0International Journal of Infectious Diseases (2009) 13, 46—58
CI CC\/ICD
http://intl.elsevierhealth.com/journals/ijid
Rapid speciation of 15 clinically relevant mycobacteria with simultaneous detection of resistance to rifampin, isoniazid, and streptomycin in Mycobacterium tuberculosis complex
Shubhada Shenai, Camilla Rodrigues*, Ajita Mehta
P.D. Hinduja National Hospital and Medical Research Centre, Veer Savarkar Marg, Mahim (West), Mumbai — 400 016, Maharashtra, India
Received 27 September 2007; received in revised form 3 March 2008; accepted 17 March 2008 Corresponding Editor: William Cameron, Ottawa, Canada
KEYWORDS
MDR-TB; Mycobacteria; Reverse line blot hybridization; Drug resistance; Species identification
Summary
Objective: To design and standardize an in-house reverse line blot hybridization (RLBH) assay for the accurate identification of 15 clinically relevant species of mycobacteria and for the detection of drug resistance to rifampin (RIF), isoniazid (INH), and streptomycin (STR) in Mycobacterium tuberculosis complex (MTB).
Material and methods: Oligonucleotides specific for 15 different species of mycobacteria and wild type and mutant alleles of selected codons in the rpob, inhA, katG, rpsL, and rrs genes were designed and immobilized on a membrane. A multiplex PCR was standardized to amplify all target genes. The assay was optimized using ATCC and known mutant strains. Three hundred MTB isolates, 85 non-tuberculous mycobacteria (NTM) isolates, and 48 smear-positive specimens were analyzed. Results were confirmed by PCR restriction enzyme assay and sequencing. Results: Upon RLBH analysis, among the NTM, 14% were identified as Mycobacterium fortuitum, 16% were identified as Mycobacterium abscessus, 20% showed 99% homology with Mycobacterium intracellulare, and 31% showed 98% homology with Mycobacterium simiae. Of the 300 MTB isolates analyzed, 75% RIF-resistant isolates had Ser531Leu mutation in the rpob gene. Of the INH-resistant isolates, 89% showed Ser315Thr mutation in the katG gene, whereas 16% showed -15 C!T mutation in the promoter region of the inhA gene. Among STR-resistant isolates, 75% had A! mutation in the rpsL gene at codon 43. RLBH results showed 96-99% concordance with phenotypic culture results.
Conclusion: This is a first attempt at combining speciation with detection of drug resistance to RIF, INH, and STR in MTB for accurate and rapid management of mycobacterial infections as well as for compiling genotypic epidemiological data.
© 2008 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved.
* Corresponding author. Tel.: +91 22 24447795/96. E-mail address: dr_crodrigues@hindujahospital.com (C. Rodrigues).
1201-9712/$36.00 © 2008 International Society for Infectious Diseases. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.ijid.2008.03.025
Introduction
The World Health Organization (WHO) has declared an alarming increase in the number of tuberculosis (TB) patients in the Indian subcontinent, with India being singled out as having the greatest burden.1 Data published from our hospital (P.D. Hinduja National Hospital and Medical Research Centre), a tertiary care center in central Mumbai, showed that in 2006, 68% of the patients for whom drug susceptibility testing (DST) was requested had multidrug-resistant tuberculosis (MDR-TB).2 Sampling bias cannot be ruled out, as Mumbai is a hot-spot for MDR-TB.2"4 As a result, there is an urgent need for rapid and accurate detection of MDR-TB in order to achieve patient cure and, from a public health perspective, to prevent transmission.
Although the most important human pathogenic species is Mycobacterium tuberculosis, there are increasing reports of infections caused by non-tuberculous mycobacteria (NTM).5'6 Diseases due to NTM, though uncommon, provide a serious diagnostic and therapeutic challenge. Most of the NTM are resistant to common anti-TB drugs, and they can be falsely labeled as MDR-TB if not identified correctly.
In India, the accurate identification of mycobacteria to the species level is not routine, and the most commonly used method for TB diagnosis is smear microscopy, which has low sensitivity and moderate specificity.7 Cultivation of mycobacteria is generally performed using conventional Lowen-stein—Jensen (L.J.) medium, which is time-consuming and sometimes negative if the disease is paucibacillary or if the patient is receiving anti-TB therapy. Biochemical tests used for the identification of mycobacteria are laborious, time-consuming, and often fail in precise identification.8 DST is generally performed on L.J. medium by an absolute concentration method. The whole process of detection of drug-resistant phenotypes of mycobacteria takes at least 4—6 weeks, thus making it hard to meet the demand of initiating a prompt and effective clinical chemotherapy in patients. Therefore, most current clinical treatments are often prescribed empirically. In recent years, several new methods have been reported for reducing the recovery time to 3 weeks or less.6'9—13 However, in disease endemic developing nations, these methods require considerable technical expertise and financial input in a routine laboratory set up. The development of a rapid, sensitive, and accurate diagnostic test with prediction of drug susceptibility or resistance would make a substantial impact on TB control activities.
In the present study, an in-house reverse line blot hybridization (RLBH) assay was developed and evaluated for accurate identification of 15 clinically relevant species of mycobacteria with simultaneous prediction of mutations causing drug resistance in M. tuberculosis complex (MTB). In this PCR-based multi-probe assay, 54 specific probes were used to screen five different genes (rpob (codons 302—534 for speciation and rifampin resistance), katG (codons 315 and 463), irihA (orf1—inhA promoter region), rpsL (codons 43 and 88), and rrs (nucleotide positions 491, 513, and 516)), which were amplified by a multiplex PCR. This was the first attempt at combining simultaneous species identification of mycobacteria with drug resistance testing to isoniazid (INH), rifampin (RIF), and streptomycin (STR) in MTB.
Material and methods Setting
This study was set at a tertiary care center located in central Mumbai with a referral bias towards non-responding cases.
Subjects
Twenty-two ATCC and CAP-QC strains of mycobacteria and 20 known mutant MTB strains were used to standardize the assay. In addition, standard ATCC strains of Staphylococcus aureus, Escherichia coli, Pseudomonas aeruginosa, Salmonella typhi, and Nocardia spp were also tested as negative controls.
On standardization, the reproducibility of the assay was checked by analyzing 300 consecutive MTB isolates reported for susceptibility testing over a period of 7 months (January 2005—July 2005) and 85 consecutive NTM isolates over a period of 18 months (Jan 2005—June 2006). All these isolates were cultivated using MGIT 960 and L.J., differentiated as MTB or NTM using p-nitroben-zoic acid (PNBA) on MGIT and p-nitro-a-acetylamino-ß-hydroxypropiophenone (NAP) on BACTEC 460 TB system. Anti-mycobacterial DST of all MTB isolates was performed on BACTEC 460 TB system (1% modified proportion method).
The applicability of this in-house RLBH assay was assessed by processing 48 AFB smear-positive (1+ and above) clinical specimens suspected to have a minimum of 10—99 bacilli/100 oil immersion fields.
Definitions
Phenotypic drug resistance was defined as greater than 1% growth in the presence of 0.1 mg/ml of INH, 2 mg/ml of RIF, 2 mg/ml of STR, and 2.5 mg/ml of ethambutol (EMB).14—16 Multidrug resistance was defined as resistance to at least INH and RIF. Pan-susceptible isolates were defined as isolates susceptible to all drugs tested.
DNA extraction
Extraction of genomic DNA from ATCC, CAP-QC, known mutant strains, and mycobacterial culture isolates was carried out by the CTAB-NACL method described previously.17 DNA extraction from smear-positive clinical specimens was carried out using the Qiagen DNA extraction kit.
Reverse line blot hybridization (RLBH)
Preparation of membrane for RLBH
Of the 54 specific amino labeled oligonucleotide probes used for species identification and DST, 10 probes were newly designed; 44 probes were selected from the previously published literature and modified (whenever required) to adjust length and Tm so that they could be processed under the same hybridization, washing conditions. All these probes were covalently linked to a nylon membrane in parallel lines.18—20 For species identification, oligonucleotide probes specific for the genus Mycobacterium, as well as 15 clinically
important species within this genus including M. tuberculosis complex, Mycobacterium avium, Mycobacterium intracellular, Mycobacterium scrofulaceum, Mycobacterium kansasii, Mycobacterium gastri, Mycobacterium fortuitum complex, Mycobacterium chelonae, Mycobacterium abscessus, Mycobacterium ulcerans, Mycobacterium gordonae, and Mycobacterium szulgai were synthesized. Similarly for susceptibility testing the following gene targets were included: rpob (codons 302—534), katG (codons 315 and 463), inhA (orfl— inhA promoter region), rpsL (codons 43 and 88), and rrs (position 491, 513, and 516). Wild type (wt) sequence as well as the most common mutant (mt) alleles were synthesized to screen the above-mentioned regions. Preliminary work done on the ahpC gene at our center did not show any mutations in the ahpC gene among INH-resistant strains (unpublished data). As mutations causing pyrazinamide (PZA) resistance are scattered throughout the pncA gene, it was not convenient to analyze such an extended region of approximately 600 bp by RLBH assay.21 Hence probes for analysis of ahpC and pncA genes were not incorporated in the RLBH assay. Forty probes were blotted onto a single membrane (membrane I) for identification of MTB and DST (Table 1). Sixteen probes for species identification (Table 2) were blotted onto the second membrane (membrane II). The uniqueness of the newly designed and modified probes was analyzed with a BLAST search (http://www.ncbi.nlm.nih.gov).
In vitro amplification of target genes
Primers were designed with special care to encompass biotin labeling. The length of the biotin unlabeled primer was increased in such a way that the minimum difference between the Tm of the biotin labeled and unlabeled primer was 8—10 °C. A novel primer pair was designed to amplify two overlapping regions specific for species identification and RIF resistance (codons 302—534) of the rpob gene in a single reaction (Table 3). In the case of rpob, inhA, and rrs genes, the reverse primer was biotin labeled as the probes were designed in the forward direction, whereas in the case of katG and rpsL genes the forward primer was biotin labeled as the probes were designed in the reverse direction. These modifications were done as the previous attempts made by Mokrousov et al. to design probes for the katG gene and the rpsL gene did not show any hybridization signals due to the secondary structure simulation for minus strand of katG and rpsL genes.20 Hence the plus strand was targeted and probes were designed in the reverse direction.
All the different target gene fragments were amplified separately and the region of interest was confirmed by sequencing. Further, to make this assay more rapid and cost effective, a touchdown multiplex PCR was standardized (Figure 1), wherein a total of 10 important loci on five target genes associated with species identification of mycobacteria and resistance testing to INH, RIF, and STR were amplified in a single reaction using Hotstart Taq polymerase master mix (Qiagen, Genetix). Positive and negative controls were run at every step. The amplified products were verified on 2.5% agarose gel electrophoresis and used to perform the RLBH assay.
Hybridization with amplified products
Heat denatured PCR products were applied on the membrane in the miniblotter and hybridized at 52 °C (membrane I) and
Table 1 Probes labeled on membrane I, for identification of
Mycobacterium tuberculosis complex and mutations causing
resistance to rifampin (RIF), isoniazid (INH), and streptomy-
cin (STR)
No. Probe Gene Description
1 Identification: rpoß Common for all
MYC mycobacteria
2 MTB rpoß M. tuberculosis complex
RIF resistance:
3 RIF wt 1 rpoß Codons 509-514
4 RIF wt 2 rpoß Codons 514-520
5 RIF wt 3 rpoß Codons 521-525
6 RIF wt 4 rpoß Codons 524-529
7 RIF wt 5 rpoß Codons 530-534
8 RIF mt 1 rpoß Codon 533 CTG^CCG
9 RIF mt 2 rpoß Codon 531 TCG^TTG
10 RIF mt 3 rpoß Codon 531 TCG!TGG
11 RIF mt 4 rpoß Codon 526 CAC!AC
12 RIF mt 5 rpoß Codon 526 CAC^GAC
13 RIF mt 6 rpoß Codon 526 CAC^TGC
14 RIF mt 7 rpoß Codon 526 CAC^CGC
15 RIF mt 8 rpoß Codon 526 CAC^CTC
16 RIF mt 8 rpoß Codon 526 CAC^AAC
17 RIF mt 9 rpoß Codon 522TCG!TGG
18 RIF mt 10 rpoß Codon 522TCG!TTG
19 RIF mt 11 rpoß Codon 516 deletion
20 RIF mt 12 rpoß Codon 516 GAC!AC
21 RIF mt 13 rpoß Codon 516 GAC^GTC
22 RIF mt 14 rpoß Codon 516 GAC^GTG
23 RIF mt 15 rpoß Codon 516 GAC^GGC
24 RIF mt 16 rpoß Codon 513 CAA^CCA
25 RIF mt 17 rpoß Codon 511 CTG^CCG
INH resistance:
26 inhA wt inhA MabA—inhA promoter
region
27 inhA mt inhA C!T transition at
codon 15
28 katG 315 wt katG Codon 315 AGC
29 katG 315 mt katG Codon 315 AGC!CC
30 katG 463 wt katG Codon 463
31 katG 463 mt katG Codon 463
polymorphism C^A
STR resistance:
32 rpsL 43 wt rpsL Codon 43 AAG
33 rpsL 43 mt rpsL Codon 43 AAG^AGG
34 rpsL 88 wt rpsL Codon 88 AAG
35 rpsL 88 mt rpsL Codon AAG^AGG
36 rrs 491 wt rrs Codon 491
37 rrs 491 mt rrs Codon 491-T
38 rrs 513 wt rrs Codon 513 CAG and
516 CCG
39 rrs 513 mt 1 rrs Codon 513 CAG!CCG
40 rrs 513 mt 2 rrs Codon 516 CCG!CTG
wt, wild type; mt, mutant.
50 °C (membrane II) for 60 min. The membrane was washed twice with gentle shaking in 250 ml 2 x SSPE/0.5% SDS for 10 min at 60 °C (membrane I) and 56 °C (membrane II). The membrane was subsequently incubated at 42 °C with 1:4000
Table 2 Probes labeled on membrane II, for identification of clinically relevant species of mycobacteria
No. Probe Description
1 MYC Common for all Mycobacterium species
2 MTB Mycobacterium tuberculosis complex
3 AVI Mycobacterium avium
4 INT Mycobacterium intracellulare
5 SCR Mycobacterium scrofulaceum
6 KAN I Common probe for Mycobacterium
kansasii types I and IV
7 KAN II Common probe for Mycobacterium
kansasii types II, III, and V
8 GAS Mycobacterium gastri
9 FORC Mycobacterium fortuitum complex
10 CHE Mycobacterium chelonae
11 ABS Mycobacterium abscessus
12 ULC Mycobacterium ulcerans
13 GEN/SIM Common probe for Mycobacterium
simiae and Mycobacterium genavense
14 GOR I Common probe for Mycobacterium
gordonae types I, III, and IV
15 GOR II Common probe for Mycobacterium
gordonae type II
16 SZU Mycobacterium szulgai
diluted streptavidin—peroxidase conjugate in 2 x SSPE/0.5% SDS for 60 min, washed twice with 250 ml 2 x SSPE/0.5% SDS at 42 °C for 10 min, rinsed twice with 2 x SSPE at RT for 5 min, and subjected to luminescent detection of hybrids with enhanced chemiluminescence (ECL) detection system followed by exposure to the ECL Hyperfilm (Amersham Biosciences). The presence of a clearly visible black signal was considered as a positive hybridization reaction. The results were easy to interpret (Figures 2 and 3). For reuse, the membranes were stripped in 1% SDS solution at 80 °C (2 x 40 min) and rinsed in 20 mM EDTA, pH 8.0 at RT. The membranes were re-used up to 7—8 times.
Figure 1 Amplification of five target genes by touchdown multiplex PCR (M = molecular weight marker; lanes 1 and 2 = patient sample; PC = positive control; NC = negative control).
Standardization and validation
Standard ATCC strains, CAP-QC strains, and M. tuberculosis complex strains with known mutations were used to standardize the probe concentrations and assay conditions for species identification and drug resistance testing. Over a period of 2 years, nearly 2500 PCR reactions were performed, and series of experiments with varying probe concentrations (1—5000 pmol), different hybridization temperatures (40— 65 °C), washing temperatures (40—60 °C), and different exposure times (7—45 min) were performed to standardize the assay conditions. The RLBH assay was successfully standardized after 85 attempts. Reproducibility of the RLBH assay was verified by testing 385 clinical isolates and 48 smear-positive clinical specimens (+1 and above), and results were compared with phenotypic culture methods as well as with genotypic methods. All the species identified by RLBH were confirmed by restriction enzyme (RE) analysis as per work done by Lee et al.13 Representative strains of all species, wt/
Table 3 Primer sequences and PCR conditions to amplify target genes
Seq. ID No. Gene
Primers
Tm °C Conditions
P-1 rpob codons Rpo F 5'-TCA AGG AGA AGC GCT ACG ACC TGG-3'
302—534 Rpo R 5'-biotin-GGG TCT CGA TCG GGC ACAT-3' (Reverse primer biotin labeled) P-2 inhA TB 92M 5'-CCT CGC TGC CCA GAA AGG GAT CC-3'
TB 93M 5'-biotin-CCG GGT TTC CTC CGG T-3' (Reverse primer biotin labeled) P-3 katG KG F biotin-5'-AGA GCT CGT ATG GCA CCG GA-3'
(Forward primer biotin labeled) KG R 5'-GCG AAT GAC CTT GCG CAG ATC-3' P-4 rpsL STR 52 5'-biotin-GTC AAG ACC GCG GCT CTG A-3'
(Forward primer biotin labeled) rpsL R 50-ACG CTT GGG CGC GGG CCC C-30 P-5 rrs STR 53 5'-TCA CCA TCG ACG AAG GTC CG GG-3'
STR 31 5 '-CTA GAC GCG TCC TGT GCA TG-3' 64
(Reverse primer biotin labeled)
95 °C 15 min
(95 °C 1 min 66 °C 1 min 72 °C 1 min) x 2c (95 °C 1 min 65 °C 1 min 72 °C 1 min) x 2
(95 °C 1 min 64 °C 1 min 72 °C 1 min) x 2
(95 °C 1 min 63 °C 1 min 72 °C 1 min) x 2
(95 °C 1 min 62 °C 1 min 72 °C 1 min) x 30/35
72 °C 5 min Hold at 15 °C
PROBES
RpoiS wild
► RIF --
Rpûli mutant Probe
InhA kat G
RIF 616 TAC
RIF 531 TTG
Pan susceptible
E ■ ■ • ■ ■ a
G ■ ■ • * ■ ■
• m i * * ■
■i ■ ■ ■ ■
7 ■ • • •
S ■ ■ ■ • > a
9 s •
1» a ■ ■ ■ ■ ■ a
11 ■ -> . ■ a
12 a ■
1J a ■ ■ a a ■
14 ■ ■ ■ ■
1í • M ■ a*
16 ■ ■
17 t m
18 m m
2« m a
Î1 m m
22 • m
?J IL - ■
ÍS * y* *
» fl t V
27 « a
28 • m
29 m m
30 31 • ■
i 5 6> a 9101112 13 141516 1718 15 20 212223 24 2S26 27 28 29 30 31 51^36 37 38 39 ¿
« ■
a • a
■ a a
■ ■
• ■
t * ■ ■
a • ■
a • ■
• ■
• ■ ■
• • ■
■ « •
» ■
■ ■
■ MYC
■ MTB com • RIF WT1
■ RIF WT2
■ RIF WT3 RIF WT4 RIF WT5
8 - 533 CCG
9 - 531 TTG 10-531 TGG 11 - 526 GAC
12-526 TAC
13- 526 CGC
14- 526 CTC
15 - 526 TGC
16 - 526 AAc 17- 522 TTG 18 - 522 TGG
19-516 TAC
20-516 GAC 21 - 516 GTC
22-516 GGC
23-Del 516
24- 513CCA
25-511 CCG 26 - InhA WT 27-lnhAMT
28 - katG 315 WT 29-katG 315 MT
30 - katG 463 WT
31 - katG 463 P
32 - RpsL 43 WT
33 - RpsL 43 MT
34 - RpsL 88 WT
35 - RpsL 88 MT
36 - Rrs 491 WT
37 - Rrs 491 MT
38 - Rrs 513 WT
39 - Rrs 513 MT 40- Rrs 516 MT
Figure 2 Membrane I for identification of Mycobacterium tuberculosis complex and drug resistance testing. On membrane I, all amino labeled probes are blotted vertically and then biotin labeled heat denatured PCR product is applied horizontally. Lane 1 corresponds to probe common for all mycobacteria. Lane 2 has probe for M. tuberculosis complex. Lanes 3—7 are five wt probes for detection of rifampin (RIF)-susceptible strains. Lanes 8—25 are 18 mt probes for detection of rifampin resistance. Lanes 26 and 27 are inhA wt and mt probes, respectively, for isoniazid (INH) resistance. Lanes 28 and 29, 30 and 31 represent wt and mt probes for codons 315 and 463 in the katG gene, respectively. Lanes 32—35 correspond to codons 43 and 88 of rpsL wt and mt probes, respectively, for streptomycin (STR) resistance. Similarly lanes 36—40 correspond to wt and mt probes for nucleotide positions 491, 513, and 516 in the rrs gene. Horizontal lanes 1—31 represent heat-denatured PCR products, i.e., patient samples or isolates. To illustrate one example, patient 10 has mycobacterial infection due to M. tuberculosis complex and hybridization signal with all RIF wt probes and none of the mt probes; therefore it is RIF-susceptible. Similarly it shows hybridization with inhA, katG, rpsL, and rrs — all wt type probes — hence it is sensitive to INH and STR. In the case of patient 2, there is mycobacterial infection due to M. tuberculosis complex, however the strain is resistant to RIF as there is no hybridization signal with wt probe 4, but hybridization with probe 19, which is a 516 TAC mutation. Similarly in the same patient there is hybridization signal with inhA wt and katG mt indicating INH resistance due to mutation in katG gene codon 315. In the case of STR, the same patient's sample showed hybridization signal with rpsL codon 43 and 88 wt type probes and rrs gene codons 491 wt and 516 mt probes. So the strain is resistant to STR due to mutation at codon 516 of the rrs gene. Hence the patient's strain is M. tuberculosis complex resistant to RIF, INH, and STR.
mt alleles, and all the mutants not binding to wt and mt probes were further confirmed by sequencing.
Fingerprinting
Fingerprinting was carried out by spoligotyping to check the clonality and mutation patterns in different clusters.18
Results
The different species identified by RLBH assay are described in Table 4. All 300 MTB isolates were correctly identified by RLBH. Among 85 NTM, 26 were rapidly growing mycobacteria (RGM), identified as M. fortuitum (46%) and M. abscessus (54%). Of the 59 slowly growing NTM, 17 (29%) showed 99%
homology with M. intracellulare and 26 (44%) showed 98% homology with M. simiae. Other NTM identified included M. kansasii (5%), M. gordonae (5%), and M. szulgai (1.7%), and one isolate (1.7%) showed 95% homology with M. avium. Mixed infections (MTB + NTM) were noticed in three cases. Interestingly, all patient isolates identified as M. intracellulare by RLBH showed the standard RE pattern of 175—100—80 with MspI digest; however with HaeIII enzyme they showed a slightly different pattern of 180—100 instead of 180—90. Further sequencing of all these strains showed insertion of 12 nucleotides in the rpob gene. There were five patient strains identified as NTM by phenotypic NAP and PNBA results, however they were confirmed as MTB by RLBH.
As described in Table 5, of the 300 MTB isolates, 179 were isolated from pulmonary specimens and 121 from extrapul-
Figure 3 Membrane II for species identification of mycobacteria. All the different amino labeled probes are blotted vertically. Then heat-denatured biotin labeled PCR products are applied horizontally. If the infection is due to mycobacteria, hybridization will be observed with probe 1, which is common for all mycobacteria. Further, depending upon the species, hybridization will be observed with the corresponding probes representing that species, e.g., patient 1 has mycobacterial infection due to M. tuberculosis complex, whereas in the case of patient 3 there is an infection due to M. abscessus as it shows hybridization with probe 11 corresponding to M. abscessus. In the case of mixed infection one can get the hybridization signal with two different species. Similarly if the infection is due to some other mycobacterial species that is not clinically relevant or not targeted, there will be a hybridization signal with probe 1 and there will not be any hybridization signal with any of the other probes.
monary specimens. Of the total, 63% were resistant to one or more anti-TB drugs phenotypically. Resistance to INH was found to be 60%, whereas resistance to RIF was 53% and STR was 46%. Among pulmonary isolates, the MDR-TB rate was found to be 70%, whereas in the case of extrapulmonary isolates the MDR-TB rate was only 26%.
Mutations in the rpoß gene were detected in 169 cases by RLBH assay (Table 6), of which 158 MTB isolates were resistant to RIF phenotypically. Of the 11 isolates showing mutations by RLBH, 10 were found to be resistant to INH phenotypically. We observed 22 different types of mutations. The codons most frequently involved in mutations were codon 531 (76%), codon 526 (7%), and codon 516 (9%). The most common mutation, Ser531Leu (TCG^TTG), was found in 126 of the 169 RIF-resistant isolates (75%). Double mutations were observed in five strains and deletion of codon 516 was noticed in two strains. No mutations were detected in two RIF-resistant strains. Compared to phenotypic DST, the concordance of the RLBH assay for the detection of RIF resistance was 96%.
As summarized in Table 7, of the 179 INH-resistant cases, 82% of resistant strains showed the mutation G^C at codon 315 in the katG gene, 9% showed the -15 C^T mutation in
the in inhA gene promoter region, while 7% isolates showed both the inhA -15 + katG 315 mutations together. Overall 89% of the INH-resistant isolates had a Ser315Thr mutation at codon 315. Seventy nine percent of our strains showed C—A polymorphism at codon 463 in the katG gene. The katG deletion was not detected in any of the INH-resistant isolates. Overall 99% concordance was observed with phenoty-pic DST for INH.
Among STR-resistant isolates (Table 8), the most prevalent mutation was the single substitution Lys43Arg in the rpsL gene found in 75% of the isolates analyzed. The other muta-tioninthe rpsL gene was substitution Lys88Arg found in 12% of the STR-resistant isolates. Five percent had the Gln513Pro mutation at nucleotide position 513 and 6% had Pro516Leu mutation at position 516 in the rrs gene. Not a single STR-resistant isolate showed mutation at position 491 in the rrs gene. Two percent of our clinical isolates did not show any mutations in the targeted region. It is also worth noting that none of the tested strains had rpsL and rrs gene mutations simultaneously. The RLBH results for STR resistance showed 98% concordance with the phenotypic DST.
Finally, the applicability of this assay was analyzed in 48 smear-positive (1+ and above, i.e., 10—99 bacilli/100 oil immersion fields) clinical specimens and the results obtained are summarized in Table 9. All were identified as MTB by RLBH. Of the 48, 14 were identified as wt to all the target genes and hence susceptible to RIF, INH, and STR. The remaining 34 were resistant to INH (Table 9). Twenty-four had the Ser315Thr mutation in katG, three had the -15 C—T mutation at the inhA promoter region, whereas seven specimens showed the presence of both these mutations, i.e., Ser315Thr in katG gene and —15 C—T in inhA gene together. Thirty-two showed mutations at codons 531, 516, and 513 in the rpob gene, hence they were resistant to RIF. Twenty-six RIF-resistant specimens had the Ser531Leu mutation in the rpob gene. Twenty-three specimens were identified as STR-resistant by RLBH assay. Twenty showed mutation Lys43Arg in the rpsL gene, two had the Lys88Arg mutation in the rpsL gene, whereas one specimen showed Pro516Leu change in the rrs gene. When applied directly to the smear-positive clinical specimen results of both species, identification and resistance testing to INH, RIF, and STR were available within 4 working days (Table 10).
Discussion
Rapid and accurate identification of mycobacteria with prompt anti-mycobacterial susceptibility testing is the cornerstone for the treatment of infected patients and for preventing pathogen spread to others. We have described an in-house RLBH method for simultaneous species identification and detection of mutations associated with multiple drug resistance in MTB.
The 360-bp region of the rpob gene targeted for speciation (codons 302—340) is located near the rpob mutation sites for RIF resistance (codons 509—534), so both these regions can be amplified together for both speciation and RIF resistance. A novel primer pair was designed to amplify both these overlapping regions (total 564 bp) simultaneously as a single reaction. We further standardized a touchdown multiplex PCR wherein all five target genes viz. rpob, katG, inhA, rpsL, and rrs were amplified together.
Table 4 Analysis of mycobacterial isolates (N = 385) for species identification by RLBH and comparison with culture, PCR-RE, and sequencing
No. of AFB RLBH PCR-RE analysis Sequencing
isolates culture
MspI HaelII Species
6 RGM M. fortuitum 175-100-70 120- -90- 80 M. fortuitum type I M. fortuitum type I (4 strains)
6 RGM M. fortuitum 105-95-80-70 120- -90- 80 M. fortuitum type II M. fortuitum type II (2 strains)
14 RGM M. abscessus 105-95-80-45 130- -100 -90 M. abscessus M. abscessus (4 strains)
17 NTM M. intracellulare 175-100-80 180- -100 M. intracellulare M. intracellulare (4 strains)
26 NTM GEN/SIM 175-100-70-45 -35 180- -110 M. simiae M. simiae (6 strains)
3 NTM M. kansasii 175-60-40 205 -90 M. kansasii type I M. kansasii type I (1 strain)
3 NTM M. gordonae 145-95-40-30 210- -95- 90 M. gordonae type I M. gordonae type I (1 strain)
1 NTM M. avium 120-90-40 210- -100 Could not be interpreted 95% homology with M. avium
1 NTM M. szulgai 175-80-40 100- -115 M. szulgai M. szulgai (1 strain)
1 NTM MTB + M. gordonae 145-95-80-60- 40- 30 250 -95- 90 Indeterminate (mixed RE pattern) MTB + M. gordonae type I (1 strain)
1 NTM MTB + M. simiae 175-100-80-70 -60 -40 250 -180 -100 Indeterminate (mixed RE pattern) MTB complex + M. simiae (1 strain)
1 NTM MTB + M. fortuitum 175-100-95-80 -60 -40 250 -120 -100-80 Indeterminate (mixed RE pattern) MTB complex + M. fortuitum (1 strain)
5 NTM MTB 175-80-60-40 250 -100 MTB complex MTB complex (3 strains)
300 MTB complex MTB complex (300 strains) 175-80-60-40 250 -100 MTB complex (25 strains) MTB complex (25 strains)
RLBH, reverse line blot hybridization; PCR-RE, PCR-restriction enzyme assay; AFB, acid-fast bacillus; RGM, rapidly growing non-tuberculous mycobacteria; NTM, non-tuberculous mycobacteria; MTB, Mycobacterium tuberculosis complex; GEN/SIM, Mycobacterium simiae and Mycobacterium genavense.
All newly designed as well as modified probes showed good hybridization results. Probes common for the genus Mycobacterium and MTB were incorporated on membrane I along with different wt and mt probes for resistance testing. Probes for different genes were immobilized on the membrane side by side, and the hybridization was performed simultaneously with five amplified PCR fragments. This did not influence the strength of the signals
and did not show any cross-reaction with other PCR products. The signal strength decreased slightly on reuse of membrane up to 7—8 times but this did not affect evaluation. As in more than 95% of the cases infection was due to MTB, maximum information was obtained using this single membrane. PCR products not showing any hybridization signal with an MTB probe were further processed on membrane II for speciation.
Table 5 Analysis of Mycobacterium tuberculosis complex isolates (N = 300) and drug susceptibility by phenotypic method
Pulmonary specimens Extrapulmonary specimens Total
179 (60%) 121 (40%)
Pan-susceptible (sensitive to all drugs) 34 (19%) 78 (64%) 112 (37%)
Resistant to one/more drugs 145 (81%) 43 (36%) 188 (63%)
Resistant to INH 140 (78%) 39 (32%) 179 (60%)
Resistant to RIF 127 (71%) 31 (26%) 158 (53%)
Resistant to STR 113 (63%) 25 (21%) 138 (46%)
Resistant to EMB 91 (51%) 28 (23%) 119 (40%)
MDR (resistant to INH and RIF) 125 (70%) 31 (26%) 156 (52%)
INH, isoniazid; RIF, rifampin; STR, streptomycin; EMB, ethambutol; MDR, multidrug-resistant.
Table 6 Mutations observed in the rpob gene by RLBH assay (N = 169)
Mutation No. % In-house RLBH assay Change confirmed by sequencing
Serine^leucine (S531L) 126 74.6 531 TCG^TTG 531 TCG- TTG
Serine^leucine (S531L) 2 1.2 531 TCG!TGG 531 TCG- TGG
Histidine^aspartate (H526D) 2 1.2 526 CAC^GAC 526 CAC- »GAC
Histidine^tyrosine (H526Y) 3 1.8 526 CAC!AC 526 CAC- TAC
Histidine^cysteine (H526C) 4 2.4 526 CAC^TGC 526 CAC- »TGC
Histidine! (H526R) 1 0.6 526 CAC^CTC 526 CAC- CTC
Histidine! (H526) 2 1.2 526 CAC^CGC 526 CAC- CGC
Aspartate!tyrosine (D516Y) 5 3 516 GAC!AC 516 GAC- TAC
Serine S522Q 1 0.6 No hybridization signal with wild type probe 3 and none of the mutant probes 522 TCG- CAG
Aspartate!Glu (D516V) 7 4.1 516 GAC^GTC 516 GAC- GTC
Aspartate!glycine (D516G) 2 1.2 516 GAC^GGC 516GAC- GGC
Deletion of codon 516 2 1.2 516 DGAC 516 DGAC
Gln!leucine (Q513L) 1 0.6 No hybridization signal 513 CAA-CTA
with wild type probe 1 and
none of the mutant probes
Gln!lysine (Q513K) 1 0.6 No hybridization signal 513 CAA-AAA
with wild type probe 1 and
none of the mutant probes
Gln!lysine (Q513K) 1 0.6 513 CAA^CCA 513 CAA- CCA
Leucine!proline (L511P) 1 0.6 511 CTG^CCG 511 CTG- CCG
Deletion of codons 509-511 1 0.6 No hybridization signal D AGC, CAG, CTG
with wild type probe 1
and none of the mutant probes
Methionine!leucine (M515L) 1 0.6 No hybridization signal 515 ATG-CTG
with wild type probes 2 and 4
Histidine!cysteine (H526C) Mutant 526 CAC^TGC 526 CAC- »TGC
Asparagine!aspartate (H518D) 1 0.6 No hybridization signal with wild type probes 2 and 4 518 AAC- GAC
Histidine!asparagine (H526N) Mutant 526 CAC^AAC 526 CAC- AAC
M515V 1 0.6 No hybridization signal with wild type probes 2 and 4 515 ATG- GTG
L511P Mutant 511 CTG^CCG 511 CTG- CCG
M511I 1 0.6 No hybridization signal with wild type probes 1 and 4 and hybridization with mutant probe 511 ATG- ATT
Histidine!asparagine (H526N) 526 CAC^AAC 526 CAC- AAC
Aspartate!tyrosine (D516Y) 1 0.6 516 GAC!AC 516 GAC- TAC
Asparagine!aspartate (H518D) No hybridization signal with wild type probes 1 and 3 518 AAC- GAC
Undetected 2 1.2
RLBH, reverse line blot hybridization.
As demonstrated in Table 5, in the case of 300 MTB isolates, 100% concordance was observed between RLBH results and phenotypic PNBA and NAP results. However of the 85 NTM detected by phenotypic methods, five were identified as MTB by RLBH. Further analysis of these five isolates by RE and sequencing confirmed that these isolates were falsely identified as NTM by phenotypic methods. Amongst the RGM, M. fortuitum complex and M. abscessus were predominant, whereas among the slowly growing NTM, M. intracellulare and M. simiae were predominant. Mixed infections were noticed in three cases by RLBH. The advantage is that both species can be identified to the species level using this assay, which is sometimes difficult using phenotypic
culture methods due to overgrowth of NTM over the slower growing MTB. All patient isolates identified as M. intracellu-lare showed insertion of 12 nucleotides in the rpob gene. This has not been reported before.
Since NTM are ubiquitous in nature, the isolation of these organisms from specimens should meet certain criteria to confirm their etiological significance, such as: (a) repeated isolation of the same organism from a patient, (b) associated positive clinical and radiological evidence and (c) histopatho-logical confirmation, (d) presence of any predisposing factors/underlying diseases, and (e) the immune status of the patient — this aids in assessing the pathogenesis of NTM when isolated.22,23 In the present study, 49% of patients met the
Table 7 Mutations observed in isoniazid-resistant strains by RLBH assay (N =179)
Region Number Amino acid change
(%) by sequencing
inhA gene promoter 16 (9%) -15 C!T
region
katG gene 147 (82%) S315T G!
inhA promoter 13 (7%) -15 C!Tand S315T
region + katG
Undetected 3 (2%)
RLBH, reverse line blot hybridization.
Table 8 Mutations observed in streptomycin-resistant strains
by RLBH assay (N = 138)
Region Number (%) Amino acid change
by sequencing
rpsL 43 mt 103 (75%) AAGLys!GGArg
rpsL 88 mt 17 (12%) AAGLys!GGArg
rrs 513 mt 7 (5%) CAGGln^CCGPro
rrs 516 C—T 8 (6%) CCGPro^CTGLeu
Undetected 3 (2%)
RLBH, reverse line blot hybridization. mt (mutant) denotes resistant strains.
American Thoracic Society (ATS) criteria and were considered as true infections. The prevalence of AIDS in our patient population was less than 2%, indicating that the infections due to NTM are increasing even in non-HIV patients.
Although there have been reports from India, the exact disease burden of NTM infections in India remains unclear. These infections are under-diagnosed in many laboratories due to lack of facilities and expertise. Our findings clearly emphasize the need for speciating all mycobacterial isolates. As most NTM are not susceptible to conventional anti-TB treatment, it is desirable to identify these isolates to the species level, particularly from AIDS and drug-resistant cases to avoid them being misinterpreted as MDR-TB.
For resistance testing, the presence of mutations was demonstrated either by hybridization to the corresponding probe representing the mt probe or by loss of hybridization to the respective wt genotype only. In nine cases, we observed loss of hybridization to the respective wt genotype only. This hybridization pattern could not tell the accurate mutation genotype, but could still detect the mutations associated with RIF resistance. Further confirmation of these nine isolates by sequencing showed deletion of three codons — 509, 510, and 511 — in one isolate. Non-targeted mutations at codons 522, 518, 515, and 513 were observed in the remaining eight isolates with double mutations in five cases. Interestingly the mutations AAC at codon 526, GAC at codon 518,
Table 9 Analysis of smear-positive clinical specimens by RLBH (N = 48):a detailed screening of all 34 isoniazid-resistant specimens
Drug Gene Susceptible Resistant mutants
Rifampin (RIF) rpoß 2 32 Mutations identified: S531L (26), S531W (1), D516V (2), D516Y (2), Q513K (1)
Isoniazid (INH) inhA katG inhA + katG 0 34 Mutations identified: -15 C—T (3) S315T (24) -15 C—T + S315T (7)
Streptomycin (STR) rpsL codon 43 rpsL codon 88 rrs codon 516 11 23 Mutations identified: Lys43Arg (20) Lys88Arg (2) Pro516Leu (1)
RLBH, reverse line blot hybridization.
a M. tuberculosis complex = 48; wild types to all target genes (pan-susceptible) = 14; resistant to one or more drugs (mutation detected in one or more genes) = 34.
Table 10 Comparison of in-house RLBH assay with phenotypic methods
Identification and recovery
Susceptibility testing for firstline drugs
Total time for identification and susceptibility testing
Efficiency
L.J. Media BACTEC 460 MGIT 960 RLBH assay
30-50% 50-70% 50-70% 95-98%
2—8 weeks 2—3 weeks 1—2 weeks 4 days
80-99% 92-100% 98.6-100% 98%-100%
2—4 weeks 8—10 days 6—10 days 4 days
4-12 weeks 3-5weeks 2-4 weeks 4 days
RLBH, reverse line blot hybridization.
Probe RIF1W (codons 509-514) prabe R|F2W (codons 514-520)
AGC CAG CTG AGC CAA TTC TTC ATG GAC CAG GAA CAA CCG
AGC CAG CTG AGC CAA TTC TTC ATG GAC CAG GAA CAA CCG CTG
509 510 511 512 513 514 514 515 516 517 518 518 520 521
rpoß gene
Insertion TTC
Figure 4 Diagrammatic representation of a false susceptible result obtained by reverse line blot hybridization (RLBH).
and ATT or CTG at codon 515 did not occur alone and were present in isolates with more than one mutation. As summarized in the Table 6, screening of the rpob gene for RIF-resistant strains revealed that 98.8% of the strains had mutations in the 80 bp core region. Various investigators have reported a rate of 20—70% for mutation at codon Ser531Leu (TCG to TTG) in the rpob region for RIF-resistant strains, followed by 0—30% at codon 526 (CAC to GAC).24,25 Indian studies carried out so far have reported rates of 39— 53% for mutation at codon 531,26—28 whereas we found this mutation at a much higher rate (75%) in RIF-resistant isolates.
Mutations were not detected in two phenotypically RIF-resistant cases. Further, sequencing of these isolates showed an insertion of codon 514 in the rpob gene in one of the isolates, whereas no mutation was observed in the RIF resistance determining region (RRDR) of the rpob gene in the remaining isolate. The occurrence of 514 TTC insertion in the rpob gene is rare and has been reported previously.29 Of the five wt probes designed for screening the RRDR of the rpob gene, RIF1W covers the codons 509—514 whereas RIF2W covers the region of codons 514—520 (Figure 4). As codon 514 is located close to the 3' end of the RIF1W probe (codons 509—514), and it is the starting point of the 5' end of the RIF2W probe (codons 514—520), both the probes showed good hybridization signals. Therefore it was falsely identified as a RIF-susceptible strain instead of resistant. This problem can be overcome by increasing the length of the RIF1W probe.
Conversely, 11 phenotypically RIF-susceptible isolates were revealed to harbor a mutation in the rpob gene. Interestingly 10/11 were found to be resistant to INH phenotypically. In seven of these 11 cases, mutations were found to be present at codons 522, 518, 516, 515, and 511, which have been reported to cause low level resistance, whereas in the remaining four cases the mutations were at codons 531 and 526, which have been reported to cause high level resistance. All these isolates were rechecked by both the methods and finally confirmed by sequencing. Phenotypic DST was performed using BACTEC 460 TB system, which cannot determine the actual proportions of resistant bacteria in culture.30'31 In addition, there is evidence that the BACTEC method may not be a reliable tool for determining resistance in cultures containing less than 10% drug-resistant bacteria,30,31 hence all these cases were identified as susceptible phenotypically.
INH is a key component of the anti-tuberculosis, short course chemotherapy. Various studies have shown that the mutation Ser315Thr in the katG gene is most abundant and present in the 33—97% of INH-resistant strains reported worldwide.22,32—35 The prevalence of this mutation is found to be highest in isolates recovered from patients of South Asian origin and lowest in isolates from patients of Middle Eastern origin. This mutation has specifically been reported
to be predominant (>90%) amongst the MDR isolates from Russia and Latvia.35,36 In our study, overall 89% of the INH-resistant isolates had this mutation. Variation in katG 463 has no relation to INH resistance and was included on the membrane as a marker frequently used in evolutionary studies. In particular, katG 463-Leu is indicative of group I (presumably the most ancient).37 In the present study, this allele was found in the majority (79%) of the isolates. The C^T mutation at the 5' end of a presumed ribosome-binding site of inhA promoter was found in overall 16% of our patient population.
In the present study, 87% of STR-resistant clinical isolates showed mutations in the rpsL gene at codons 43 and 88 (Table 8). Eleven percent showed mutations in the rrs gene at positions 513 and 516. Mutations in rpsL and rrs genes were shown to confer STR resistance in up to 80% of STR-resistant MTB isolates.38,39 In comparison, our results showed higher mutations (87%) in the rpsL gene and 11% in the rrs gene. Three (2%) of the STR-resistant culture isolates did not show any mutations in the rpsL or rrs genes. These isolates probably acquired resistance by other means. None of the STR-susceptible isolates examined had alterations in the rpsL or rrs genes.
Application of the RLBH assay on culture isolates is still time-consuming, as mycobacterial growth takes 2—6 weeks. To further reduce the time, the applicability of this assay was analyzed in 48 smear-positive (1+ and above, i.e., 10—99 bacilli/100 oil immersion fields) clinical specimens and the results obtained are summarized in Table 9. Of these 48 smear-positive specimens, four were from patients who had been receiving anti-TB treatment for various periods of time, and culture was negative in all these patients even after6weeks. In contrast, RLBH results were available within 4 days of specimen collection (including the time required for DNA extraction, amplification, RLBH, and results analysis) and correlated well with the clinical picture of all these patients. This highlights the importance of RLBH analysis in cases where the culture is negative and DST cannot be performed. The technique was found to be rapid and reproducible in smear-positive clinical specimens with a total turnaround time of only 4 working days for both identification and susceptibility testing.
Comparison of the RLBH assay with commercially available kits for rapid detection of MDR-TB is shown in Table 11. Commercial assays have separate kits for species identification of mycobacteria and separate kits for detection of drug resistance. Tests like the InnoLiPA for RIF resistance can detect MTB targeting 23SrRNA—16SRNA and cover only the four most frequently observed mutations viz. D516V, H526Y, H526L, and S531L in the rpob gene. Of the two commercially available line probe assays, the Hain test is used to detect katG gene mutations, however it is not available in India. They do not target any mutation for STR resistance. In the
Table 11 Comparison of in-house RLBH assay with commercially available line probe assays
In-house RLBH assay Genotype MTBDR InnoLiPA Rif.TB (Innogenetics)
(Hain Life Sciences)
Species identification M. tuberculosis complex M. tuberculosis complex M. tuberculosis complex
+ 14 clinically relevant
species of NTM
Gene targeted for rpob 23S-rRNA/16-RNA 23S-rRNA
species identification
DNA amplification Multiplex PCR for five Multiplex PCR for two PCR
target genes target genes
Direct application to Yes Yes Yes (new version)
clinical specimens
Mutations targeted
rpob 17 mutations (25 probes) 4 mutations (9 probes) 4 mutations (9 probes)
katG 4 mutations 4 mutations Not included
inhA 1 mutation (2 probes) 1 mutation (2 probes) Not included
rpsL 2 mutations (4 probes) Not included Not included
rrs 3 mutations (5 probes) Not included Not included
Assay description Blot assay Strip assay Strip assay
Cost per test (Rs, Indian rupee) 1500/- Not available in India 5000/-
RLBH, reverse line blot hybridization; NTM, non-tuberculous mycobacteria. RLBH has been approved as a new diagnostic test in our hospital.
present study 22 different types of mutations were observed in RIF-resistant isolates. Of the total mutations observed in RIF-resistant cases, only 82% can be detected through the resistant probes targeted by commercial assays. In our RLBH assay probes for detection of mutations on rpsL, rrs are also included to detect resistance to STR. The advantage of the in-house RLBH assay is that more probes can be designed and incorporated as per the regional variations depending on the epidemiological data.
Commercial assays are strip-based assays hence use one strip per patient. In contrast this in-house RLBH assay is a blot assay. At times, 40 different patient samples can be screened using a single membrane and common reagents like washing solutions, hybridization solutions, and detection reagents. It is easy to handle, and as for the spoligotyping membrane, once prepared the membrane can be used at least 7—8 times. In resource-limited settings, approximately 320 patients can be screened using a single membrane. Analysis of multiple samples and reuse of the membranes further reduces the cost of the assay without affecting the quality of results. The cost of the current commercial LiPA kit is $45 per sample tested. When additional costs for import and transport are taken into account, the actual cost per sample is as high as $116, i.e., approximately 5000 Indian rupees (Rs).40 This high price precludes the wide application of this assay in the geographic areas where it is most urgently needed; our in-house assay costs approximately Rs 1500 per sample for speciation with detection of MDR and SM resistance. Hence it is suitable for application in the routine clinical laboratory, especially in developing countries.
Results of this study indicate a high predominance of rpob 531, katG 315, and rpsL 43 mutations in our MDR-TB isolates. Further analysis of all these MDR isolates by spoligotyping showed the prevalence of Beijing genotype; 54% of MDR isolates belong to Beijing genotype showing an increase from the previously reported 35%.41 Among this Beijing group, 84% of RIF-resistant strains showed Ser531Leu mutation in the
rpob gene, 96% Ser315Thr at codon 315 in the katG gene, and 83% STR-resistant strains showed Lys43Arg at codon 43 in the rpsL gene. As reported previously,35'41—43 we also observed the predominance of katG codon 315 mutations in Beijing genotype. This high resistance should be interpreted with caution as it is from a tertiary care center with a referral bias towards non-responding cases and so is not representative of the prevalence of resistance in the nation.
To summarize, we have successfully standardized an in-house RLBH assay, the first of its kind for species identification with simultaneous drug resistance testing to INH, RIF, and STR in MTB. Results of this study indicate a high predominance of rpob 531, katG 315, and rpsL 43 mutations among Beijing genotype and MDR-TB isolates. Therefore, the in-house RLBH assay can be a practical tool for rapid detection of multiple point mutations in the routine diagnostic workup. The method is rapid, requires the same equipment as that used for spoligotyping, and allows for the analysis of up to 40 samples simultaneously. Multiple point mutations can be screened in this single experiment and results are available within 4 working days when applied directly to the smear-positive clinical specimens. Besides being cost effective, this assay enables earlier treatment and prompt implementation of infection control procedures to reduce the morbidity and mortality and the spread of MDR-TB. The present RLBH is an adequate tool for the direct analysis of MDR-TB in clinical specimens and fulfills many of the attributes of an ideal TB diagnostic test as proposed by the WHO for developing countries.
Acknowledgements
This study was supported by the National Health and Education Society of P.D. Hinduja National Hospital and Medical Research Centre. We thank Dr Max Salfinger, Wadsworth Center, New York Health Department for sending us the known mutant strains for standardization of this assay.
Conflict of interest: No conflict of interest to declare. References
1. Cohn D, Bustreo LF, Raviglione MC. Drug-resistant tuberculosis: review of the worldwide situation and the WHO/IULATD global surveillance project. Clin Infect Dis 1997;24(Suppl l):S121—30.
2. Rodrigues C, Shenai S, Sadani M, Thakkar P, Sodha A, Soman R, et al. Multidrug-resistant tuberculosis in Mumbai: it's only getting worse. Int J Tubercle Lung Dis 2006;10:1421-2.
3. Almeida D, Rodrigues C, Udwadia ZF, Lalvani A, Gothi GD, Mehta P, et al. Incidence of multidrug-resistant tuberculosis in urban and rural India and implications for prevention. Clin Infect Dis 2003;36:e152—4.
4. Almeida D, Udwaida ZF, Rodrigues C, Mehta A. Drug-resistant tuberculosis in India. Int J Tuberc Lung Dis 2002;6:1-2.
5. Rosenzweig DY. Non-tuberculous mycobacterial disease in the immunocompetent adult. Semin Respir Infect 1996;11:252-81.
6. Kim BJ, Lee LJ, Park BN, Kim SJ, Bai GH, Kim SJ, et al. Differentiation of mycobacterial species by PCR- restriction analysis of DNA (342 base pairs) of RNA polymerase gene (rpoß). J Clin Microbiol 2001;39:2102-9.
7. Gordin F, Slutlin G. The validity of acid-fast smears in the diagnosis of tuberculosis. Arch Pathol Lab Med 1990; 114: 1025-7.
8. Paramasivan CN, Govindan D, Prabhakar R, Doumsundaran PR, Subbammal S, Tripathy SP. Species level identification of non-tuberculous mycobacteria from South Indian BCG trial area during 1981. Tubercle 1985;66:9-15.
9. Hines N, Payeur JB, Hoffman LJ. Comparison of the recovery of Mycobacterium bovis isolates using the BACTEC MGIT 960, BAC-TEC 460 system, and Middlebrook 7H10 and 7H11 solid media. J Vet Diagn Invest 2006;18:243-50.
10. Springer B, Stockman L, Teschner K, Roberts GD, Bottger EC. Two-laboratory collaborative study on identification of mycobacteria: molecular versus phenotypic methods. J Clin Microbiol 1996;34:296-303.
11. Kirschner P, Rosenau J, Springer B, Teschner K, Feldmann K, Böttger EC. Diagnosis of mycobacterial infections by nucleic acid amplification: 18-month prospective study. J Clin Microbiol 1996;34:304-12.
12. Johansen IS. Rapid diagnosis of mycobacterial diseases, and their implications on clinical management. Dan Med Bull 2006; 53:2845.
13. Lee H, Park HJ, Cho SN, Bai GH, Kim SJ. Species identification of mycobacteria by PCR-restriction fragment length polymorphism of the rpoß gene. J Clin Microbiol 2000;38:2966-71.
14. Collins CH, Grange JM, Yates MD. Organization and practice in tuberculosis bacteriology. 2nd ed. London: Butterworth Heinemann; 1997.
15. World Health Organization/International Union Against Tuberculosis and Lung Disease. Guidelines for surveillance of drug resistance in tuberculosis. WHO/TB/96,216. Geneva, Switzerland: WHO/IUATLD; 1997.
16. Siddiqi SH. BACTEC TB product and procedure manual, revision D. Sparks, MD, USA: Becton Dickinson; 1995.
17. van Soolingen D, Hermans PW, Dehaas PE, Soll DR, van Embden JD. Occurrence and stability of insertion sequences in Mycobacterium tuberculosis complex strains: evaluation of an insertion sequence-dependent DNA polymorphism as a tool in the epidemiology of tuberculosis. J Clin Microbiol 1991;29:2578-86.
18. Kremer K, van Zetten M, van Embden JD, Schouls L, van Soolingen D. 'RIFO ASSAY' a PCR + reverse line blot hybridization method to detect rifampicin resistance. Bilthoven, The Netherlands: National Institute of Public Health and Environment; 2003
19. Kamerbeek J, Schouls L, Kolk A, van Agterveld M, van Soolingen D, Kuijper S, et al. Simultaneous detection and strain differ-
entiation of Mycobacterium tuberculosis for diagnosis and epidemiology. J Clin Microbiol 1997;35:907-14.
20. Mokrousov I, Bhanu VN, Suffy PN, Kadival GV, Yap SF, Cho SN, et al. Multicentric evaluation of reverse line blot assay for detection of drug resistance in Mycobacterium tuberculosis clinical isolates. J Microbiol Methods 2004;57:323-35.
21. Rodrigues C, Shenai S, Sadani M, Sukhadia N, Mehta A. Comparison of phenotypic and genotypic methods for pyrazinamide susceptibility testing. Abstract for oral presentation. Abstracts of the 1st Asia Pacific Region Conference of the IUATLD, Malaysia, August 2007.
22. Wallace RJ, Glassroth J, Griffith DE, Olivier KN, Cook JL, Gordin F. Diagnosis and treatment of diseases caused by nontuberculous mycobacteria. Am J Respir Crit Care Med 1997;156:S1-25.
23. Griffith DE, Aksamit T, Brown-Elliott BA. An official ATS/IDSA statement: Diagnosis, treatment, and prevention of nontuber-culous mycobacterial diseases. Am J Respir Crit Care Med 2007;175:367-416.
24. Pozzi G, Melone M, Iona E, Orru G, Thoresen OF, Ricci ML, et al. rpoß mutations in multidrug-resistant strains of Mycobacterium tuberculosis isolated in Italy. J Clin Microbiol 1999;37:1197-9.
25. Zhang SL, Shen JG, Xu PH, Li DX, Sun ZQ, Li L, et al. A novel genotypic test for rapid detection of multi-drug resistant Mycobacterium tuberculosis isolates by a multiplex probe array. J Appl Microbiol 2007;103:1262-71.
26. Mani C, Selcakumar N, Narayanan S, Narayanan PR. Mutations in the rpoB gene of multidrug-resistant Mycobacterium tuberculosis clinical isolates from India. J Clin Microbiol 2001;39:2987-90.
27. Siddiqi N, Shamim M, Hussain S, Choudhary RK, Ahmed N, Prachee S, et al. Molecular characterization of multidrug-resis-tant isolates of Mycobacterium tuberculosis from patients in North India. Antimicrob Agents Chemother 2002;46:443-50.
28. Mani C, Selvakumar N, Narayanan S, Narayanan PR. Comparison of DNA sequencing. PCR-SSCP and phaB assays with indirect sensitivity testing for detection of rifampicin resistance in Mycobacterium tuberculosis. Int J Tuberc Lung Dis 2003;7:652-9.
29. Hirano K, Abe C, Takashi M. Mutations in the rpoB gene of rifampin-resistant Mycobacterium tuberculosis strains isolated mostly in Asian countries and their rapid detection by line probe assay. J Clin Microbiol 1999;37:2663-6.
30. Heifets LB, Cangelosi GA. Drug susceptibility testing of Myco-bacterium tuberculosis: a neglected problem at the turn of the century. Int J Tuberc Lung Dis 1999;3:564-81.
31. National Committee for Clinical Laboratory Standards. Antimy-cobacterial susceptibility testing for M. tuberculosis. Tentative standard M2-T. Villanova, PA: NCCLS; 1995.
32. Kiepiela P, Bishop KS, Smith AN, Roux I, York DF. Genomic mutations in the katG, inhA, and ahpC genes are useful for the prediction of isoniazid resistance in Mycobacterium tuberculosis isolates from Kwazulu Natal, South Africa. Tubercle Lung Dis 2000;80:47-56.
33. Wu XQ, Lu Y, Zhang JX, Liang JQ, Li HM, Zhang GY, et al. Detection of the mutations in katG 315 and inhA-15 of Myco-bacterium tuberculosis strains isolated from Chinese patients. Chin Med J (Engl) 2006;119:230-3.
34. TracevskaT, Jansone I, BorkaL, MargaO, Baumanis V. Mutation in rpoß and katG gene leading to drug resistance in Mycobacterium tuberculosis in Latvia. J Clin Microbiol 2002;40:3789-92.
35. Mokrousov I, Navskaya O, Otten T, Limeschenko E, Steklova L, Vyshnevskiy B. High prevalence of katG Ser315Thr substitution among isoniazid-resistant Mycobacterium tuberculosis clinical isolates from Northwestern Russia, 1996 to 2001. Antimicrob Agents Chemother 2002;46:1417-24.
36. Shemyakin IG, Stepanshina VN, Ivanov IY, Lipin MY, Anisimova VA, OnasenkoAG, etal. Characterization of drug-resistant isolates of Mycobacterium tuberculosis derived from Russian inmates. Int J Tuberc Lung Dis 2004;8:1194-203.
37. Sreevatsan S, Pan X, Stockbauer K, Connell N, Krieiswirth B, Whittam T, et al. Restricted structural gene polymorphism in the Mycobacterium tuberculosis complex indicates evolutionary recent global dissemination. Proc Natl Acad Sci USA 1997;97:9869-74.
38. Meier A, Sander P, SchaperKJ, Scholz M, BottgerEC. Correlation of molecular resistance mechanism and phenotypic resistance levels in streptomycin-resistant Mycobacterium tuberculosis. Antimicrob Agents Chemother 1996;40:2452-4.
39. Finken M, Kirschner P, Meier A, Wrede A, Bottger C. Molecular basis of streptomycin resistance in Mycobacterium tuberculosis: alterations of the ribosomal protein S12 gene and point mutations within a functional 16S ribosomal RNA pseudoknot. Mol Microbiol 1993;9:1239-46.
40. Morgan M, Kakantri S, Flores L, Pai M. A commercial line probe assay for the rapid detection of rifampicin resistance in Mycobacterium tuberculosis: a systematic review and meta-analysis. BMC Infect Dis 2005;5:62-70.
41. Almeida D, Rodrigues C, Ashavaid TF, Lalvani A, Udwadia Z, Mehta A. High incidence of the Beijing genotype among multidrug resistant isolates of Mycobacterium tuberculosis in a tertiary care centre in Mumbai, India. Clin Infect Dis 2005;40:881-6.
42. Espinal MA. The global situation of MDR-TB. Tuberculosis 2003;83:44-51.
43. Glynn JR, Whiteley J, Bifani PJ, Kremer K, van Soolingen D. Worldwide occurrence of Beijing/W strains of Mycobacterium tuberculosis. A systematic review. Emerg Infect Dis 2002;8: 843-9.
