|Year : 2017 | Volume
| Issue : 2 | Page : 156-161
Characterization of pyrazinamide resistance in consecutive multidrug-resistant mycobacterium tuberculosis isolates in sweden between 2003 and 2015
Mikael Mansjo1, Jim Werngren1, Sven Hoffner2
1 Department of Microbiology, Public Health Agency of , Solna, Sweden
2 Department of Microbiology, Public Health Agency of ; Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden
|Date of Web Publication||19-May-2017|
Department of Microbiology, Public Health Agency of Sweden, Solna
Source of Support: None, Conflict of Interest: None
Background: The first line anti-tuberculosis drug pyrazinamide (PZA) is important when treating PZA susceptible multidrug-resistant tuberculosis (MDR-TB). Several drug resistance surveys have however reported PZA resistance among a significant proportion of multidrug-resistant (MDR) cases and this undoubtedly highlights the need for accurate and reliable detection of PZA resistance. Unfortunately, the testing of PZA susceptibility is associated with technical difficulties and even though the introduction of pncA sequencing has helped to address this issue, misclassification may still occur. In this study, we determined the prevalence and characteristics of PZA resistance in Swedish MDR-TB strains. Materials and Methods: 153 MDR-TB strains isolated in Sweden between 2003 and 2015 were analyzed for PZA resistance by considering both phenotypic and genotypic data. Results: The phenotypic test showed that 58% of the multidrug-resistant isolates were PZA resistant and the correlation between phenotype and genotype was solid, although a small number of isolates deviate from the expected phenotypic-genotypic pattern. Conclusion: The results indicate that the prevalence of pyrazinamide resistance among Swedish MDR cases is increasing.
Keywords: Drug resistance, Mycobacterium tuberculosis, pncA, pyrazinamide
|How to cite this article:|
Mansjo M, Werngren J, Hoffner S. Characterization of pyrazinamide resistance in consecutive multidrug-resistant mycobacterium tuberculosis isolates in sweden between 2003 and 2015. Int J Mycobacteriol 2017;6:156-61
|How to cite this URL:|
Mansjo M, Werngren J, Hoffner S. Characterization of pyrazinamide resistance in consecutive multidrug-resistant mycobacterium tuberculosis isolates in sweden between 2003 and 2015. Int J Mycobacteriol [serial online] 2017 [cited 2019 Dec 8];6:156-61. Available from: http://www.ijmyco.org/text.asp?2017/6/2/156/206596
| Introduction|| |
Multidrug-resistant tuberculosis (MDR-TB) is an increasing threat to public health not least in the former USSR where a very high prevalence of resistant TB has been repeatedly reported., To cure patients from MDR-TB and to stop transmission of resistant Mycobacterium tuberculosis strains, effective combinations of anti-TB drugs must be identified and administrated as early as possible.
The drug regimens are based on remaining first-line drugs and a limited number of additional compounds. Among these agents, pyrazinamide (PZA), a prodrug which inside the bacteria is activated to pyrazinoic acid by the enzyme pyrazinamidase (PZase), has a special and unique role. PZA is a potent first-line agent for the treatment of TB with activity also against a significant part of drug-resistant TB strains. Unfortunately, many clinical TB laboratories in countries with high MDR-TB numbers do not perform drug susceptibility testing (DST) to PZA due to technical difficulties and poor reproducibility. Thus, in most MDR-TB, high burden areas information on PZA resistance is lacking. Globally, the knowledge of PZA resistance is in fact much more limited than for any other important anti-TB drug, and reliable data that could predict the clinical usefulness of adding PZA to the therapeutic regimen are very rare. Still, PZA is routinely included in the MDR regimens when patients are at low risk for hepatotoxicity.
In Sweden, TB is a rare disease with an incidence of 8.5/100,000 in 2015. Drug-resistant TB is most often seen in our immigrant population, and an increase to around 15–20 newly diagnosed MDR-TB patients per year has been noted during the last 10-year period.
The aim of this study was to determine the prevalence of PZA resistance among Swedish MDR-TB strains and its association with different patient origins as well as M. tuberculosis genotypes. We also investigated how the pncA genotype correlated with the results from the phenotypic PZA susceptibility testing. An abstract from the 1st AASM Congress based on parts of this study has previously been published in The International Journal of Mycobacteriology.
| Materials and Methods|| |
Mycobacterium tuberculosis isolates
A total of 153 clinical MDR-TB isolates from the national strain collection at the Public Health Agency of Sweden, representing all new Swedish cases of MDR-TB registered at the Public Health Agency of Sweden between January 2003 and December 2015, were analyzed in this study. The first isolate from each patient was used in all but one case (SEA200700283; the second sample was isolated 2 months after the initial one) and two of the strains were isolated from the same patient, but belonged to different genotypes, and were therefore classified as two different cases (SEA200700170 and SEA200700300).
Spoligotyping was performed as previously described and used to categorize the isolates according to their different genotypes. Genotypes were assigned through SITVIT WEB.
Pyrazinamide susceptibility testing
Data on drug susceptibility was compiled from the phenotypic DSTs performed at the Public Health Agency of Sweden or at one of the five regional clinical TB laboratories in Sweden. During the initial part of the study period, between 2003 and 2008, the phenotypic DST was performed with the Bactec 460TB system and since 2009 with the Bactec MGIT 960 system according to the standard protocol provided by the manufacturer (Becton Dickinson Biosciences, Sparks, MD, USA). Statistical analysis was performed with VassarStats (http://www.vassarstats.net/clin1.html).
Sequencing of the pncA gene
Sequencing of the pncA gene was performed to detect mutations related to PZA resistance and according to our earlier described protocol. Briefly, the 561 base pair long pncA gene was amplified together with approximately 200 nucleotides up and downstream of the gene using the pncA_F3 (AAGGCCGCGATGACACCTCT) and pncA_R4 (GTGTCGTAGAAGCGGCCGAT) as primers. ExoSAP-IT (Affymetrix, USA) was used to purify the polymerase chain reaction products, and the sequencing reactions were performed with the BigDye Terminator v3.1 Cycle Sequencing Kit (Applied Biosystems, USA) together with the pncA_F3, pncA_R4, pncA_P3-F (ATCAGCGACTACCTGGCCGA), and pncA_P4-R (GATTGCCGACGTGTCCAGAC) primers, respectively. The sequencing reactions were purified with the BigDye Xterminator Purification Kit (Applied Biosystems, USA), and the sequencing data obtained from the 3500xL Genetic Analyzer (Applied Biosystems, USA) was analyzed with the CLC Main Workbench software (Qiagen, Hilden, Germany) using the pncA sequence from H37Rv ATCC 25618 as reference. Whole genome sequencing (WGS) was used to further investigate the samples for which we did not obtain any Sanger data. Shortly, DNA extracted with the column-based QIAamp DNA Mini Kit (Qiagen, Hilden, Germany) was sequenced on the IonTorrent PGM platform (Thermo Fisher Scientific Inc., Waltham, MA, USA) according to the manufacturer's instructions. The obtained sequence reads were later mapped against the M. tuberculosis H37Rv reference genome (GenBank accession no. NC_000962.3) in CLC Genomics Workbench 8 (Qiagen, Hilden, Germany).
A modified Wayne's test was performed on strains with discordant phenotypic and genotypic results. In short, one loop of the bacterial cells, resuspended in 200 μl of phosphate-buffered saline (PBS), was used to inoculate the surface of a Middlebrook 7H9 agar medium tube containing PZA (400 mg/l). The tubes were incubated at 37°C, and 7 days later, 1 ml of fresh 1% ferrous ammonium sulfate (Sigma-Aldrich, Germany) was added to each tube. The samples were left in room temperature for 30 min and checked for the appearance of a pink band in the subsurface agar (indicating the presence of an active PZase). Negative tubes were subsequently incubated at 4°C overnight for the confirmation of the previous result. The PZA susceptible H37Rv ATCC 25618 was included as a PZase-positive control, and a clinical Mycobacterium bovis isolate (SEA12261) was used as a PZase-negative control. A medium control, to which PBS without bacteria was added, was included in the test to exclude false positive results.
| Results|| |
Phenotypic testing of pyrazinamide resistance in Swedish multidrug-resistant strains
The phenotypic testing for PZA resistance revealed that 88 of the 153 Swedish MDR-TB strains (58%) isolated between January 2003 and December 2015 were PZA resistant [Figure 1]a and [Figure 1]b. Among the MDR strains of Beijing genotype, a higher prevalence of PZA resistance was seen corresponding to 37 resistant samples among the 53 Beijing strains (70%). Among the other major genotypes, T (n = 35) and CAS (n = 21), the prevalence was determined to 63% and 33%, respectively, and among all, non-Beijing genotypes (n = 100) the prevalence of PZA resistance was 51%.
|Figure 1: (a) Pyrazinamide resistance in new Swedish cases of multidrug-resistant tuberculosis between 2003 and 2015. Samples were categorized according to the year of isolation at the regional tuberculosis laboratories. (b) Prevalence of pyrazinamide resistance among new Swedish cases of multidrug-resistant tuberculosis between 2003 and 2015.|
Click here to view
The WHO Eastern Mediterranean Region was the most prevalent WHO region of origin, representing 41% (n = 62) of the MDR cases detected in Sweden between 2003 and 2015 [Table 1]. Among the different WHO regions, excluding the WHO Region of the Americas and the WHO South-East Asia Region which only contributed with one MDR case each, the prevalence of PZA resistance varied between 53% and 63% [Table 1]. The highest prevalence (63%) was found in patients with their origin in the WHO African Region. Overall, patients from a total of 39 countries are represented in our study.
|Table 1: Distribution of pyrazinamide resistance among the different WHO regions of origin|
Click here to view
pncA genotypes in Swedish multidrug-resistant strains
Sequencing of the pncA gene was performed to identify polymorphisms within the gene or its putative promoter. Sanger sequences were obtained for 151 isolates, and WGS was performed on the remaining two samples to get a complete set of pncA genotypes. The results showed that 98% (n = 86) of the strains classified as PZA resistant by our phenotypic test also had a nonsynonymous mutation within the pncA gene or its putative promoter. Conversely, 69% of the PZA susceptible strains (n = 45) had a wild-type nucleotide sequence in the pncA gene [Table 2]. The silent mutation Ser65Ser (C195T) was found to be the only nucleotide variation in 17% (n = 11) of the PZA susceptible strains, whereas 14% (n = 9) carried a nonsynonymous mutation in the pncA gene [Table 3]. Although classified as susceptible, two of these isolates did grow in the MGIT tube, indicating an increased minimum inhibitory concentration (MIC) and an “intermediate” resistance, but they were still below the level to be classified as resistant (typically, the PZA susceptible strains presented zero growth [Growth Units = 0] in the PZA tube at the time of reading the result). In two of the PZA susceptible strains carrying a nonsynonymous mutation, we found the silent mutation Ser65Ser in addition to the nonsynonymous mutation (Ser65Ser + Trp119Cys for SEA200600467 and Ser65Ser + His82Asp for SEA201300156). Moreover, none of the PZA-resistant strains harbored this silent mutation only, and spoligotyping confirmed that all strains with the Ser65Ser mutation belonged to the CAS family.
|Table 2: Pyrazinamide phenotype in relation to pncA genotype in Swedish multidrug-resistant tuberculosis isolates|
Click here to view
|Table 3: Multidrug.resistant isolates with discordant phenotype and genotype|
Click here to view
The sensitivity and specificity of nonsynonymous pncA mutations for the detection of PZA resistance were found to be 98% (95% confidence interval [CI]: 91.3–99.6) and 86% (95% CI: 74.8–93.1), respectively, when using the phenotypic testing as reference method [Table 2]. Similarly, the positive and negative predictive values were determined to 91% (95% CI: 82.3–95.3) and 97% (95% CI: 87.0–99.4).
The sequencing results also indicated the presence of heteroresistance, i.e., the presence of both susceptible and resistant bacteria within the population, in at least two of the samples, SEA200300144 and SEA201000097. Both these strains were PZA resistant in the MGIT tests while the pncA sequences were interpreted as wild type by the sequencing analysis software [Table 3]. A manual investigation of the chromatograms did, however, reveal the presence of smaller nonwild-type peaks in both samples (nucleotide position 425 in SEA200300144 and nucleotide position 545 in SEA201000097).
In total, more than 60 different nonsynonymous pncA mutations (including point mutations, insertions, and deletions) were identified in this study and four of them were found to be located upstream of the pncA start codon (A-11G, T-7, A-4T, and a deletion from nucleotide-1 to nucleotide 14). In addition, we identified more complex variations in three isolates (SEA201400002, SEA201400248, and SEA201500560) and for two of those, SEA201400002 and SEA201400248, WGS was needed to obtain the pncA genotype. In detail, WGS results revealed a complete deletion of pncA in SEA201400002 and a deletion from nucleotide-1 to nucleotide 14 in combination with the point mutation A-4T in SEA201400248.
The most common nonsynonymous mutation, A-11G, was found in 7 isolates, whereas 53 (79%) of the nonsynonymous mutations were only seen once in our material. A part of the material has been included in a previous study, and to the best of our knowledge, most nonsynonymous mutations have been described earlier. A complete list of the mutations is presented in [Supplementary Table 1 [Additional file 1]].
Enzymatic testing of strains with discordant phenotypic and genotypic results
To further characterize the eleven strains with discordant phenotypic and genotypic results, a modified Wayne's test was performed [Table 3]. Of the two strains harboring a wild-type pncA gene while still being phenotypically resistant to PZA, the Waynes's test supported the MGIT result in one case and the sequencing result in the other. Conversely, six out of nine PZA susceptible strains carrying a nonsynonymous mutation in the pncA gene were negative in the PZase test, indicating the absence of an active PZase, and in turn, PZA resistance. Among the three remaining samples (pncAmut/PZA susceptible) with a positive PZase test, we found SEA201200211 an isolate harboring the Thr47Ala mutation reported as a nonresistance conferring mutation.,, The PZase test performed on this strain resulted in a weak pink band in the subsurface agar, indicating the presence of an at least partially active PZase.
| Discussion|| |
The DST results obtained in this study showed that 58% of the Swedish MDR-TB strains isolated between 2003 and 2015 were PZA resistant. This is in line with results from similar studies performed in other settings where the prevalence of PZA resistance among MDR strains typically has been in the interval of 40%–60%,,,, and in a meta-analysis published in 2011, the median prevalence of PZA resistance in culture isolates was determined to 51%. Not surprisingly though, both higher and lower figures have been published,, and the level of PZA resistance clearly varies between different geographical settings, a finding which presumably depends on the usage and handling of PZA in the different areas (different DST methodologies might also contribute to the discrepancies).
During the studied 13-year period, we have witnessed an increased number of PZA-resistant MDR-TB cases in Sweden, and the prevalence of PZA resistance among the MDR-TB strains does also seem to increase [Figure 1]b. This suggests a general increase of PZA resistance, at least in areas where these patients have their origin. The highest prevalence of PZA resistance was as previously mentioned found in patients originating from the WHO African Region; however, only three of the strains isolated from these patients belonged to the Beijing genotype (the major genotype with the highest prevalence of PZA resistance in our study).
Moreover, the present study serves as an evaluation of the PZA algorithm used at the reference laboratory at the Public Health Agency of Sweden. Clinical MDR and extensively drug-resistant strains from the regional TB laboratories in Sweden are routinely tested for PZA resistance in the Bactec MGIT 960 system and submitted to pncA sequencing. If discordant results are obtained, the MGIT test is repeated, and a PZase test is considered.
Overall, the present study confirms the strong correlation between phenotypic PZA resistance and mutations in the pncA gene already reported; 98% of the PZA-resistant strains had a nonwild-type pncA gene whereas 69% of the PZA susceptible strains carried a wild-type pncA gene, a figure which increases to 86% if the silent mutation Ser65Ser is included in this group. 14% (n = 9) of the PZA susceptible strains harbored a nonsynonymous pncA mutation, and this is fairly coherent with the numbers presented in other, comparable studies.
Even though we use three different methods to test for PZA resistance, data can still be hard to interpret and the clinical relevance of the results shown in [Table 3] might not be very clear. Among the PZA-resistant strains, two were classified as pncAwt, and these results suggest the presence of an alternative resistance mechanism independent of pncA or the presence of a heterogeneous population. In line with the latter hypothesis, we were able to detect additional peaks below the wild-type nucleotide peak in the sequencing chromatogram for both these strains (SEA200300144 and SEA201000097), clearly indicating the presence of a mixed population (pncAwt and pncAmut).
Regarding the nine PZA susceptible strains carrying a nonsynonymous pncA mutation, three were classified as susceptible by the PZase test. SEA201200211 harbored the Thr47Ala mutation known to not confer full PZA resistance,,, while Ala102Val was found in both SEA201400128 and SEA201500032. Ala102Val has previously been reported as a mutation not being associated with PZA resistance and that conclusion is coherent with the phenotypic susceptibility and the positive enzymatic test reported here.
The remaining six PZA susceptible strains with nonsynonymous pncA mutations were all negative in the PZase test and two of these, SEA200700070 and SEA201300156, did actually grow in the presence of PZA in the MGIT system, but below the level to be classified as resistant. The other four strains (SEA200600467, SEA200800022, SEA200900184, and SEA201400384) did not grow at all in the presence of PZA, and the sequencing data did not indicate any heterogeneity in the bacterial population, but these isolates might possibly have a MIC value lower than the current critical concentration for PZA 100 mg/L (no MIC determinations were performed in this study). The nonsynonymous mutations found in these four samples have been reported earlier and were in a previous paper classified into categories ranging from “Mutations conferring PZA r at very high confidence,” “Mutations conferring PZA r at high confidence,” and “Mutations not involved in phenotypic resistance.” To explain this, one might argue that isolates with MIC values close to the critical concentration are more vulnerable to inter- and intra-laboratory variations which could in turn explain the inconsistent results in these cases. The results do highlight the need for further studies which determine the MIC values of specific pncA mutations and a more comprehensive knowledge could hopefully contribute to a reduced risk of reporting false PZA resistance in present and future molecular tests.
| Conclusions|| |
This study reports a solid correlation between the PZA phenotype and the pncA genotype in Swedish MDR strains and that the prevalence of PZA resistance among these strains has increased since 2003.
We acknowledge the five Swedish clinical TB laboratories who kindly provided the strains to the reference laboratory at the Public Health Agency of Sweden. We also thank Ylva Lidén and Juan Carlos Toro for excellent technical assistance.
Financial support and sponsorship
Financial support from the Swedish Research Council, grant no 540-2013-8797, is gratefully acknowledged.
Conflicts of interest
There are no conflicts of interest.
| References|| |
Skrahina A, Hurevich H, Zalutskaya A, Sahalchyk E, Astrauko A, van Gemert W, et al.
Alarming levels of drug-resistant tuberculosis in Belarus: Results of a survey in Minsk. Eur Respir J 2012;39:1425-31.
Zignol M, Dara M, Dean AS, Falzon D, Dadu A, Kremer K, et al.
Drug-resistant tuberculosis in the WHO European Region: An analysis of surveillance data. Drug Resist Updat 2013;16:108-15.
Scorpio A, Zhang Y. Mutations in pncA, a gene encoding pyrazinamidase/nicotinamidase, cause resistance to the antituberculous drug pyrazinamide in tubercle bacillus. Nat Med 1996;2:662-7.
Zhang Y, Mitchison D. The curious characteristics of pyrazinamide: A review. Int J Tuberc Lung Dis 2003;7:6-21.
World Health Organization. Global Tuberculosis Report 2014. Geneva: World Health Organization; 2014.
World Health Organization. Companion Handbook to the WHO Guidelines for the Programmatic Management of Drug-Resistant Tuberculosis. Geneva: World Health Organization; 2014.
Mansjö M, Verngren J, Hoffner S. Pyrazinamide resistance in Swedish multidrug-resistant tuberculosis 2003-2013. Int J Mycobacteriol 2015;4 Suppl 1:125.
Kamerbeek J, Schouls L, Kolk A, van Agterveld M, van Soolingen D, Kuijper S, et al.
Simultaneous detection and strain differentiation of Mycobacterium tuberculosis
for diagnosis and epidemiology. J Clin Microbiol 1997;35:907-14.
Demay C, Liens B, Burguière T, Hill V, Couvin D, Millet J, et al.
SITVITWEB – A publicly available international multimarker database for studying Mycobacterium tuberculosis
genetic diversity and molecular epidemiology. Infect Genet Evol 2012;12:755-66.
Juréen P, Werngren J, Toro JC, Hoffner S. Pyrazinamide resistance and pncA gene mutations in Mycobacterium tuberculosis
. Antimicrob Agents Chemother 2008;52:1852-4.
Singh P, Wesley C, Jadaun GP, Malonia SK, Das R, Upadhyay P, et al.
Comparative evaluation of Löwenstein-Jensen proportion method, BacT/ALERT 3D system, and enzymatic pyrazinamidase assay for pyrazinamide susceptibility testing of Mycobacterium tuberculosis
. J Clin Microbiol 2007;45:76-80.
Streicher EM, Bergval I, Dheda K, Böttger EC, Gey van Pittius NC, Bosman M, et al. Mycobacterium tuberculosis
population structure determines the outcome of genetics-based second-line drug resistance testing. Antimicrob Agents Chemother 2012;56:2420-7.
Miotto P, Cabibbe AM, Feuerriegel S, Casali N, Drobniewski F, Rodionova Y, et al. Mycobacterium tuberculosis
pyrazinamide resistance determinants: A multicenter study. Mbio 2014;5:e01819-14. doi: 10.1128/mBio.01819-14.
Sreevatsan S, Pan X, Zhang Y, Kreiswirth BN, Musser JM. Mutations associated with pyrazinamide resistance in pncA of Mycobacterium tuberculosis
complex organisms. Antimicrob Agents Chemother 1997;41:636-40.
Morlock GP, Crawford JT, Butler WR, Brim SE, Sikes D, Mazurek GH, et al.
Phenotypic characterization of pncA mutants of Mycobacterium tuberculosis
. Antimicrob Agents Chemother 2000;44:2291-5.
Dormandy J, Somoskovi A, Kreiswirth BN, Driscoll JR, Ashkin D, Salfinger M. 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. Chest 2007;131:497-501.
Louw GE, Warren RM, Donald PR, Murray MB, Bosman M, Van Helden PD, et al.
Frequency and implications of pyrazinamide resistance in managing previously treated tuberculosis patients. Int J Tuberc Lung Dis 2006;10:802-7.
Mphahlele M, Syre H, Valvatne H, Stavrum R, Mannsåker T, Muthivhi T, et al.
Pyrazinamide resistance among South African multidrug-resistant Mycobacterium tuberculosis
isolates. J Clin Microbiol 2008;46:3459-64.
Ando H, Mitarai S, Kondo Y, Suetake T, Sekiguchi JI, Kato S, et al.
Pyrazinamide resistance in multidrug-resistant Mycobacterium tuberculosis
isolates in Japan. Clin Microbiol Infect 2010;16:1164-8.
Stoffels K, Mathys V, Fauville-Dufaux M, Wintjens R, Bifani P. Systematic analysis of pyrazinamide-resistant spontaneous mutants and clinical isolates of Mycobacterium tuberculosis
. Antimicrob Agents Chemother 2012;56:5186-93.
Chang KC, Yew WW, Zhang Y. Pyrazinamide susceptibility testing in Mycobacterium tuberculosis
: A systematic review with meta-analyses. Antimicrob Agents Chemother 2011;55:4499-505.
Perdigão J, Macedo R, João I, Fernandes E, Brum L, Portugal I. Multidrug-resistant tuberculosis in Lisbon, Portugal: A molecular epidemiological perspective. Microb Drug Resist 2008;14:133-43.
Stavrum R, Myneedu VP, Arora VK, Ahmed N, Grewal HM. In-depth molecular characterization of Mycobacterium tuberculosis
from New Delhi – Predominance of drug resistant isolates of the 'modern' (TbD1) type. PLoS One 2009;4:e4540.
Zhang Y. Rapid molecular detection of pyrazinamide resistance: The way forward. Int J Tuberc Lung Dis 2015;19:128.
[Table 1], [Table 2], [Table 3]