rapid diagnosis of extensively drug resistant tuberculosis using a
TRANSCRIPT
Rapid diagnosis of Extensively Drug Resistant tuberculosis using a Reverse Line 1
Blot Hybridization assay. 2
Kanchan Ajbani, 3
PhD student, Dept of Research, P.D. Hinduja National Hospital & Medical Research 4
Centre, Mumbai, India 5
Anjali Shetty, 6
Consultant Microbiologist, Dept of Microbiology, P.D. Hinduja National Hospital & 7
Medical Research Centre, Mumbai, India 8
Ajita Mehta Late*, (Passed away on 6th Feb 2010) 9
Consultant Microbiologist, Dept of Microbiology, P.D. Hinduja National Hospital & 10
Medical Research Centre, Mumbai, India 11
Camilla Rodrigues 12
Consultant Microbiologist, Dept of Microbiology, P.D. Hinduja National Hospital & 13
Medical Research Centre, Mumbai, India 14
Conflict of interest: There are no conflicts of interest for this manuscript. 15
Source of financial support: The study was funded by National 20 Health and 16
Education Society, P. D. Hinduja National hospital and Medical Research Centre, 17
Mumbai, India. 18
Address for Correspondence: 19
Dr. Camilla Rodrigues 20
Consultant Microbiologist, 21
Dept of Microbiology, Research Laboratories, P. D. Hinduja National Hospital and 22
Medical Research Centre, Veer Savarkar Marg, Mahim, Mumbai – 400016, India. 23
Copyright © 2011, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.J. Clin. Microbiol. doi:10.1128/JCM.02511-10 JCM Accepts, published online ahead of print on 25 May 2011
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
Tel: 0091 – 22 – 24447795 24
Fax No.; 099 – 22 – 24442318 25
E-mail address: [email protected] 26
Short title: Rapid diagnosis of XDR-TB 27
Article type: Major article 28
Key words: XDR-TB, fluoroquinolones, aminoglycosides, RLBH 29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
ABSTRACT 47
BACKGROUND 48
Drug resistance in Tuberculosis (TB) is a matter of grave concern for TB control 49
programs as there is currently no cure for some extensively drug resistant strains. There is 50
concern that this resistance could transmit, stressing the need for additional control 51
measures, as rapid diagnostic method, and newer drugs for treatment. We developed an 52
in house assay that can rapidly detect resistance to drugs involved in the definition of 53
extensively drug resistant tuberculosis (XDR-TB) directly from smear positive 54
specimens. 55
Methods 56
215 phenotypically XDR-TB and 50 pan susceptible isolates were analyzed using 57
the Reverse Line Blot Hybridization (RLBH) assay. The assay was also successfully 58
applied to 73 smear positive clinical specimens. 59
Results and discussion 60
The RLBH assay exhibited good sensitivity for the detection of resistance to 61
isoniazid (99%), rifampicin (99%), fluoroquinolones (95.3%) and second line 62
aminoglycosides (94.8%). The application of this assay on direct smear positive clinical 63
specimen revealed 93% concordance with the phenotypic drug susceptibility test (DST) 64
for the above mentioned drugs. Time to accurate DST results was significantly reduced to 65
3 days from weeks. 66
Conclusion: 67
This molecular assay is a highly accurate screening tool for XDR-TB, which achieves a 68
substantial reduction in diagnostic delays. 69
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
INTRODUCTION 70
The estimate of tuberculosis (TB) infections continues to increase globally and in 71
1993 the WHO declared TB a global public health emergency, being the only disease so 72
far to warrant that designation. In many countries where the incidence of TB is high, the 73
major driving force of the epidemic is transmission. The foremost reason for this is that 74
there are prolonged delays from the onset of the TB disease in individuals to the time of 75
actual diagnosis. During this period an active TB case may infect numerous other people 76
thereby contributing to the increase of the epidemic [20]. The emergence of multidrug 77
resistant tuberculosis (MDR-TB), defined as resistance to both rifampicin (RIF) and 78
isoniazid (INH) which is a compounding factor for the control of the disease. Patients 79
harboring MDR-TB need alternative treatment regimens involving second-line drugs that 80
are more expensive, toxic, and less effective. Moreover, extensively Drug Resistant 81
tuberculosis (XDR-TB) (resistance to INH, RIF, fluoroquinolones (FQ) and any one 82
second line injectable i.e. kanamycin (KAN), amikacin (AM), capreomycin (CM)) 83
constitutes an emerging threat for the control of the disease over the period [4]. Accurate 84
diagnosis and initiation of early treatment are essential to reduce transmission and 85
improve the outcome for patients with tuberculosis. Timely identification of drug 86
resistance is especially important to avoid subjecting patients with resistance to 87
inadequate first line treatment for months while awaiting drug susceptibility testing 88
(DST) results. This exposure to inadequate therapy results in amplified drug resistance 89
[21]. 90
Drug resistance in M. tuberculosis is attributed to random mutations in the 91
mycobacterium genome. Mutations in the specific codons can therefore be exploited to 92
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
rapidly detect drug resistance [8, 9, 12, 19, 22]. Resistance to first line drugs (INH, RIF) 93
is attributed to mutations in inhA and katG for INH and rpoB for RIF [18, 19, 22, 23]. 94
Resistance to fluoroquinolones and second line aminoglycosides is most frequently 95
associated with mutations in gyrA, gyrB and rrs respectively. Studies have demonstrated 96
that by targeting mutations in codon 90, 91 and 94 in gyrA approximately 70- 90% of all 97
fluoroquinolone resistance can be accurately predicted [2, 5, 17]. Similarly, mutations in 98
the rrs gene particularly the hot spot region from 1401- 1484 can accurately predict 99
resistance to second line aminoglycosides [10, 16]. The reverse line blot hybridization 100
(RLBH) [11, 24] assay is a probe based method where multiple oligonucleotide probes 101
carrying the mutant sequence and wild type sequence are immobilized on nitrocellulose 102
strips and hybridized with biotin labeled PCR products. This technique had been 103
successfully applied for the identification of mutations related to the resistance of 104
isoniazid, rifampicin and streptomycin at our centre. The assay was optimized to 105
additionally detect mutations conferring FQ and second line aminoglycosides resistance 106
(KAN, AM, CM), with the same hybridization conditions such that a single blot could be 107
used to detect resistance to the anti tubercular drugs included in the XDR-TB definition. 108
This enabled us to reduce the time to accurate DST. 109
110
111
112
113
114
115
116
117
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
MATERIALS AND METHODS 118
Setting :- 119
The study was carried at a mycobacteriology laboratory at P. D. Hinduja National 120
Hospital & MRC (PDHNH) a tertiary care centre in Central Mumbai with a referral bias 121
towards complicated and non- responding cases. The study was approved by the internal 122
institutional review board (IRB). The laboratory receives about 10,000 cultures annually 123
with a culture positivity rate of 40%. Being private tertiary laboratory susceptibility is 124
only done on request wherein of the 3609 requests for DST from Jan 08-Dec 09, 65% of 125
isolates are MDR-TB and of these 9.1% are XDR-TB. 126
A] Bacterial isolates 127
Two hundred and fifteen consecutive XDR-TB isolates tested by Mycobacterial Growth 128
Indicator Tube (MGIT 960) TB system were collected over a period of 24 months (Jan 08 129
- Dec 09). 50 consecutive pan susceptible isolates tested by MGIT 960 were also tested 130
using RLBH assay. All the collected strains were stored at -700C. DNA extraction from 131
mycobacterial culture isolates was carried out by the cetyl trimethyl ammonium bromide- 132
sodium chloride (CTAB-NaCl) method [15]. 133
B] Clinical specimens 134
73 consecutive acid fast bacilli (AFB) smear positive clinical specimens (1+ and above) 135
collected over a period of 5 weeks were also tested by RLBH method. Of the 73 136
specimens, 31 specimens were 1+ on acid fast bacilli smear, 15 were 2+ and 27 were 3+ 137
AFB smear. Smear positive specimens were subjected to decontamination by NaLC- 138
NaOH [6] method and the sediment was used for DNA extraction that was carried out 139
using Qiagen DNA extraction kit. 140
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
Amplification of all targets in multiplex PCR 141
Enlisted in table 1 are the PCR primer sequences wherein the primers for inhA, 142
katG and rpoB are the same as previous [14]. Novel primers were designed for 143
amplifying gyrA, gyrB and rrs that are also listed in table1. The cycling condition for the 144
touch down multiplex for the isolates and direct clinical specimen are provided as 145
supplemental material. 146
Designing the specific sequence probes for the fluoroquinolones and second line 147
aminoglycosides 148
The initial 150 XDR-TB isolates were amplified as single targets and sequenced 149
[1]. Based on the mutations specific amino labeled oligonucleotide probes for 150
fluoroquinolones and second line aminoglycosides were designed so that they can be 151
processed under same hybridization and washing conditions as the MDR blot and 152
covalently linked to a nylon membrane in parallel lines. 6 mutant probes and 2 wild type 153
probes were designed for gyrA and 1 wild and 1 mutant probe for gyrB. For 154
aminoglycosides the rrs region was targeted with 2 wild type probes and 3 mutant probes 155
correlating with high level resistance to KAN, AM and CM (Table 2). The uniqueness 156
and cross reactivity of the all newly designed and modified probes was analyzed with the 157
BLAST search (http://www.ncbi.nlm.nih.gov). 158
Hybridization with amplified products 159
Heat denatured PCR products were applied on the membrane in the miniblotter and 160
hybridized at 53 0C for 60 min. The membrane was washed twice with gentle shaking in 161
250 ml 2X SSPE/0.5% SDS for 10 min at 600C. The membrane was subsequently 162
incubated at 420C with 1:4000 diluted streptavidin-peroxidase conjugate in 2X 163
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
SSPE/0.5% SDS for 60 min, washed twice with 250 ml 2X SSPE/0.5% SDS at 420C for 164
10 min, rinsed twice with 2X SSPE at room temperature (RT) for 5 min, and subjected to 165
luminescent detection of hybrids with enhanced chemiluminescence (ECL) detection 166
system followed by exposure to the ECL Hyperfilm (Amersham Biosciences). The 167
presence of a clearly visible black signal was considered as a positive hybridization 168
reaction. The results were easy to interpret (Figure 1). For reuse, the membranes were 169
stripped in 1% SDS solution at 800C (twice for 40 min) and rinsed in 20 mM EDTA, pH 170
8.0 at RT. The membrane can be stripped and re-used up to 7 times without 171
compromising the results. 172
Reproducibility of the RLBH assay was validated by testing 215 clinical XDR-TB 173
isolates (of these 150 were also sequenced and results were compared with sequencing 174
[1] as well as RLBH), 50 susceptible isolates and 73 smear positive clinical specimens 175
(+1 and above). The results of the clinical specimen were compared with phenotypic 176
susceptibility performed on MGIT. 177
Statistical analysis: 178
A 2 × 2 table was used to calculate Sensitivity, specificity of the RLBH assay in 179
comparison with the phenotypic DST i.e. MGIT. A p value (significance) of < 0.05 is 180
deemed statistically significant. To compare two tests, e.g. RLBH & MGIT, kappa was 181
used as a measure of agreement with 95% confidence interval (CI). 182
183
184
185
186
187
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
RESULTS 188
Probe Designing: 189
Initial 150 XDR- TB isolates and 50 pan susceptible isolates were sent for sequencing for 190
screening mutations in gyrA, gyrB and rrs and based on the mutations seen; the 191
respective wild and mutant probes were designed for the fluoroquinolones and second 192
line aminoglycosides that were used for standardizing the RLBH assay conditions. 193
Evaluation of the RLBH assay: 194
The 150 sequenced isolates were then used as known controls for the respective 195
mutations. The latter 65 XDR- TB isolates were only screened by RLBH assay. 196
Performance of the assay: 197
A] Rifampicin resistance: 198
Of the total 215 XDR-TB and 50 pan susceptible isolates screened by RLBH assay, one 199
isolate showed RIF resistance by MGIT (phenotypic assay) but did not show any 200
mutation with the RLBH assay. The diagnostic sensitivity of the assay to detect RIF 201
resistance was calculated to be 99% with the specificity being 100% as none of the 50 202
pan susceptible isolates showed any mutation with RLBH. The kappa value was 0.99 203
(95% CI) indicating good agreement between RLBH and phenotypic susceptibility. 204
B] Isoniazid resistance: 205
Of the total 215 XDR-TB and 50 pan susceptible isolates, one isolate showed INH 206
resistance by MGIT but did not bind with any mutant probes on the RLBH membrane. 207
The diagnostic sensitivity of the assay to detect INH resistance was calculated to be 99% 208
with the specificity being 100% as none of the 50 pan susceptible isolates showed any 209
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
mutation with RLBH. The kappa value was 0.99 (95% CI) indicating good agreement 210
between RLBH and phenotypic susceptibility. 211
C] Fluoroquinolone resistance: 212
Of the total 215 XDR-TB and 50 pan susceptible isolates, 10 isolates phenotypically 213
resistant to fluoroquinolones did not reveal any mutation by the RLBH assay. These 214
samples were sent for sequencing to check for the mutations in gyrA and gyrB. 215
Sequencing did not show any mutations, indicating mutations outside the targeted region 216
or other mechanisms i.e. efflux mechanism. The diagnostic sensitivity of the RLBH assay 217
to detect fluoroquinolone resistance was calculated to be 95.34% with a specificity of 218
100% as none of the 50 pan susceptible isolates showed any mutation with RLBH. The 219
kappa value was 0.89 (95% CI) indicating good agreement between RLBH and 220
phenotypic susceptibility. 221
D] Second line aminoglycosides (kanamycin & amikacin): 222
Of the 215 XDR-TB and 50 pan susceptible isolates 11 isolates were phenotypically 223
resistant to KAN and AM but did not reveal any mutation by the RLBH assay. These 224
samples were sent for sequencing for the rrs region to check for the mutations. 225
Sequencing did not show any mutations, indicating mutations outside the targeted region 226
or other mechanisms of resistance. The diagnostic sensitivity for the detection of 227
resistance to second line aminoglycosides by RLBH assay was 94.8% with specificity of 228
100%. The kappa value was 0.87 (95% CI) indicating good agreement between RLBH 229
and phenotypic susceptibility. 230
The various mutations seen in the 215 XDR-TB isolates are listed in table 3. Of all the 231
XDR-TB isolates 213/215 (99%) showed katG S315T substitution correlating with 232
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
isoniazid resistance. 199/215 (92%) showed S531L substitution in the rpoB gene. The 233
mutations in the gyrA were seen on the 90th
, 91st and 94
th codon. The frequency of the 234
D94G substitution was seen in 101/215 (46%) of XDR-TB isolates. The other 235
substitutions seen on the 94th
codon were D94A and D94N seen in 31/215 (14%) and 236
27/215(12.5%) respectively. 106/215 XDR-TB isolates showed A1401G mutation in rrs 237
region and were phenotypically resistant to KAN and AMK but susceptible to CM. A 238
further 56/215 XDR-TB isolates with the same mutation showed phenotypic resistance to 239
KAN, AMK and CM. The G1484T mutation in rrs resulted in resistance to all three 240
second line injectables by phenotypically method. 241
Applicability on Direct smear positive clinical specimen: 242
The results obtained on isolates were promising but they were time consuming, thus the 243
assay was tested on direct clinical specimens. 73 specimens were tested, of these 8 244
culture proven to be mycobacteria other than tuberculosis (MOTT) showed only binding 245
with Mycobacterial complex and no binding with Mycobacterium tuberculosis probe on 246
the membrane indicating atypical infection which was in concordance with the 247
phenotypic MGIT method. There were 7 samples that showed inhibition with RLBH. The 248
percent of agreement for susceptibility testing between the phenotypic test and RLBH 249
assay was calculated to be 93% for all the targeted drugs. With the application of the 250
assay on direct clinical specimens the turn around time to results was reduced to 3 days 251
(assay takes about 4-5 hours but DNA extraction and PCR amplification totals the time to 252
results as 3 days) as compared to 2- 4 weeks as in case of the phenotypic liquid culture 253
system. 254
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
The RLBH membrane was used 7 times without compromising the signal strength but the 255
strength of the signal was reduced from 8th
use thereby causing difficulty to accurately 256
read the results. 257
Time to DST 258
On an average the time to culture positivity for a smear 3+ positive is within 7 days [13] 259
and thereafter another 10-14 days for DST results. The total time to DST for a smear 3+ 260
positive specimen would therefore be 17- 21 days. This is increased with smear 2+ 261
positive specimen wherein time to identification is 8-10 days and thereafter another 10-14 262
days for DST total time being 18- 24 days. Similarly for smear 1+ positive specimen the 263
time to identification is 11 days and thereafter 10-14 days for DST the total time being 264
21- 25 days. In comparison the RLBH assay takes only 3 days from direct 1+ smear 265
positive specimen (including the DNA extraction, PCR amplification and assay 266
procedure) as compared to phenotypic method. 267
268
269
270
271
272
273
274
275
276
277
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
DISCUSSION 278
The present study is a retrospective study that shows the rapidity of the RLBH assay 279
using a probe based technique over the phenotypic methods with a sensitivity for 280
detection of INH, RIF, FQ and second line aminoglycosides resistance as 99%, 99%, 281
95.34% and 94.8% respectively. These performance characteristics are indicative of 282
equivalence with conventional DST. In most developing countries, DST is usually 283
performed on solid media, such as Lowenstein Jensen wherein results are available after 284
8-18 weeks during which time many patients with MDR-TB or XDR-TB may have died, 285
transmitted their disease or both. Conventional DST for second-line drug testing is far 286
less standardized than first-line DST. The critical concentrations are not as clear cut, 287
often as that the minimum inhibitory concentration (MIC) values are close to the assumed 288
critical concentrations. The RLBH assay detects the presence of MTB with resistant 289
pattern to INH, RIF, FQ, KAN, AM and CM within a short span of 3 days directly from a 290
smear positive clinical specimen by detect point mutations present in rpoB, katG, inhA, 291
rrs, gyrA and gyrB. This could help clinicians in initiating appropriate therapy and 292
implement infection control measures to curb the spread of resistance. Compared with 293
PCR based sequencing and DNA microarray techniques, RLBH can be performed 294
without expensive devices. Though mutations to the first line drugs are well studied and 295
among the Indian XDR-TB isolates in a study conducted at Delhi [7] the key mutations 296
observed in the rpoB were S531L and S315T in katG which is similar to our findings. In 297
a previous study from this centre [14] done on MDR isolates revealed an array of point 298
mutations from 511- 531 in rpoB, which seems to have been selected out in case of 299
XDR-TB isolates to be Ser531Leu. Among rifampicin resistance isolates our findings are 300
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
in concordance with other studies [1] that are dominated by Ser531Leu mutation among 301
the XDR-TB. There are limited mutation studies for the second line drugs, the FQ 302
resistance is correlated with mutations in the Quinolone Resistance Determining Region 303
(QRDR) in gyrA majorly and gyrB and 90% prediction of the FQ resistance was achieved 304
with mutations at 90th
and 94th
codon which is similar to the findings at Delhi [7] 305
XDR-TB isolates. Studies from out of India also show mutation in the QRDR region of 306
gyrA similar to the mutations observed among our XDR-TB isolates [1]. The resistance to 307
second line injectables can be predicted by the nucleotide changes in the 3’ part of the 308
16S rRNA (rrs) gene. With this assay 89% of the second line aminoglycoside resistance 309
could be detected. The key mutations were seen at A1401G and G1484T similar to 310
characterization of XDR-TB isolates from Delhi [7]. This study was dominated by the 311
A1401G and G1484T mutations in the rrs whereas other studies [1] have reported several 312
other point mutations in the rrs gene. 313
Commercial assays MTBDRplus and MTBDRsl (Hain Lifesciences) [8] is strip based 314
and uses two strips per patient that is priced approximately at USD 49 (separate strip for 315
MDR and FQ plus injectible) excluding the equipment cost. In contrast RLBH uses a 316
single membrane to identify mutations for XDR-TB and the membrane once prepared can 317
be stripped and reused for at least 7 times without affecting the signal strength. 40 318
samples along with positive control and negative control can be detected simultaneously 319
at one time and this reduces the cost of the assay making it suitable for application in 320
routine clinical laboratory in developing countries. Though the GeneXpert [3] also 321
evaluated at our centre is the new state of the art technology, it can only detect resistance 322
to rifampicin and the cost of the test at reduced price for India is USD17 excluding the 323
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
cost of the instrument whereas our assay targets all the drugs involved in XDR-TB 324
definition. Thus, the present assay is an adequate tool to rapidly detect these point 325
mutations that confer resistance to INH, RIF, FQ, KAN, AM, and CM. Besides being 326
cost effective, this assay enables initiation of early treatment which could curb the spread 327
of resistant strains thus helping in control of tuberculosis. 328
The assay can detect only those mutations that are covered by the wild type or 329
mutant probe. Novel mutations with locations other than the probe region or outside the 330
amplified region would be a shortcoming. But the system is an open system wherein 331
newer probes can be added. Though molecular assays have a short turn around time and are 332
the state of the art they cannot completely replace the conventional methods as there are 333
multiple genes involved in resistance and the molecular assays target only known mutations. 334
335
336
337
338
339
340
341
342
343
344
345
346
347
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
CONCLUSION 348
In this study we have attempted to characterize the genetic mechanism of the XDR 349
isolates wherein we saw good concordance with the phenotypic resistance in case of 350
isoniazid, rifampicin and fluoroquinolones and second line aminoglycosides. As XDR-TB is 351
emerging as a highly threatening pathogen in a pervasive association with the acquired 352
immunodeficiency syndrome epidemic, rapid tests to promptly identify resistance to first- 353
and second-line antitubercular agents are urgently required. It has been observed that point 354
mutations in rpoB, inhA, katG, gyrA, gyrB and rrs can accurately predict resistance to these 355
drugs. RLBH assay proposes to be an adequate tool for direct analysis of XDR-TB from 356
clinical specimens thus fulfilling many attributes of an ideal TB diagnostic test as proposed 357
by the WHO for developing countries. 358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
Table 1: Primer sequences for inhA, katG, rpoB, gyrA, gyrB and rrs and cycling 374
condition. 375
Target region Primer sequences
inhA TB 92M- 5’ CCT CGC TGC CCA GAA AGG GA TCC -3’
TB 93M- 5’ biotin-CCG GGT TTC CTC CGG T-3’
katG KGF-biotin-5 AGA GCT CGT ATG CCG GA-3’
KGR-5’ GCG AAT GAC CTT GCG CAG ATC-3’
rpoB Rpo F - 5’-TCA AGG AGA AGC GCT ACG ACC TGG-3
Rpo R- 5’ biotin- GGG TCT CGA TCG GGC ACA T-3’
gyrA gyAF 5’CGA ACC GGT TGA CAT CGA GCA GGA G 3’
gyA R 5’ biotin CAG CAT CTC CAT CGC CAA CG 3’
gyrB gyrB F 5’ GAA GCC AAC CCC ACC GAC GCG A 3’
gyrB R (5’ biotin CTG CGC TGC CAC TTG AGT TTG 3’)
rrs rrs F 5’CGT TCC CTT GTG GCC TGT GTG CAG 3’
rrs R 5’ biotin GTT GGG GCG TTT TCG TGG TGC 3’
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
Table 2: Probe sequences for fluoroquinolone and aminoglycosides 394
No Probe Description Sequences
gyrA gene
1 90th codon Wild type
5'- ACC ACC CGC ACG GCG ACG CGT C-3'
2 90th codon Mutant type
5'- ACCACC CGC ACG GCG ACG TGT C-3'
3 91st codon Mutant type
5'-ACC CGC ACG GCG ACG CGC C-3'
4 94th codon Wild type
5'-CG ATC TAC GAC AGC CTG GTG-3'
5 94th codon Mutant type I
5'-CG ATC TAC GCC AGC CTG GTG-3'
6 94th codon Mutant type II
5'-CG ATC TAC GGC AGC CTG GTG-3'
7 94th codon Mutant type III
5’- CGA TCT ACT ACA CCC TGG TG
8 94th codon Mutant type IV
5’ CGA TCT ACA ACA CCC TGG TG- 3’
9 95th codon Polymorphism
5'-CGA TCT ACG ACA CCC TGG TG-3'
gyrB gene
10 510 Wild type
5’ GTG CTA AAG AAC ACC GAA G 3’
11 510 Mutant type
5’ CGA CCG GGT GCT AAA GAG CA 3’
rrs gene
12 1401 Wild type
5'-ACC GCC CGT CAC GTC ATG AA-3'
13 1401 Mutant type
5'-ACC GCC CGT CGC GTC ATG AA-3'
14 1402 Mutant type
5’ ACC GCC CGT CAA GTC ATG AA 3
15 1484 Wild type
5'-CGG CGA TTG GGA CGA AGT C-3'
16 1484 Mutant type
5'-CGG CGA TTG GGA CTA AGT C-3'
395
396
397
398
399
400
401
402
403
404
405
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
Table 3: Mutations seen in XDR patients 406
Drug Gene Mutation seen
Amino
acid
change
No. of
isolates
Isoniazid katG G315C S315T 213
Isoniazid inhA+katG C-15T 44
Rifampicin rpoB C531T S531L 199
Rifampicin rpoB A526G H526Y 7
Rifampicin rpoB G516T +
A526G
D516 T 2
Rifampicin rpoB A516T +
A526G
D516E 2
Rifampicin rpoB A516T D516E 2
Rifampicin rpoB G516T D516T 1
Fluoroquinolone i.e. ofloxacin gyrA C90T+ C95G
C90T
A90V 23
21
Fluoroquinolones ofloxacin &
moxifloxacin
gyrA T91C S91P 2
Fluoroquinolone ofloxacin &
moxifloxacin
gyrA A94G+ C95G
A94G
D94G
83
18
Fluoroquinolones ofloxacin &
moxifloxacin
gyrA A94C + C95G
A94C
D94A
19
12
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
Fluoroquinolone ofloxacin &
moxifloxacin
gyrA G94A + C95G
G94A
D94N
19
8
Aminoglycoside i.e.
kanamycin, amikacin
rrs
A1401G
106
Aminoglycoside i.e.
kanamycin, amikacin,
capreomycin
rrs
A1401G
56
Aminoglycoside i.e.
kanamycin, amikacin,
capreomycin
rrs G1484T 42
407
408
409
410
411
412
413
414
415
416
417
418
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
Figure 1: RLBH performed for rpoB, katG, inhA, gyrA, gyrB and rrs. 419
420
421
422 423
All the different probes are blotted vertically and patient’s sample (PCR product) are applied horizontally. 424
Probe in lane no 1 is common probe for all mycobacteria then there is probe for MTB complex. Then 5 425
wild type probes for rpoB gene to check rifampicin susceptible cases and 16 mutant probes for detecting 426
rifampicin resistance. Similarly we targeted inhA wild type and mutant probe for promoter region followed 427
by katG gene wild type and mutant probes for codons 315. This is further followed by the wild and mutant 428
probes for the gyrA mutations in codon 90, 91, 94 and 95. Similarly we also targeted the gyrB 510 region 429
with wild type and mutant probe. For the second line aminoglycoside rrs region 1401, 1402, 1484 were 430
used. To illustrate an example, patient sample marked by the arrow has mycobacterial infection due to M. 431
1 2 3 4 5 6 7 8 9 10 1112 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 363738 39 40 41 4243
1 - MYC (Common for
Mycobacteria)
2 – MTB (M.
tuberculosis) complex
3 - RIF WT1 (codons
509 -514)
4 - RIF WT2 (codons
514 – 520)
5 - RIF WT3 (codons
521 -525)
6 - RIF WT4 (codons
524-529)
7 - RIF WT5 (codons
530-534)
8 - 533 CCG
9 - 531 TTG
10- 531 TGG
11 - 526 GAC
12 - 526 TAC
13 - 526 CGC
14- 526 TGC
15 - 526 AAC
16- 522 TTG
17 - 522 TGG
18 - Del 516
19 - 516 TAC
20 - 516 GAC
21 - 516 GTC
22 - 516 GGC
23- 513 CCA
24 – inh A wt
25- inhA mt
26-katG 315wt
27- katG 315 mt
28- gyA90 wt
29- gyA 90mt
30- gyA 91 mt
31- gyA94 wt
32- gyA 94mt AAC
33-gyA94mt GCC
34- gyA 94 mt GGC
35- gyA94 mt TAC
36- gyA 95 mt
37- gyB wt
38- gyBmt
39-rrs 1401 wt
40-rrs 1401 mt
41-rrs 1402 mt
42-rrs 1484 wt
43-rrs 1484 mt
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
tuberculosis complex and shows binding with probes 3-6 which are the rpoB wild type codon 509- 529. It 432
does not show binding with probe 7 but shows binding with probe 9 indicating RIF resistance due to 531 433
TTG mutation. Further on the same sample shows binding with probe 24 which is inhA promoter wild type 434
and shows binding with probe 27 which is katG 315 mutant probe, thereby indicating INH resistance due to 435
katG mutation. Same sample shows binding with 28th
, 31st, 36th
, 37th
, 39th
and 42nd
probe which correspond 436
to gyrA 90 wild type, gyrA 94 wild type, gyrA 95 polymorphism, gyrB wild type, rrs 1401 wild type and rrs 437
1484 wild type respectively, thus giving the result for that particular sample to be resistant to INH, RIF and 438
susceptible to FQ and KAN, AM, CM. 439
440
Acknowledgement: 441
We sincerely thank the National Health Education Society, P D Hinduja National hospital 442
and Medical Research Centre for funding this study and also the entire staff of the 443
Mycobacteriology Department at P D Hinduja National hospital and Medical Research 444
Centre. I would also like to thank Dr. Mankeshwar for his esteemed guidance in the 445
statistical analysis. 446
447
448
449
450
451
452
453
454
455
456
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
References: 457
1. Ajbani K, Rodrigues C, Shenai S, Mehta A. Mutation detection and accurate diagnosis 458
of Extensively Drug Resistant Tuberculosis: Report from a tertiary care centre in India. 459
Journal of Clinical Microbiology. 2011; 49(4): 1588- 90. 460
2. Alangaden G, Manavathu E, Vakulenko S, Zvonok N, Lerner S. 461
Characterization of fluoroquinolone resistant mutant strains of Mycobacterium 462
tuberculosis selected in the laboratory and isolated from patients. Antimicrobial 463
agents and Chemotherapy. 1995; 39: 1700- 1703. 464
3. Boehme CC, Nabeta P, Hillemann D, Nicol MP, Shenai S,Krapp F et al. Rapid 465
molecular detection of tuberculosis and Rifampicin resistance. The New England Journal 466
of Medicine. 2010; 363 (11): 1005- 15. 467
4. Gandhi N, Nunn P, Dheda K, Schaaf H, et al. Multidrug resistant and extensively 468
drug resistant tuberculosis: a threat to global control of tuberculosis. Lancet. 2010; 469
375: 1830- 43. 470
5. Ginsburg. A, Grosset. J, Bishai. W. Fluoroquinolones, tuberculosis, and resistance. 471
Lancet Infect Dis, 2003; 3; 432 – 442. 472
6. Kent and Kubina GP. Public health mycobacteriology a guide for level III laboratory. 473
Atlanta, CA: U.S. Department of Health and Human Services, Public health services. 474
Center for disease control. 1985. P 64-8. 475
7. Khanna A, Raj V S, Tarai B, Sood R, et al. Emergence of Extensive Drug resistant 476
(XDR) Mycobacterium tuberculosis and molecular characterization of these clinical 477
isolates from Delhi region, India. Antimicrob. Agents Chemother. 2010; 54 (11): 4789- 478
4793. 479
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
8. Kiet V S, Lan NT M, An D D, Dung N et al. Evaluation of the MTB DRsl test for 480
detection of second line drug resistance in Mycobacterium tuberculosis. Journal of 481
Clinical Microbiology. 2010; 48: 2934- 39. 482
9. Martin- Casabona N, Mimo DX, Gonzalez T, Rossello J, Arcalis L. Rapid method 483
for testing susceptibility of Mycobacterium tuberculosis by using DNA probes. 484
Journal of Clinical Microbiology. 1997; 35: 2521- 2525. 485
10. Maus C, Plikaytis B, Shinnick T. Molecular analysis of cross resistance to Capreomycin, 486
Kanamycin, Amikacin and Viomycin in Mycobacterium tuberculosis. Antimicrobial 487
agents and Chemotherapy. 2005; 49: 319-397. 488
11. Mokrousov I, Bhanu VN, Suffys PN et al. Multicentre evaluation of reverse line 489
blot assay for detection of drug resistance in Mycobacterium tuberculosis clinical 490
isolates. Journal of Microbiological Methods. 2004; 57:323-335. 491
12. Musser JM. Antimycobacterial agent resistance in mycobacteria: molecular genetic 492
insights. Clin Microbiol Rev. 1995; 8:496-514. 493
13. Rodrigues C, Shenai S, Sadani M, Sukhadia N, et al. Evaluation of the BACTEC MGIT 494
960 system for the recovery and identification of Mycobacterium tuberculosis complex in 495
a high throughput tertiary care centre. IJMM. 2009; 27(3): 217- 21. 496
14. Shenai. S, Rodrigues. C, Mehta. A. Rapid speciation of 15 clinically relevant 497
mycobacteria with simultaneous detection of resistance to rifampin, isoniazid, and 498
streptomycin in Mycobacterium tuberculosis. International Journal of Infectious 499
Diseases, 2009; 13:46-58. 500
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
15. Somerville W, Thibert L, Schwartzman K, Behr M. Extraction of Mycobacterium 501
tuberculosis DNA: a question of containment. Journal of Clinical Microbiology. 502
2005; 43: 2996- 2997. 503
16. Suzuki et al. Detection of Kanamycin resistant Mycobacterium tuberculosis by 504
identifying mutations in the 16S rRNA gene. Journal of Clinical Microbiology. 1998; 505
36:1220- 1225. 506
17. Takiff H, Salazar L, Guerrero C, Philipp W, Huang W, et al. Cloning and 507
nucleotide Sequence of Mycobacterium tuberculosis gyrA and gyrB Genes and 508
detection of quinolone resistance mutations. Antimicrobial agents and 509
Chemotherapy. 1994; 38: 773- 780. 510
18. Telenti A, Honore N, Bernasconi C, March J, Ortega A, Heym B, Takiff HE and 511
Cole ST. Genotypic assessment of isoniazid and rifampicin resistance in 512
Mycobacterium tuberculosis: a blind study at reference laboratory level. Journal of 513
Clinical Microbiology. 1997b; 35:719-723. 514
19. Torres MJ, Criado A, Palomares JC and Aznar J. Use of real-time PCR and 515
fluorimetry for rapid detection of rifampin and isoniazid resistance-associated 516
mutations in Mycobacterium tuberculosis. Journal of Clinical Microbiology. 2000; 517
38:3194-3199. 518
20. Uys P, Warren R, Helden P. A threshold value for the time delay to TB diagnosis. 519
PLOS One. 2007; 2(8): 1- 5. 520
21. Yagui M, Perales M, Asencios L, Vergara L et al. Timely diagnosis of MDR- TB 521
under program conditions: is rapid drug susceptibility testing sufficient? Int J Tuberc 522
Lung Dis. 2006; 10(8): 838- 843. 523
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from
22. Yam WC, Tam DM, Leung CC, Tong HI, Chan KH, Leung ET, Wong KC, Yew 524
WW ,et al. Direct detection of rifampicin resistant Mycobacterium tuberculosis in 525
respiratory specimens by PCR-DNA sequencing. Journal of Clinical Microbiology. 526
2004; 42, 4438-4443. 527
23. Zhang M, Yue J, Yang YP, Zhang HM, Lei JQ, Jin RI, Zhang SL and Wang 528
HH. Detection of mutations associated with isoniazid resistance in Mycobacterium 529
tuberculosis isolates from China. Journal of Clinical Microbiology.2005; 43: 5477-530
5482. 531
24. Zhang Y, Coyne MY, Will SG, Levenson CH and Kawasaki ES. Single base 532
mutational analysis of cancer and genetic diseases using membrane bound modified 533
oligonucleotides. Nucleic Acids Res. 2000; 19:3929:3933. 534
535
536
537
538
539
540
541
542
543
544
545
546
on April 7, 2018 by guest
http://jcm.asm
.org/D
ownloaded from