Journal of Antimicrobial Chemotherapy Advance Access published May 22, 2014
J Antimicrob Chemother doi:10.1093/jac/dku161
Detection of mutations associated with isoniazid resistance in multidrug-resistant Mycobacterium tuberculosis clinical isolates Tomasz Jagielski1*, Zofia Bakuła1, Katarzyna Roeske1, Michał Kamin´ski1, Agnieszka Napio´rkowska2, Ewa Augustynowicz-Kopec´2, Zofia Zwolska2 and Jacek Bielecki1 Department of Applied Microbiology, Institute of Microbiology, Faculty of Biology, University of Warsaw, I. Miecznikowa 1, 02-096 Warsaw, Poland; 2Department of Microbiology, National Tuberculosis and Lung Diseases Research Institute, Płocka 26, 01-138 Warsaw, Poland *Corresponding author. Tel: +48 (0) 22 55 41 312; Fax: +48 (0) 22 55 41 402; E-mail: [email protected]
Received 27 January 2014; returned 24 March 2014; revised 7 April 2014; accepted 14 April 2014 Objectives: To determine the prevalence of isoniazid resistance-conferring mutations among multidrug-resistant (MDR) isolates of Mycobacterium tuberculosis from Poland. Methods: Nine genetic loci, including structural genes (katG, inhA, ahpC, kasA, ndh, nat and mshA) and regulatory regions (i.e. the mabA-inhA promoter and oxyR-ahpC intergenic region) of 50 MDR M. tuberculosis isolates collected throughout Poland were PCR-amplified in their entirety and screened for mutations by direct sequencing methodology. Results: Forty-six (92%) MDR M. tuberculosis isolates had mutations in the katG gene, and the katG Ser315Thr substitution predominated (72%). Eight (16%) isolates (six with a mutated katG allele) had mutations in the inhA promoter region and two such isolates also had single inhA structural gene mutations. Mutations in the oxyR-ahpC locus were found in five (10%) isolates, of which all but one had at least one additional mutation in katG. Mutations in the remaining genetic loci (kasA, ndh, nat and mshA) were detected in 12 (24%), 4 (8%), 5 (10%) and 17 (34%) MDR isolates, respectively. All non-synonymous mutants for these genes harboured mutations in katG. One isolate had no mutations in any of the analysed loci. Conclusions: This study accentuates the usefulness of katG and inhA promoter mutations as predictive markers of isoniazid resistance. Testing only for katG 315 and inhA – 15 mutations would detect isoniazid resistance in 84% of the MDR M. tuberculosis sample. This percentage would increase to 96% if the sequence analysis was extended to the entire katG gene. Analysis of the remaining genetic loci did not contribute greatly to the identification of isoniazid resistance. Keywords: M. tuberculosis, tuberculosis, TB, multidrug resistance, sequence analysis, spoligotyping, first-line anti-TB drugs
Introduction Isoniazid is a key component of drug regimens used worldwide for the treatment of tuberculosis (TB). However, the future clinical utility of the drug has been jeopardized by the increasing prevalence of isoniazid resistance among Mycobacterium tuberculosis strains. As many as 13.3% of all TB cases are resistant to isoniazid, either alone or in combination with other drugs.1 Although the molecular mechanisms of isoniazid resistance are not fully understood, numerous studies have linked them to distinct mutations in various genetic loci of the M. tuberculosis genome. Three loci most commonly affected are the katG gene, encoding catalaseperoxidase, which transforms isoniazid to its pharmacologically active form,2 – 8 the inhA gene (with the mabA-inhA promoter region), coding for enoyl-acyl carrier protein (ACP) reductase, a
mycolic acid biosynthetic pathway enzyme2 – 8 and the ahpC locus, including the structural gene of alkyl hydroperoxide reductase (ahpC), involved in the cellular response to oxidative stress, and the upstream regulatory region (oxyR-ahpC).5 – 9 Other genes that have been shown to be associated with isoniazid resistance include kasA, coding for b-ketoacyl ACP synthase, another enzyme involved in mycolic acid biosynthesis,5,10 ndh, coding for NADH dehydrogenase,5,11,12 and, more recently, nat and mshA, coding for arylamine N-acetyltransferase (NAT),13 which, through acetylation of isoniazid, renders the drug therapeutically inactive, and the glycosyltransferase involved in the biosynthesis of mycothiol.14 The aim of this study was to investigate the prevalence of mutations in a panel of genetic loci implicated in conferring isoniazid resistance on MDR M. tuberculosis clinical isolates from Poland.
# The Author 2014. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail: [email protected]
1 of 7
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
1
Jagielski et al.
Materials and methods Strains and drug susceptibility testing
Primer Locus
designation
nucleotide sequence (5′ 3′ )
Product size (bp)
ahpC
Fw: AhpC-F Rev: AhpC-R AhpC-seqFa AhpC-seqRa
AGCGACTTCACGGCACGATG TGATGTCTTTGGCGTACTCG GGAATGTCGCAACCAAATG GGCCTTGAGCTTTTCTATAC
732
kasA
Fw: kasA1F Rev: kasA1R Fw: kasA2F Rev: kasA2R Fw: kasA3F Rev: kasA3R
CGTTCAGGCGAGGCTTGAGG CCGGTCTGGATCGACCTCCG GGACAGCTATGGGAGTCCGC ACCCAGCAATCGGGCCAACG GCACGCCAAAGCCCGTGGCG GGGCCTCGCGACCCGCGATG
351
Fw: mshA-F Rev: mshA-R mshAFMa mshARMa
CTTCGGTTCCTGCAAGGATGG CGAGCGCATTCTGGATCACAC AGATCGACACCGGGATGGAC CGTCAACACCGACGATGAAG
mshA
528 523 1505
Fw, forward; Rev, reverse. All primers used for amplification were also used for sequencing. a Additional intragenic primers used for sequencing.
DNA extraction Genomic DNA was extracted by the cetyl-trimethyl ammonium bromide (CTAB) method.17
gene, KF549665 and KF549666 for the ndh gene, KF549668 – KF549670 for the nat gene and KF564643 and KF564644 for the mshA gene.
Sequencing strategy
Statistical analysis
Seven structural genes (katG, inhA, ahpC, kasA, ndh, nat, mshA) and two regulatory regions (the mabA-inhA promoter region and the oxyR-ahpC intergenic region) were screened for mutations by direct sequencing of each locus in its entirety. The oligonucleotide primers used for PCR amplification of the respective genetic loci were either based on previously reported studies or newly designed in the present study [Table 1 and Table S1 (available as Supplementary data at JAC Online)]. The PCR mixtures were prepared with a TopTaq Master Mix Kit (Qiagen) and run under optimized thermocycling conditions on a LabCycler (SensoQuest). Amplification products were analysed by electrophoresis in 1.5% agarose gels, purified with a Clean-Up Kit (A&A Biotechnology) and sequenced by using a BigDye ver. 3.1 Terminator Cycle Sequencing Kit (Applied Biosystems) in an ABI 3130xl Genetic Analyzer (Applied Biosystems). To confirm the presence of mutations, sequencing was done in both directions with the same primer pairs as those used in the amplification reactions and, occasionally, with additional, internal primers (Table 1). Sequence data were assembled and analysed with ChromasPro (ver. 1.7.1) software (Technelysium), and the resulting consensus sequences were aligned against the wild-type sequences of the respective genetic loci of M. tuberculosis reference strain H37Rv (TubercuList; http:// genolist.pasteur.fr/TubercuList/) using the blastn algorithm (http://blast.ncbi. nlm.nih.gov/).
The associations between individual mutations and between mutations and MICs of isoniazid or spoligotyping families were assessed with the x2 test and ANOVA, respectively. A P value of ,0.05 was considered statistically significant.
Nucleotide sequence accession numbers The sequences with novel mutations were deposited in GenBank (National Center for Biotechnology Information; http://www.ncbi.nlm.nih.gov/) under the following accession numbers: KF539901–KF539910 for the katG gene, KF563064 for the inhA gene, KF563062 and KF563063 for the oxyR-aphC regulatory region, KF549667 for the ahpC gene, KF563065 for the kasA
2 of 7
Results Of the 50 MDR TB isolates examined, 46 (92%) had mutations within the katG gene. The katG codon 315 was affected in 36 (72%) isolates, with a GC change at nucleotide position 944, conferring a Ser315Thr amino acid substitution, found most frequently (33 isolates; 66%). Mutations at other katG codons were distributed among 13 (26%) isolates (Table S2, available as Supplementary data at JAC Online). Forty-four (88%) isolates had a single point mutation in the katG gene, whereas five (10%) isolates had a double katG mutation and one isolate had a triple katG mutation. Two isolates with two and three katG mutations, albeit with a wild-type 315 codon, had unique termination mutations, resulting in a truncated and non-functional catalase peptide. Both these isolates presented the highest MICs of isoniazid (80 and 100 mg/L). For the remaining 44 (88%) isolates with katG mutations, the isoniazid MICs were within the range 0.5 –10 mg/L. Eight (16%) isolates (six with a mutated katG allele) had mutations in the mabA-inhA locus (MICs 1 – 10 mg/L). Transitions at nucleotide positions –15 (CT) and –8 (TC) of the inhA promoter region were observed in seven (14%) isolates and one isolate,
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
A total of 50 MDR M. tuberculosis isolates, collected at the National Tuberculosis and Lung Diseases Institute in Warsaw, were included in the study. The isolates were recovered from 46 unrelated pulmonary TB patients (40 men and 6 women; age range, 31 – 79 years; median age, 50.5 years) from across Poland. These patients represented all bacteriologically confirmed MDR TB cases reported in Poland in 2004 [TB cases registered (total), n ¼9493; TB notification rate, n¼24.9/100000 population]. Previously, isolates from these patients had been genotyped by means of spoligotyping.15 Seven (15.2%) of the patients had not previously been treated for TB, and were thus classified as new MDR TB cases, whereas 39 (84.8%) had received anti-TB treatment for at least 1 month in the past and were defined as acquired MDR TB cases. Primary isolation, culturing and species identification were carried out by standard mycobacteriological procedures, essentially as described elsewhere.16 Initial drug susceptibility testing against first-line anti-TB drugs was performed using the 1% proportion method on Lo¨wenstein –Jensen (LJ) medium, with the following critical drug concentrations: isoniazid, 0.2 mg/L; rifampicin, 40 mg/L; streptomycin, 4 mg/L; and ethambutol, 2 mg/L.16 Subsequently, the MICs of isoniazid were determined in LJ medium containing 2-fold incremental concentrations of the drug ranging from 0.05 to 100 mg/L. The MIC was defined as the lowest drug concentration that inhibited .99% of the bacterial population in the drug-free control culture.
Table 1. Primers used for PCR amplification and sequencing of the ahpC, kasA and mshA genes
JAC
Isoniazid resistance mutations in MDR TB isolates
ST53, none of the eight isolates, including those from the same patients, had identical mutational patterns, differing by at least one nucleotide polymorphism. Cluster ST891 comprised three isolates with an identical mutational pattern and two isolates that differed by the presence of additional mutations in the mshA gene. Another cluster (ST1557) contained four isolates, of which only one was different by a single polymorphism in the oxyR-ahpC locus. Of the four Beijing genotype isolates, two had identical mutational profiles, whereas the other two differed by single nucleotide polymorphisms at the katG, mabA-inhA and inhA loci. Finally, two isolates of cluster ST37 had three different polymorphisms at either the katG or the mabA-inhA locus. No association was found between specific mutation frequency and type of resistance (primary versus acquired). Also, no association was observed between any mutation and a spoligotype family, except that mutations Arg463Leu (GT at position 1388) and Ala187Val (CT at position 560) in the katG and mshA genes, respectively, were found only in Beijing genotype isolates. Most of the MDR TB isolates classified in the T, Latin-American Mediterranean (LAM) and Beijing families presented additional resistance to streptomycin (P ¼ 0.03).
Discussion This work provides a description of mutations in a large set of genetic loci associated or speculated to be associated with isoniazid resistance in M. tuberculosis. The results reported here are in many respects consistent with and comparable to those of previous studies. For instance, mutations in the katG gene predominated (92% of all MDR isolates) and mutations at codon 315 in the katG gene were detected in nearly two-thirds (72%) of the MDR strains. High detection rates of katG 315 mutations in MDR strains have been reported from China (72.7%),18 Mexico (67.6%)19 and countries neighbouring Poland, including Germany (88.4%),20 Lithuania (89.7%),3 Latvia (94.1%)21 and Russia (93.5%).2 Mutations at katG codon 315, in particular the Ser315Thr substitution, have been shown to confer a selective advantage by reducing the ability of KatG to activate isoniazid, while preserving substantial catalaseperoxidase activity to protect isoniazid-resistant tubercle bacilli from oxidative stress.22 Indeed, all but two (34/36; 94.4%) MDR isolates in this study, carrying the katG Ser315Thr mutation, were catalase-positive (data not shown).23 Mutations at katG 315 have repeatedly been associated with a high level of isoniazid resistance (≥4 mg/L).8,24 In this study, the MICs of isoniazid for isolates with the katG 315 mutation were within the range 1 – 10 mg/L, with the majority of isolates (27/36; 75%) exhibiting MICs of 2.5 mg/L (P,0.01), a value corresponding to a moderate level of resistance. Apart from codon 315, mutations in the katG gene were identified at 11 other codons. Mutations at eight of these codons had not been reported previously (Table S2). The relevance of the newly described katG mutations to isoniazid resistance can hardly be established as they were accompanied by other mutations in katG and/or other genetic loci. The only exception was substitution Leu101Pro, observed as the sole mutation in one MDR strain. That this mutation confers isoniazid resistance seems likely, given that the affected amino acid residue is located in the close vicinity of the KatG active site.25 Isoniazid resistance was also associated with termination mutations (Lys46STOP, Glu454STOP) in two
3 of 7
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
respectively. Two isolates with a CT change at the –15 position of the inhA promoter also contained single substitutions in the inhA structural gene, resulting in Ile194Thr (TC at position 649) and Ser94Ala (TG at position 289) replacements. Mutations in the oxyR-ahpC intergenic region were found in five (10%) isolates; two isolates had a single nucleotide change at position – 48 (GA); one was at position – 54 (CT) and the other was at position –57 (CT) upstream of the transcriptional initiation site of the ahpC gene. A three-nucleotide change (TCAATC) in the ahpC promoter region was demonstrated in one isolate (MIC 10 mg/L). Except for the latter, all isolates with mutations in the oxyR-ahpC intergenic region had at least one additional mutation in the katG gene. All isolates with mutations in the oxyR-ahpC region had no alterations in the mabA-inhA operon. A silent mutation, Ile114Ile (AC at position 342), was the sole ahpC structural gene mutation detected and was present in only one isolate (MIC 5 mg/L). This isolate also carried single mutations in the katG gene and the inhA promoter. Two types of substitution mutation were identified in the kasA gene. They were both GA transitions at positions 805 (Gly269Ser) and 961 (Ala321Thr), found in 11 (22%) isolates and one isolate, respectively. All 12 kasA mutants had additional mutations in the katG gene but no mutations in the mabA-inhA and oxyR-ahpC loci. The MICs of isoniazid for these isolates were 1 or 2.5 mg/L. Single ndh gene mutations were detected in four (8%) isolates. The mutations occurred at nucleotide position 53 (TC; Val18Ala) in two isolates, 843 (GC; Ser281Ser) in one isolate and 898 (GC; Ala300Pro) in another isolate. Three isolates with an altered ndh gene sequence contained mutations in the katG gene and two isolates, including that with the wild-type katG allele, contained mutations in the ahpC promoter region. Mutations in the nat gene were detected in five (10%) isolates. Two isolates had the same termination mutation at codon 173 (GA at position 518), two had Gly207Arg replacements (GA at position 619) and one isolate had an Ala103Gly replacement (CG at position 308). All of the nat mutants carried single katG mutations. In addition, mutations in the ndh gene and the ahpC promoter were present in three and two of these isolates, respectively. Four types of point mutations were identified in the mshA gene of 17 (34%) isolates (MICs ≥1 mg/L). The mutated sites were at positions 332 (AG; Asn111Ser) in 11 (22%) isolates, 560 (CT; Ala187Val) in 4 (8%) isolates and 1085 (CT; Ala362Val) in 1 isolate. One isolate had an insertion of a cytosine at position 1283, producing a frameshift mutation. All isolates carrying mshA mutations had an altered katG gene sequence. The mshA mutants had additional changes in one (kasA, three isolates; ahpC promoter, two isolates; nat, one isolate) or two (inhA promoter and coding region, two isolates; ndh and nat, one isolate) loci. Of all MDR TB isolates tested, 15 (30%) had mutations in only one genetic locus (i.e. katG in 13 isolates and mabA-inhA in 2 isolates). Twenty-one (42%) isolates harboured mutations in two, 11 (22%) in three, and 2 (4%) in four genetic loci. One isolate (MIC 1 mg/L) had the wild-type sequences at all nine loci analysed. Comparison of spoligotyping results with mutational profiles at the isoniazid resistance-associated loci showed that only two of six spoligotype clusters [sequence types (STs) 1051 and F] were entirely homogeneous with regard to their mutational patterns (Figure 1). In the largest cluster, represented by spoligotype
E A sh
5895 101*
A 775
1377
2333
H1
2233
47
H1
U H
1885 T1
T
708
X1
5325*
1558
T1
X
3832
442
T3
456* 1233
40 50
T4 H3
9310
180
H3
1114
253
T1
2688
42
456* H T
1233 9310 1114 2688 1612
891
LAM9
LAM
446 2086
874
874
524*
524* B
U
4991 17170*
3298
3298 1051
T1
T
434
1746
S
S
434
469
C
U
469
647
280
T
647
124
D
U
124
2575
E
U
2575
947
947
T1_RUS2
12489*
F
U
1334
3020* Beijing
BEIJING
1334 K131 3430 STOP -46 Val 68 Gly Trp 91 Arg Leu 101 Pro Met 126 Ile Arg128 Gln Pro 131 Arg Pro 131 Gln Asp 194 Ty Ala 234 Gly Ser 315 Ile Ser 315 Asn Ser 315 Th Ser 315 Th Glu 454 STOP Arg 463 Leu NA NA Ser 94 Ala Pro 136 Pro Ile 194 Th NA NA NA NA Ile 114 Ile Gly 269 Se Ala 321 Th Val 18 Ala Ser 281 Se Ala 300 Pro Ala 103 Gly Trp 173 STOP Gly 207 Arg Asn 111 Se Ala 187 Va Ala 362 Va Frameshif
3430
11420 1324
G
3020* K131
B
535 3832
LAM9
446
1324
R
2497
1612
0.1
103 11844
T3
11420
ST
4619 T
37
12489*
F
794
T1
103
17170*
EM
6679 53
11844
4991
RI
3312
6679
2086
73 5325*
3312
535
4202 590
73
2497
2233
4365
590
4619
101* 1377
692 1557
4365
794
5895
1885
692 4202
IN H
m
na t
nd h
ox
yR ah ahp p C ka C sA
F
135 insT 203 T>G 271 T>C 302 T>C 378 G>A 383 G>A 392 C>G 392 C>A 580 G>T 701 C>G 944 G>T 944 G>A 944 G>C 945 C>T 1360 G>T 1388 G>T -15 C>T -8 T>C 289 T>G 408 G>A 581 T>C -48 G>A -54 C>T -57 C>T -82/-80 TCA>ATC 342 A>C 805 G>A 961 G>A 53 T>C 843 G>C 898 G>C 308 C>G 518 G>A 619 G>A 332 A>G 560 C>T 1085 C>T 1283 insC
ka t
ab Ain hA hA
A
in
D
m
C
G
B
Jagielski et al.
4 of 7
A
Figure 1. A dendrogram based on spoligotype patterns (DendroUPGMA, Jaccard index, UPGMA) along with 38 mutational changes within the nine genetic loci of the MDR M. tuberculosis strains investigated. A, strain no.; B, spoligotype pattern (binary); C, spoligotype signature (shared international type, SIT); D, phylogenetic clade; E, mutational pattern (data were coded in binary format, with black and white circles representing the presence and absence of an individual mutation, respectively; mutations at the nucleotide and amino acid level are depicted at the top and the bottom of the panel, accordingly); F, drug resistance pattern (resistance and susceptibility to a given drug is presented as black and white rectangles, respectively). INH, isoniazid; RIF, rifampicin; EMB, ethambutol, STR, streptomycin. Asterisks next to the strain numbers indicate new MDR TB cases (primary MDR resistance).
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
Isoniazid resistance mutations in MDR TB isolates
resistance is unlikely as this mutation was accompanied by an alteration in katG. The mechanism behind the involvement of the ndh mutations in resistance to isoniazid is probably an accumulation of intracellular NADH, which competitively inhibits the binding of the isoniazid – NAD adduct to InhA or the peroxidation of isoniazid by KatG.12 Two non-synonymous ndh mutations were detected in three MDR isolates in this study. One mutation (Val18Ala), shared by two isolates, has already been reported in both isoniazid-resistant and -susceptible tubercle bacilli.5,10 – 12 Both these Val18Ala mutants had a Ser315Thr substitution mutation in katG, and isoniazid resistance was rather due to the latter mutation. Likewise, mutation in codon 128 of katG could be more responsible for the isoniazid resistance in one isolate with a novel ndh mutation (Ala300Pro) than this mutation itself. Although the role of nat mutations in isoniazid resistance remains obscure, it has been speculated that they might increase the affinity of the NAT enzyme for the drug, thereby accelerating its inactivation by acetylation.13 Three nat mutations in five MDR isolates were observed in this study. The previously identified Gly207Arg mutation was shown to reduce the activity of the NAT enzyme towards isoniazid.29 Reduced NAT activity must also be expected as a result of the newly discovered termination mutation (Trp173STOP). Therefore, the only nat mutation that may somehow contribute to isoniazid resistance is the Ala103Gly substitution, found in one MDR isolate. Yet the effect of this mutation is masked by the co-occurrence, in this isolate, of two other mutations, at the katG and mshA loci. In order to verify the recently proposed concept that mutations in mycothiol biosynthesis genes may contribute to isoniazid resistance, the MDR isolates were screened for mutations in mshA, a gene encoding one of the mycothiol biosynthesis enzymes. Although mshA mutations occurred in one-third of the MDR isolates (17/50), the relevance of these mutations to isoniazid resistance is unclear, since they were always accompanied by mutations in other loci (e.g. all the mshA mutants carried a mutated katG allele). Among the mshA mutations, greater interest should be paid to a novel frameshift mutation (insC 1283) found in an isolate with a high MIC of isoniazid (10 mg/L). This isolate also had Ser315Thr and Val18Ala substitutions in the katG and ndh genes, respectively, but another isolate with these two mutations and with no other mutation exhibited an MIC of 2.5 mg/L. It is thus plausible that a severely defective mshA gene (e.g. due to a frameshift mutation) contributes to the increase in isoniazid resistance. No correlation was observed between any mutation and spoligotype family. Noticeably, however, mutations Arg463Leu and Ala187Val in the katG and mshA genes, respectively, were found exclusively in Beijing genotype isolates. Importantly, these isolates were demonstrated to be unrelated upon mycobacterial interspersed repetitive unit-variable-number tandem repeat (MIRU-VNTR) analysis and IS6110 restriction fragment length polymorphism (RFLP) profiling (data not shown). Whether these polymorphisms are associated with the Beijing family requires larger-scale studies. It should be noted that the specificity of the Arg463Leu mutation for the MDR TB isolates belonging to the Beijing genotype has been reported previously.30,31 This study explored the prevalence of mutations in nine genetic loci reported to be associated with isoniazid resistance in 50 MDR TB clinical isolates. Knowledge of the nature and frequency of isoniazid-resistance conferring mutations guides the development
5 of 7
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
isolates. These isolates had the highest MICs (80 and 100 mg/L) and had no catalase activity. This is in agreement with previous studies showing a clear link between a high level of isoniazid resistance and the loss of catalase activity with severe mutation events, such as frameshift mutations, nonsense point mutations and complete gene deletions.8,10,26 After mutations in katG, mutations in the mabA-inhA regulatory region, leading to the overexpression of InhA, a primary target of isoniazid, represent the second most common mechanism of resistance to this drug in M. tuberculosis.6,7,26 In this study, mutations in the inhA promoter were found in 16% (8/50) of MDR isolates. The observed frequency of inhA promoter mutations among MDR isolates is similar to that reported from Russia (15.4%)2 but much higher than that from Germany (2.9%)20 and much lower than that from Portugal (91.4%).27 In two isolates, mutations upstream of the inhA initiation codon (–15CT) were the only mutations found, confirming their contribution to isoniazid resistance. Two isolates in which an inhA regulatory region mutation co-occurred with a Ser315Thr substitution in katG possessed high MICs of isoniazid (5 and 10 mg/L). This is in line with previous findings showing inhA promoter mutations alone to be associated with low-level resistance to isoniazid, whereas when coupled with katG 315 mutations they conferred a higher level of isoniazid resistance.4,10 The katG 315 and inhA promoter mutant with a high MIC of isoniazid (10 mg/L) was exceptional in that it contained an additional mutation, which was the inhA –8TC substitution, instead of –15CT, as seen in all other inhA promoter mutants. This illustrates a cumulative effect of multiple mutations on the increase in isoniazid resistance. Nevertheless, the low frequency of strains simultaneously harbouring mutations at katG 315 and the inhA promoter was reflected by a strong negative association between these two types of mutation (P,0.01). One possible explanation for this is that inhA promoter mutations exert an attenuating effect on MDR strains possessing a fitness advantage conferred by katG 315 mutations, and that this type of double mutant is selected against during strain evolution.5 Another hot-spot for mutations in isoniazid-resistant M. tuberculosis isolates is the oxyR-ahpC locus. Mutations in the ahpC promoter of katG-defective isoniazid-resistant M. tuberculosis strains have been described which result in AhpC overexpression, and thus compensate for the loss of catalase-peroxidase activity.9,10,24,28 However, mutations in the oxyR-ahpC regulatory region have also been identified in isoniazid-susceptible M. tuberculosis isolates, implying that these mutations have little impact on the isoniazid resistance phenotype.5 In this study, ahpC promoter mutations were detected in five (10%) MDR isolates, of which all but one had katG mutations. The low frequency of ahpC promoter mutations can partly be explained by the fact that they have rarely been observed in strains already bearing katG 315 mutations (72% of the sample studied).5,11,18,28 This is because the katG 315 mutants preserve enough catalase activity to allow in vivo survival of mycobacteria in the absence of compensatory ahpC mutations.9 Analysis of the kasA gene showed 12 (24%) MDR isolates to contain substitution mutations. All except one of these isolates had the missense mutation Gly269Ser, which had an Ala321Thr replacement. The kasA Gly269Ser substitution has been reported in both isoniazid-susceptible and -resistant M. tuberculosis isolates and has thus been concluded to be a gene polymorphism, not involved in isoniazid resistance.5,26 The Ala321Thr substitution in kasA had not yet been described, but its role in isoniazid
JAC
Jagielski et al.
Acknowledgements Some of the results of this study were presented as a poster at the Twenty-third European Congress of Clinical Microbiology and Infectious Diseases, Berlin, Germany, 2013 (P2374) and published previously (identification and analysis of mutations in the katG gene).23
Funding This work was supported by the «Iuventus Plus» grant from the Polish Ministry of Science and Higher Education (IP2011018771).
Transparency declarations
7 Kiepiela P, Bishop KS, Smith AN et al. Genomic mutations in the katG, inhA and aphC genes are useful for the prediction of isoniazid resistance in Mycobacterium tuberculosis isolates from Kwazulu Natal, South Africa. Tuber Lung Dis 2000; 80: 47– 56. 8 Zhang M, Yue J, Yang YP et al. Detection of mutations associated with isoniazid resistance in Mycobacterium tuberculosis isolates from China. J Clin Microbiol 2005; 43: 5477 –82. 9 Kelley CL, Rouse DA, Morris SL. Analysis of ahpC gene mutations in isoniazid-resistant clinical isolates of Mycobacterium tuberculosis. Antimicrob Agents Chemother 1997; 41: 2057–8. 10 Ramaswamy SV, Reich R, Dou SJ et al. Single nucleotide polymorphisms in genes associated with isoniazid resistance in Mycobacterium tuberculosis. Antimicrob Agents Chemother 2003; 47: 1241–50. 11 Cardoso RF, Cardoso MA, Leite CQ et al. Characterization of ndh gene of isoniazid resistant and susceptible Mycobacterium tuberculosis isolates from Brazil. Mem Inst Oswaldo Cruz 2007; 102: 59– 61. 12 Lee AS, Teo AS, Wong SY. Novel mutations in ndh in isoniazid-resistant Mycobacterium tuberculosis isolates. Antimicrob Agents Chemother 2001; 45: 2157 –9. 13 Coelho MB, Costa ER, Vasconcellos SE et al. Sequence and structural characterization of tbnat gene in isoniazid-resistant Mycobacterium tuberculosis: identification of new mutations. Mutat Res 2011; 712: 33– 9. 14 Brossier F, Veziris N, Truffot-Pernot C et al. Molecular investigation of resistance to the antituberculous drug ethionamide in multidrugresistant clinical isolates of Mycobacterium tuberculosis. Antimicrob Agents Chemother 2011; 55: 355–60. 15 Jagielski T, Augustynowicz-Kopec´ E, Zozio T et al. Spoligotype-based comparative population structure analysis of multidrug-resistant and isoniazid-monoresistant Mycobacterium tuberculosis complex clinical isolates in Poland. J Clin Microbiol 2010; 48: 3899– 909.
None to declare.
16 Augustynowicz-Kopec´ E, Zwolska Z, Jaworski A et al. Drug-resistant tuberculosis in Poland in 2000: second national survey and comparison with the 1997 survey. Int J Tuberc Lung Dis 2003; 7: 645–51.
Supplementary data
17 van Embden JDA, Cave MD, Crawford JT et al. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J Clin Microbiol 1993; 31: 406–9.
Table S1 and Table S2 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).
References 1 World Health Organization and International Union Against Tuberculosis and Lung Disease (WHO/IUATLD). Global Project on Anti-Tuberculosis Drug Resistance Surveillance 2002 – 2007: Anti-Tuberculosis Drug Resistance in the World: 4th Global Report. Geneva, Switzerland: WHO, 2008.
18 Luo T, Zhao M, Li X et al. Selection of mutations to detect multidrug-resistant Mycobacterium tuberculosis strains in Shanghai, China. Antimicrob Agents Chemother 2010; 54: 1075 –81. 19 Ramaswamy SV, Dou SJ, Rendon A et al. Genotypic analysis of multidrug-resistant Mycobacterium tuberculosis isolates from Monterrey, Mexico. J Med Microbiol 2004; 53: 107– 13.
2 Afanas’ev MV, Ikryannikova LN, Il’ina EN et al. Molecular characteristics of rifampicin- and isoniazid-resistant Mycobacterium tuberculosis isolates from the Russian Federation. J Antimicrob Chemother 2007; 59: 1057 –64.
20 Hillemann D, Kubica T, Ru¨sch-Gerdes S et al. Disequilibrium in distribution of resistance mutations among Mycobacterium tuberculosis Beijing and non-Beijing strains isolated from patients in Germany. Antimicrob Agents Chemother 2005; 49: 1229231.
3 Bakonyte D, Baranauskaite A, Cicenaite J et al. Molecular characterization of isoniazid-resistant Mycobacterium tuberculosis clinical isolates in Lithuania. Antimicrob Agents Chemother 2003; 47: 2009–11.
21 Tracevska T, Jansone I, Broka L et al. Mutations in the rpoB and katG genes leading to drug resistance in Mycobacterium tuberculosis in Latvia. J Clin Microbiol 2002; 40: 3789– 92.
4 Guo H, Seet Q, Denkin S et al. Molecular characterization of isoniazidresistant clinical isolates of Mycobacterium tuberculosis from the USA. J Med Microbiol 2006; 55: 1527 –31.
22 Pym AS, Saint-Joanis B, Cole ST. Effect of katG mutations on the virulence of Mycobacterium tuberculosis and the implication for transmission in humans. Infect Immun 2002; 70: 4955– 60.
5 Hazbo´n MH, Brimacombe M, Bobadilla del Valle M et al. Population genetics study of isoniazid resistance mutations and evolution of multidrug-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother 2006; 50: 2640– 9.
23 Jagielski T, Grzeszczuk M, Kamin´ski M et al. Identification and analysis of mutations in the katG gene in multidrug-resistant Mycobacterium tuberculosis clinical isolates. Pneumonol Alergol Pol 2013; 81: 298– 307.
6 Herrera L, Valverde A, Saiz P et al. Molecular characterization of isoniazid-resistant Mycobacterium tuberculosis clinical strains isolated in the Philippines. Int J Antimicrob Agents 2004; 23: 572–6.
24 Clemente WT, Soares Lima SS, Palaci M et al. Phenotypic and genotypic characterization of drug-resistant Mycobacterium tuberculosis strains. Diagn Microbiol Infect Dis 2008; 62: 199–204.
6 of 7
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
of new molecular assays for rapid detection of isoniazid-resistant strains in clinically and geographically diverse settings. These methods are based on a limited number of selected mutational sites that are predictive of isoniazid resistance. Here, testing only for katG 315 and inhA –15 mutations would detect isoniazid resistance in 84% of the MDR TB sample. Extension of the analysis to the entire katG gene would increase the proportion of identified isoniazid-resistant isolates to 96%. This underscores the usefulness of mutations in katG and mabA-inhA regulatory region as genetic markers of isoniazid resistance. However, methods that employ such markers still cannot fully replace conventional drug susceptibility profiling. This is because the genetic mechanisms of isoniazid resistance in many TB isolates remain unknown. This is also the case for one isolate in this study, for which no mutation was found in any of the nine genetic loci analysed.
Isoniazid resistance mutations in MDR TB isolates
JAC
25 Bertrand T, Eady NA, Jones JN et al. Crystal structure of Mycobacterium tuberculosis catalase-peroxidase. J Biol Chem 2004; 279: 38991–9.
organisms recovered from diseased humans and animals in diverse localities. Antimicrob Agents Chemother 1997; 41: 600–6.
26 Cardoso RF, Cooksey RC, Morlock GP et al. Screening and characterization of mutations in isoniazid-resistant Mycobacterium tuberculosis isolates obtained in Brazil. Antimicrob Agents Chemother 2004; 48: 3373– 81.
29 Sholto-Douglas-Vernon C, Sandy J, Victor TC et al. Mutational and expression analysis of tbnat and its response to isoniazid. J Med Microbiol 2005; 54: 1189–97.
27 Perdiga˜o J, Macedo R, Joa˜o I et al. Multidrug-resistant tuberculosis in Lisbon, Portugal: a molecular epidemiological perspective. Microb Drug Resist 2008; 14: 133–43. 28 Sreevatsan S, Pan X, Zhang Y et al. Analysis of the oxyR-ahpC region in isoniazid-resistant and -susceptible Mycobacterium tuberculosis complex
30 Samper S, Iglesias MJ, Rabanaque MJ et al. Systematic molecular characterization of multidrug-resistant Mycobacterium tuberculosis complex isolates from Spain. J Clin Microbiol 2005; 43: 1220– 7. 31 Ma¨kinen J, Marjama¨ki M, Haanpera¨-Heikkinen M et al. Extremely high prevalence of multidrug resistant tuberculosis in Murmansk, Russia: a population-based study. Eur J Clin Microbiol Infect Dis 2011; 30: 1119–26.
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
7 of 7
J Antimicrob Chemother doi:10.1093/jac/dku161
Detection of mutations associated with isoniazid resistance in multidrug-resistant Mycobacterium tuberculosis clinical isolates Tomasz Jagielski1*, Zofia Bakuła1, Katarzyna Roeske1, Michał Kamin´ski1, Agnieszka Napio´rkowska2, Ewa Augustynowicz-Kopec´2, Zofia Zwolska2 and Jacek Bielecki1 Department of Applied Microbiology, Institute of Microbiology, Faculty of Biology, University of Warsaw, I. Miecznikowa 1, 02-096 Warsaw, Poland; 2Department of Microbiology, National Tuberculosis and Lung Diseases Research Institute, Płocka 26, 01-138 Warsaw, Poland *Corresponding author. Tel: +48 (0) 22 55 41 312; Fax: +48 (0) 22 55 41 402; E-mail: [email protected]
Received 27 January 2014; returned 24 March 2014; revised 7 April 2014; accepted 14 April 2014 Objectives: To determine the prevalence of isoniazid resistance-conferring mutations among multidrug-resistant (MDR) isolates of Mycobacterium tuberculosis from Poland. Methods: Nine genetic loci, including structural genes (katG, inhA, ahpC, kasA, ndh, nat and mshA) and regulatory regions (i.e. the mabA-inhA promoter and oxyR-ahpC intergenic region) of 50 MDR M. tuberculosis isolates collected throughout Poland were PCR-amplified in their entirety and screened for mutations by direct sequencing methodology. Results: Forty-six (92%) MDR M. tuberculosis isolates had mutations in the katG gene, and the katG Ser315Thr substitution predominated (72%). Eight (16%) isolates (six with a mutated katG allele) had mutations in the inhA promoter region and two such isolates also had single inhA structural gene mutations. Mutations in the oxyR-ahpC locus were found in five (10%) isolates, of which all but one had at least one additional mutation in katG. Mutations in the remaining genetic loci (kasA, ndh, nat and mshA) were detected in 12 (24%), 4 (8%), 5 (10%) and 17 (34%) MDR isolates, respectively. All non-synonymous mutants for these genes harboured mutations in katG. One isolate had no mutations in any of the analysed loci. Conclusions: This study accentuates the usefulness of katG and inhA promoter mutations as predictive markers of isoniazid resistance. Testing only for katG 315 and inhA – 15 mutations would detect isoniazid resistance in 84% of the MDR M. tuberculosis sample. This percentage would increase to 96% if the sequence analysis was extended to the entire katG gene. Analysis of the remaining genetic loci did not contribute greatly to the identification of isoniazid resistance. Keywords: M. tuberculosis, tuberculosis, TB, multidrug resistance, sequence analysis, spoligotyping, first-line anti-TB drugs
Introduction Isoniazid is a key component of drug regimens used worldwide for the treatment of tuberculosis (TB). However, the future clinical utility of the drug has been jeopardized by the increasing prevalence of isoniazid resistance among Mycobacterium tuberculosis strains. As many as 13.3% of all TB cases are resistant to isoniazid, either alone or in combination with other drugs.1 Although the molecular mechanisms of isoniazid resistance are not fully understood, numerous studies have linked them to distinct mutations in various genetic loci of the M. tuberculosis genome. Three loci most commonly affected are the katG gene, encoding catalaseperoxidase, which transforms isoniazid to its pharmacologically active form,2 – 8 the inhA gene (with the mabA-inhA promoter region), coding for enoyl-acyl carrier protein (ACP) reductase, a
mycolic acid biosynthetic pathway enzyme2 – 8 and the ahpC locus, including the structural gene of alkyl hydroperoxide reductase (ahpC), involved in the cellular response to oxidative stress, and the upstream regulatory region (oxyR-ahpC).5 – 9 Other genes that have been shown to be associated with isoniazid resistance include kasA, coding for b-ketoacyl ACP synthase, another enzyme involved in mycolic acid biosynthesis,5,10 ndh, coding for NADH dehydrogenase,5,11,12 and, more recently, nat and mshA, coding for arylamine N-acetyltransferase (NAT),13 which, through acetylation of isoniazid, renders the drug therapeutically inactive, and the glycosyltransferase involved in the biosynthesis of mycothiol.14 The aim of this study was to investigate the prevalence of mutations in a panel of genetic loci implicated in conferring isoniazid resistance on MDR M. tuberculosis clinical isolates from Poland.
# The Author 2014. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For Permissions, please e-mail: [email protected]
1 of 7
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
1
Jagielski et al.
Materials and methods Strains and drug susceptibility testing
Primer Locus
designation
nucleotide sequence (5′ 3′ )
Product size (bp)
ahpC
Fw: AhpC-F Rev: AhpC-R AhpC-seqFa AhpC-seqRa
AGCGACTTCACGGCACGATG TGATGTCTTTGGCGTACTCG GGAATGTCGCAACCAAATG GGCCTTGAGCTTTTCTATAC
732
kasA
Fw: kasA1F Rev: kasA1R Fw: kasA2F Rev: kasA2R Fw: kasA3F Rev: kasA3R
CGTTCAGGCGAGGCTTGAGG CCGGTCTGGATCGACCTCCG GGACAGCTATGGGAGTCCGC ACCCAGCAATCGGGCCAACG GCACGCCAAAGCCCGTGGCG GGGCCTCGCGACCCGCGATG
351
Fw: mshA-F Rev: mshA-R mshAFMa mshARMa
CTTCGGTTCCTGCAAGGATGG CGAGCGCATTCTGGATCACAC AGATCGACACCGGGATGGAC CGTCAACACCGACGATGAAG
mshA
528 523 1505
Fw, forward; Rev, reverse. All primers used for amplification were also used for sequencing. a Additional intragenic primers used for sequencing.
DNA extraction Genomic DNA was extracted by the cetyl-trimethyl ammonium bromide (CTAB) method.17
gene, KF549665 and KF549666 for the ndh gene, KF549668 – KF549670 for the nat gene and KF564643 and KF564644 for the mshA gene.
Sequencing strategy
Statistical analysis
Seven structural genes (katG, inhA, ahpC, kasA, ndh, nat, mshA) and two regulatory regions (the mabA-inhA promoter region and the oxyR-ahpC intergenic region) were screened for mutations by direct sequencing of each locus in its entirety. The oligonucleotide primers used for PCR amplification of the respective genetic loci were either based on previously reported studies or newly designed in the present study [Table 1 and Table S1 (available as Supplementary data at JAC Online)]. The PCR mixtures were prepared with a TopTaq Master Mix Kit (Qiagen) and run under optimized thermocycling conditions on a LabCycler (SensoQuest). Amplification products were analysed by electrophoresis in 1.5% agarose gels, purified with a Clean-Up Kit (A&A Biotechnology) and sequenced by using a BigDye ver. 3.1 Terminator Cycle Sequencing Kit (Applied Biosystems) in an ABI 3130xl Genetic Analyzer (Applied Biosystems). To confirm the presence of mutations, sequencing was done in both directions with the same primer pairs as those used in the amplification reactions and, occasionally, with additional, internal primers (Table 1). Sequence data were assembled and analysed with ChromasPro (ver. 1.7.1) software (Technelysium), and the resulting consensus sequences were aligned against the wild-type sequences of the respective genetic loci of M. tuberculosis reference strain H37Rv (TubercuList; http:// genolist.pasteur.fr/TubercuList/) using the blastn algorithm (http://blast.ncbi. nlm.nih.gov/).
The associations between individual mutations and between mutations and MICs of isoniazid or spoligotyping families were assessed with the x2 test and ANOVA, respectively. A P value of ,0.05 was considered statistically significant.
Nucleotide sequence accession numbers The sequences with novel mutations were deposited in GenBank (National Center for Biotechnology Information; http://www.ncbi.nlm.nih.gov/) under the following accession numbers: KF539901–KF539910 for the katG gene, KF563064 for the inhA gene, KF563062 and KF563063 for the oxyR-aphC regulatory region, KF549667 for the ahpC gene, KF563065 for the kasA
2 of 7
Results Of the 50 MDR TB isolates examined, 46 (92%) had mutations within the katG gene. The katG codon 315 was affected in 36 (72%) isolates, with a GC change at nucleotide position 944, conferring a Ser315Thr amino acid substitution, found most frequently (33 isolates; 66%). Mutations at other katG codons were distributed among 13 (26%) isolates (Table S2, available as Supplementary data at JAC Online). Forty-four (88%) isolates had a single point mutation in the katG gene, whereas five (10%) isolates had a double katG mutation and one isolate had a triple katG mutation. Two isolates with two and three katG mutations, albeit with a wild-type 315 codon, had unique termination mutations, resulting in a truncated and non-functional catalase peptide. Both these isolates presented the highest MICs of isoniazid (80 and 100 mg/L). For the remaining 44 (88%) isolates with katG mutations, the isoniazid MICs were within the range 0.5 –10 mg/L. Eight (16%) isolates (six with a mutated katG allele) had mutations in the mabA-inhA locus (MICs 1 – 10 mg/L). Transitions at nucleotide positions –15 (CT) and –8 (TC) of the inhA promoter region were observed in seven (14%) isolates and one isolate,
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
A total of 50 MDR M. tuberculosis isolates, collected at the National Tuberculosis and Lung Diseases Institute in Warsaw, were included in the study. The isolates were recovered from 46 unrelated pulmonary TB patients (40 men and 6 women; age range, 31 – 79 years; median age, 50.5 years) from across Poland. These patients represented all bacteriologically confirmed MDR TB cases reported in Poland in 2004 [TB cases registered (total), n ¼9493; TB notification rate, n¼24.9/100000 population]. Previously, isolates from these patients had been genotyped by means of spoligotyping.15 Seven (15.2%) of the patients had not previously been treated for TB, and were thus classified as new MDR TB cases, whereas 39 (84.8%) had received anti-TB treatment for at least 1 month in the past and were defined as acquired MDR TB cases. Primary isolation, culturing and species identification were carried out by standard mycobacteriological procedures, essentially as described elsewhere.16 Initial drug susceptibility testing against first-line anti-TB drugs was performed using the 1% proportion method on Lo¨wenstein –Jensen (LJ) medium, with the following critical drug concentrations: isoniazid, 0.2 mg/L; rifampicin, 40 mg/L; streptomycin, 4 mg/L; and ethambutol, 2 mg/L.16 Subsequently, the MICs of isoniazid were determined in LJ medium containing 2-fold incremental concentrations of the drug ranging from 0.05 to 100 mg/L. The MIC was defined as the lowest drug concentration that inhibited .99% of the bacterial population in the drug-free control culture.
Table 1. Primers used for PCR amplification and sequencing of the ahpC, kasA and mshA genes
JAC
Isoniazid resistance mutations in MDR TB isolates
ST53, none of the eight isolates, including those from the same patients, had identical mutational patterns, differing by at least one nucleotide polymorphism. Cluster ST891 comprised three isolates with an identical mutational pattern and two isolates that differed by the presence of additional mutations in the mshA gene. Another cluster (ST1557) contained four isolates, of which only one was different by a single polymorphism in the oxyR-ahpC locus. Of the four Beijing genotype isolates, two had identical mutational profiles, whereas the other two differed by single nucleotide polymorphisms at the katG, mabA-inhA and inhA loci. Finally, two isolates of cluster ST37 had three different polymorphisms at either the katG or the mabA-inhA locus. No association was found between specific mutation frequency and type of resistance (primary versus acquired). Also, no association was observed between any mutation and a spoligotype family, except that mutations Arg463Leu (GT at position 1388) and Ala187Val (CT at position 560) in the katG and mshA genes, respectively, were found only in Beijing genotype isolates. Most of the MDR TB isolates classified in the T, Latin-American Mediterranean (LAM) and Beijing families presented additional resistance to streptomycin (P ¼ 0.03).
Discussion This work provides a description of mutations in a large set of genetic loci associated or speculated to be associated with isoniazid resistance in M. tuberculosis. The results reported here are in many respects consistent with and comparable to those of previous studies. For instance, mutations in the katG gene predominated (92% of all MDR isolates) and mutations at codon 315 in the katG gene were detected in nearly two-thirds (72%) of the MDR strains. High detection rates of katG 315 mutations in MDR strains have been reported from China (72.7%),18 Mexico (67.6%)19 and countries neighbouring Poland, including Germany (88.4%),20 Lithuania (89.7%),3 Latvia (94.1%)21 and Russia (93.5%).2 Mutations at katG codon 315, in particular the Ser315Thr substitution, have been shown to confer a selective advantage by reducing the ability of KatG to activate isoniazid, while preserving substantial catalaseperoxidase activity to protect isoniazid-resistant tubercle bacilli from oxidative stress.22 Indeed, all but two (34/36; 94.4%) MDR isolates in this study, carrying the katG Ser315Thr mutation, were catalase-positive (data not shown).23 Mutations at katG 315 have repeatedly been associated with a high level of isoniazid resistance (≥4 mg/L).8,24 In this study, the MICs of isoniazid for isolates with the katG 315 mutation were within the range 1 – 10 mg/L, with the majority of isolates (27/36; 75%) exhibiting MICs of 2.5 mg/L (P,0.01), a value corresponding to a moderate level of resistance. Apart from codon 315, mutations in the katG gene were identified at 11 other codons. Mutations at eight of these codons had not been reported previously (Table S2). The relevance of the newly described katG mutations to isoniazid resistance can hardly be established as they were accompanied by other mutations in katG and/or other genetic loci. The only exception was substitution Leu101Pro, observed as the sole mutation in one MDR strain. That this mutation confers isoniazid resistance seems likely, given that the affected amino acid residue is located in the close vicinity of the KatG active site.25 Isoniazid resistance was also associated with termination mutations (Lys46STOP, Glu454STOP) in two
3 of 7
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
respectively. Two isolates with a CT change at the –15 position of the inhA promoter also contained single substitutions in the inhA structural gene, resulting in Ile194Thr (TC at position 649) and Ser94Ala (TG at position 289) replacements. Mutations in the oxyR-ahpC intergenic region were found in five (10%) isolates; two isolates had a single nucleotide change at position – 48 (GA); one was at position – 54 (CT) and the other was at position –57 (CT) upstream of the transcriptional initiation site of the ahpC gene. A three-nucleotide change (TCAATC) in the ahpC promoter region was demonstrated in one isolate (MIC 10 mg/L). Except for the latter, all isolates with mutations in the oxyR-ahpC intergenic region had at least one additional mutation in the katG gene. All isolates with mutations in the oxyR-ahpC region had no alterations in the mabA-inhA operon. A silent mutation, Ile114Ile (AC at position 342), was the sole ahpC structural gene mutation detected and was present in only one isolate (MIC 5 mg/L). This isolate also carried single mutations in the katG gene and the inhA promoter. Two types of substitution mutation were identified in the kasA gene. They were both GA transitions at positions 805 (Gly269Ser) and 961 (Ala321Thr), found in 11 (22%) isolates and one isolate, respectively. All 12 kasA mutants had additional mutations in the katG gene but no mutations in the mabA-inhA and oxyR-ahpC loci. The MICs of isoniazid for these isolates were 1 or 2.5 mg/L. Single ndh gene mutations were detected in four (8%) isolates. The mutations occurred at nucleotide position 53 (TC; Val18Ala) in two isolates, 843 (GC; Ser281Ser) in one isolate and 898 (GC; Ala300Pro) in another isolate. Three isolates with an altered ndh gene sequence contained mutations in the katG gene and two isolates, including that with the wild-type katG allele, contained mutations in the ahpC promoter region. Mutations in the nat gene were detected in five (10%) isolates. Two isolates had the same termination mutation at codon 173 (GA at position 518), two had Gly207Arg replacements (GA at position 619) and one isolate had an Ala103Gly replacement (CG at position 308). All of the nat mutants carried single katG mutations. In addition, mutations in the ndh gene and the ahpC promoter were present in three and two of these isolates, respectively. Four types of point mutations were identified in the mshA gene of 17 (34%) isolates (MICs ≥1 mg/L). The mutated sites were at positions 332 (AG; Asn111Ser) in 11 (22%) isolates, 560 (CT; Ala187Val) in 4 (8%) isolates and 1085 (CT; Ala362Val) in 1 isolate. One isolate had an insertion of a cytosine at position 1283, producing a frameshift mutation. All isolates carrying mshA mutations had an altered katG gene sequence. The mshA mutants had additional changes in one (kasA, three isolates; ahpC promoter, two isolates; nat, one isolate) or two (inhA promoter and coding region, two isolates; ndh and nat, one isolate) loci. Of all MDR TB isolates tested, 15 (30%) had mutations in only one genetic locus (i.e. katG in 13 isolates and mabA-inhA in 2 isolates). Twenty-one (42%) isolates harboured mutations in two, 11 (22%) in three, and 2 (4%) in four genetic loci. One isolate (MIC 1 mg/L) had the wild-type sequences at all nine loci analysed. Comparison of spoligotyping results with mutational profiles at the isoniazid resistance-associated loci showed that only two of six spoligotype clusters [sequence types (STs) 1051 and F] were entirely homogeneous with regard to their mutational patterns (Figure 1). In the largest cluster, represented by spoligotype
E A sh
5895 101*
A 775
1377
2333
H1
2233
47
H1
U H
1885 T1
T
708
X1
5325*
1558
T1
X
3832
442
T3
456* 1233
40 50
T4 H3
9310
180
H3
1114
253
T1
2688
42
456* H T
1233 9310 1114 2688 1612
891
LAM9
LAM
446 2086
874
874
524*
524* B
U
4991 17170*
3298
3298 1051
T1
T
434
1746
S
S
434
469
C
U
469
647
280
T
647
124
D
U
124
2575
E
U
2575
947
947
T1_RUS2
12489*
F
U
1334
3020* Beijing
BEIJING
1334 K131 3430 STOP -46 Val 68 Gly Trp 91 Arg Leu 101 Pro Met 126 Ile Arg128 Gln Pro 131 Arg Pro 131 Gln Asp 194 Ty Ala 234 Gly Ser 315 Ile Ser 315 Asn Ser 315 Th Ser 315 Th Glu 454 STOP Arg 463 Leu NA NA Ser 94 Ala Pro 136 Pro Ile 194 Th NA NA NA NA Ile 114 Ile Gly 269 Se Ala 321 Th Val 18 Ala Ser 281 Se Ala 300 Pro Ala 103 Gly Trp 173 STOP Gly 207 Arg Asn 111 Se Ala 187 Va Ala 362 Va Frameshif
3430
11420 1324
G
3020* K131
B
535 3832
LAM9
446
1324
R
2497
1612
0.1
103 11844
T3
11420
ST
4619 T
37
12489*
F
794
T1
103
17170*
EM
6679 53
11844
4991
RI
3312
6679
2086
73 5325*
3312
535
4202 590
73
2497
2233
4365
590
4619
101* 1377
692 1557
4365
794
5895
1885
692 4202
IN H
m
na t
nd h
ox
yR ah ahp p C ka C sA
F
135 insT 203 T>G 271 T>C 302 T>C 378 G>A 383 G>A 392 C>G 392 C>A 580 G>T 701 C>G 944 G>T 944 G>A 944 G>C 945 C>T 1360 G>T 1388 G>T -15 C>T -8 T>C 289 T>G 408 G>A 581 T>C -48 G>A -54 C>T -57 C>T -82/-80 TCA>ATC 342 A>C 805 G>A 961 G>A 53 T>C 843 G>C 898 G>C 308 C>G 518 G>A 619 G>A 332 A>G 560 C>T 1085 C>T 1283 insC
ka t
ab Ain hA hA
A
in
D
m
C
G
B
Jagielski et al.
4 of 7
A
Figure 1. A dendrogram based on spoligotype patterns (DendroUPGMA, Jaccard index, UPGMA) along with 38 mutational changes within the nine genetic loci of the MDR M. tuberculosis strains investigated. A, strain no.; B, spoligotype pattern (binary); C, spoligotype signature (shared international type, SIT); D, phylogenetic clade; E, mutational pattern (data were coded in binary format, with black and white circles representing the presence and absence of an individual mutation, respectively; mutations at the nucleotide and amino acid level are depicted at the top and the bottom of the panel, accordingly); F, drug resistance pattern (resistance and susceptibility to a given drug is presented as black and white rectangles, respectively). INH, isoniazid; RIF, rifampicin; EMB, ethambutol, STR, streptomycin. Asterisks next to the strain numbers indicate new MDR TB cases (primary MDR resistance).
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
Isoniazid resistance mutations in MDR TB isolates
resistance is unlikely as this mutation was accompanied by an alteration in katG. The mechanism behind the involvement of the ndh mutations in resistance to isoniazid is probably an accumulation of intracellular NADH, which competitively inhibits the binding of the isoniazid – NAD adduct to InhA or the peroxidation of isoniazid by KatG.12 Two non-synonymous ndh mutations were detected in three MDR isolates in this study. One mutation (Val18Ala), shared by two isolates, has already been reported in both isoniazid-resistant and -susceptible tubercle bacilli.5,10 – 12 Both these Val18Ala mutants had a Ser315Thr substitution mutation in katG, and isoniazid resistance was rather due to the latter mutation. Likewise, mutation in codon 128 of katG could be more responsible for the isoniazid resistance in one isolate with a novel ndh mutation (Ala300Pro) than this mutation itself. Although the role of nat mutations in isoniazid resistance remains obscure, it has been speculated that they might increase the affinity of the NAT enzyme for the drug, thereby accelerating its inactivation by acetylation.13 Three nat mutations in five MDR isolates were observed in this study. The previously identified Gly207Arg mutation was shown to reduce the activity of the NAT enzyme towards isoniazid.29 Reduced NAT activity must also be expected as a result of the newly discovered termination mutation (Trp173STOP). Therefore, the only nat mutation that may somehow contribute to isoniazid resistance is the Ala103Gly substitution, found in one MDR isolate. Yet the effect of this mutation is masked by the co-occurrence, in this isolate, of two other mutations, at the katG and mshA loci. In order to verify the recently proposed concept that mutations in mycothiol biosynthesis genes may contribute to isoniazid resistance, the MDR isolates were screened for mutations in mshA, a gene encoding one of the mycothiol biosynthesis enzymes. Although mshA mutations occurred in one-third of the MDR isolates (17/50), the relevance of these mutations to isoniazid resistance is unclear, since they were always accompanied by mutations in other loci (e.g. all the mshA mutants carried a mutated katG allele). Among the mshA mutations, greater interest should be paid to a novel frameshift mutation (insC 1283) found in an isolate with a high MIC of isoniazid (10 mg/L). This isolate also had Ser315Thr and Val18Ala substitutions in the katG and ndh genes, respectively, but another isolate with these two mutations and with no other mutation exhibited an MIC of 2.5 mg/L. It is thus plausible that a severely defective mshA gene (e.g. due to a frameshift mutation) contributes to the increase in isoniazid resistance. No correlation was observed between any mutation and spoligotype family. Noticeably, however, mutations Arg463Leu and Ala187Val in the katG and mshA genes, respectively, were found exclusively in Beijing genotype isolates. Importantly, these isolates were demonstrated to be unrelated upon mycobacterial interspersed repetitive unit-variable-number tandem repeat (MIRU-VNTR) analysis and IS6110 restriction fragment length polymorphism (RFLP) profiling (data not shown). Whether these polymorphisms are associated with the Beijing family requires larger-scale studies. It should be noted that the specificity of the Arg463Leu mutation for the MDR TB isolates belonging to the Beijing genotype has been reported previously.30,31 This study explored the prevalence of mutations in nine genetic loci reported to be associated with isoniazid resistance in 50 MDR TB clinical isolates. Knowledge of the nature and frequency of isoniazid-resistance conferring mutations guides the development
5 of 7
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
isolates. These isolates had the highest MICs (80 and 100 mg/L) and had no catalase activity. This is in agreement with previous studies showing a clear link between a high level of isoniazid resistance and the loss of catalase activity with severe mutation events, such as frameshift mutations, nonsense point mutations and complete gene deletions.8,10,26 After mutations in katG, mutations in the mabA-inhA regulatory region, leading to the overexpression of InhA, a primary target of isoniazid, represent the second most common mechanism of resistance to this drug in M. tuberculosis.6,7,26 In this study, mutations in the inhA promoter were found in 16% (8/50) of MDR isolates. The observed frequency of inhA promoter mutations among MDR isolates is similar to that reported from Russia (15.4%)2 but much higher than that from Germany (2.9%)20 and much lower than that from Portugal (91.4%).27 In two isolates, mutations upstream of the inhA initiation codon (–15CT) were the only mutations found, confirming their contribution to isoniazid resistance. Two isolates in which an inhA regulatory region mutation co-occurred with a Ser315Thr substitution in katG possessed high MICs of isoniazid (5 and 10 mg/L). This is in line with previous findings showing inhA promoter mutations alone to be associated with low-level resistance to isoniazid, whereas when coupled with katG 315 mutations they conferred a higher level of isoniazid resistance.4,10 The katG 315 and inhA promoter mutant with a high MIC of isoniazid (10 mg/L) was exceptional in that it contained an additional mutation, which was the inhA –8TC substitution, instead of –15CT, as seen in all other inhA promoter mutants. This illustrates a cumulative effect of multiple mutations on the increase in isoniazid resistance. Nevertheless, the low frequency of strains simultaneously harbouring mutations at katG 315 and the inhA promoter was reflected by a strong negative association between these two types of mutation (P,0.01). One possible explanation for this is that inhA promoter mutations exert an attenuating effect on MDR strains possessing a fitness advantage conferred by katG 315 mutations, and that this type of double mutant is selected against during strain evolution.5 Another hot-spot for mutations in isoniazid-resistant M. tuberculosis isolates is the oxyR-ahpC locus. Mutations in the ahpC promoter of katG-defective isoniazid-resistant M. tuberculosis strains have been described which result in AhpC overexpression, and thus compensate for the loss of catalase-peroxidase activity.9,10,24,28 However, mutations in the oxyR-ahpC regulatory region have also been identified in isoniazid-susceptible M. tuberculosis isolates, implying that these mutations have little impact on the isoniazid resistance phenotype.5 In this study, ahpC promoter mutations were detected in five (10%) MDR isolates, of which all but one had katG mutations. The low frequency of ahpC promoter mutations can partly be explained by the fact that they have rarely been observed in strains already bearing katG 315 mutations (72% of the sample studied).5,11,18,28 This is because the katG 315 mutants preserve enough catalase activity to allow in vivo survival of mycobacteria in the absence of compensatory ahpC mutations.9 Analysis of the kasA gene showed 12 (24%) MDR isolates to contain substitution mutations. All except one of these isolates had the missense mutation Gly269Ser, which had an Ala321Thr replacement. The kasA Gly269Ser substitution has been reported in both isoniazid-susceptible and -resistant M. tuberculosis isolates and has thus been concluded to be a gene polymorphism, not involved in isoniazid resistance.5,26 The Ala321Thr substitution in kasA had not yet been described, but its role in isoniazid
JAC
Jagielski et al.
Acknowledgements Some of the results of this study were presented as a poster at the Twenty-third European Congress of Clinical Microbiology and Infectious Diseases, Berlin, Germany, 2013 (P2374) and published previously (identification and analysis of mutations in the katG gene).23
Funding This work was supported by the «Iuventus Plus» grant from the Polish Ministry of Science and Higher Education (IP2011018771).
Transparency declarations
7 Kiepiela P, Bishop KS, Smith AN et al. Genomic mutations in the katG, inhA and aphC genes are useful for the prediction of isoniazid resistance in Mycobacterium tuberculosis isolates from Kwazulu Natal, South Africa. Tuber Lung Dis 2000; 80: 47– 56. 8 Zhang M, Yue J, Yang YP et al. Detection of mutations associated with isoniazid resistance in Mycobacterium tuberculosis isolates from China. J Clin Microbiol 2005; 43: 5477 –82. 9 Kelley CL, Rouse DA, Morris SL. Analysis of ahpC gene mutations in isoniazid-resistant clinical isolates of Mycobacterium tuberculosis. Antimicrob Agents Chemother 1997; 41: 2057–8. 10 Ramaswamy SV, Reich R, Dou SJ et al. Single nucleotide polymorphisms in genes associated with isoniazid resistance in Mycobacterium tuberculosis. Antimicrob Agents Chemother 2003; 47: 1241–50. 11 Cardoso RF, Cardoso MA, Leite CQ et al. Characterization of ndh gene of isoniazid resistant and susceptible Mycobacterium tuberculosis isolates from Brazil. Mem Inst Oswaldo Cruz 2007; 102: 59– 61. 12 Lee AS, Teo AS, Wong SY. Novel mutations in ndh in isoniazid-resistant Mycobacterium tuberculosis isolates. Antimicrob Agents Chemother 2001; 45: 2157 –9. 13 Coelho MB, Costa ER, Vasconcellos SE et al. Sequence and structural characterization of tbnat gene in isoniazid-resistant Mycobacterium tuberculosis: identification of new mutations. Mutat Res 2011; 712: 33– 9. 14 Brossier F, Veziris N, Truffot-Pernot C et al. Molecular investigation of resistance to the antituberculous drug ethionamide in multidrugresistant clinical isolates of Mycobacterium tuberculosis. Antimicrob Agents Chemother 2011; 55: 355–60. 15 Jagielski T, Augustynowicz-Kopec´ E, Zozio T et al. Spoligotype-based comparative population structure analysis of multidrug-resistant and isoniazid-monoresistant Mycobacterium tuberculosis complex clinical isolates in Poland. J Clin Microbiol 2010; 48: 3899– 909.
None to declare.
16 Augustynowicz-Kopec´ E, Zwolska Z, Jaworski A et al. Drug-resistant tuberculosis in Poland in 2000: second national survey and comparison with the 1997 survey. Int J Tuberc Lung Dis 2003; 7: 645–51.
Supplementary data
17 van Embden JDA, Cave MD, Crawford JT et al. Strain identification of Mycobacterium tuberculosis by DNA fingerprinting: recommendations for a standardized methodology. J Clin Microbiol 1993; 31: 406–9.
Table S1 and Table S2 are available as Supplementary data at JAC Online (http://jac.oxfordjournals.org/).
References 1 World Health Organization and International Union Against Tuberculosis and Lung Disease (WHO/IUATLD). Global Project on Anti-Tuberculosis Drug Resistance Surveillance 2002 – 2007: Anti-Tuberculosis Drug Resistance in the World: 4th Global Report. Geneva, Switzerland: WHO, 2008.
18 Luo T, Zhao M, Li X et al. Selection of mutations to detect multidrug-resistant Mycobacterium tuberculosis strains in Shanghai, China. Antimicrob Agents Chemother 2010; 54: 1075 –81. 19 Ramaswamy SV, Dou SJ, Rendon A et al. Genotypic analysis of multidrug-resistant Mycobacterium tuberculosis isolates from Monterrey, Mexico. J Med Microbiol 2004; 53: 107– 13.
2 Afanas’ev MV, Ikryannikova LN, Il’ina EN et al. Molecular characteristics of rifampicin- and isoniazid-resistant Mycobacterium tuberculosis isolates from the Russian Federation. J Antimicrob Chemother 2007; 59: 1057 –64.
20 Hillemann D, Kubica T, Ru¨sch-Gerdes S et al. Disequilibrium in distribution of resistance mutations among Mycobacterium tuberculosis Beijing and non-Beijing strains isolated from patients in Germany. Antimicrob Agents Chemother 2005; 49: 1229231.
3 Bakonyte D, Baranauskaite A, Cicenaite J et al. Molecular characterization of isoniazid-resistant Mycobacterium tuberculosis clinical isolates in Lithuania. Antimicrob Agents Chemother 2003; 47: 2009–11.
21 Tracevska T, Jansone I, Broka L et al. Mutations in the rpoB and katG genes leading to drug resistance in Mycobacterium tuberculosis in Latvia. J Clin Microbiol 2002; 40: 3789– 92.
4 Guo H, Seet Q, Denkin S et al. Molecular characterization of isoniazidresistant clinical isolates of Mycobacterium tuberculosis from the USA. J Med Microbiol 2006; 55: 1527 –31.
22 Pym AS, Saint-Joanis B, Cole ST. Effect of katG mutations on the virulence of Mycobacterium tuberculosis and the implication for transmission in humans. Infect Immun 2002; 70: 4955– 60.
5 Hazbo´n MH, Brimacombe M, Bobadilla del Valle M et al. Population genetics study of isoniazid resistance mutations and evolution of multidrug-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother 2006; 50: 2640– 9.
23 Jagielski T, Grzeszczuk M, Kamin´ski M et al. Identification and analysis of mutations in the katG gene in multidrug-resistant Mycobacterium tuberculosis clinical isolates. Pneumonol Alergol Pol 2013; 81: 298– 307.
6 Herrera L, Valverde A, Saiz P et al. Molecular characterization of isoniazid-resistant Mycobacterium tuberculosis clinical strains isolated in the Philippines. Int J Antimicrob Agents 2004; 23: 572–6.
24 Clemente WT, Soares Lima SS, Palaci M et al. Phenotypic and genotypic characterization of drug-resistant Mycobacterium tuberculosis strains. Diagn Microbiol Infect Dis 2008; 62: 199–204.
6 of 7
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
of new molecular assays for rapid detection of isoniazid-resistant strains in clinically and geographically diverse settings. These methods are based on a limited number of selected mutational sites that are predictive of isoniazid resistance. Here, testing only for katG 315 and inhA –15 mutations would detect isoniazid resistance in 84% of the MDR TB sample. Extension of the analysis to the entire katG gene would increase the proportion of identified isoniazid-resistant isolates to 96%. This underscores the usefulness of mutations in katG and mabA-inhA regulatory region as genetic markers of isoniazid resistance. However, methods that employ such markers still cannot fully replace conventional drug susceptibility profiling. This is because the genetic mechanisms of isoniazid resistance in many TB isolates remain unknown. This is also the case for one isolate in this study, for which no mutation was found in any of the nine genetic loci analysed.
Isoniazid resistance mutations in MDR TB isolates
JAC
25 Bertrand T, Eady NA, Jones JN et al. Crystal structure of Mycobacterium tuberculosis catalase-peroxidase. J Biol Chem 2004; 279: 38991–9.
organisms recovered from diseased humans and animals in diverse localities. Antimicrob Agents Chemother 1997; 41: 600–6.
26 Cardoso RF, Cooksey RC, Morlock GP et al. Screening and characterization of mutations in isoniazid-resistant Mycobacterium tuberculosis isolates obtained in Brazil. Antimicrob Agents Chemother 2004; 48: 3373– 81.
29 Sholto-Douglas-Vernon C, Sandy J, Victor TC et al. Mutational and expression analysis of tbnat and its response to isoniazid. J Med Microbiol 2005; 54: 1189–97.
27 Perdiga˜o J, Macedo R, Joa˜o I et al. Multidrug-resistant tuberculosis in Lisbon, Portugal: a molecular epidemiological perspective. Microb Drug Resist 2008; 14: 133–43. 28 Sreevatsan S, Pan X, Zhang Y et al. Analysis of the oxyR-ahpC region in isoniazid-resistant and -susceptible Mycobacterium tuberculosis complex
30 Samper S, Iglesias MJ, Rabanaque MJ et al. Systematic molecular characterization of multidrug-resistant Mycobacterium tuberculosis complex isolates from Spain. J Clin Microbiol 2005; 43: 1220– 7. 31 Ma¨kinen J, Marjama¨ki M, Haanpera¨-Heikkinen M et al. Extremely high prevalence of multidrug resistant tuberculosis in Murmansk, Russia: a population-based study. Eur J Clin Microbiol Infect Dis 2011; 30: 1119–26.
Downloaded from http://jac.oxfordjournals.org/ at West Virginia University, Health Science Libraryy on May 26, 2014
7 of 7
