Pyrazinamide resistance among South African multi-drug resistant Mycobacterium tuberculosis ACCEPTED

Transcription

1 JCM Accepts, published online ahead of print on 7 August 008 J. Clin. Microbiol. doi:0.8/jcm Copyright 008, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved. Pyrazinamide resistance among South African multi-drug resistant Mycobacterium tuberculosis isolates Running title: Pyrazinamide resistant M. tuberculosis isolates Matsie Mphahlele,, Heidi Syre, 5, Håvard Valvatne, Ruth Stavrum, 5, Turid Mannsåker 3, Tshilidzi Muthivhi, Karin Weyer, P Bernard Fourie 4, and Harleen M. S. Grewal, 5 *. The Gade Institute, Section of Microbiology and Immunology, University of Bergen, N-50, Bergen, Norway Medical Research Council, Pretoria, South Africa 3 National Reference Laboratory for Mycobacteria, Norwegian Institute of Public Health, Oslo, Norway 4 Medicine in Need, MRC Building, Pretoria, South Africa. 5 Department of Microbiology and Immunology, Haukeland University Hospital, N-50, Bergen, Norway. *Corresponding author. Mailing address: Section of Microbiology and Immunology, The Gade Institute, University of Bergen and Haukeland University Hospital, N-50 Bergen, Norway. Phone Fax Harleen.Grewal@Gades.uib.no. 3 Key words: tuberculosis, drug resistance, pyrazinamide

2 4 5 ABSTRACT Pyrazinamide is important in tuberculosis treatment, as it is bactericidal to semi-dormant mycobacteria not killed by other anti-tuberculosis drugs. Pyrazinamide is also one of the cornerstone drugs retained in the treatment of multi-drug resistant tuberculosis (MDR-TB). However, due to technical difficulties, routine drug susceptibility testing of Mycobacterium tuberculosis for pyrazinamide is in many laboratories not performed. The objective of our study was to generate information on pyrazinamide susceptibility among South African multi-drug resistant (MDR) and susceptible M. tuberculosis isolates from pulmonary tuberculosis patients. Seventy-one MDR and 59 fully susceptible M. tuberculosis isolates collected during the national surveillance study (00-00, by the Medical Research Council, South Africa) were examined for pyrazinamide susceptibility by the radiometric BACTEC 460 TB system, pyrazinamidase activity (Wayne s assay) and sequencing of the pnca gene. The frequency of pyrazinamide resistance (by BACTEC) among the MDR M. tuberculosis isolates was 37 of 7 (5.%) and 6 of 59 (0.%) among fully sensitive isolates. A total of 5 unique mutations in the pnca gene were detected. The majority of these were point mutations that resulted in amino acid substitutions. Twenty-eight isolates had identical mutations in the pnca gene, but could be differentiated from each other by a combination of spoligotype pattern and MIRU loci. 45

3 A high proportion of South African MDR M. tuberculosis isolates were resistant to pyrazinamide, suggesting a reconsideration of its role in patients treated previously for tuberculosis as well as the treatment of MDR-TB. INTRODUCTION South Africa is ranked 4 th among the high tuberculosis (TB) burden countries designated by the World Health Organization (WHO) with an incidence rate of 940 cases per citizens in 006 (). TB-HIV/AIDS co-infection rates are high, with more than 50% of adult TB patients also being HIV-positive (). Multi-drug resistant TB (MDR-TB) is further worsening South Africa s TB epidemic. National studies of MDR-TB conducted by the Medical Research Council (MRC) of South Africa in 00 found that.6% of new TB cases, and 6.7% of cases in re-treatment, had MDR-TB (i.e. resistance to isoniazid and rifampicin) [, 5]. A total of 5866 M. tuberculosis isolates were collected during the survey. Information on previous treatment was available for 97.4% of the patients. Overall, 70.3% of the patients had no history of previous treatment, while 7.% indicated that they had received treatment for TB before. Drug resistance to any of the four first line drugs, isoniazid, rifampicin, ethambutol and streptomycin was detected in 7.8% cultures from new TB patients and in 5.5% cultures from re-treatment cases [, 3]. One hundred and seventy-nine isolates were MDR M. tuberculosis isolates of which 7 were available for this study (). 67 3

4 According to the drug resistance survey conducted in () by the MRC, South Africa, isolates with resistance to isoniazid were detected in 5.7% of new patients and in.8% of patients with a previous history of TB treatment. Rifampicin resistance was found in.8% of new and 7.5% of re-treatment cultures. Ethambutol resistance was very low at 0.8% in new patients and.4% in previously treated cases. Isolates with resistance to streptomycin were found in 4.3% of new and 8.% of re-treatment patients (). Pyrazinamide is an important first-line drug used in the shortcourse treatment of TB in combination with isoniazid, rifampicin and ethambutol (). Pyrazinamide is also one of the cornerstone drugs retained in the treatment of MDR-TB. Pyrazinamide can kill semi-dormant tubercle bacilli that persist in acidic inflammatory environments (7, ), and are not affected by other anti-tb drugs. The intracellular sterilizing activity of pyrazinamide allows the treatment duration to be reduced from the earlier norm of 9 months to the current standard of 6 months, aiding compliance and decreasing the risk of developing MDR-TB. Routine drug susceptibility testing of pyrazinamide is complicated to perform as the growth of the bacilli is impeded by the acidic conditions required for optimal drug activity (0). A recent study (9), describes a high (53.5%) frequency of pyrazinamide resistance among isolates from previously treated patients from the Western Cape, South Africa. However, no studies have addressed the countrywide frequency of pyrazinamide resistance among South African MDR M. tuberculosis isolates. DNA sequencing studies of the pnca gene from pyrazinamide resistant and pyrazinamide susceptible isolates of M. tuberculosis have established an association between mutations in this gene and pyrazinamide resistance. Mutations in the pnca gene are known to be a major mechanism of pyrazinamide resistance (3, 7) although some studies have shown resistant isolates which have 4

5 a wild type gene, suggesting alternative mechanism for pyrazinamide resistance (, 6). Earlier observations by Butler and Kilburn (), suggest that the resistance of M. tuberculosis isolates to pyrazinamide correlates with the absence of an amidase activity (pyrazinamidase). Since most pyrazinamide resistant M. tuberculosis isolates have lost pyrazinamidase activity, enzyme activity assays are sometimes used to replace pyrazinamide susceptibility testing (). Thus, the aim of this study was to generate information on pyrazinamide susceptibility among South African MDR and susceptible M. tuberculosis isolates from pulmonary TB patients through the use of phenotypic and genotypic methods. In-vitro susceptibility to pyrazinamide by BACTEC 460 TB method was correlated with pyrazinamidase activity and the frequency of mutations in the pnca gene of M. tuberculosis isolates. Isolates with identical mutations were further analyzed by spoligotyping and variable number tandem repeat of mycobacterial interspersed repetitive unit (MIRU-VNTR) typing. MATERIALS AND METHODS Mycobacterial isolates A total of 5866 M. tuberculosis isolates were collected during the national drug resistance survey conducted by the MRC, South Africa, in (). Of the 79 MDR M. tuberculosis isolates collected in the survey, 7 were tested for pyrazinamide susceptibility by the BACTEC 460 TB method, Wayne s assay and pnca gene sequencing. Of the 584 fully susceptible (for isoniazid, rifampicin, ethambutol, streptomycin) isolates that were collected during the survey, 59 were also studied for drug susceptibility to pyrazinamide by the above mentioned methods. Isolates were excluded due to contamination, growth problems, and mislabelling. The characteristics of the 7 5

6 MDR M. tuberculosis isolates included in this study are provided in Table. The identity of all isolates was confirmed by the AccuProbe Culture Identification Tests (Accuprobe; Gen-Probe Inc., San Diego, CA) and spoligotyping. Isolates identified as mycobacteria other than M. tuberculosis were excluded from the study. DNA extraction M. tuberculosis isolates were grown on Lowenstein-Jensen media at 37 o C for 3 to 4 weeks. DNA from mycobacterial cells was extracted using a boiling technique (7). A µl loopful of cells was suspended in 00 µl of TE buffer ph 8.0 (0mM Tris-Cl, mm EDTA), and heat-killed by incubation at 95 o C for 5-0 min. The supernatant containing the extracted DNA was collected by centrifugation at 000 rpm for 7 min. Drug susceptibility testing Pyrazinamide susceptibility testing was performed at the MRC Supranational Tuberculosis Reference Laboratory, in South Africa, by the radiometric BACTEC 460 TB system (Beckton Dickinson Sparks, MD) at pyrazinamide concentrations of 00 µg/ml, as recommended by the manufacturer (8). A clinical M. bovis isolate was included as a pyrazinamide resistant control and M. tuberculosis H37Rv (ATCC 794) was included as a pyrazinamide susceptible control. The recommended interpretive criteria for the ratio of the growth index (GI) of the drug containing vial to the GI of the control vial were followed: GI ratio less than 9%, susceptible; 9 to %, borderline; and greater than %, resistant. 35 6

7 Wayne s assay Pyrazinamidase activity was assayed according to Wayne s procedure (0). A heavy loopful of growth from an actively growing culture was inoculated onto the surface of Middlebrook 7H9 agar (Difco, Detroit, MI), containing 00 µg/ml pyrazinamide (Sigma-Aldrich, Steinheim, Germany) and µg/ml sodium pyruvate (Merck, Darmstadt, Germany). The test tubes were incubated at 37 o C, for 7 days. After incubation, ml of freshly prepared % ferrous ammonium sulphate (Fe(NH 4 ) (SO 4 ), 6H O) (Fluka BioChemika, Buchs, Switzerland) solution was added to each tube and the tubes were kept at 4 o C for 4 hours. A tube inoculated with M. bovis was included as a negative control and M. tuberculosis H37Rv (ATCC 794) was used as a positive control. The pyrazinamidase activity assay was considered positive if a pink band was seen in the upper part of the butt. Amplification and sequencing of pnca gene The pnca gene was amplified from each isolate using primers P and P6, as described previously (6). The expected size of the PCR product was 70 bp, which included the full length of the pnca gene (56 bp) as well as 04 bp of the upstream sequence and 55 bp of the downstream sequence. PCR amplifications were carried out in a GeneAmp PCR system 9700 thermocycler (Perkin Elmer, Applied Biosystems, CA). Each reaction mixture (5 µl) contained.5 µl of the crude DNA extraction, 0.05 pmol of each PCR primer, and.5 µl Hot Start Mastermix kit (Qiagen, Germany). The reaction mixtures were subjected to 5 min at 95 o C, followed by 30 cycles of 40 7

8 57 58 sec at 94 o C, 60 sec 55 o C, 40 sec at 7 o C and terminated by 5 min at 7 o C. Successful gene amplifications were confirmed by the Bioanalyzer DNA 000 chip kit (Agilent Technologies, US) PCR products from each strain were purified with the QIAquick PCR purification kit (Qiagen GmbH Hilden, Germany), according to the manufacture s instructions. PCR products were subjected to a sequencing reaction using a BigDye Terminator Cycle Sequencing Kit and primer P. The amplification profile consisted of an initial 5 min for denaturation at 96 o C, 30 cycles of 96 o C for 30 sec, 50 o C for 5 sec, 60 o C for 4 min and elongation time of 7 o C for 5 min in an ABI 377 automatic DNA sequencer (Applied Biosystems Inc., Foster, CA). Nucleotide sequences were analyzed using CHROMAS software (Technelysium Ltd). Genotyping Spoligotyping Spoligotyping was performed to further characterize M. tuberculosis isolates with identical mutations in the pnca gene, according to an internationally standardized protocol (6) to detect the presence or absence of 43 variable spacers in the direct repeat (DR) region of M. tuberculosis. DNA was amplified by PCR using primer DRb and biotinylated DRa, complementary to either side of the DR unit. The resulting amplification products were then hybridized to a commercially available membrane (Isogen Bioscience BV, Bilthoven, the Netherlands), with covalently linked parallel rows of 43 synthetic oligonucleotides representing the unique spacer sequences between DR units in the DR region. 78 8

9 The conditions used for PCR were 3 min at 96 o C, followed by 0 cycles of min at 96 o C, min at 55 o C, and 30 sec at 7 o C. An MS-Excel spreadsheet was used to analyze the spoligotyping results and the spoligotyping image was converted into an octal digital format (3). The data was further analyzed by comparison with the 4 th version of an international spoligotyping database () MIRU-VNTR typing VNTR typing was used to further discriminate between strains that had identical pnca mutations and spoligotype patterns. As previously described (8), primer sets were used to individually amplify regions with known tandem repeats. Amplifications for all MIRU-PCR reactions, except MIRU-4, were performed in a total volume of 0 µl containing 0,5 µl bacterial DNA, 0.5 µl (5 µm) of each PCR primer, 4 µl HotMasterMix kit (Eppendorf, Germany) and 5 µl ddh O. An initial denaturation of min at 94 o C was followed by 35 cycles of denaturation at 94 o C for 30 sec, annealing at 58 o C for min and extension at 65 o C for 30 sec, followed by an final extension at 65 o C for 5 min. Amplifications with the MIRU-4 primers were performed using, µl DNA, 0.5 µl of the MIRU-4 forward and reverse primers (0µM), 6 µl Hot Start Mastermix kit (Qiagen, Germany),.5 µl Q-solution (Qiagen, Germany) and,5 µl ddh O. An initial denaturation of 5 min at 95 o C was followed by 40 cycles of denaturation at 94 o C for min, annealing at 58 o C for.5 min and extension at 7 o C for.5 min, followed by a final extension at 7 o C for 5 min. 00 9

10 Samples were visualized and analyzed with a Bioanalyzer DNA 000 chip kit (Agilent Technologies, US). Amplicon sizes were used to estimate the number of complete repeats at each locus as described by Frothingham and Meeker-O Connell (5). Analysis of strains with discrepant pyrazinamide susceptibility results MDR and fully susceptible M. tuberculosis isolates that had discordant results in any of the three methods applied (BACTEC 460 TB system, Wayne s and sequencing method), were sent to a second laboratory (National Institute for Public Health, Oslo, Norway) for validation of susceptibility to pyrazinamide by re-testing pyrazinamide susceptibility by BACTEC 460 TB system. The results from the arbiter laboratory were considered conclusive. RESULTS Pyrazinamide susceptibility testing by phenotypic methods Of 7 MDR M. tuberculosis isolates tested by three methods (BACTEC 460 TB system, Wayne s assay and sequencing of pnca gene), 49 isolates had concordant results and had discordant results in all three methods. Of 49 MDR M. tuberculosis isolates with concordant results, 30 were resistant and 9 were susceptible to pyrazinamide by BACTEC 460 TB method. Of MDR M. tuberculosis isolates with discordant results in all three methods applied, 8 were resistant and 4 were sensitive to pyrazinamide by the BACTEC 460 TB method. 0

11 The MDR M. tuberculosis isolates with discordant results were re-tested for pyrazinamide susceptibility by the BACTEC 460 TB system in a second laboratory (Norwegian Institute of Public Health, Oslo, Norway). Thus, 7 were found to be resistant (one isolate had borderline resistance), and 5 were sensitive to pyrazinamide. Therefore, on incorporation of re-tested results, 37 of 7 (5.%) MDR M. tuberculosis isolates were resistant to pyrazinamide by the BACTEC method (Table ). Among 59 isolates susceptible for rifampicin, isoniazid, ethambutol, and streptomycin, the pyrazinamide susceptibility results of 54 isolates agreed in all 3 methods. Five isolates had discordant results in the methods applied and were re-tested for pyrazinamide susceptibility by the BACTEC 460 TB system in a second laboratory. Of 54 isolates with concordant results in all three methods, 5 were sensitive and 3 were resistant to pyrazinamide by the BACTEC 460 TB method. Of 5 isolates with discordant results, 3 were resistant ( isolate showed borderline resistance) and were sensitive to pyrazinamide on re-testing in a second laboratory. On incorporation of re-tested BACTEC results, 6 of 59 (0.%) fully susceptible M. tuberculosis isolates were resistant to pyrazinamide (Table ). Of the 37 pyrazinamide resistant MDR-TB isolates by BACTEC (36 pyrazinamide resistant and borderline resistant isolate), 3 isolates showed no pyrazinamidase activity of which 30 isolates had mutations in the pnca gene (Table ). Four of 6 fully susceptible isolates, which were pyrazinamide resistant by BACTEC (5 pyrazinamide resistant and borderline resistant isolate) showed no pyrazinamidase activity and had mutations in the pnca gene. Overall, using BACTEC 460 TB as

12 the reference method, and 7 isolates had discordant results in Wayne s assay and sequencing of pnca gene, respectively (Table ). Mutations in the pnca gene Twenty-five unique mutations out of a total of 44 different mutations were detected among 4 (36 MDR and 6 fully susceptible) M. tuberculosis isolates that had mutations in the pnca gene (Table 3). The majority of these were point mutations that resulted in amino acid substitutions. Point mutations included nucleotide substitution (37 of 44), deletions (3 of 44), and insertions (4 of 44) in the pnca gene (Table 3). Two of the isolates had synonymous (silent) nucleotide substitutions, confirmed by pyrazinamide sensitivity in BACTEC and pyrazinamidase activity (Table 3). The most frequently encountered change was the A- to- G point mutation at position 35 (9 of 44). Isolates that shared identical mutations could be assigned to ten different groups (Table 3). Spoligotyping Twenty-eight (Table 3) M. tuberculosis isolates distributed in 0 different groups had identical mutations in the pnca gene and were characterized by spoligotyping. On the basis of spoligotyping, 4 M. tuberculosis isolates were distributed in 8 clusters (a cluster was defined as two or more isolates from different patients with identical spoligotype patterns and mutations in the pnca gene). Four of the 8 M. tuberculosis isolates had unique spoligotyping patterns. The 8 clusters comprised isolates belonging to previously described shared-type (ST): 8,, 34, 7, 9, 37, 79 and 96 (). 66

13 The largest cluster was ST 37 comprising 9 isolates which had the same nucleotide substitution of A- to- G at position 35. Three isolates belonged to ST 96. The remaining clusters had isolates each. To determine the relatedness of all isolates in clusters, additional molecular typing by MIRU- VNTR typing was performed. MIRU-VNTR typing Twenty-four isolates distributed in 8 clusters with identical mutations in the pnca gene and spoligotype patterns in each cluster, were subjected to analysis of MIRU loci. MIRU-VNTR had a greater resolving power than spoligotyping. Thus, no isolates were found to have the same MIRU-VNTR codes. However, isolates within each cluster displayed related MIRU-VNTR profiles. Of the MIRU-VNTR loci tested, MIRU-3 produced 6 alleles and was the most polymorphic. The least polymorphic loci were MIRU-6 and MIRU-4. At MIRU-6, tandem repeats were absent in 45.8% of the isolates. All isolates had one copy of the MIRU locus 4. DISCUSSION Global surveillance has shown that drug resistant TB is widespread and is now a threat to TB control programs in many countries. Drug-resistance complicates the treatment of TB cases often leading to unsuccessful outcomes. In addition, MDR-TB causes high morbidity and mortality, 3

14 particularly in immuno-compromised populations. The treatment of drug resistant and MDR-TB is expensive and has many problems. Therefore, strategies to detect, treat and prevent drug resistance are urgently needed. Treatment guidelines for MDR-TB patients in South Africa were implemented as national policy in 00. Pyrazinamide is routinely used in first line therapy for TB, but is also one of the cornerstone drugs used in the treatment of MDR-TB and is included in the standardized MDR-TB treatment regimen in South Africa. However, due to technical difficulties routine drug susceptibility testing for pyrazinamide is not performed and consequently little is known about the actual rates of pyrazinamide resistance in South Africa, as opposed to similar data existing for other first-line anti- TB drugs. A number of laboratory methods have been used to assess the susceptibility of M. tuberculosis to pyrazinamide. Conventional mycobacterial susceptibility testing methods, depending on growth of the organisms when exposed to drugs, are impeded by performance difficulties (requirement of a low ph, standardization of inoculum size and clumping effect [0, 9]). Other approaches such as the Wayne s assay involves the detection of the presence or absence of pyrazinamidase activity as a marker of resistance, with resistant strains presumed to be lacking the enzyme (0). Mutations in the pnca gene encoding pyrazinamidase involved in activation of pyrazinamide to the active form pyrazinoic acid provides a possible molecular marker for resistance determination (4) Thus, in this study, we tested the susceptibility of 7 MDR and 59 fully susceptible M. tuberculosis isolates collected during the national drug resistance survey conducted by the MRC, South Africa, 4

15 in (). Susceptibility to pyrazinamide was tested by the radiometric BACTEC 460 TB system, Wayne s assay, and sequencing of the pnca gene. Thirty-seven of 7 (5.%) MDR M. tuberculosis isolates were resistant to pyrazinamide by the BACTEC 460 TB system. Similarly, there were 6 of 59 (0.%) fully susceptible M. tuberculosis isolates that were resistant to pyrazinamide. Seven of 43 (6.3%) BACTEC pyrazinamide resistant M. tuberculosis isolates showed pyrazinamidase activity when tested by Wayne s assay. This suggests participation of other M. tuberculosis genomic regions in conferring resistance to pyrazinamide and is supported by alternative mechanisms leading to pyrazinamide resistance: active efflux of bactericidal pyrazinoic acid from the organism (4) and defects in pyrazinamide uptake by the organism (5). Thirty-nine of 43 (90.7%) BACTEC pyrazinamide resistant M. tuberculosis isolates showed pnca mutations in this study. The finding of pyrazinamide resistant isolates that lack mutations in the pnca gene suggests that alternative mechanisms for drug resistance may exist. In addition, we found three isolates which were pyrazinamide susceptible but had mutations in the pnca gene implying that mutations in this region do not affect the bacterium s susceptibility to pyrazinamide. It is generally considered that mutations leading to pyrazinamide resistance are scattered along the pnca gene (4). The mutations in the pnca gene identified in this study, were found scattered outside the regions suggested to have some degree of clustering of mutations (Gly3-Thr4, Pro69-Leu85, and Ile5-Asp), as previously described (7). Our findings of diverse and scattered mutations along the pnca gene support similar observations in other studies (9, 4). 33 5

16 The study shows that the presence of mutation in the pnca gene correlates well with pyrazinamide resistance in BACTEC (for 39 of 4 isolates) and with a loss of pyrazinamidase activity (for 34 of 4 isolates). An A -to- G point mutation at position 35 in.4% of the isolates was the most common type of pnca mutation in the study. The isolates with identical mutations in the pnca gene could be differentiated from each other by a combination of spoligotype pattern and analysis of MIRU loci. Thus, all isolates containing mutation in the pnca gene were unique. In agreement with previous suggestions (9, 7), the detection of pnca mutations by direct sequencing of the pnca gene by PCR could provide a rapid method for the diagnosis of pyrazinamide resistant M. tuberculosis, thereby contributing to more effective management of TB patients and, potentially limiting the spread of drug-resistant M. tuberculosis isolates. Although, the strains studied may not be fully representative of all MDR M. tuberculosis strains from the National survey, the (alarmingly) high (5.%) proportion of strains that were resistant to pyrazinamide highlights the need for further systematic testing of pyrazinamide during drug resistance surveys. Susceptibility tests for pyrazinamide are difficult to perform and need careful standardization. Furthermore, the clinical relevance of in vitro resistance to pyrazinamide also needs urgent investigation, as the drug is widely used for both drug susceptible TB re-treatment regimens and MDR-TB treatment. In conclusion, we show that a high proportion of South African MDR-TB isolates tested are resistant to pyrazinamide, suggesting a reconsideration of its role in patients treated previously for TB as well as in MDR-TB treatment

17 ACKNOWLEDGEMENTS The authors thank Jeanette Brand (MRC, South Africa), Hilde Jacobsen (University of Bergen) and staff at the Norwegian Institute of Public Health, Oslo for their technical assistance. The study was supported by Helse Vest, Norway, the University of Bergen, NORAD (scholarship to Matsie Mphahlele) and the Medical Research Council, South Africa. TRANSPARENCY DECLARATIONS None to declare. REFERENCES. Brudey, K., J. R. Driscoll, L. Rigouts, W. M. Prodinger, A. Gori, S. A. Al-Hajoj, C. Allix, L. Aristimuno, J. Arora, and V. Baumanis Mycobacterium tuberculosis complex genetic diversity: mining the fourth international spoligotyping database (SpolDB4) for classification, population genetics and epidemiology. BMC Microbiol. 6:3.. Butler, W. R., and J. O. Kilburn Susceptibility of Mycobacterium tuberculosis to pyrazinamide and its relationship to pyrazinamidase activity. Antimicrob. Agents and Chemother. 4:

18 Dale, J.W., D. Brittain, A. A. Cataldi, D. Cousins, J. T. Crawford, J. Driscoll, H. Heersma, T. Lillebaek, T. Quitugua, N. Rastogi, R.A Skuce, C, Sola, D. van Soolingen, V. Vincent. 00. Spacer oligonucleotide typing of bacteria of the Mycobacterium tuberculosis complex: recommendations for standardized nomenclature. Int. J.Tuber. Lung Dis. 5: Davies, A. P., O. J. Billington, T. D. McHugh, D. A. Mitchison, and S. H. Gillespie Comparison of phenotypic and genotypic methods for pyrazinamide susceptibility testing with Mycobacterium tuberculosis. J. Clin. Microbiol. 38: Frothingham, R., and W. A. Meeker-O Connell Genetic diversity in the Mycobacterium tuberculosis complex based on variable numbers of tandem DNA repeats. Microbiol. 44: Groenen, P. M., A. E. Bunschoten, D. van Soolingen, and J. D.van Embden Nature of DNA polymorphism in the direct repeat cluster of Mycobacterium tuberculosis, application for strain differentiation by a novel typing method. Molecular Microbiology 0: Heifets, L., and P. Lindholm-Levy. 99. Pyrazinamide sterilizing activity in vitro against semidormant Mycobacterium tuberculosis bacterial populations. Am. Rev. Respir. Dis. 45:3-5. 8

19 Kam, K. M., W. C. Yip, W. L. Tse, K. L. Wong, K. T. Lam, K. Kremer, B. K. Au, and D. van Soolingen Utility of mycobacterial interspersed repetitive unit typing for differentiating multidrug-resistant Mycobacterium tuberculosis isolates of the Beijing family. J. Clin. Microbiol. 43: Louw, G.E., R. M. Warren, P. R. Donald, M. B. Murray, M. Bosman, P. D. van Helden, D. B. Young, and T. C. Victor Frequency and implications of pyraminazide resistance in managing previously treated tuberculosis patients. Int J Tuberc Lung Dis. 0: McDermott, W., and R. Tompsett Activation of pyrazinamide and nicotinamide in acidic environment in vitro. Am. Rev. Respir. Dis. 70: Mitchison, D. A The action of antituberculosis drugs in short-course chemotherapy. Tubercle. 66:9-5.. Morlock, G. P., J. T. Crawford, W. R. Butler, S. E. Brim, D. Sikes, G. H. Mazurek, C. L. Woodley, and R. C. Cooksey Phenotypic characterization of pnca mutants of Mycobacterium tuberculosis. Antimicrob. Agents. Chemother. 44: Portugal, I., L. Barreiro, J. Moniz-Pereira, and L. Brum pnca mutations in pyrazinamide-resistant Mycobacterium tuberculosis isolates in Portugal. Antimicrob Agents Chemother. 48:

20 Ramaswamy, S., and J. M. Musser Molecular genetic basis of antimicrobial agent resistance in Mycobacterium tuberculosis: 998 update. Tuber. Lung. Dis. 79: Raynaud, C., M. A Laneelle, R. H. Senaratne, P. Draper, G. Laneelle, and M. Daffe Mechanisms of pyrazinamide resistance in mycobacteria: importance of lack of uptake in addition to lack of pyrazinamidase activity. Microbiol.45: Scorpio, A., and Y. Zhang Mutations in pnca, a gene encoding pyrazinamidase/nicotinamidase, cause resistance to the antituberculous drug pyrazinamide in tubercle bacillus. Nat. Med. : Scorpio, A., P. Lindholm-Levy, L. Heifets, R. Gilman, S. Siddiqi, M. Cynamon, and Y. Zhang Characterization of pnca mutations in pyrazinamide-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother. 4: Siddiqi, S. H BACTEC TB system, product and procedure manual. Towson, Md: Becton Dickinson Diagnostic Instruments Systems. 9. Stottmeier, K. D., R. E. Beam, and G. P. Kubica Determination of drug susceptibility of mycobacteria to pyrazinamide in 7H0 agar. Am. Rev. Respir. Dis. 96: Wayne, L. G Simple pyrazinamidase and urease tests for routine identification of mycobacteria. Am. Rev. Respir. Dis. 09:

21 Weyer, K., J. Lancaster, J. Brand, M. van der Walt, and J. Levin Survey of tuberculosis drug resistance in South Africa Final report. World Health Organization Global tuberculosis control: surveillance, planning, financing. World Health Organization, Geneva, Switzerland. WHO/HTM/TB/ World Health Organization/IUATLD Antituberculosis drug resistance in the world. Third Global Report. World Health organization, Geneva, Switzerland. WHO/HTM/TB/ Zhang, Y., A. Scorpio, H. Nikaido, and Z. Sun Role of acid ph and deficient efflux of pyrazinoic acid in unique susceptibility of Mycobacterium tuberculosis to pyrazinamide. J. Bacteriol. 8: Zignol, M., M. S. Hosseini, A. Wright, C. Lambregts-van Weezenbeek, P. Nunn, C. J. Watt, and B. G. Williams Global incidence of multirug-resistant tuberculosis. J Infect Dis. 94:

22 Table. MDR M. tuberculosis isolates categorised by province, age, gender, HIV status, and treatment category. Characteristics Province : Eastern Cape Mpumalanga Limpopo North West Western Cape Gauteng Free State KwaZulu Natal Age: >59 Unknown Sex: Female Male Isolates included % (n=7) 4.3 () 45. (9) 7.6 (3) 6.9 (7) 46. (6) 47. (8) 7.3 (3) 53.8 (4) 36.7 () 39.7 (5) 4.9 (8) 4.3 () 40.0 (6) (3) 4.4 (48) Isolates not tested % (n=08) 57.7 (5) 54.8 (3) 8.4 (4) 74. (0) 53.8 (7) 5.9 (9) 7.7 (8) 46. () 63.3 (9) 60.3 (38) 58. (5) 57.7 (5) 60.0 (9) 00.0 () 63.5 (40) 58.6 (68) Total (n=79)* HIV : Status Positive Unknown 36.4 (0) 4.7 (35) 37.5 (6) 63.6 (35) 58.3 (49) 6.5 (4) Treatment: New cases Previously treated Unknown 30.8 (4) 47.4 (46) 5.0 () 69. (54) 5.6 (5) 75.0 (3) * A total of 79 MDR M. tuberculosis isolates were collected during the National drug resistance survey (00-00) by the Medical Research Council of South Africa

23 Table. Results* from pyrazinamide susceptibility testing of 7 MDR and 59 susceptible M. tuberculosis isolates by BACTEC 460 TB, pnca gene sequencing and pyrazinamidase activity. Isolates (total) MDR (7) Fully susceptible (59) BACTEC Results 36 Susceptible 34 Borderline 5 Susceptible 53 Borderline PZase- Mut PZase - Mut PZase + Mut PZase + Mut - Abbreviations: Mut, mutation; +, present; -, absent; PZase, pyrazinamidase. * Isolates with discordant results in the methods applied were retested for pyrazinamide susceptibility by BACTEC 460 TB in a second laboratory and the results from the arbiter laboratory were considered conclusive and are incorporated in the final results

24 Table 3. Overview of twenty-five different mutations in the pnca gene of 4 (36 MDR and 6 fully susceptible) M. tuberculosis isolates from South Africa. No. of isolates 9 Mutation group a Change in nucleotide b (mutation number) T C pos - () T C pos -0 () G A pos (3) G T pos (4), T G pos 04 (5) A G pos 35 (6) Ins G pos 35 (7) T C pos 40 (8) Del T pos 40 (9) T G pos 04 (5) T C pos 75 (0) C T pos 95 () C T pos () C T pos 36 (3), Ins T pos 360 (4) A T pos 86 (5) G T pos 89 (6) G A pos 90 (7) Amino acid substitution Putative regulatory area mutation Putative regulatory area mutation D8N D8Y, L34R DG Frameshift C4R A0STOP L35R S59P None H7T A79V, Frameshift K96STOP G97C G97D BACTEC 460 TB results resistant*, sensitive Sensitive Wayne s assay results Positive 8 negative, positive Positive Positive negative, positive 9 C T pos 305 (8) A0V negative, positive Ins G pos 35 (9) Frameshift G A pos 394 (0) G3STOP 3 0 T C pos 45 () L5S A G pos 460 () R54G T C pos 476 (3) L59P Del T pos 55 (4) Frameshift Del T pos 58 (5) Frameshift a Isolates that shared identical mutations were assigned to ten different groups and further characterized by spoligotyping. b Abbreviations: Del, deletion; Ins, insertion. *: Borderline resistant.