Original Research LUNG INFECTION Discrepant Results Between Pyrazinamide Susceptibility Testing by the Reference BACTEC 460TB Method and pnca DNA Sequencing in Patients Infected With Multidrug-Resistant W-Beijing Mycobacterium tuberculosis Strains* Jillian Dormandy, BS; Akos Somoskovi, MD, PhD; Barry N. Kreiswirth, PhD; Jeffrey R. Driscoll, PhD; David Ashkin, MD; and Max Salfinger, MD Background: Mycobacterium tuberculosis strains belonging to the W-Beijing family have received broad clinical and public health attention because of their rapid worldwide spread and their frequent association with outbreaks, multidrug resistance, and treatment failures and relapses. Methods: The present study examined a large number of multidrug-resistant strain-w isolates (isolates of 29 patients) by susceptibility testing for pyrazinamide (PZA) using the reference BACTEC 460TB method (Becton Dickinson Diagnostic Instrument Systems; Sparks, MD) and also by DNA sequencing of the pnca gene. Results: We found that despite of the presence of a strain W-specific Thr47Ala in the pnca gene, all strains showed susceptibility to PZA in the reference BACTEC 460TB system due to their higher minimum inhibitory concentrations (relative to BACTEC 460TB PZA-susceptible strains). Conclusions: Our results suggest that the current radiometric reference method cannot reproducibly detect PZA resistance in patients infected with W-Beijing strains. Therefore, PZA susceptibility testing should instead be based on analysis of the pnca gene for resistanceassociated mutations. (CHEST 2007; 131:497 501) Key words: multidrug resistant; Mycobacterium tuberculosis; pnca; pyrazinamide; susceptibility testing; tuberculosis; W-Beijing genotype Abbreviations: INH isoniazid; MDR multidrug resistant; MIC minimal inhibitory concentration; PZA pyrazinamide; PZase pyrazinamidase; RFLP restriction fragment length polymorphism; RIF rifampin; XDR extensively drug resistant The worldwide increase of multidrug-resistant (MDR) tuberculosis (resistance at least to isoniazid [INH] and rifampin [RIF]) and extensively *From the Wadsworth Center (Drs. Somoskovi, Driscoll, and Salfinger and Ms. Dormandy), New York State Department of Health, Albany, NY; Public Health Research Institute (Dr. Kreiswirth), Newark, NJ; and A.G. Holley State Hospital (Dr. Ashkin), Lantana, FL. This work was performed at the Wadsworth Center, New York State Department of Health, Albany, NY, and Public Health Research Institute, Newark, NJ. The authors have no financial relationship with a commercial entity that has an interest in the subject of this article. drug-resistant tuberculosis (XDR) [resistance to INH and RIF and at least three of the six main classes of second-line drugs] is one of the greatest threats to achieving control of global tuberculosis. 1 Manuscript received August 12, 2006; revision accepted September 13, 2006. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians (www.chestjournal. org/misc/reprints.shtml). Correspondence to: Max Salfinger, MD, Wadsworth Center, New York State Department of Health, PO Box 509, Albany, NY 12201-0509; e-mail: salfinger@wadsworth.org DOI: 10.1378/chest.06-1899 www.chestjournal.org CHEST / 131 / 2/ FEBRUARY, 2007 497
Following multiple nosocomial outbreaks in New York City in the 1990s, a unique strain with specific phenotypic (MDR) and genetic characteristics was identified and named as strain W. 2 Another outbreak strain with similar molecular epidemiologic characteristics emerged in the Beijing province of China. 3 Genetically, these strains can be characterized by multiple (15 to 26) copies of IS6110 that have similar restriction fragment length polymorphism (RFLP) genotyping patterns ( 65%); a unique IS6110 insertion in the origin of replication; and a specific spacer oligotype (spoligotype) lacking spacers 1 through 34 in the direct-repeat chromosomal locus of Mycobacterium tuberculosis. 2 4 Recently, strains bearing these characteristics have been classified into a large phylogenetic lineage called the W- Beijing lineage. 4 Since their discovery, members of the W-Beijing family have received broad clinical and public health attention because of their rapid worldwide spread and their frequent association with outbreaks, multidrug resistance, treatment failures, and relapses. 5,6 Immunologic studies 7 with animals have also identified specific pathogenic properties for these strains. Specifically, W-Beijing strains exhibit a higher virulence through their elicitation of a less sustained tumor necrosis factor- response from host macrophages, a weaker protective T-helper type 1 cell response, a more advanced and extensive pneumonia, and significantly higher mortality compared to M tuberculosis H37Rv in a BALB/c mouse model. 7 In addition, patients with W-Beijing strain infections often show fever that is unrelated to disease severity and toxicity. 8 Materials and Methods Between 1998 and 2005, a total of 211 MDR tuberculosis cases were reported in New York City. M tuberculosis strains obtained from 29 patients (13.7%) with MDR tuberculosis that were sent for characterization and drug susceptibility testing to the Clinical Mycobacteriology Laboratory at the Wadsworth Center, New York State Department of Health, during this period were included in the present study. Spoligotyping, a polymerase chain reaction-based genotyping method that detects 43 known spacer sequences that are interspersed between the direct repeats in the genomic repeat region of M tuberculosis, was performed with a commercially available kit (Isogen Bioscience BV; Maarssen, the Netherlands) according to the instructions of the manufacturer. 9 IS6110 RFLP fingerprinting was performed according to a standardized protocol as described previously. 10 The IS6110 fingerprint patterns of the examined strains were analyzed using software (Bionumerics version 3.0; Applied Maths; Austin, TX) as described earlier. 10 Routine testing for drug susceptibility by the radiometric BACTEC 460TB system (Becton Dickinson Diagnostic Instrument Systems; Sparks, MD) was performed on all isolates as described earlier. 11 In order to determine the level of pyrazinamide (PZA) susceptibility or resistance, on a subset (n 5) of the study isolates, the minimal inhibitory concentrations (MICs) for PZA were also determined in duplicates. In addition to the concentration of 100 g/ml at ph 6.0 for determination of drug resistance by the standard BACTEC protocol, the following PZA concentrations were also tested: 6.25 g/ml, 12.5 g/ml, 25 g/ml, and 50 g/ml. Drug resistance to first-line agents, with the exception of PZA, was confirmed by the agar proportion method using Middlebrook 7H10 agar. 12 In addition, resistance to INH, RIF, and PZA was confirmed by katg, rpob, and pnca sequencing, as reported earlier. 13 15 Primers Tb86 (5 -GAAACAGCGGCGCTGATCGT-3 ) and Tb87 (5 -GTTGTCCCATTTCGTCGGGG-3 ) flanking the region encoding amino acid Ser315 of katg were used to amplify a 209-base-pair product. 13 Primers rpo95. (5 -CCACCCAG- GACGTGGAGGCGATCACACCG-3 ) and rpo397 (5 -GT- CAACCCGTTCGGGTTCATCGAAACG-3 ) were used to amplify a 329-base-pair product, which included the relevant segment of rpob 14. Using two primers pnca-p1 5 -GCT-GGT- CAT-GTT-CGC-GAT-CG-3 and pnca-p6 5 - GCT-TTG-CGG- CGA-GCG-CTC-CA 3, which flanked the entire pnca gene and its upstream promoter, we generated a 700-base-pair product from each isolate as described previously. 15 The same primers were used for DNA sequencing of both strands of the three genes, with an automated DNA sequencer (Applied Biosystems 3700; Applied Biosystems; Foster City, CA). The sequencing results were compared with the wild-type katg, prob, and pnca sequence from M tuberculosis H37Rv, for detection of mutations. The DNA sequencing was performed by the Molecular Genetics Core Facility at the Wadsworth Center. Results The genotyping results demonstrated that all 29 isolates were strain-w isolates (belonged to the Table 1 Characteristics of the 29 MDR Isolates Belonging to the W-Beijing Family Laboratory Tests Results BACTEC 460TB susceptibility testing results for INH, RIF, Resistant ethambutol, and streptomycin (ph 6.8) BACTEC 460TB susceptibility testing results for PZA (ph 6.0) Susceptible Spoligotyping (octal code) 000000000003771 IS6110 RFLP genotyping W (15 isolates) and W-family* (14 isolates) rpob gene DNA sequencing His526Tyr (CAC to TAC) katg gene DNA sequencing Ser315Thr (ACC to ACA) pnca gene DNA sequencing Thr47Ala (ACC to GCC) *Including strains W1, W12, W33, W665, and W769. 498 Original Research
W-Beijing family) [Table 1]. All of these isolates showed resistance to INH, RIF, ethambutol, and streptomycin, while all demonstrated susceptibility to PZA (Table 1). DNA sequencing identified mutations, resulting in amino acid change Ser315Thr (ACC to ACA) in the katg gene and amino acid change His526Tyr (CAC to TAC) in the rpob gene, in all 29 isolates (Table 1). Interestingly, sequencing of the pnca gene revealed a rare Thr47Ala (ACC to GCC) mutation in all isolates that had been associated with strain-w isolates in earlier studies (Table 1). 16 With regard to MIC testing, three of the five strains tested showed a MIC of 50 g/ml, and two of the five strains showed a MIC of 100 g/ml. Discussion PZA, a nicotinamide analog thought to target an enzyme involved in fatty acid synthesis, is a pro-drug; it is converted to its active form, pyrazinoic acid, by the mycobacterial enzyme pyrazinamidase (PZase). 17 PZA has an excellent sterilizing effect on semidormant tubercle bacilli and, when used in combination with RIF, shortens the treatment of tuberculosis patients from 1 year to 6 months. 17 Several previous studies have suggested that susceptibility testing of M tuberculosis to PZA is not always reliable. 17 Testing was initially hindered by the requirement that the medium have a ph of 5.5 in order for the drug to be active. 18 However, many M tuberculosis isolates will not grow at a low ph. Subsequently, a more simplified broth-based radiometric assay (BACTEC 460TB) using ph 6.0 was described. 19 This method resulted in well-defined differences between PZA-susceptible and PZA-resistant strains, and became the currently recommended assay. 12 However, poor growth at ph 6.0 can lead to inconclusive results in up to 3.5% of the strains with this method, as well. 20 Other studies, 21 which measured and compared PZase activity of PZA-susceptible and PZA-resistant strains with results obtained by the radiometric method, have shown that at least 0.8% of the strains that grew in the presence of PZA were PZase positive and, thus, were reported as falsely resistant in the radiometric assay. More recently, DNA sequencing of the gene encoding PZase (pnca) has proven to be a reliable predictor: 72 to 98% of the PZA-resistant clinical isolates tested by the radiometric assay carry a mutation in the structural gene or in the putative promoter region of the gene. 17 DNA sequencing of the pnca gene has been found to be an accurate assay (99.4% sensitivity) for predicting PZA susceptibility or resistance also in our laboratory, based on the testing of 236 PZA-susceptible and 172 PZA-resistant (in the BACTEC 460TB system) isolates. 22 Taken together, these various observations suggest that for some strains the radiometric assay is inaccurate. The issue of inconsistent and nonreproducible PZA susceptibility testing results can be even more problematic in patients with polyresistant tuberculosis; for them, the inclusion or exclusion of PZA frequently plays an important role in the therapy. 23 For the reasons listed above, combined with our detection of a Thr47Ala mutation in the strain-w isolates, we decided to revisit the susceptibility to PZA of this family. Using the reference radiometric method on a subset (n 5) of the study isolates, we determined the MICs for PZA. Interestingly, all five strains showed higher MICs than BACTEC 460TB PZA-susceptible strains but lower MICs than BACTEC 460TB PZA-resistant strains. 19 Similar results had been observed for three MDR strains that also carried the rare Thr47Ala pnca mutation. 24 In that study 24 two of the strains exhibited resistance to 100 g/ml of PZA, and the third strain was resistant to 800 g/ml of PZA. Thus, in two independent studies, 24 (including this study), a Thr47Ala mutation in pnca has been found to correlate with increased resistance to PZA, suggesting that the Thr47Ala allele results in a less active form of PZase and, consequently, resistance to PZA. Conclusions First, PZA is a key drug in the treatment of tuberculosis. PZA shortens the duration of treatment when used in combination with RIF, and patients receiving regimens that contain PZA have a lower bacteriologic relapse rate compared to patients treated with regimens without PZA. 25 However, it has been shown that PZA has a limited role in preventing drug resistance, 26 and recent studies 27,28 revealed that PZA resistance is common in MDR isolates from previously treated patients. Therefore, the accurate prediction of PZA resistance is essential when managing treatment failure of patients with MDR and/or XDR tuberculosis who were already exposed to this drug (Fig 1). Secondly, this is particularly true in the case of infection with a drug-resistant W-Beijing strain; for these strains, transmission of the disease is more rapid, treatment failures and relapses are more common, and the clinical presentation may be more severe than with non W-Beijing strains. Our results suggest that a more accurate and rapid PZA drug resistance assay is warranted: the current radiometric reference method may not reproducibly detect PZA resistance in patients infected with W-Beijing strains. Instead, PZA susceptibility testing should be www.chestjournal.org CHEST / 131 / 2/ FEBRUARY, 2007 499
Figure 1. Turnaround times for detection of W-Beijing strains and their drug resistance patterns in comparison with the rate of spread of tuberculosis if transmission is not interrupted by rapid detection of these strains. DNA sequencing of the katg and rpob genes enables the rapid identification of MDR and/or XDR strains. DNA sequencing of the pnca gene enables the rapid identification of MDR and or XDR strains W. Stick figures indicate the number of individuals with infection if drug resistance is not detected and/or treatment is not optimal based on analysis of the pnca gene for resistanceassociated mutations (Fig 1). To shorten the turnaround time for detection of PZA resistance in these patients, the clinical mycobacteriology laboratory community will need to implement the pnca sequencing assay for direct use on clinical specimens; such an adaptation will permit identification of PZA drug-resistant mutants within 24 to 48 h (Fig 1). Moreover, this novel screening approach will immediately identify strain-w isolates given that they carry the unique Thr47Ala allele. Finally, successful treatment of cases with MDR and/or XDR tuberculosis relies on the use of drugs that clinicians need to have confidence. The implication that a mutation may be present in certain strains that may render a drug inactive clinically is important in that the clinician may want to add other drugs that the strain demonstrates susceptibility to rather than rely on one that may have reduced or no activity. The clinical significance of the presence of such a mutation needs to be further explored in animal models and in clinical trials. However, clinicians should be aware of the presence of such mutations from a diagnostic standpoint (more rapid diagnosis of potential W-Beijing strains) as well as potential clinical implications. ACKNOWLEDGMENT: The authors wish to thank Keith Derbyshire for critical review of the manuscript, and the Molecular Genetics Core Facility of the Wadsworth Center for performing the sequencing analyses. References 1 Emergence of Mycobacterium tuberculosis with extensive resistance to second-line drugs worldwide, 2000 2004. MMWR Morb Mortal Wkly Rep 2006; 55:301 305 2 Bifani PJ, Plikaytis BB, Kapur V, et al. Origin and interstate spread of a New York City multidrug-resistant Mycobacterium tuberculosis clone family. JAMA 1996; 275:452 457 3 van Soolingen D, Qian L, de Haas PE, et al. Predominance of a single genotype of Mycobacterium tuberculosis in countries of east Asia. J Clin Microbiol 1995; 33:3234 3238 4 Kremer K, Glynn JR, Lillebaek T, et al. Definition of the Beijing/W lineage of Mycobacterium tuberculosis on the basis of genetic markers. J Clin Microbiol 2004; 42:4040 4049 5 Glynn JR, Whiteley J, Bifani PJ, et al. Worldwide occurrence of Beijing/W strains of Mycobacterium tuberculosis: a systematic review. Emerg Infect Dis 2002; 8:843 849 6 Lan NT, Lien HT, Tung le B, et al. Mycobacterium tuberculosis Beijing genotype and risk for treatment failure and relapse, Vietnam. Emerg Infect Dis 2003; 9:1633 1635 7 Lopez B, Aguilar D, Orozco H, et al. A marked difference in pathogenesis and immune response induced by different Mycobacterium tuberculosis genotypes. Clin Exp Immunol 2003; 133:30 37 8 van Crevel R, Nelwan RH, de Lenne W, et al. Mycobacterium tuberculosis Beijing genotype strains associated with febrile response to treatment. Emerg Infect Dis 2001; 7:880 883 9 Kremer K, van Soolingen D, Frothingham R, et al. Comparison of methods based on different molecular epidemiological markers for typing of Mycobacterium tuberculosis complex 500 Original Research
strains: interlaboratory study of discriminatory power and reproducibility. J Clin Microbiol 1999; 37:2607 2618 10 Kurepina N, Likhoshvay E, Shashkina E, et al. Targeted hybridization of IS6110 fingerprints identifies the W-Beijing Mycobacterium tuberculosis strains among clinical isolates. J Clin Microbiol 2005; 43:2148 2154 11 Inderlied CB, Salfinger M. Antimycobacterial agents and susceptibility tests. In: Murray PR, Baron EJ, Pfaller MA, et al eds. Manual of clinical microbiology. Washington, DC: American Society for Microbiology, 1999; 1601 1623 12 National Committee for Clinical Laboratory Standards. Susceptibility testing of mycobacteria, nocardia and other aerobic actinomycetes: approved standard. NCCLS document M24-A. Wayne, PA: National Committee for Clinical Laboratory Standards, 2003 13 Somoskovi A, Song Q, Mester J, et al. Use of molecular methods to identify the Mycobacterium tuberculosis complex (MTBC) and other mycobacterial species and to detect rifampin resistance in MTBC isolates following growth detection with the BACTEC MGIT 960 system. J Clin Microbiol 2003; 41:2822 2826 14 Parsons LM, Salfinger M, Clobridge A, et al. Phenotypic and molecular characterization of Mycobacterium tuberculosis isolates resistant to both isoniazid and ethambutol. Antimicrob Agents Chemother 2005; 49:2218 2225 15 Scorpio A, Lindholm-Levy P, Heifets L, et al. Characterization of pnca mutations in pyrazinamide-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother 1997; 41: 540 543 16 Sreevatsan S, Pan X, Zhang Y, et al. Mutations associated with pyrazinamide resistance in pnca of Mycobacterium tuberculosis complex organisms. Antimicrob Agents Chemother 1997; 41:636 640 17 Zhang Y, Telenti A. Genetics of drug resistance in Mycobacterium tuberculosis. In: Hatful GF, Jacobs WR Jr, eds. Molecular genetics of mycobacteria. Washington, DC: ASM Press, 2000; 235 254 18 Butler WR, Kilburn JO. Susceptibility of Mycobacterium tuberculosis to pyrazinamide and its relationship to pyrazinamidase activity. Antimicrob Agents Chemother 1983; 24: 600 601 19 Salfinger M, Reller LB, Demchuk B, et al. Rapid radiometric method for pyrazinamide susceptibility testing of Mycobacterium tuberculosis. Res Microbiol 1989; 140:301 309 20 Miller MA, Thibert L, Desjardins F, et al. Growth inhibition of Mycobacterium tuberculosis by polyoxyethylene stearate present in the BACTEC pyrazinamide susceptibility test. J Clin Microbiol 1996; 34:84 86 21 Miller MA, Thibert L, Desjardins F, et al. Testing of susceptibility of Mycobacterium tuberculosis to pyrazinamide: comparison of BACTEC method with pyrazinamidase assay. J Clin Microbiol 1995; 33:2468 2470 22 Dormandy J, Forbes S, Kaswa MK, et al. Is pnca sequencing a valid assay to replace the reference BACTEC 460TB method for PZA susceptibility testing [abstract]? Paper presented at: 106th General Meeting of the American Society for Microbiology; May 22 24, 2006; Orlando, FL; C-334:199 23 Hewlett D Jr, Horn DL, Alfalla C. Drug-resistant tuberculosis: inconsistent results of pyrazinamide susceptibility testing. JAMA 1995; 273:916 917 24 Morlock GP, Crawford JT, Butler WR, et al. Phenotypic characterization of pnca mutants of Mycobacterium tuberculosis. Antimicrob Agents Chemother 2000; 44:2291 2295 25 Controlled trial of 4 three-times-weekly regimens and a daily regimen all given for 6 months for pulmonary tuberculosis: second report; the results up to 24 months. Hong Kong Chest Service/BMJ Research Council. Tubercle 1982; 63:89 98 26 Mitchison DA. Mechanisms of the action of drugs in the short-course chemotherapy. Bull Int Union Tuberc 1985; 60:36 40 27 Louw GE, Warren RM, Donald PR, et al. Frequency and implications of pyrazinamide resistance in managing previously treated tuberculosis patients. Int J Tuberc Lung Dis 2006; 10:802 807 28 Saravia JC, Appleton SC, Rich ML, et al. Retreatment management strategies when first-line tuberculosis therapy fails. Int J Tuberc Lung Dis 2005; 9:421 429 www.chestjournal.org CHEST / 131 / 2/ FEBRUARY, 2007 501