Characteristics of compensatory mutations in the rpoC gene and their association with compensated transmission of Mycobacterium tuberculosis

Abstract

The aim of this study was to characterize rpoC gene mutations in Mycobacterium tuberculosis (MTB) and investigate the factors associated with rpoC mutations and the relation between rpoC mutations and tuberculosis (TB) transmission. A total of 245 MTB clinical isolates from patients with TB in six provinces and two municipalities in China were characterized based on gene mutations through DNA sequencing of rpoC and rpoB genes, phenotyping via standard drug susceptibility testing, and genotypic profiling by mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) typing. Approximately 36.4% of the rifampin-resistant isolates harbored nonsynonymous mutations in the rpoC gene. Twenty-nine nonsynonymous single mutations and three double mutations were identified. The rpoC mutations at locus 483 (11.3%) were predominant, and the mutations at V483G, W484G, I491V, L516P, L566R, N698K, and A788E accounted for 54.5% of the total detected mutations. Fifteen new mutations in the rpoC gene were identified. Rifampin resistance and rpoB mutations at locus 531 were significantly associated with rpoC mutations. MIRU-VNTR genotype results indicated that 18.4% of the studied isolates were clustered, and the rpoC mutations were not significantly associated with MIRU-VNTR clusters. A large proportion of rpoC mutation was observed in the rifampicin-resistant MTB isolates. However, the findings of this study do not support the association of rpoC mutation with compensated transmissibility.

Keywords

tuberculosis / drug resistance / compensatory mutations / transmission

Introduction

Tuberculosis (TB), especially drug-resistant TB, poses a serious challenge to disease control in China. An effective TB control program should be based on a comprehensive understanding of the circulating Mycobacterium tuberculosis (MTB) in the country. Previous studies have indicated that the recent transmission of MTB strains, including multidrug-resistant (MDR) ones (resistance to at least isoniazid and rifampin), contributes substantially to the TB disease burden in China; about one-third of MDR-TB cases are attributed to this recent transmission [¹,²]. The factors that contribute to the development and transmission of MTB are complex, and previous studies have focused on inadequate therapy, poor implementation of national TB programs, and host susceptibility. However, MTB-related characteristics, including the ability to infect the host, persist, proliferate, and transmission have often been ignored. Gene mutations that confer drug resistance are commonly associated with fitness costs [³,⁴], and drug-resistant bacteria are usually considered to be less fit than susceptible strains and less likely to cause epidemics than drug-sensitive ones. However, a previous study revealed that several drug resistance mutations concomitantly lower or obviate fitness costs and that low average fitness costs of mutations are associated with increased severity of epidemics caused by drug-resistant TB [⁵]. Mathematical model results have also shown that the relative fitness of drug-resistant strains may play an important role in spreading MDR-TB [⁶,⁷].

Rifampin (RMP) is an important first-line anti-TB drug, and resistance to RMP poses a serious challenge to the effective treatment of TB or MDR-TB. RMP resistance results from mutations in the RMP resistance-determining region (RRDR) of the rpoB gene, which encodes the b subunit of RNA polymerase. RMP resistance can be caused by different mutations in the rpoB gene, but clinical strains with rpoB mutation S531L are highly prevalent [⁸]. The success of S531L mutations in clinical MTB isolates may depend not only on a low initial fitness cost but also on the ability to acquire compensatory mutations for further reducing the fitness cost; this compensatory evolution can reduce initial fitness defects caused by particular drug resistance-conferring mutations and is potentially important for the long-term persistence of drug resistance [⁹–¹²].

Among fitness compensatory mutations with the rpoB gene, rpoA and rpoC genes, which encode a and b′ subunits of RNA polymerase, respectively, have been extensively investigated; a study on MTB has revealed that mutations in rpoA and rpoC are common among clinical strains with rpoB S531L mutations [¹³]. Other studies assessed the possible link between rpoC mutations and MTB transmission by comparing the proportion of compensatory rpoC mutations in clustered or nonclustered strains; these studies found that compensatory mutations are more common in clustered than in nonclustered strains and concluded that rpoC mutations may restore the fitness cost caused by resistance-conferring mutations, thereby enhancing the ability of MDR-TB strains to spread [¹⁴,¹⁵]. However, MDR-TB strains with compensatory mutations are unlikely to cluster and do not promote the influence of compensatory mutations on MDR-TB transmission [¹⁶]. These contradicting conclusions highlight the need for future studies. In the current study, we used strains collected from four provinces and two municipalities to describe the profile of mutations in the rpoC gene and determine if compensatory mutations facilitate the spreading of MTB.

Materials and methods

Study population

The study sample comprised MTB cultures isolated from patients with smear-positive pulmonary TB who visited local TB hospitals or dispensaries between February 2012 and October 2013 in four provinces (Heilongjiang, Henan, Sichuan, and Zhejiang) and two municipalities (Chongqing and Tianjin) in China. Only one isolate from each patient was included in this study.

Drug susceptibility testing

Acid-fast bacillus (AFB) positive cultures recovered from a Löwenstein–Jensen (L–J) medium were used for identification and drug susceptibility testing (DST). MTB identification was performed by testing susceptibility to p-nitrobenzoic acid. Susceptibility to RMP, isoniazid, streptomycin, ethambutol, kanamycin, and ofloxacin was examined by applying the proportion method in L–J medium in accordance with the World Health Organization’s standard proportional method.

DNA extraction

DNA was extracted from MTB isolates grown on L–J medium via a simple method. Bacteria were suspended in 400 mL of 1×TE buffer, heat killed at 80 °C for 30 min, boiled at 100 °C for 15 min, and centrifuged at 12 000 g for 15 min to remove cell debris. The supernatant was stored at −20 °C for further use.

rpoB and rpoC gene amplification and sequence analysis

Polymerase chain reaction (PCR) amplification was performed, followed by DNA sequencing of the rpoB RMP RRDR region, which is an 81 bp hotspot region (codons 507 to 533) of the rpoB gene. The primers for rpoB were rpoB F(5′-TACGGTCGGCGAGCTGATCC-3′) and rpoB R(5′-TACGGCGTTTCGATGAACC-3′). The entire rpoC locus (Rv0668) was amplified and sequenced. The primer sequence used for rpoC was obtained from a previous work [¹⁴]. PCR amplification was performed under the following conditions: 5 min of denaturation at 94 °C, 30 cycles of denaturation at 94 °C for 30 s, annealing at 58 °C for 30 s, elongation at 72 °C for 30 s, and final extension of 7 min at 72 °C. The amplicons were purified and sequenced with the ABI DNA sequencer model 377, and sequences were analyzed using DNAstar and BioEdit software.

Molecular genotyping

MTB Beijing genotype was identified via RD-105 multiplex PCR [¹⁷]. Mycobacterial interspersed repetitive unit-variable number tandem repeat (MIRU-VNTR) typing was performed on chromosomal DNA extracted from MTB isolates. PCR was conducted to amplify a standard panel of 15 loci [¹⁸], and the PCR products were separated on 2% agarose gel with 50 or 100 bp DNA markers. DNA extracted from the reference strain H37Rv was used as positive control. A MIRU-VNTR cluster was defined when two or more isolates from different patients showed the same 15-locus MIRU-VNTR genotype profiles; otherwise, the isolates were considered unique.

Statistical analyses

The cluster of MIRU-VNTR genotyping was performed with the MIRU-VNTRplus web application (http://www.miru-vntrplus.org/MIRU/index.faces). Statistical analyses were performed with SAS9.1 (SAS Institute Inc.). Univariate and multivariate logistic regression were used to identify the risk factors for rpoC mutation or factors for MIRU-VNTR cluster. Variables with P<0.05 were considered statistically significant.

Results

Characteristics of the study population and isolated mycobacteria

A total of 257 patients were included in this study. Among the mycobacterial isolates, eight were identified as nontuberculous mycobacteria. Among the 249 MTB isolates, 245 had available results on DST, genotyping, and gene sequencing. The median age of the study patients was 44.2±16.3 years (14–82 years), and 183 (75.0%) patients were male. Among the 243 patients with available treatment history, 125 (51.4%) were new. According to the DST results, 94 (38.4%) were RMP-sensitive isolates, and 151 (61.6%) were RMP-resistant isolates, among which 103 (42.0%) were RMP monoresistant or MDR isolates and 48 (19.6%) were pre-extensively drug-resistant (pre-XDR, defined as strains resistant to isoniazid and RMP and to at least one fluoroquinolone or one second-line injectable agent [SLI]: kanamycin, amikacin, or capreomycin) or XDR (defined as strains with resistance to at least isoniazid, RMP, a fluoroquinolone, and at least one SLI) isolates. RD-105 multiplex PCR revealed that 198 (80.8%) were Beijing family strains.

Characteristics of mutations in the rpoC gene

Gene sequence analyses revealed that 55 (36.4%) RMP-resistant isolates harbored 29 nonsynonymous single mutations and 3 double mutations in the rpoC gene. The rpoC gene loci frequently involved in the mutations were codon 483 (11.3%), followed by codons 484 (2.0%), 491 (2.0%), 698 (2.0%), 516 (1.3%), 332 (1.3%), 521 (1.3%), 566 (1.3%), and 788 (1.3%). Seven nonsynonymous mutations at V483G, W484G, I491V, L516P, L566R, N698K, and A788E accounted for 54.5% of the total detected mutations. Fifteen new mutations, which have not been previously reported, were found in the rpoC gene (Table 1, Fig. 1). For the 94 RMP-susceptible isolates, 4 (4.26%) nonsynonymous (R703C, I800V, D279G, and A466E) and 3 synonymous (2 isolates with P1040P and one isolate with G109G) mutations were identified. The results were confirmed by repeated phenotypic DST and sequencing of the RRDR of rpoB and the entire rpoC gene.

Factors associated with rpoC mutation

Logistic regression models were used to identify the possible risk factors associated with rpoC mutation. The analyzed factors included age, gender, treatment history, strain family, RMP susceptibility, and rpoB mutation. Univariate and multivariate analyses revealed that host factors, such as age, gender, and treatment history, were not associated with mutations in the rpoC gene. The strain background Beijing family was not associated with mutations in rpoC either, whereas RMP resistance (monoresistant or MDR, pre-XDR, or XDR) and rpoB mutation at locus 531 were statistically associated with mutations in the rpoC gene (Table 2).

The sequencing of the RRDR of the rpoB gene of 151 RMP-resistant strains revealed that 144 (95.4%) of the isolates carried nonsynonymous mutations, 71 (47.0%) harbored point mutations at locus 531, 69 (45.7%) showed mutation S531L, and 2 (1.3%) had mutation S531W. Among the 71 strains with mutations at locus 531 of the rpoB gene, 36 (50.7%) also harbored rpoC mutation, which was significantly higher than that of isolates with no mutation or mutations at loci other than locus 531 of the rpoB gene (P<0.0001, Table 2).

Factors associated with MIRU-VNTR clusters

According to results of MIRU-VNTR genotyping in this study, 45 (18.4%) isolates were distributed across 20 clusters, 17 of which were composed of 2 strains each, 1 cluster was formed by 3 strains, and 2 clusters were composed of 4 strains each.

To determine possible factors associated with MTB transmissibility, we compared the proportion of relative factors between isolates belonging to MIRU-VNTR clusters and isolates with a unique profile. Host factors, such as age, gender, and treatment history, and bacterial factors, including strain family, RMP phenotypic drug resistance profile, and rpoB and rpoC mutations, were included in the logistic regression model. Multivariate logistic regression analysis revealed that host factors, including age, gender, and treatment history, were not associated with MIRU-VNTR clusters (Table 3). In the MIRU-VNTR clusters, the proportion of Beijing family strains was higher than that of non-Beijing family ones. However, this difference was not statistically significant in the multivariate regression model (P = 0.1238). Compared with RMP-susceptible isolates, the proportions of isolates in the clusters increased with high levels of drug resistance but were not statistically significant (P = 0.4091, P = 0.1613). Similarly, we did not find any association between rpoB mutation and MIRU-VNTR clusters (P = 0.8095). We investigated the link between rpoC mutations and MTB transmissibility by comparing the proportion of isolates with and without rpoC mutations in the clusters. Approximately 25.4% of the isolates harbored rpoC mutation, which was slightly higher than that of isolates without rpoC mutation (16.1%). However, this value was not statistically significant (P = 0.4519, Table 3).

We used RMP-resistant isolates as a subset to assess if rpoC mutations facilitate the transmission of RMP-resistant strains and found that the proportions of isolates with and without rpoC mutation in the MIRU-VNTR clusters were 27.3% and 18.8%, respectively. This difference was not statistically significant (OR, 1.625, 95% CI, 0.742–3.560, P = 0.2248).

Discussion

This study examined the profile of compensatory mutations in the rpoC gene of MTB isolates from patients with pulmonary TB in two municipalities and four provinces of China and investigated the possible factors associated with the transmission of MTB. The results showed that 36.4% of the RMP-resistant MTB isolates harbored nonsynonymous rpoC mutations. Of the RMP-resistant isolates with nonsynonymous mutations in the rpoC gene, 98.2% (54/55) harbored mutations in the RRDR of the rpoB gene. A recent study on MTB revealed that more than 30% of MDR clinical strains isolated from high-burden countries have compensatory mutations [¹⁹], and the proportion of compensatory mutations found in the present study is slightly larger than that reported in high-burden countries. Previous studies have revealed that RMP-resistant strains harbor rpoC mutations, but these mutations do not appear in any RMP-sensitive isolates [¹⁴,¹⁹–²¹]. In the current study, four nonsynonymous and three synonymous mutations were identified in RMP-sensitive isolates. These results are in accordance with those of a previous study conducted in China [¹⁵]. Such mutations should be studied in the future.

In this study, we identified 29 nonsynonymous single mutations and 3 double mutations in the rpoC gene. Although most of the mutations in rpoC have been previously reported [¹⁴–¹⁶,¹⁹,²¹–²⁴], we identified 12 new point mutations and 3 double mutations in rpoC.

We analyzed the diversities of rpoC mutations in different codons of the rpoB gene and found that different loci (511, 516, 522, 526, and 531) of the rpoB gene exhibited varying patterns of compensatory mutations of rpoC, except for V483G of rpoC mutation, which was identified at loci 516 and 531 of the rpoB gene. We identified 9 and 18 different types of rpoC mutations at locus 526 (including double mutation H526N and M515V) and locus 531 (including double mutation S531L and L511P) of rpoB, respectively. Mutation with S531L had a different type of rpoC mutation and a high level of diversity.

Multivariate logistic regression analysis revealed a statistically significant association between rpoC mutation and RMP resistance and mutations at locus 531 of the rpoB gene. This result is in accordance with those of previous studies [¹⁴,¹⁵,²²]. In the current study, strains carrying mutations at locus 531 of rpoB were frequently found in the isolated strains. Approximately 50.0% of isolates with point mutations at locus 531 of the rpoB gene harbored rpoC mutation, which was significantly higher than that of no mutation or mutation(s) at loci other than locus 531. An explanation for this result is compensatory mutation [¹⁹,²⁵,²⁶]. Previous studies have strongly implied that rpoC and rpoA mutations in RMP-resistant MTB are fitness compensatory mutations, and strains with these compensatory mutations show high competitive fitness not only in vitro but also in vivo [²⁷,²⁸]. Another possible reason may be related to the structure of DNA-dependent RNA polymerase. In bacteria, this enzyme comprises multiple subunits (a2bb′w), among which the two largest subunits (b′ and b) that contain collinearly arranged segments of a conserved sequence are encoded by rpoC and rpoB. The evolutionary relationship between b′ and b subunits may be particularly strong. Polymorphisms in the subunits of an enzyme may interact closely with one other to reduce initial fitness defects caused by drug-resistant conferring mutation [¹⁹,²⁹,³⁰].

The influence of bacterial factors, such as genetic, clinical, and demographic characteristics of patients, on MTB clustering was analyzed in this study. In contrast with a previous study, we did not find patients infected with the Beijing family strain to be clustered [³¹]. Our study did not identify any risk factors, such as RMP resistance profile, rpoB mutation, or treatment history of patients that were statistically significant associated with MIRU-VNTR clusters. rpoC mutation was not a risk factor significantly associated with MIRU-VNTR clusters, a result that is consistent with that of a recent report [¹⁶,²³] but contrary to those in other studies [¹⁴,¹⁵]. These heterogeneous results may be due to differences in study setting, study object, methodologies applied, and quality of the TB control program.

Our study has several limitations. First, the 15 loci for the MIRU-VNTR analysis of MTB may have decreased the discrimination power and thus overestimated the cluster rate in this study. Second, according to previous studies, rpoC mutations are more frequent than rpoA gene mutations [¹⁴].Thus, rpoC was sequenced for mutation detection, and rpoA gene mutation was not. Third, according to the findings of a previous study [³²], 95% of RMP-resistant mutations are present in the rpoB gene, and the majority of the mutations in the rpoB gene are within RRDR. Thus, this study focused on sequencing the RRDR of the rpoB gene and may have missed mutations located outside of the RRDR of the rpoB gene.

In summary, this study found a high proportion of rpoC mutations in RMP-resistant MTB isolates prevalent in China. The results showed a significant association between rpoC and rpoB mutations at locus 531, suggesting that rpoC mutation may reduce the fitness cost of RMP-resistant MTB, leading to compensated transmissibility. However, the finding on the lack of significant association of rpoC mutation with the clustering of MTB isolates does not support the role of rpoC mutation in transmissibility.

References

[1] Yang C, Shen X, Peng Y, Lan R, Zhao Y, Long B, Luo T, Sun G, Li X, Qiao K, Gui X, Wu J, Xu J, Li F, Li D, Liu F, Shen M, Hong J, Mei J, DeRiemer K, Gao Q. Transmission of Mycobacterium tuberculosis in China: a population-based molecular epidemiologic study. Clin Infect Dis 2015; 61(2): 219–227

[2] Yang C, Luo T, Shen X, Wu J, Gan M, Xu P, Wu Z, Lin S, Tian J, Liu Q, Yuan Z, Mei J, DeRiemer K, Gao Q. Transmission of multidrug-resistant Mycobacterium tuberculosis in Shanghai, China: a retrospective observational study using whole-genome sequencing and epidemiological investigation. Lancet Infect Dis 2017; 17(3): 275–284

[3] Gagneux S, Long CD, Small PM, Van T, Schoolnik GK, Bohannan BJ. The competitive cost of antibiotic resistance in Mycobacterium tuberculosis. Science 2006; 312(5782): 1944–1946

[4] Mariam DH, Mengistu Y, Hoffner SE, Andersson DI. Effect of rpoB mutations conferring rifampin resistance on fitness of Mycobacterium tuberculosis. Antimicrob Agents Chemother 2004; 48(4): 1289–1294

[5] Knight GM, Colijn C, Shrestha S, Fofana M, Cobelens F, White RG, Dowdy DW, Cohen T. The distribution of fitness costs of resistance-conferring mutations is a key determinant for the future burden of drug-resistant tuberculosis: a model-based analysis. Clin Infect Dis 2015; 61(Suppl 3): S147–S154

[6] Cohen T, Murray M. Modeling epidemics of multidrug-resistant M. tuberculosis of heterogeneous fitness. Nat Med 2004; 10(10): 1117–1121

[7] Blower SM, Chou T. Modeling the emergence of the ‘hot zones’: tuberculosis and the amplification dynamics of drug resistance. Nat Med 2004; 10(10): 1111–1116

[8] Tang K, Sun H, Zhao Y, Guo J, Zhang C, Feng Q, He Y, Luo M, Li Y, Sun Q. Characterization of rifampin-resistant isolates of Mycobacterium tuberculosis from Sichuan in China. Tuberculosis (Edinb) 2013; 93(1): 89–95

[9] Brandis G, Pietsch F, Alemayehu R, Hughes D. Comprehensive phenotypic characterization of rifampicin resistance mutations in Salmonella provides insight into the evolution of resistance in Mycobacterium tuberculosis. J Antimicrob Chemother 2015; 70(3): 680–685

[10] Hughes D, Brandis G. Rifampicin resistance: fitness costs and the significance of compensatory evolution. Antibiotics (Basel) 2013; 2(2): 206–216

[11] Brandis G, Wrande M, Liljas L, Hughes D. Fitness-compensatory mutations in rifampicin-resistant RNA polymerase. Mol Microbiol 2012; 85(1): 142–151

[12] Maisnier-Patin S, Andersson DI. Adaptation to the deleterious effects of antimicrobial drug resistance mutations by compensatory evolution. Res Microbiol 2004; 155(5): 360–369

[13] Brandis G, Hughes D. Genetic characterization of compensatory evolution in strains carrying rpoB Ser531Leu, the rifampicin resistance mutation most frequently found in clinical isolates. J Antimicrob Chemother 2013; 68(11): 2493–2497

[14] de Vos M, Müller B, Borrell S, Black PA, van Helden PD, Warren RM, Gagneux S, Victor TC. Putative compensatory mutations in the rpoC gene of rifampin-resistant Mycobacterium tuberculosis are associated with ongoing transmission. Antimicrob Agents Chemother 2013; 57(2): 827–832

[15] Li QJ, Jiao WW, Yin QQ, Xu F, Li JQ, Sun L, Xiao J, Li YJ, Mokrousov I, Huang HR, Shen AD. Compensatory mutations of rifampin resistance are associated with transmission of multidrug-resistant Mycobacterium tuberculosis Beijing genotype strains in China. Antimicrob Agents Chemother 2016; 60(5): 2807–2812

[16] Liu Q, Zuo T, Xu P, Jiang Q, Wu J, Gan M, Yang C, Prakash R, Zhu G, Takiff HE, Gao Q. Have compensatory mutations facilitated the current epidemic of multidrug-resistant tuberculosis? Emerg Microbes Infect 2018; 7(1): 98–105

[17] Reed MB, Pichler VK, McIntosh F, Mattia A, Fallow A, Masala S, Domenech P, Zwerling A, Thibert L, Menzies D, Schwartzman K, Behr MA. Major Mycobacterium tuberculosis lineages associate with patient country of origin. J Clin Microbiol 2009; 47(4): 1119–1128

[18] Supply P, Allix C, Lesjean S, Cardoso-Oelemann M, Rüsch-Gerdes S, Willery E, Savine E, de Haas P, van Deutekom H, Roring S, Bifani P, Kurepina N, Kreiswirth B, Sola C, Rastogi N, Vatin V, Gutierrez MC, Fauville M, Niemann S, Skuce R, Kremer K, Locht C, van Soolingen D. Proposal for standardization of optimized mycobacterial interspersed repetitive unit-variable-number tandem repeat typing of Mycobacterium tuberculosis. J Clin Microbiol 2006; 44(12): 4498–4510

[19] Comas I, Borrell S, Roetzer A, Rose G, Malla B, Kato-Maeda M, Galagan J, Niemann S, Gagneux S. Whole-genome sequencing of rifampicin-resistant Mycobacterium tuberculosis strains identifies compensatory mutations in RNA polymerase genes. Nat Genet 2011; 44(1): 106–110

[20] Farhat MR, Shapiro BJ, Kieser KJ, Sultana R, Jacobson KR, Victor TC, Warren RM, Streicher EM, Calver A, Sloutsky A, Kaur D, Posey JE, Plikaytis B, Oggioni MR, Gardy JL, Johnston JC, Rodrigues M, Tang PK, Kato-Maeda M, Borowsky ML, Muddukrishna B, Kreiswirth BN, Kurepina N, Galagan J, Gagneux S, Birren B, Rubin EJ, Lander ES, Sabeti PC, Murray M. Genomic analysis identifies targets of convergent positive selection in drug-resistant Mycobacterium tuberculosis. Nat Genet 2013; 45(10): 1183–1189

[21] Casali N, Nikolayevskyy V, Balabanova Y, Ignatyeva O, Kontsevaya I, Harris SR, Bentley SD, Parkhill J, Nejentsev S, Hoffner SE, Horstmann RD, Brown T, Drobniewski F. Microevolution of extensively drug-resistant tuberculosis in Russia. Genome Res 2012; 22(4): 735–745

[22] Casali N, Nikolayevskyy V, Balabanova Y, Harris SR, Ignatyeva O, Kontsevaya I, Corander J, Bryant J, Parkhill J, Nejentsev S, Horstmann RD, Brown T, Drobniewski F. Evolution and transmission of drug-resistant tuberculosis in a Russian population. Nat Genet 2014; 46(3): 279–286

[23] Feng JY, Jarlsberg LG, Salcedo K, Rose J, Janes M, Lin SG, Osmond DH, Jost KC, Soehnlen MK, Flood J, Graviss EA, Desmond E, Moonan PK, Nahid P, Hopewell PC, Kato-Maeda M. Clinical and bacteriological characteristics associated with clustering of multidrug-resistant tuberculosis. Int J Tuberc Lung Dis 2017; 21(7): 766–773

[24] Yun YJ, Lee JS, Yoo JC, Cho E, Park D, Kook YH, Lee KH. Patterns of rpoC mutations in drug-resistant Mycobacterium tuberculosis isolated from patients in South Korea. Tuberc Respir Dis (Seoul) 2018; 81(3): 222–227

[25] Levin BR, Perrot V, Walker N. Compensatory mutations, antibiotic resistance and the population genetics of adaptive evolution in bacteria. Genetics 2000; 154(3): 985–997

[26] Moura de Sousa J, Balbontín R, Durão P, Gordo I. Multidrug-resistant bacteria compensate for the epistasis between resistances. PLoS Biol 2017; 15(4): e2001741

[27] Gagneux S, Long CD, Small PM, Van T, Schoolnik GK, Bohannan BJ. The competitive cost of antibiotic resistance in Mycobacterium tuberculosis. Science 2006; 312(5782): 1944–1946

[28] Gagneux S. Fitness cost of drug resistance in Mycobacterium tuberculosis. Clin Microbiol Infect 2009; 15(Suppl 1): 66–68

[29] Lane WJ, Darst SA. Molecular evolution of multisubunit RNA polymerases: structural analysis. J Mol Biol 2010; 395(4): 686–704

[30] Opalka N, Brown J, Lane WJ, Twist KA, Landick R, Asturias FJ, Darst SA. Complete structural model of Escherichia coli RNA polymerase from a hybrid approach. PLoS Biol 2010; 8(9): e1000483

[31] Chen YY, Chang JR, Wu CD, Yeh YP, Yang SJ, Hsu CH, Lin MC, Tsai CF, Lin MS, Su IJ, Dou HY. Combining molecular typing and spatial pattern analysis to identify areas of high tuberculosis transmission in a moderate-incidence county in Taiwan. Sci Rep 2017; 7(1): 5394

[32] Zaw MT, Emran NA, Lin Z. Mutations inside rifampicin-resistance determining region of rpoB gene associated with rifampicin-resistance in Mycobacterium tuberculosis. J Infect Public Health 2018; 11(5): 605–610

RIGHTS & PERMISSIONS

Higher Education Press and Springer-Verlag GmbH Germany, part of Springer Nature