacterium bovis among HIV-infected patients. A case-control study. AIDS 7:1453–1460. 3. Eisenach, K. D., M. D. Cave, J. H. Bates, and J. T. Crawford. 1990. Polymerase chain reaction amplification of a repetitive DNA sequence specific for Mycobacterium tuberculosis. J. Infect. Dis. 161:977–981.

VOL. 35, 1997

GENETIC CHARACTERIZATION OF MDR M. BOVIS OUTBREAK

4. Groenen, P. M. A., A. E. Bunschoten, D. van Soolingen, and J. van Embden. 1993. Nature of DNA polymorphism in the direct repeat cluster of Mycobacterium tuberculosis; application for strain differentiation by a novel typing method. Mol. Microbiol. 10:1057–1065. 5. Guerrero, A., J. Cobo, J. Fortun, E. Navas, C. Quereda, J. Bla ´zquez, A. Ortega, and E. Go ´mez-Mampaso. 1995. The emergence of drug-resistant tuberculosis in a Spanish hospital, abstr. I89, p. 221. In Program and abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy. American Society for Microbiology, Washington, D.C. 5a.Guerrero, A., et al. Submitted for publication. 6. Gutierrez, M., S. Samper, J. A. Gavigan, J. F. Garcıá Marıń, and C. Martıń. 1995. Differentiation by molecular typing of Mycobacterium bovis strains causing tuberculosis in cattle and goats. J. Clin. Microbiol. 33:2953–2956. 7. Imboden, I., S. Cole, T. Bodmer, and A. Telenti. 1993. Detection of rifampin resistance mutations in Mycobacterium tuberculosis and M. leprae. In D. H. Persing, T. F. Smith, F. C. Tenover, and J. White (ed.), Diagnostic molecular microbiology. Mayo Foundation, Rochester, Minn. 8. Kamerbeek, J. L., L. Schouls, A. Kolk, M. van Agterveld, D. van Soolingen, S. Kuijper, A. Bunschoten, H. Molhuizen, R. Shaw, M. Goyal, and J. van Embden. 1997. Simultaneous detection and strain differentiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin. Microbiol. 35:907–914. 9. Martıń, C., S. Samper, I. Otal, P. Asensio, R. Gomez-Lus, G. Torrea, and B. Gicquel. 1994. New methods for diagnosis and epidemiological studies of tuberculosis based on PCR and RFLP, p. 105–113. In F. G. Priest (ed.), Bacterial diversity and systematics. Plenum Press, New York, N.Y. 10. Otal, I., C. Martıń, V. Vicent-Le´vy-Fre´bault, D. Thierry, and B. Gicquel. 1991. Restriction fragment length polymorphism analysis using IS6110 as epidemiological marker in tuberculosis. J. Clin. Microbiol. 29:1252–1254. 11. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 12. Sjo ¨bring, U., M. Mecklenburg, A. B. Andersen, and H. Mio ¨rner. 1990.

13. 14. 15. 16.

17.

18.

19.

20.

1393

Polymerase chain reaction for detection of Mycobacterium tuberculosis. J. Clin. Microbiol. 28:2200–2204. Small, P. M., and J. D. A. van Embden. 1994. Molecular epidemiology of tuberculosis, p. 569–582. In B. R. Bloom (ed.), Tuberculosis. American Society for Microbiology, Washington, D.C. Snider, D. E. J., M. Raviglione, and A. Kochi. 1994. Global burden of tuberculosis, p. 3–12. In B. R. Bloom (ed.), Tuberculosis. American Society for Microbiology, Washington, D.C. Telenti, A., P. Imboden, F. Marchesi, D. Lowrie, S. T. Cole, J. Colston, L. Matter, K. Schopfer, and T. Bodmer. 1993. Detection of rifampicin-resistance mutation in Mycobacterium tuberculosis. Lancet 341:647–650. Telenti, A., P. Imboden, F. Marchesi, T. Schmidheini, and T. Bodmer. 1993. Direct, automated detection of rifampin-resistant Mycobacterium tuberculosis by polymerase chain reaction and single-strand conformation polymorphism analysis. Antimicrob. Agents Chemother. 37:2054–2058. van Embden, J. D. A., M. D. Cave, J. T. Crawford, J. W. Dale, K. D. Eisenach, B. Gicquel, P. Hermans, C. Martıń, R. McAdam, T. Shinnick, and P. M. Small. 1993. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J. Clin. Microbiol. 31:406–409. van Soolingen, D., P. E. W. de Haas, P. W. M. Hermans, P. M. A. Groenen, and J. D. A. van Embden. 1993. Comparison of various repetitive DNA elements as genetic markers for strains differentiation and epidemiology of Mycobacterium tuberculosis. J. Clin. Microbiol. 31:1987–1995. van Soolingen, D., P. E. W. Hass, J. Haagsma, T. Eger, P. W. M. Hermans, V. Ritacco, A. Alito, and J. D. A. van Embden. 1994. Use of various genetic markers in differentiation of Mycobacterium bovis strains from animals and humans and for studying epidemiology of bovine tuberculosis. J. Clin. Microbiol. 32:2425–2433. World Health Organization. 1994. Zoonotic tuberculosis (Mycobacterium bovis): memorandum from WHO meeting (with participation of FAO). Bull. W. H. O. 72:851–857.