Journal of Applied Microbiology ISSN 1364-5072
ORIGINAL ARTICLE
Multiplex PCR assay for simultaneous detection and
differentiation of Mycobacterium tuberculosis,
Mycobacterium avium complexes and other Mycobacterial
species directly from clinical specimens
K. Gopinath and S. Singh
Division of Clinical Microbiology, Department of Laboratory Medicine, All India Institute of Medical Sciences, Ansari Nagar, New Delhi, India
Keywords
HIV, Mycobacterium avium, mixed infection,
multiplex PCR, tuberculosis.
Correspondence
Sarman Singh, Division of Clinical
Microbiology, Department of Laboratory
Medicine, All India Institute of Medical
Sciences, Ansari Nagar, New Delhi 110 029,
India. E-mail: sarman_singh@yahoo.com
2008 ⁄ 1183: received 9 July 2008, revised 4
December 2008 and accepted 13 December
2008
doi:10.1111/j.1365-2672.2009.04218.x
Introduction
India has far more cases of tuberculosis than any other
country in the world and it accounts for one-fifth of the
global TB burden (WHO 2008). A total of 9Æ2 million
new cases and 1Æ7 million deaths from tuberculosis (TB)
occurred in 2006, of which 0Æ7 million had human
immunodeficiency virus (HIV) co-infection. There are
two main concerns about TB control in India and
ª 2009 The Authors
Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
Abstract
Aims: Polymerase chain reaction (PCR) is the most rapid and sensitive method
for diagnosing mycobacterial infections and identifying the aetiological Myco-
bacterial species in order to administer the appropriate therapy and for better
patient management.
Methods and Results: Two hundred and thirty-five samples from 145 clinically
suspected cases of tuberculosis were processed for the detection of Mycobacte-
rial infections by ZN (Ziehl Neelsen) smear examination, L-J & BACTECTM
MGIT-960 culture and multiplex PCR tests. The multiplex PCR comprised of
genus-specific primers targeting hsp65 gene, Mycobacterium tuberculosis com-
plex-specific primer targeting cfp10 (Rv3875, esxB) region and Mycobacterium
avium complex-specific primer pairs targeting 16S–23S Internal Transcribed
Spacer sequences. The multiplex PCR developed had an analytical sensitivity of
10 fg (3–4 cells) of mycobacterial DNA. The multiplex PCR test showed the
highest (77Æ24%) detection rate, while ZN smear examination had the lowest
(20%) detection rate, which was bettered by L-J culture (34Æ4%) and BAC-
TECTM MGIT-960 culture (50Æ34%) methods. The mean isolation time for M.
tuberculosis was 19Æ03 days in L-J culture and 8Æ7 days in BACTECTM MGIT-
960 culture. Using the multiplex PCR, we could establish M. tuberculosis + M.
avium co-infection in 1Æ3% HIV-negative and 2Æ9% HIV-positive patients. The
multiplex PCR was also highly useful in diagnosing mycobacteraemia in
38Æ09% HIV-positive and 15Æ38% HIV-negative cases.
Conclusions: The developed in-house multiplex PCR could identify and differ-
entiate the M. tuberculosis and M. avium complexes from other Mycobacterial
species directly from clinical specimens.
Significance and Impact of the Study: The triplex PCR developed by us could
be used to detect and differentiate M. tuberculosis, M. avium and other myco-
bacteria in a single reaction tube.
elsewhere: the rapid and sensitive diagnosis of the infec-
tion and identification of the pathogenic species. The
latter has gained significant importance after AIDS epi-
demic, where several nontubercular species of the myco-
bacterium cause human diseases. Because of these two
major concerns and limitations, progress in case detection
had slowed down globally in 2006 and began to stall in
China and India (WHO 2008). After AIDS epidemic,
nontubercular mycobacteria, especially Mycobacterium
425
Multiplex PCR assay for M. tuberculosis and M. avium complex
avium complex, have increasingly been reported in these
severely immuno-compromised patients (Benson et al.
2004; Steinbrook 2007). Not infrequently, such patients
will also have multiple infections with Mycobacterium
tuberculosis, M. avium and others. Several reports of iso-
lating M. avium concomitantly with M. tuberculosis as
mixed infection in HIV-positive patients have recently
been published from all parts of the world including the
Netherlands (Kox et al. 1995), Colombia (Murcia-
Aranguren et al. 2001), Thailand (Mahaisavariya et al.
2003), Brazil (Ruiz et al. 2001; Rosas et al. 2003) and
India (Gopinath et al. 2008). Prophylaxis and manage-
ment of disease caused by nontubercular mycobacteria
differ greatly from that of M. tuberculosis. Therefore, an
early detection and identification of the infecting myco-
bacterial species are most desirable for early and specific
therapy and better patient management (Montessori et al.
1996; Powderly 1996), especially after HIV epidemic.
In developing countries, the diagnosis of mycobacterial
infections is not made timely leading to mismanagement
and empirical therapeutic trials, which give rise to devel-
opment of drug resistance (CDC 2002). The major reason
for this delayed and inaccurate diagnosis is reliance of
laboratories on least sensitive detection method i.e.
microscopical demonstration of acid-fast bacilli (AFB).
This method has the poorest detection rate in HIV posi-
tive patients (Gopinath et al. 2007). Some tertiary care
hospital laboratories in these countries carry out specia-
tion but only after culturing the organisms by the
conventional Lowenstein–Jensen (L-J) culture, which is a
cumbersome and time-consuming procedure (Hanna
et al. 1999). The whole exercise of isolation and specia-
tion sometime takes too long time and by that time most
often the patient has almost completed his treatment.
With the employment of automated culture detection
system, the time taken for detecting the bacilli has
reduced and the sensitivity has also increased (Hanna
et al. 1999), but not to the desired level, especially for
extrapulmonary samples like cerebrospinal fluid (CSF),
pleural fluid and urine.
Despite the limitation of detecting dead bacilli, Poly-
merase Chain Reaction (PCR)-mediated DNA amplifica-
tion methods allow rapid and precise identification of
mycobacteria from all samples including blood and pauci-
bacillary clinical specimens irrespective of their HIV
status. Even though M. tuberculosis-specific monoplex
PCRs are being commercially marketed and also done in
some research and referral laboratories, these monoplex
PCRs are bound to miss out the coinfections and infec-
tions with nontubercular mycobacteria. Therefore, in this
study, a multiplex PCR method was developed, by using
specific gene targets, to detect and differentiate M. tubercu-
losis complex, M. avium complex and other Mycobacterial
426 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
K. Gopinath and S. Singh
species directly from clinical samples. The sensitivity and
specificity of the multiplex PCR were evaluated and com-
pared with Ziehl Neelsen (ZN) microscopy, L-J culture
and BACTECTM MGIT-960 automated culture system.
Materials and methods
This study was carried out in the Department of Labora-
tory Medicine, All India Institute of Medical Sciences
(AIIMS), New Delhi, India. The study was approved by
Institutional Research Ethics Committee and a written
consent was obtained from all patients who consented to
provide 5 ml of blood specimen for multiplex PCR-based
detection of mycobacteriosis.
Optimization of the multiplex PCR assay
Mycobacterial strains used
A total of six mycobacterial reference strains, 128 clinical
isolates of mycobacteria and other bacteria were used to
assess the specificity of the multiplex PCR. The six refer-
ence mycobacterial strains (M. tuberculosis H37Rv
(TMC-102), M. avium (NCTC-8551), M. intracellulare
(TMC1406), Mycobacterium terrae (TMC-1450), Mycobac-
terium duvalii (NCTC-358) and Mycobacterium smegmatis
mc2155) were obtained as a kind gift from Dr V.M. Kat-
och, Director, National JALMA Institute for Leprosy and
other Mycobacterial Diseases, Agra, India. The mycobac-
terial clinical isolates used for the standardization were
confirmed by multiple molecular methods including the
species-specific PCR, AccuprobeTM MTB complex detec-
tion test (Gen-Probe Inc., San Diego, CA, USA) and
DNA sequencing, details of which are published elsewhere
(Singh et al. 2007; Gopinath et al. 2008). These well-
characterized isolates included M. tuberculosis H37Rv
(n = 62), M. avium (n = 11), Mycobacterium scrofulaceum
(n = 5), M. smegmatis (n = 4), Mycobacterium intracellu-
lare (n = 4), Mycobacterium fortuitum (n = 4), M. terrae
(n = 4), Mycobacterium kansasii (n = 3), M. bovis
(n = 2), Mycobacterium chelonae (n = 2) and M. bovis
BCG (n = 1). Five other bacterial species included as neg-
ative controls were: Escherichia coli (n = 19), Staphylococ-
cus aureus (n = 3), Proteus mirabilis (n = 2), Acinetobacter
sp. (n = 1) and Pseudomonas aeruginosa (n = 1). All
mycobacterial strains were grown and maintained on L-J
medium (Kent and Kubica 1985), whereas other bacterial
species were maintained on appropriate bacteriological
solid media such as Muller–Hinton Agar, McConkey
Agar, etc. For DNA isolation from nonmycobacterial iso-
lates, two loopful of bacteria was taken in a sterile 1Æ5 ml
eppendorf tube, DNA was isolated using rapid Phenol–
Chloroform extraction method (Sambrook et al. 1989)
and the isolated DNA was dissolved in 50 ll of TE buffer,
ª 2009 The Authors
K. Gopinath and S. Singh
before quantification. To determine the analytical sensiti-
vity of multiplex PCR, the DNA of M. tuberculosis H37Rv
(TMC-102), M. avium (NCTC-8551) and M. smegmatis
mc2155 were serially diluted in TE buffer from 1 lg
to 1 fg per microlitre. The DNA concentration was
measured spectrophotometrically (Thermo Scientific,
Wilmington, DE, USA); 1 ll of the diluted DNA was
used as a template for PCR.
Optimization of the PCR protocol
For optimizing the multiplex PCR to detect mycobacteria
directly from clinical specimens, a total of 31 sputa that
were mycobacteriologically negative by both culture and
monoplex PCR were taken out from our archival collec-
tion stored at )70C, defrosted to room temperature.
Fourteen of these were spiked with M. tuberculosis H37
Rv, eight with M. avium, seven with E. coli and the
remaining two with Proteus mirabilis. The inoculum used
for spiking the clinical specimens was made with a loop-
ful of the respective culture suspended in 1Æ0 ml of sterile
distilled water equivalent to No.1 McFarland turbidity
standard (Difco ⁄ BBLTM), and 50 ll of this suspension
was used for spiking the sputum specimen (c. 3 ml). After
brief vortexing, 0Æ5 ml of the spiked and neat specimen
was aliquoted and DNA was extracted by Phenol–Chloro-
form method (Sambrook et al. 1989). The ethanol precip-
itated DNA was further solubilized in 20 ll of TE buffer
and 10 ll of the same was used for PCR reaction.
Evaluation of the m-PCR assay using clinical specimens
After standardizing the DNA isolation and PCR protocols,
a prospective study was conducted from April 2005 to
March 2006. In this, 145 patients referred from various
clinics of our own institute and other hospitals in and
around Delhi for TB diagnosis were included. All the
patients were aware of their HIV status during the
enrolment in the study. The clinical suspicion of tubercu-
losis was based on their clinico-radiological findings, e.g.
fever with or without cough; weight loss or night sweats
lasting longer than 2 weeks; and ⁄ or cavitary lesions in the
lung fields on radiological examination; or suppurating or
nonsuppurating cold single or mated lymph nodes (ATS
1999; Reves et al. 2003). All patients were counselled
prior to the test and who consented for providing
clinical specimen, viz. sputum, lymph node aspirate
(LNA), ascitic fluid, stool and urine, were included in the
study. From 145 patients, a total of 235 clinical
specimens (169 pulmonary and 66 extra pulmonary) were
obtained and processed for various in vitro diagnostic
methods including the multiplex PCR under evaluation.
Of the 145 patients, 68 were HIV positive and 77 HIV
negative.
ª 2009 The Authors
Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
Multiplex PCR assay for M. tuberculosis and M. avium complex
Of the 68 HIV infected patients, 49 had suspected pul-
monary tuberculosis (PTB), while 19 had extrapulmonary
tuberculosis (EPTB). From 68 HIV-positive patients, 92
clinical specimens were obtained (72 sputa from 49 HIV-
PTB patients and 20 specimens from 19 EPTB patients)
and processed using the protocol described below. From
77 HIV-negative patients, 143 clinical specimens were
obtained, which included 97 pulmonary (95 sputum and
2 BAL) specimens and 46 extra pulmonary specimens (21
LNAs, 7 pleural fluids, 6 urine, 4 CSF, 4 peritoneal fluid,
2 endometrial aspirate and 2 faecal samples).
For assessing the specificity of multiplex PCR assay on
other unrelated specimens, 39 specimens from patients
having no past history of TB and having confirmed diag-
nosis of nonmycobacterial diseases were included. These
included 12 LNAs ⁄ biopsies from patients of malignant
lymphoma, 10 sputa from patients with pyogenic respira-
tory tract infections, 9 bronchioalveolar lavage from
patients with intrathoracic malignancy and 8 urine
samples from patients with bacterial urinary tract infec-
tions were included.
Mycobacteraemia has been reported more frequently in
HIV infected patients from all parts of the world inclu-
ding India (Gopinath et al. 2008). To detect mycobactera-
emia using the multiplex PCR, 42 HIV-positive and 52
HIV-negative cases with a clinical suspicion of PTB (ATS
1999; Reves et al. 2003) were also included, as a satellite
study. For detecting mycobacteraemia, 5 ml of intrave-
nous blood was collected under aseptic conditions in
sterile containers having 1 ml acid citrate dextrose.
Culture and identification
The sterile body fluids (CSF and LNA) were directly inocu-
lated into the MGIT-960 culture bottles and part of the
specimens was stored for the DNA isolation at )70C. The
contaminated specimens such as sputum, stool, urine and
ascitic fluid were first decontaminated with NALC-NaOH
(modified Petroff’s method) as reported earlier (Kent and
Kubica 1985). In brief, about 5–6 ml of contaminated
specimen (sputum, urine or stool) was mixed with an equal
volume of 0Æ5% NALC – 4% NaOH mixture in 50 ml ridge
capped round bottom processing tube, the mixture was
vortexed and incubated at 37C for 10 min. After incuba-
tion, mixture was neutralized with phosphate buffer (pH
6Æ8) up to a total volume of 50 ml followed by centrifuga-
tion at 8000 g for 10 min. The supernatant was decanted
and the pellet was resuspended in 2 ml of the phosphate
buffer. From the decontaminated sample, 0Æ5 ml of the
suspension was inoculated into the BACTECTM MGIT
tubes and 0Æ1 ml in L-J slants under sterile conditions in
the biosafety cabinet type 2 (Kartos, India) and from the
rest of the sample, smears were made for the ZN staining
from both the direct and decontaminated specimens.
427
Multiplex PCR assay for M. tuberculosis and M. avium complex
The inoculated MGIT-960 tubes were loaded in the
BACTECTM MGIT-960 system, and the growth was
continuously monitored up to 42 days in the fluorescence
units and tubes flash positive after reaching a cut-off
growth. The L-J slants were incubated at 37C for
6 weeks (Kent and Kubica 1985) and checked every week
for the growth. Positive signalled BACTECTM MGIT-960
tubes and colonies on L-J medium were confirmed for
the presence of mycobacteria by microscopical examina-
tion of the smears prepared from these cultures. Flashed
positive cultures with no demonstrable AFB were checked
with Gram’s staining for other bacteria. Cultures showing
other bacteria on Gram’s staining were decontaminated as
per the protocol mentioned above. If negative in both
Gram and ZN staining, these were further incubated for a
maximum period of 60 days or till the growth appeared
in the medium. The Mycobacterium species isolated from
the clinical specimens were identified phenotypically and
by biochemical tests including heat stable catalase, nitrate,
niacin and aryl-sulfatase test (Kent and Kubica 1985)
Other bacteria were also identified and speciated in accor-
dance with standard reference manuals (Holt 1994; Col-
lier et al. 1998) in practice in our laboratory.
Speciation of the isolates by PCR
For speciation of mycobacteria by multiplex PCR, DNA
was isolated from BACTECTM MGIT-960 tubes (1 ml
growth) or from the L-J medium (one loopful). The
aliquots were kept in a water bath at 80C for 20 min to
heat kill the mycobacteria. The suspension was centri-
fuged at 6100 g for 10 min and the pellet was resus-
pended in 1 ml TE buffer (Sigma, St Louis, MO, USA).
To this suspension, 0Æ5% Sodium Dodecyl Sulfate and
1Æ2% Triton X-100 were added followed by boiling for
1 h. As this method does not warrant any further purifi-
cation of the DNA and, hence, was directly used for PCR
(Kocagoz et al. 1993). We have been using the same
protocol in our laboratory for the past several years.
DNA isolation directly from the Clinical specimens
Extraction of DNA from the clinical specimens and
PBMCs was performed as described previously (Ausubel
et al. 1998). In brief, essentially, cell wall was lysed with
lysozyme, followed by proteinase K digestion and sodium
dodecyl sulfate treatment of proteins. Proteins and mac-
romolecules were precipitated using NaCl and hexadecyl-
trimethlyammonium bromide ⁄ NaCl solutions. Nucleic
acids were recovered from the aqueous phase after extrac-
tion with chloroform and isoamyl alcohol. The DNA was
precipitated overnight with isopropanol at )20C. The
pellet was washed with ethanol, reconstituted in 50 ll of
TE buffer, and 10 ll of the same was used in the PCR.
428 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
K. Gopinath and S. Singh
For extracting the DNA from whole blood, 5–10 ml
blood from suspected patients was collected aseptically as
described above and peripheral blood mononuclear cells
(PBMCs) were separated using Ficoll Hypaque (density
1Æ007 grams ⁄ ml; Sigma). These PBMCs were subjected for
DNA extraction method for PCR, as described in the
beginning of this section. .
Genus- and species-specific PCR primers
In the multiplex PCR we used three primers sets,
including the Mycobacterium genus specific primers
(hsp 65) (Telenti et al. 1993), M. avium complex
specific (MAC) primers (Park et al. 2000) and a novel
M. tuberculosis (MTB) complex-specific set of primers
for the first time. The latter targets the cfp10 (Rv3875,
esxB) and the upstream primer was named cfp10F
(5¢GGCAGAGATGAAGACCGATG3¢) and downstream
primer as cfp10R (5¢CCAAGAAGCAGCCAATAAGC3¢).
All oligonucleotides used in the study were synthesized by
Microsynth Inc. (Balgach, Switzerland). The DNA samples
from clinical samples and culture bottles were subjected
to multiplex PCR and the constituents of mixtures were
as follows: 500 mmol l)1 KCl, 100 mmol l)1 Tris–HCl
(pH 9Æ0), 1% Triton X-100, 0Æ2 mmol l)1 each deoxynu-
cleoside triphosphate (dATP, dGTP, dTTP and dCTP),
1Æ5 mmol l)1 MgCl2, 20 pmol of each primer and 1 U of
Taq DNA polymerase (Promega, USA). Each PCR was
started with a ‘hot start’ for 10 min at 94C followed by
30 cycles each of 1 min at 94C, 1 min at 60C and
1 min at 72C, and a final extension for 10 min at 72C
in a thermal cycler (MJ Research, USA). DNA isolated
from cultured M. tuberculosis H37Rv and M. avium
was included as positive controls with sterile water as
negative control in each run. Precautions to prevent
cross contamination were also taken. Amplified products
were resolved through 2% agarose gel in Tris-acetate
buffer. Mycobacterium genus specific, 191 bp for the
M. tuberculosis complex specific and 144 bp for
M. avium complex specific were visualized by staining
with ethidium bromide (Fig. 1).
Purified PCR products were sequenced directly with
the primers Tb11 and Tb12 (Mycobacterium sp.), cfp10F
and cfp10R (M. tuberculosis), and MACF and MACR
(M. avium complex) in a commercial sequencing facility.
The sequences were aligned using the multiple alignment
algorithms in the ClustalW and Blast tool to deter-
mine the species using the reference sequences available
in GenBank.
The results of culture, AFB smear and PCR were com-
pared with the final diagnosis of TB using individual
patients as the unit of analysis and separately using indi-
vidual specimens as the unit of analysis. Significance was
determined by the Chi-square test. The PCR results were
ª 2009 The Authors
K. Gopinath and S. Singh
M 1 2 3 4 5
441 bp
191 bp
144 bp
Figure 1 Multiplex PCR amplification of genus-specific 441 bp (hsp
65), MTB-specific 191 bp (cfp 10), M. avium complex-specific (MAC)
144 bp. (M – 100 bp marker (MBI, Fermentas); 1,4 – M. tuberculosis;
2,5 – M. avium complex; 3,6 – M. tuberculosis + M. avium complex;
7 ) Negative control).
classified as true positives (Tp), true negatives (Tn), false
positives (Fp) and false negatives (Fn). Sensitivity was
calculated as: (Tp ⁄ Tp) + (Fn) X-100, and specificity was
calculated as: (Tn ⁄ Tn) + (Fp) X-100.
Results
Optimization of the multiplex PCR assay
The multiplex PCR for the detection and differentiation
between the members of M. tuberculosis complex, M. avi-
um complex and other Mycobacterium species was stan-
dardized with the DNA isolated from the mycobacterial
and other bacterial strains. The multiplex PCR precisely
amplified the corresponding targets viz, Mycobacterium
genus specific (441 bp), M. tuberculosis complex specific
(191 bp) and M. avium complex specific (144 bp)
(Fig. 1). A total of eight amplified products were ran-
domly selected and sequenced. The assembled sequences
were compared with the reference sequences by Blastn
analysis, which showed 98–100% homology with their
target sequences indicating high specificity of our PCR.
Furthermore, no nonspecific amplification was seen
among the other 26 DNA samples from nonmycobacterial
isolates used in this study (details not shown here). The
multiplex PCR developed was further optimized with the
clinical specimens spiked with mycobacterial and other
bacterial species. All the 14 M. tuberculosis and eight
M. avium spiked sputum specimens were precisely identi-
fied. In addition, the nine clinical specimens spiked with
other nonmycobacterial species and the specimens used as
neat did not show any nonspecific amplification (details
not shown here). While evaluating analytical sensitivity of
PCR, it was found to be able to detect as low as 10 fg
ª 2009 The Authors
Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
Multiplex PCR assay for M. tuberculosis and M. avium complex
6 7 (a) M 1 2 3 4 5 6 7
(b) M 1 2 3 4 5 6 7
(c) M 1 2 3 4 5 6 7
Figure 2 Analytical sensitivity of multiplex PCR for detecting Myco-
bacterium sp., Mycobacterium tuberculosis, Mycobacterium avium
complexes. Figure shows the analytical sensitivity of multiplex PCR
from the DNA isolated from M. tuberculosis, M. avium and other
mycobacterial species. Amplified products were visible only with a
DNA concentration exceeding 10 fg (4–5 bacilli). (a) amplified from 1,
10 ng; 2, 100 pg; 3, 10 pg; 4, 1 pg; 5, 100 fg; 6, 10 fg; 7, 1 fg DNA
of Mycobacterium smegmatis (b) amplified from 1, 10 ng; 2, 100 pg;
3, 10 pg; 4, 1 pg; 5, 100 fg; 6, 10 fg; 7, 1 fg DNA of M. tuberculosis
(c) amplified from 1, 100 fg; 2, 1 pg; 3, 10 pg; 4, 100 pg; 5, 1 ng; 6,
10 ng of DNA of M. avium. The DNA concentration was measured
spectrophotometrically at 260 nm.
(3–4 mycobacteria) of DNA (Fig. 2). These preliminary
results prompted and gave us a confidence to evaluate the
application of multiplex PCR system in clinical specimens
along with the other tests such as microscopy, L-J and
BACTECTM MGIT-960 culture.
Patients and specimens
A total of 145 patients, 122 (84Æ13%) men and 23
(15Æ86%) women with an age range of 13–58 (31 ± 6Æ71)
years, were included in this study. Of these, 77 (53Æ1%)
were HIV negative and 68 (46Æ85%) were HIV positive.
Of the 68 HIV positive patients [aged between 28 and
46 (mean 30 ± 4Æ3) years] included in the study, 53
(77Æ94%) were men and 15 (22Æ05%) were women. The
common clinical manifestations in these patients were
weight loss (87Æ12%), fever (78Æ2%), cough with or with-
out expectoration (71Æ4%) and raised ESR (68Æ6%). The
429
Multiplex PCR assay for M. tuberculosis and M. avium complex K. Gopinath and S. Singh

Table 1 Detection of mycobacterial infections by multiplex PCR from both pulmonary and extrapulmonary specimens of HIV-positive patients

Culture positive* Multiplex PCR positive

Sputum Extrapulmonary Total Sputum Extrapulmonary Total

(n = 72) (n = 20) (n = 92) (n = 72) (n = 20) (n = 92)

Mycobacterium tuberculosis 34 (47Æ22) 8 (40) 42 (45Æ65) 48 (66Æ66) 12 (60) 60 (65Æ21)

Mycobacterium avium 4 (5Æ55) 1 (5) 5 (5Æ43) 3 (4Æ16) 1 (5) 4 (4Æ34)
MTB + M. avium 2 (2Æ77) – 2 (2Æ17) 3 (4Æ16) – 3 (3Æ26)
Mycobacterium spp. 2 (2Æ77) 1 (5) 3 (3Æ26) 4 (5Æ55) 1 (5) 5 (5Æ43)

Total 42 (58Æ33) 10 (50) 52 (56Æ52)§ 58 (80Æ55) 14 (70) 72 (78Æ26)

Values given in parenthesis are percentages.

*Data include combined result of L-J and MGIT-960 culture.
PCR from one of the specimen is positive for both M. tuberculosis and M. avium complex.
Both BACTEC and L-J Culture yielded M. avium in one of the specimen.
§18 (19Æ56%) were positive by ZN smear examination, 34 (36Æ95%) were positive for mycobacteria in L-J and 48 (52Æ17%) in MGIT-960 culture.

pattern of clinical manifestations was similar in the 77
HIV-negative patients, i.e. fever (88Æ88%), cough with or
without expectoration (81Æ6%) and raised ESR (61Æ5%).
Of the 68 clinically suspected patients of HIV-TB
co-infection, 31 had disseminated TB, 16 had PTB, and
seven had tuberculous lymphadenitis, abdominal TB, and
genitourinary TB. Of 77 HIV-negative patients, 34 had
PTB, 21 had tuberculous lymphadenitis, 11 tuberculous
meningitis, seven genitourinary and four had abdominal TB.
Culture positivity rate by L-J and BACTEC MGITTM 960
The culture isolation rate of Mycobacteria from pulmo-
nary specimen was 58Æ33% from the HIV-positive
(Table 1) and 53Æ6% HIV-negative patients (Table 2),
whereas the isolation rate from extrapulmonary specimens
was found to be higher among the HIV-positive cases as
compared with HIV-negative cases (50% vs 34Æ78%,
P = 0Æ01) (Tables 1 and 2). From 68 HIV-positive
patients, 52 isolates were obtained of which 42 (80Æ7%)
were M. tuberculosis, 5 (9Æ61%) M. avium, 2 (3Æ84%)
M. chelonae, and 1 (1Æ53%) M. terrae. In one patient,
mixed infection of M. tuberculosis + M. avium complex
was detected (Tables 2 and 3). From 77 HIV-negative
patients, we had 68 isolates of which 63 (92Æ6%) were
M. tuberculosis, two were M. avium and one each was
M. terrae, M. scrofulaceum and M. xenopi (Table 2).
Positivity rate of multiplex PCR
The positivity rate of the in-house multiplex PCR was
found to be the highest of all the diagnostic tests used in
this study (P = 0Æ01). As mentioned in the materials and
methods section, from 68 HIV positive patients, 92 sam-
ples were obtained (PTB, 72 and EPTB 20). From 72 pul-
monary samples, 48 (66Æ7%) were detected with
M. tuberculosis, while 12 of the 20 (60%) EPTB specimens
were found positive for M. tuberculosis. The details of
these findings are given in Table 1. Hence the overall
positivity rate of the in-house m-PCR in samples from
HIV-positive patients was 78Æ2% (72 ⁄ 92). The detection
rate of the m-PCR was not significantly different in

Table 2 Detection of mycobacterial infections by multiplex PCR from both pulmonary and extrapulmonary specimens of HIV-negative patients

Culture positive* Multiplex PCR positive

Sputum Extrapulmonary Total Sputum Extrapulmonary Total

(n = 97) (n = 46) (n = 143) (n = 97) (n = 46) (n = 143)

Mycobacterium tuberculosis 48 (49Æ48) 15 (32Æ6) 63 (44Æ05) 61 (62Æ88) 26 (56Æ52) 87 (60Æ83)

Mycobacterium avium 2 (2Æ06) – 2 (1Æ3) 3 (3Æ09) – 3 (2Æ1)
MTB + M. avium – – – – 1 (2Æ17) 1 (0Æ69)
Mycobacterium spp. 2 (2Æ06) 1 (2Æ17) 3 (2Æ09) 4 (4Æ12) 2 (4Æ34) 6 (4Æ19)

Total 52 (53Æ6) 16 (34Æ78) 68 (47Æ55) 68 (70Æ1) 29 (63Æ04) 97 (67Æ83)

Values given in parenthesis are percentages.

*Data include combined result of L-J and MGIT-960 culture.
Endometrial aspirate sample from 23-year-old female with a history of infertility, negative by culture.
Twenty-nine (20Æ3%) were positive by ZN smear examination, 47 (32Æ8%) were positive for mycobacteria in L-J and 60 (41Æ9%) in MGIT-960 culture.

ª 2009 The Authors

430 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
K. Gopinath and S. Singh Multiplex PCR assay for M. tuberculosis and M. avium complex

Table 3 Sensitivity of the multiplex PCR vis-á-vis other tests on pulmonary and extrapulmonary specimens of suspected HIV–tuberculosis patients

Multiplex PCR (+)

MTB MAC MTB + MAC Mycobacterium spp. Total

HIV Positive (92) 60 (65Æ2) 4 (4Æ3) 3 (3Æ3) 5 (5Æ4) 72 (78Æ3)

Bacteriologically confirmed specimens (52) 41 (78Æ8) 4 (7Æ7) 3 (5Æ8) 3 (5Æ8) 51 (98Æ07)
MTB (42) 41 (97Æ6) 0 0 0 41 (97Æ6)
MAC (5) 0 4 (80) 1 (20) 0 5 (100)
MTB+MAC (2) 0 0 2 (100) 0 2 (100)
Mycobacterium spp. (3) 0 0 0 3 (100) 3 (100)
Bacteriologically negative specimens (40) 19 (47Æ5) 0 0 2 (5) 21 (52Æ5)
HIV Negative (143) 87 (60Æ9) 3 (2Æ1) 1 (0Æ7) 6 (4Æ2) 97 (67Æ8)
Bacteriologically confirmed specimens (68) 61 (89Æ7) 2 (2Æ9) 0 3 (4Æ4) 66 (97Æ05)
MTB (63) 61 (96Æ8)* 0 0 0 61 (96Æ8)
MAC (2) 0 2 (100) 0 0 2 (100)
Mycobacterium spp. (3) 0 0 0 3 (100) 3 (100)
Bacteriologically negative specimens (75) 26 (34Æ7) 1 (1Æ3) 1 (1Æ3) 3 (4) 31 (41Æ3)

Values given in parenthesis are percentages.

*lymph node aspirate and CSF specimen from two HIV-negative cases were negative by PCR.
culture positive lymph node aspirate specimen from a HIV-positive case was negative by PCR.

samples obtained from HIV-negative patients. These rates
were 62Æ88% and 56Æ52%, respectively, for PTB and EPTB
samples and overall positivity rate was 67Æ83% (103 ⁄ 143)
(Table 2). Even this difference was statistically insignifi-
cant.
Sensitivity and specificity of the multiplex PCR
The sensitivity of the multiplex PCR in mycobacteriologi-
cally confirmed, HIV-positive cases and HIV-negative
cases was 97Æ1% (51 ⁄ 52) and 97Æ05% (66 ⁄ 68) respectively
(Table 3). LNA from one HIV positive and one negative
patient and one CSF from a HIV-negative patient
remained negative even after repeating the PCR and
DNA isolation. However, amplification of human gamma
globulin gene (578 bp) in every run as an internal control
indicated that our PCR method was accurate.
While evaluating the specificity of multiplex PCR, it
was found to be 94Æ87% and false-positive reaction was
seen only in two samples (BAL and Urine sample one
each) (Table 4). The two false-positive results remained
positive even after repeating the PCR, suggesting a
possible overlapping co-morbidity.
Detection of Mycobacteraemia by multiplex PCR
Detection of circulating Mycobacterial DNA in the blood
stream was significant finding of the study. Multiplex PCR
was positive in blood specimens from 13 (30Æ95%) of the
42 HIV-positive PTB patients included for mycobacterial

Table 4 Specificity of multiplex PCR on clinical specimens collected from the nontubercular disease controls

PCR results
Specificity
Nature of specimens Disease Positive Negative (in %)

Lymph node Aspirate ⁄ Malignant lymphoma (N = 12) – 12 100

biopsies (n = 12)
Sputa (n = 10) Chronic bacterial lung infection (N = 10) – 10 100
Bronchio Alveolar Intrathoracic malignancy (N = 3) – 3 100
Lavage (n = 9)
Other infectious aetiology (N = 6)* 1 5 83Æ3
Urine (n = 8) Other urinary tract infections (N = 8) 1 7 87Æ5

Total (n = 39) 2 37 94Æ87

*Patients with pulmonary infiltrates from different origin such as sarcodiosis ⁄ COPD ⁄ pneumonia.
Urinary tract infection caused by bacteria such as Escherichia coli, Pseudomonas sp.
Specimens were negative by smear microscopy, L-J and MGIT-960 culture methods.

ª 2009 The Authors

Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435 431
Multiplex PCR assay for M. tuberculosis and M. avium complex K. Gopinath and S. Singh

Table 5 Detection rate of mycobacteraemia by multiplex PCR from both HIV positive and negative suspected pulmonary TB cases

Detection of mycobacteraemia by multiplex PCR (n = 42)

Patient selection criteria MTB MAC MTB + MAC Total

HIV-Positive cases (42)

Bacteriologically confirmed PTB cases* (18) 9 (50) 1 (5Æ5) 1 (5Æ5) 11 (61Æ1)
MTB (13) 8 (61Æ53) – – 8 (61Æ53)
MAC (3) – 1 (33Æ3) 1 (33Æ3) 2 (66Æ6)
MTB + MAC (1) 1 – – 1
MOTT (1) – – – –
Bacteriologically negative PTB cases* (24) 4 (16Æ16) – 1 (4Æ16) 5 (20Æ83)
PCR Positive (15) 3 (20) – 1 (6Æ66) 4 (26Æ6)
PCR Negative (9) 1 (11Æ1) – – 1(11Æ1)

Total 13 (30Æ95) 1 (2Æ38) 2 (4Æ76) 16 (38Æ09)

Multiplex PCR positive from blood specimens (n = 52)

Patient selection criteria MTB MAC MTB + MAC Total

HIV-negative cases (52)

Bacteriologically confirmed PTB cases* (28) 7 (25) – – 7 (25)
Bacteriologically negative PTB cases* (24) 1 (4Æ16) – – 1 (4Æ16)
PCR Positive (11) 1 (9Æ09) – – 1 (9Æ09)
PCR Negative (13) – – – –

Total 8 (15Æ38) – – 8 (15Æ38)

Values given in parenthesis are percentages.

*Sputa collected from the cases were processed for smear L-J, MGIT-960 culture and PCR. To consider as bacteriologically confirmed cases, at
least one specimen from the patient must have been positive in smear and ⁄ or L-J or MGIT-960 culture.

DNA detection. Interestingly, eight (15Æ38%) of the 52
HIV-negative cases were also found to have circulating
mycobacterial DNA in their blood specimens. All except
two samples had M. tuberculosis amplification only. The
two patients had double amplification of M. tuberculosis
and M. avium complex specific targets and were found to
be HIV positive (Table 5).
Discussion
Multiplex PCR assays are technically more demanding as
compared with conventional monoplex PCR protocols.
These technical concerns are competition and ⁄ or homo-
logy between the primers chosen, annealing temperature
and optimum concentration of the primers used. Though
different multiplex PCR assays have been reported for
the differentiation of M. avium from M. intracellularae
complex (Ruiz et al. 2001), M. tuberculosis complex from
M. bovis (Shah et al. 2002; Pinsky and Banaei 2008) and
M. tuberculosis complex from M. avium complex (Mus-
tafa et al. 1999; Tanaka et al. 2003; Park et al. 2006),
ours was unique in that we could successfully differenti-
ate M. tuberculosis and M. avium complexes from other
mycobacterial species. Mustafa et al. (1999) developed
an m-PCR for the differentiation of MTB complex from
other mycobacterial species with a sensitivity of 88%,
whereas the present assay had sensitivity between 96Æ0%
and 100% depending on the nature of the clinical speci-
mens (Table 3). Even Park et al. (2006) had reported
m-PCR to differentiate M. tuberculosis from M. avium
complexes with a sensitivity of 96Æ4% and they used only
respiratory samples, while our system had high sensitivity
not only in the respiratory samples but also in various
other clinical specimens including sputum, CSF, LNA,
pleural fluid, blood and urine (Table 3). Our system not
only worked well with all types of clinical samples but
also equally useful in both-HIV positive (sensitivity of
98Æ07%) and HIV-negative (97Æ05%) cases.
Our Multiplex PCR was also highly useful in detecting
the mixed infections of M. avium + M. tuberculosis in
two HIV-positive cases. It is interesting that in one case,
L-J culture showed mixed growth of M. tuberculosis and
M. avium, but in the other case, we could have missed
the M. avium if the m-PCR had not been applied to the
specimen. The sputum sample from this patient showed
luxurious growth of M. tuberculosis in the primary
culture. However, after having the PCR results, the seed
culture was subcultured and we successfully differentiated
mixed growth of M. tuberculosis and M. avium on the L-J
medium slant. It is possible that due to similar culture
characteristics of M. tuberculosis and M. avium, the latter
is missed out, if not looked for carefully (Runyon 1974).

ª 2009 The Authors

432 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
K. Gopinath and S. Singh
Mycobacterium avium is an emerging pathogen in AIDS
patients and needs better attention of both clinical micro-
biologists and treating physicians. Nunez et al. (1997)
from Spain reported M. tuberculosis and M. avium from
the skin of an AIDS patient, and Damian et al. (2004)
from New York reported isolation of M. tuberculosis and
M. avium from sputum samples of an immunocompetent
patient. Recent reports indicate that not only mixed
infections in immunocompromised patients are becoming
common but also rarely two organs of the same patient
can be colonized with two different mycobacterial species
(Murcia-Aranguren et al. 2001; Gopinath et al. 2008).
To the best of our knowledge and in the literature
search published in English, ours is the first multiplex
PCR used for detecting mycobacteraemia in blood speci-
mens from both HIV-positive and HIV-negative patients
(Table 5). Using multiplex PCR system, we observed that
38Æ09% of the blood specimens were positive for myco-
bacteriosis. Hence, the utility of blood samples in the
diagnosis of tuberculosis is highly promising, at least as
an adjunct if not a primary choice of sample. Even
though some authors have used blood specimens for
diagnosing tuberculosis in HIV infected patients (Iralu
et al. 1993; Rodrigues Vde et al. 2002 and Crump et al.
2003), monoplex PCR assays have been used in all these
studies. In the present study, the multiplex PCR was
successfully applied to diagnose and differentiate myco-
bacterial infections. More interestingly, in the present
study, we could detect mycobacterial DNA in 5 (20Æ83%)
of 24 HIV negative cases using blood m-PCR (Table 5),
because the detection rate of mycobacteraemia in HIV-
negative patients using culture methods is reported to be
extremely low (von Gottberg et al. 2001; Waddell et al.
2001; Gopinath et al. 2008).
PCR-positive results without bacteriological confirma-
tion of the specimen are often doubted as false positive,
but these apprehensions are not often true (Katoch 2004).
In some cases, this may be the only evidence of infection.
Nevertheless, care should be taken while interpreting the
PCR results if considered as sole evidence. For PCR
assays, DNA extraction is crucial and for optimal DNA
quantity, one might require costly reagents such as resins
or columns to remove possible inhibitors. However, the
present study shows that the protocol for DNA extraction
and PCR used in our laboratory was highly satisfactory
and nonspecific inhibition was seen only in 1 out of 120
(0Æ83%) mycobacteriologically confirmed specimens
(Table 3). On culturing this LNA specimen, M. tuberculo-
sis grew in MGIT-960. The LNA has earlier also been
reported to have PCR inhibition (Singh et al. 2000), how-
ever, the rate of inhibition using the protocol used in this
study was practically negligible as only one out of 21
(4Æ7%) samples had PCR inhibition. The protocol did not
ª 2009 The Authors
Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
Multiplex PCR assay for M. tuberculosis and M. avium complex
require costly resins or any columns, which may not be
affordable in developing countries, if PCR is adopted for
TB-control programmes.
In this study, the primers (cfp10) selected from the
RD-1 region of M. tuberculosis were specific to M. tuber-
culosis complex except M. bovis BCG. However, the exis-
tence of similar homologous genes in M. kansassi,
Mycobacterium marinum and Mycobacterium ulcerans does
not compromise the specificity because of their rare asso-
ciation with human disease and environmental niche at
least in India. In recent years, a major advance in mole-
cular diagnostic has been real-time PCR. The real-time
technology is not only more rapid and but also quantifies
the pathogen in the given specimen. Recently, a rapid and
cost-effective two-step, multiplex real-time PCR assay
based on genomic deletions in the Mycobacterium has
been reported by Pinsky and Banaei (2008). This assay
was used to identify M. tuberculosis, M. bovis, M. bovis
BCG, and other members of the complex from both
culture and directly from clinical specimens. The real-
time PCR has an edge over the conventional PCR because
the PCR products can be visualized on the screen simul-
taneously to the amplification process using end-stage
hybridization or fluorescence probes and does not require
gel-based detection. However, conventional PCRs have
been adopted in most of the laboratories unlike real-time
PCRs which are more difficult to standardize and are
expensive.
The biggest impact of nucleic acid amplification tech-
nology is expected on TB-control programmes. With the
implementation of DOTS strategy, specimen examination
is limited to direct sputum smear microscopy, which can-
not rule out the NTM lung disease (Jeon et al. 2005;
Singh et al. 2007). In our settings, 17Æ6% of the PTB
(Sankar et al. 2008) and 12Æ4% of EPTB patients were
found to have nontubercular diseases. This has been a
contentious issue if all DOTS programmes should
continue with only smear microscopy, which misses NTM
disease and often the nonresponders are labelled as
Multi-drug resistant tuberculosis cases (Gopinath et al.
2007; Singh 2008). Therefore, we propose that the multi-
plex PCR developed in our laboratory can detect and
differentiate the causative mycobacterial species within
24–36 h and this development can also be utilized in the
national TB-control programmes.
Acknowledgements
We wish to thank Dr V.M. Katoch of Central JALMA
Institute for leprosy and other Mycobacterial Diseases,
Agra for providing us with the standard strains of
mycobacteria. Financial support from Indian Council of
Medical Research and Department of Biotechnology,
433
Multiplex PCR assay for M. tuberculosis and M. avium complex
Govt. of India to S.S. and K.G. (Senior Research Fellow-
ship) is acknowledged. The work was conducted as a part
of PhD programme of K.G.
References
ATS. (1999) Diagnostic Standards and Classification of Tuber-
culosis in Adults and Children. This official statement of
the American Thoracic Society and the Centers for Disease
Control and Prevention was adopted by the ATS Board of
Directors, July 1999. This statement was endorsed by the
Council of the Infectious Disease Society of America. Am J
Respir Crit Care Med 161, 1376–1395.
Ausubel, F.M., Brent, R., Kingston, R.E., Moore, D.D.,
Seidman, J.G., Smith, J.A. and Struhl, K. (1998) Prepara-
tion of Genomic DNA from Bacteria. Current Protocols in
Molecular Biology. pp. 1129–1215. New York: John Wiley
and Sons.
Benson, C.A., Kaplan, J.E., Masur, H., Pau, A., Holmes, K.K.,
CDC and National Institutes of Health; Infectious Diseases
Society of America. (2004) Treating opportunistic infec-
tions among HIV-exposed and infected children: recom-
mendations from CDC, the National Institutes of Health,
and the Infectious Diseases Society of America. MMWR
Recomm Rep 53, 1–112.
Centers for Disease Control and Prevention (CDC). (2002)
Disseminated infection with simiae-avium group mycobac-
teria in persons with AIDS – Thailand and Malawi, 1997.
MMWR Morb Mortal Wkly Rep 51, 501–502.
Collier, L., Balows, A. and Sussman, M.. ed. (1998). Topley
and Wilson’s Microbiology’ and Microbial Infections, 9th
edn. London: Arnold.
Crump, J.A., Tanner, D.C., Mirrett, S., McKnight, C.M. and
Reller, L.B. (2003) Controlled comparison of BACTEC
13A, MYCO ⁄ F LYTIC, BacT ⁄ ALERT MB, and ISOLATOR
10 systems for detection of mycobacteremia. J Clin Micro-
biol 41, 1987–1990.
Damian, J.B., D’Costa, C., Vinette, E., Iqbal, J., Mannheimer,
S., Schicchi, J.S. and Nachman, S. (2004) A case of
Mycobacterium tuberculosis (MTB) and Mycobacterium
avium-complex (MAC) co-infection in an immunocompe-
tent host: a pathogen and a colonizer or two pathogens in
the same host? Chest 126, 985S–986S.
Gopinath, K., Manisankar, M., Kumar, S. and Singh, S. (2007)
Controlling multidrug-resistant tuberculosis in India.
Lancet 369, 741–742.
Gopinath, K., Kumar, S. and Singh, S. (2008) Prevalence of
mycobacteremia in Indian HIV-infected patients detected
by the MB ⁄ BacT automated culture system. Eur J Clin
Microbiol Infect Dis 27, 423–431.
von Gottberg, A., Sacks, L., Machala, S. and Blumberg, L.
(2001) Utility of blood cultures and incidence of
mycobacteremia in patients with suspected tuberculosis in
a South African Infectious disease hospital. Int J Tuberc
Lung Dis 5, 80–86.
434 Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
K. Gopinath and S. Singh
Hanna, B.A., Ebrahimzadeh, A., Elliott, L.B., Morgan, M.A.,
Novak, S.M., Rusch-Gerdes, S., Acio, M., Dunbar, D.F.
et al. (1999) Multicenter evaluation of the BACTEC
MGIT-960 system for recovery of mycobacteria. J Clin
Microbiol 37, 748–752.
Holt, J.G. ed. (1994) Bergey’s Manual of Determinative Bacterio
-logy, 9th edn. Baltimore: Williams and Wilkins.
Iralu, J.V., Sritharan, V.K., Pieciak, W.S., Wirth, D.F., Maguire,
J.H. and Barker, R.H. Jr (1993) Diagnosis of Mycobacte-
rium avium bacteremia by polymerase chain reaction.
J Clin Microbiol 31, 811–814.
Jeon, K., Koh, W.J., Kwon, O.J., Suh, G.Y., Chung, M.P., Kim,
H., Lee, N.Y., Park, Y.K. et al. (2005) Recovery rate of
NTM from AFB smear-positive sputum specimens at a
medical centre in South Korea. Int J Tuberc Lung Dis 9,
1046–1051.
Katoch, V.M. (2004) Diagnostic relevance of detection of
mycobacteremia. Indian J Med Res 119, xi–xii.
Kent, P.T. and Kubica, G.P. (1985). Public Health Laboratory.
A guide for the level III laboratory. Atlanta, GA: U.S.
Department of Health and Human Services publication no
(CDC) 86-8230. Centres for Disease Control and
Prevention.
Kocagoz, T., Yilmaz, E., Ozkara, S., Kocagoz, S., Hayran, M.,
Sachedeva, M. and Chambers, H.F. (1993) Detection of
Mycobacterium tuberculosis in sputum samples by polymer-
ase chain reaction using a simplified procedure. J Clin
Microbiol 31, 1435–1438.
Kox, L.F.F., van Leeuwen, J., Knijper, S., Jansen, H.M. and
Kolk, A.H. (1995) PCR assay based on DNA coding for
16S rRNA for detection and identification of
mycobacteria in clinical samples. J Clin Microbiol 33,
3225–3233.
Mahaisavariya, P., Chaiprasert, A., Khemngern, S., Manonukul,
J., Gengviniij, N., Ubol, P.N. and Pinitugsorn, S. (2003)
Nontuberculous mycobacterial skin infections: clinical and
bacteriological studies. J Med Assoc Thai 86, 52–60.
Montessori, V., Phillips, P., Montaner, J., Haley, L., Craib, K.,
Bessuille, E. and Black, W. (1996) Species distribution in
human immunodeficiency virus-related mycobacterial
infections: implications for selection of initial treatment.
Clin Infect Dis 22, 989–992.
Murcia-Aranguren, M.I., Gomez-Marin, J.E., Alvarado, F.S.,
Bustillo, J.G., de Mendivelson, E., Gomez, B., Leon, C.I.,
Triana, W.A. et al. (2001) Frequency of tuberculous and
non-tuberculous mycobacteria in HIV infected patients
from Bogota, Colombia. BMC Infect Dis 1, 21.
Mustafa, A.S., Abal, A.T. and Chugh, T.D. (1999) Detection of
Mycobacterium tuberculosis complex and non-tuberculous
mycobacteria by multiplex polymerase chain reactions.
East Mediterr Health J 5, 61–70.
Nunez, M., Miralles, E.S., Hilara, Y., Pintado, V., Harto, A.
and Ledo, A. (1997) Concurrent cytomegalovirus,
M. tuberculosis and M. avium-intracellulare cutaneous
infection in an HIV patient. J Dermatol 24, 401–404.
ª 2009 The Authors
K. Gopinath and S. Singh
Park, H., Jang, H., Kim, C., Chung, B., Chang, C.L., Park, S.K.
and Song, S. (2000) Detection and identification of myco-
bacteria by amplification of the internal transcribed spacer
regions with genus and species-specific PCR primers.
J Clin Microbiol 38, 4080–4085.
Park, H., Kim, C., Park, K.H. and Chang, C.L. (2006) Develop-
ment and evaluation of triplex PCR for direct detection of
mycobacteria in respiratory specimens. J Appl Microbiol
100, 161–167.
Pinsky, B.A. and Banaei, N. (2008) Multiplex real-time PCR
assay for rapid identification of Mycobacterium tuberculosis
complex members to the species level. J Clin Microbiol 46,
2241–2246.
Powderly, W.G. (1996) Prophylaxis for HIV-related infections:
a work in progress. Ann Intern Med 124, 342–344.
Reves, R., Reichman, L.B., Simone, P.M., Starke, J.R. and
Vernon, A.A. (2003) American Thoracic Society ⁄ Centers
for Disease Control and Prevention ⁄ Infectious Diseases
Society of America: treatment of tuberculosis. Am J Respir
Crit Care Med 167, 603–662.
Rodrigues Vde, F., Queiroz Mello, F.C., Ribeiro, M.O., Fonse-
ca, L., Kritski, A.L., Rossetti, M.L. and Zaha, A. (2002)
Detection of Mycobacterium avium in blood samples of
patients with AIDS by using PCR. J Clin Microbiol 40,
2297–2299.
Rosas, R.C., Salomao, R., da Matta, D.A., Lopes, H.V., Pigna-
tari, A.C. and Colombo, A.L. (2003) Bloodstream infec-
tions in late-stage acquired immunodeficiency syndrome
patients evaluated by a lysis centrifugation system. Mem
Inst Oswaldo Cruz 98, 529–532.
Ruiz, M., Rodriguez, J.C., Escribano, I., Garcia-Martinez, J.,
Rodriguez-Valera, F. and Royo, G. (2001) Application of
molecular biology techniques to the diagnosis of nontuber-
culous mycobacterial infections. APMIS 109, 857–864.
Runyon, E.H. (1974) Ten mycobacterial pathogens. Tubercle
55, 235–240.
Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular
Cloning: A Laboratory Manual, NY: Cold Spring Harbor
Laboratory Press.
Sankar, M.M., Gopinath, K., Singla, R. and Singh, S. (2008)
In-vitro antimycobacterial drug susceptibility testing of
ª 2009 The Authors
Journal compilation ª 2009 The Society for Applied Microbiology, Journal of Applied Microbiology 107 (2009) 425–435
Multiplex PCR assay for M. tuberculosis and M. avium complex
non-tubercular mycobacteria by tetrazolium microplate
assay. Ann Clin Microbiol Antimicrob 7, 15.
Shah, D.H., Verma, R., Bakshi, C.S. and Singh, R.K. (2002) A
multiplex-PCR for the differentiation of Mycobacterium
bovis and Mycobacterium tuberculosis. FEMS Microbiol Lett
214, 39–43.
Singh, S. (2008) Scaling up anti-mycobacterial drug suscepti-
bility testing services in India: it is high time. Indian J Med
Microbiol 26, 209–211.
Singh, K.K., Muralidhar, M., Kumar, A., Chattopadhyaya,
T.K., Kapila, K., Singh, M.K., Sharma, S.K., Jain, N.K.
et al. (2000) Comparison of in house polymerase chain
reaction with conventional techniques for the detection
of Mycobacterium tuberculosis DNA in granulomatous
lymphadenopathy. J Clin Pathol 53, 355–361.
Singh, S., Gopinath, K., Shahdad, S., Kaur, M., Singh, B. and
Sharma, P. (2007) Nontuberculous mycobacterial infec-
tions in Indian AIDS patients detected by a novel set of
ESAT-6 polymerase chain reaction primers. Jpn J Infect Dis
60, 14–18.
Steinbrook, R. (2007) Tuberculosis and HIV in India. N Engl J
Med 356, 1198–1199.
Tanaka, I.I., Anno, I.S., Leite, S.R., Cooksey, R.C. and Leite,
C.Q. (2003) Comparison of a multiplex-PCR assay with
mycolic acids analysis and conventional methods for the
identification of mycobacteria. Microbiol Immunol 47,
307–312.
Telenti, A., Marchesi, F., Balz, M., Bally, F., Bottger, E.C. and
Bodmer, T. (1993) Rapid identification of mycobacteria to
the species level by polymerase chain reaction and enzyme
analysis. J Clin Microbiol 31, 175–178.
Waddell, R.D., Lishimpi, K., von Reyn, C.F., Chintu, C.,
Baboo, K.S., Kreiswirth, B., Talbot, E.A., Karagas, M.R.
et al. (2001) Bacteremia due to Mycobacterium tuberculosis
or M. bovis, Bacille Calmette-Guerin (BCG) among
HIV- positive children and adults in Zambia. AIDS 15,
55–60.
World Health Organization (WHO) report. (2008) Global
Tuberculosis Control: Surveillance, Planning, Financing.
Geneva: World Health Organization, 2008.
(WHO ⁄ HTM ⁄ TB ⁄ 2008.393).
435