The occurrence of rare rpoB mutations in rifampicin-resistant clinical Mycobacterium tuberculosis isolates from Kuwait

Abstract

Mutations associated with rifampicin (RIF) resistance in two regions of the rpoB gene were studied by line probe (INNO-LiPA Rif. TB) assay (LiPA) and/or DNA sequencing in 32 RIF-resistant and 21 RIF-susceptible Mycobacterium tuberculosis strains isolated from 53 tuberculosis patients in Kuwait. The LiPA identified all susceptible strains as RIF sensitive, and the DNA sequences of the 81 bp rifampicin resistance-determining region (RRDR) and the N-terminal region of the rpoB gene from five isolates were identical to the wild-type sequences from M. tuberculosis H₃₇Rv. The LiPA identified 24 of 32 (75%) phenotypically documented RIF-resistant M. tuberculosis isolates as RIF resistant, with specific detection of mutation in 19 isolates, whilst 8 strains were identified as RIF susceptible. DNA sequencing of the RRDR confirmed the results for the 19 RIF-resistant isolates and identified precise mutations in five isolates for which specific base changes could not be determined by LiPA, as well as identifying a mutation (insertion 514TTC) in three isolates and no mutation in five of the eight isolates that were detected as RIF susceptible by LiPA. Two of the latter five isolates contained a V146F mutation in the N-terminal region of the rpoB gene and were recovered from patients of Middle Eastern origin. These analyses identified 11 different mutations in two regions of the rpoB gene. The majority of the isolates carrying identical or no mutations in the rpoB gene exhibited unique DNA banding patterns in fingerprinting studies. The occurrence of rare mutations in the rpoB gene, namely insertion 514TTC in 3 (9%) and V146F in 2 (6%) of 32 RIF-resistant M. tuberculosis isolates from Kuwait, is the highest reported so far.

Introduction

Tuberculosis (TB) is a major infectious disease killing nearly two million people, mostly in developing or underdeveloped countries, each year [1]. The incidence of TB has also increased in developed countries, particularly involving HIV-positive individuals, immigrants/expatriate workers from countries in which TB is endemic, the homeless and prison inmates [1], [2]. The emergence of drug-resistant strains of Mycobacterium tuberculosis has become a major obstacle in the management of TB, and the incidence of drug-resistant TB varies considerably around the world [3]. The number of new TB cases that are resistant to rifampicin and isoniazid (the two most effective anti-TB drugs) is the most important measure of the problem of drug resistance in a given country [4]. Treatment of multidrug-resistant (MDR) (resistant to at least rifampicin and isoniazid) strains of M. tuberculosis (MDR-TB) requires replacement of the standard short-course regimens with expensive, toxic and less effective second-line drugs and often results in high clinical relapse rates [4], [5]. Phenotypic drug susceptibility testing by the radiometric method requires 4–14 days after the primary culture has been isolated [6]. This may result in inadequate treatment of persons infected with drug-resistant strains who will, therefore, remain infectious for longer periods than those infected with drug-susceptible strains.

Rifampicin (RIF) is a key first-line anti-TB drug and more than 90% of RIF-resistant M. tuberculosis isolates are also resistant to isoniazid [7]. Thus, rapid detection of RIF resistance could also identify MDR strains. The molecular basis of resistance to RIF in nearly 95% of RIF-resistant strains is due to missense mutations or, less commonly, to small in-frame deletions or insertions in or around the 81 bp rifampicin resistance-determining region (RRDR) (corresponding to codons 507–533, Escherichia coli numbering system) of the rpoB gene, encoding the β-subunit of DNA-dependent RNA polymerase [7], [8]. Thus, several molecular methods have been developed for the rapid (1–2 days) detection of rpoB mutations in this region [8], [9], [10], [11], [12], [13], [14], [15], [16], [17] and one of these, the line probe (INNO-LiPA Rif. TB) assay (LiPA) is available commercially [10], [11]. Resistance in nearly 5% of the RIF-resistant isolates either involves mutations in other regions of the rpoB gene such as codon V146 or is due to unknown mechanisms [7], [18], [19].

More than 60 mutations within or near the RRDR of the rpoB gene have been unambiguously characterised by DNA sequencing and, depending upon the region/ethnic origin of the infected TB patients, mutations at codon 531, 526 or 516 were found to be most common [20]. Although these mutations may also serve as epidemiological markers, the pattern of these mutations in countries with a low incidence of RIF-resistant TB, particularly those with large immigrant/expatriate populations originating from TB endemic countries, may vary owing to different sampling (ethnic origin/region) of the infected patients [9], [20], [21]. This is due to the fact that most TB cases, including RIF-resistant TB, in these countries occur in non-native populations originating from TB endemic countries and may involve reactivation of previously acquired infection [22].

Kuwait, an Arabian country in the Persian Gulf region with nearly 25 cases per 100 000 population, has a low incidence of TB [23]. However, Kuwait has a large expatriate workforce originating from TB endemic countries. Approximately 600 patients, nearly 80% being expatriates, are diagnosed with active disease every year. Nearly 1% of the M. tuberculosis isolates are RIF resistant or MDR-TB strains and nearly all of these are isolated from expatriate patients [20], [23], [24]. We have previously shown that 18 of 19 RIF-resistant strains isolated over a 3-year-period (1997–2000) from Kuwait could be rapidly identified by LiPA and that most of these (11 of 19) isolates contained mutations at codon 526 of the rpoB gene [20]. Since nearly all (18 of 19) RIF-resistant M. tuberculosis strains were recovered from expatriate patients and since the expatriate population is constantly changing, this study was carried out to characterise the rpoB mutations in RIF-resistant M. tuberculosis isolates recovered from TB patients in the subsequent 4-year-period (2001–2004) by LiPA and DNA sequencing. The N-terminal region of the rpoB gene was also analysed by DNA sequencing for RIF-resistant M. tuberculosis isolates that did not contain a mutation in or around the RRDR. Molecular fingerprinting of the M. tuberculosis isolates carrying identical mutation or no mutation in the rpoB gene was carried out by double-repetitive-element polymerase chain reaction (DRE-PCR) to determine strain relatedness.