| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Biotechnology, All India Institute of Medical Sciences, Ansari Nagar, New Delhi 110029, India
2 Lala Ram Sarup Institute of TB and Respiratory Diseases, New Delhi 110030, India
Correspondence
Jaya Sivaswami Tyagi
jstyagi{at}aiims.ac.in
Received 2 March 2007
Accepted 21 June 2007
Abbreviations: FAM, 6-carboxyfluorescein; ISBCN, IS6110 beacon; TB, tuberculosis; TO, target oligonucleotide; USP, universal sample processing.
Present address: Division of Infectious Diseases, Department of Medicine and the Ruy V. Lourenço Center for the Study of Emerging and Reemerging Pathogens, New Jersey Medical School, University of Medicine and Dentistry of New Jersey, Newark, NJ 07103, USA. 
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
2 million casualties annually worldwide (WHO, 2005). Conventional approaches adopted for the diagnosis of TB include medical history, tuberculin skin test, chest X-rays and bacteriological examination. The poor sensitivity of conventional smear microscopy and the delay in obtaining culture results prevent the early diagnosis of TB. We demonstrated the improved detection of direct smear-negative cases by universal sample processing (USP) smear microscopy (Chakravorty & Tyagi, 2005; Chakravorty et al., 2005a). However, a more sensitive test is required to detect samples with a low bacterial load. Since the advent of PCR, there has been an explosion in its use for TB diagnosis owing to its speed and sensitivity (Eisenach, 1998). However, these advantages are partially offset by economic and technical factors and simplification is required before PCR can be widely employed. In addition, PCR is often associated with some degree of false positivity (Beige et al., 1995). However, a positive smear result enhances the diagnostic accuracy of a positive PCR in the absence of culture (Haldar et al., 2005). PCR products are most frequently detected by electrophoresis, which is cumbersome, time-consuming and carries a risk of cross-contamination. Alternatively, PCR products can be detected by molecular beacons that fluoresce upon hybridization to complementary target DNA thereby enabling rapid detection of pathogens. However, they have been mostly used in real-time formats (Tyagi & Kramer, 1996; Tsourkas & Bao, 2003). We have previously used devR PCR in gel assays for diagnosing pulmonary and extra-pulmonary TB (Singh et al., 1999, 2000; Chakravorty et al., 2005b, 2006). Against this background, we developed an asymmetric devR PCR assay and used two 6-carboxyfluorescein (FAM)-labelled molecular beacons for end-point detection of Mycobacterium tuberculosis DNA in sputum. The devR assay was compared with the molecular-beacon-based IS6110 PCR described previously (Piatek et al., 1998). USP smear microscopy and PCR results were analysed using culture as the ‘gold’ standard. The tests were found to be robust and the judicious use of PCR along with USP smears should find use in the rapid detection of smear-negative samples in high-TB-incidence settings.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Specimen processing. Aliquots of USP-processed samples were analysed by smear microscopy, culture and PCR as described previously (Chakravorty & Tyagi, 2005; Chakravorty et al., 2005a). Briefly, 1.5–2 vols USP solution [6 M guanidinium hydrochloride, 50 mM Tris/Cl (pH 7.5), 25 mM EDTA, 0.5 % Sarcosyl, 0.1 M ß-mercaptoethanol (all from Sigma)] was added to each sputum sample. After incubation for 5–10 min at room temperature (or 15 min at 37 °C if the sample was extremely viscous), 10–15 ml sterile water was added and the sample was centrifuged at 5000–6000 g for 15–20 min at room temperature. The supernatant was discarded carefully and the sediment was washed once more with 2 ml USP solution and recentrifuged as before. The pellet was washed thoroughly with 10 ml water, resuspended in 500 µl 0.05 % Tween 80 and used for smear microscopy (20 %), culture (40 %) and DNA isolation and PCR (40 %).
The slides were subjected to Ziehl–Neelsen staining and graded by two independent readers as described previously (Akhtar et al., 2000). USP-processed deposits were inoculated in 7H9 liquid media containing albumin dextrose complex and PANTA (polymyxin B, amphotericin B, nalidixic acid, trimethoprim and azlocillin) supplement (Becton Dickinson) to maximize the isolation of bacteria from smear-negative samples. The tubes were incubated at 37 °C for up to 4 weeks. USP has been shown previously to be compatible with solid and liquid culture media and was no more toxic than N-acetyl-L-cysteine (NALC)–NaOH (Chakravorty & Tyagi, 2005). The cultures were confirmed by M. tuberculosis complex-specific devR PCR (Chakravorty et al., 2006). A definite TB diagnosis was made with a positive culture from sputum. DNA was isolated from the remaining 200 µl portion of the sample as described previously (Chakravorty & Tyagi, 2005). Briefly, 120 µl lysis solution containing 10 % Chelex-100 (Bio-Rad Laboratories), 0.3 % Tween 20 and 0.03 % Triton X-100 (both from Sigma) was added to the processed sediment, mixed well and incubated at 90 °C for 40 min. The lysate was centrifuged briefly at 12 000 r.p.m. for 10 min and used in PCR.
Design and characterization of the molecular beacons.
MB 1 and MB 2 molecular beacon oligonucleotide probes (Metabion) mapped within the M. tuberculosis devR gene (Table 1
). The annealing temperature for PCR was determined from the denaturation profiles of the molecular beacons as described by Tyagi & Kramer (1996). Briefly, the change in fluorescence of MB 1, MB 2 or the IS6110 beacon (ISBCN) either present alone or with the corresponding complementary target oligonucleotide (TO) over a temperature range was measured in the I-Cycler (Bio-Rad). For this, two tubes containing 0.3 µM each of MB 1 and MB 2 or ISBCN in 50 µl buffer (3.5 mM MgCl2 and 10 mM Tris/HCl, pH 8.0) and 0.3 µM each of MB 1 and MB 2 plus 0.6 µM devR TO 1 and TO 2 or IS6110 TO were subjected to a decrease in temperature from 96 °C to 41 °C in 1 °C steps and fluorescence was measured during a 30 s hold at each temperature. The melting curve of the molecular beacons was obtained from a –dF/dT plot. The signal to noise ratio of MB 1, MB 2 and MB 1+MB 2 was determined by measuring the fluorescence intensity of the molecular beacon in the presence or absence of a twofold excess of its corresponding TO at 43 °C (Fig. 1).
|
|
. Three detection formats were employed: molecular-beacon-based end-point detection using fluorimetric and visual methods (methods 1 and 2, respectively) and gel detection using ethidium bromide (method 3). For devR assays (methods 1 and 2), the reactions contained 0.1 µM devRf4 primer, 0.5 µM devRr3 primer (Microsynth), 1x PCR buffer (100 mM Tris/HCl, pH 8.8, 500 mM KCl, 0.8 % Nonidet P40), 4 mM MgCl2, 0.25 mM dNTPs, 2.5 U Taq DNA polymerase (GeneTaq; MBI Fermentas), 10 µl specimen DNA and 0.3 µM and 0.6 µM each of MB 1 and MB 2 for methods 1 and 2, respectively. For the devR assay (method 3), reactions contained 0.5 µM each of primers devRf4 and devRr3, 1x PCR buffer, 1.5 mM MgCl2, 0.2 mM dNTPs, 1 U Taq DNA polymerase and 10 µl specimen DNA. The thermal cycling parameters were 10 min at 94 °C, 45 cycles each of 1 min at 94 °C, 1 min at 43 °C and 30 s at 72 °C and a final extension of 7 min at 72 °C. For methods 1 and 2, this was followed by a final denaturation (94 °C for 5 min) and annealing (43 °C for 10 min).
For IS6110 assays (methods 1 and 2), the reactions contained 0.1 µM IS6110 forward primer, 0.5 µM IS6110 reverse primer (herein referred to as ISBCNF and ISBCNR primers), 1x PCR buffer, 4 mM MgCl2, 0.25 mM dNTPs, 2.5 U Taq DNA polymerase, 10 µl specimen DNA and 0.3 µM and 0.6 µM ISBCN for methods 1 and 2, respectively. IS6110 assay (method 3) reactions contained 0.5 µM each of primers ISBCNF and ISBCNR (Piatek et al., 1998), 1x PCR buffer, 1.5 mM MgCl2, 0.2 mM dNTPs, 1 U Taq DNA polymerase and 10 µl specimen DNA. The thermal cycling parameters were 10 min at 94 °C, 45 cycles each of 1 min at 94 °C, 1 min at 60 °C and 30 s at 72 °C, and a final extension of 7 min at 72 °C. For methods 1 and 2, this was followed by a final denaturation (94 °C for 5 min) and annealing (60 °C for 10 min).
Detection of amplified products. In method 1 (fluorimetric format), the reaction contents were transferred into wells of a 96-well plate containing 150 µl 20 mM Tris/Cl (pH 8.0) and 1 mM MgCl2 and the fluorescence was measured using 491 nm excitation and 515 nm emission in a spectrofluorimeter (Spectra MAX Gemini XS; Molecular Devices) at 43 °C. In method 2 (visual format), the tubes were placed over a blue light source (Dark Reader; Clare Chemical Research) for detection of fluorescence by visual inspection by two persons. In method 3 (gel format), the PCR products were detected by ethidium bromide staining after electrophoresis on a 2.3 % agarose gel and illumination with UV light.
Performance of the devR PCR assay. The limit of detection was evaluated on serial dilutions of purified M. tuberculosis DNA. PCR assays were set up as described above in fluorimetric, visual and gel formats. The specificity of the devR PCR assay was assessed using DNA in crude lysates prepared from scrapings of mycobacterial cultures grown on Löwenstein–Jensen slants by heating in 0.1 % Triton X-100 for 40 min at 90 °C.
Statistical analysis. The diagnostic performance of PCR was evaluated using culture as the ‘gold’ standard. Test results were classified as true positives (Tp), true negatives (Tn), false positives (Fp) and false negatives (Fn). Sensitivity was calculated as [Tp/(Tp +Fn)]x100 and specificity as [Tn/(Tn+Fp)]x100. The efficiency was calculated as [Tp+Tn/N]x100, where N=148. The statistical parameter Z factor was used to evaluate the quality of the fluorimetric assay (Zhang et al., 1999). The significance of the difference between direct and USP smear results was calculated using McNemar's test (Altman, 1991). The performance of various detection techniques was assessed for agreement between them by calculating the kappa agreement value (Altman, 1991).
|
RESULTS AND DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
Improved detection by USP smear microscopy
A sensitive smear test is an extremely handy and low-cost tool for diagnosing TB in endemic disease settings. The USP smear technique was expected to detect additional positive samples over and above those detected by direct smear examination. One hundred and twenty of 148 samples were positive by liquid culture. Eighty-five of 120 culture-positive samples were USP smear-positive in contrast to only 24 that were positive by direct smear examination, and not a single direct smear-positive sample was missed. USP smears could detect 61 more samples as positive than the direct smear technique. Thus direct smear microscopy had a sensitivity of only 20 % versus USP smear, which had a sensitivity of 71 % (Table 2; P <0.0001). These samples were predominantly of scanty grade and their detection is consistent with the enhanced performance of USP smear microscopy (Chakravorty et al., 2005a). We cannot explain why two USP smear-positive samples failed to grow in culture and yet were positive according to all six PCR assays.
|
). This was attributed to the preferential accumulation of the target DNA strand during asymmetric amplification.
Performance of PCR assays
The devR PCR assay was judged to be quite specific for M. tuberculosis complex organisms. DNA from both M. tuberculosis and Mycobacterium bovis was detected and in silico analysis indicated that devR is completely conserved in Mycobacterium africanum and Mycobacterium microti (http://www.sanger.ac.uk/sequencing/Mycobacterium/africanum/; http://www.sanger.ac.uk/Projects/M_microti/). The assay could not distinguish between the M. tuberculosis complex and Mycobacterium kansasii DNA (not shown). However, M. tuberculosis can be readily verified by IS6110 PCR, which does not detect M. kansasii (Hellyer et al., 1996). It should be mentioned that M. tuberculosis lacking IS6110 and M. kansasii cannot be discriminated by these two assays since they are both positive for the devR assay and negative for IS6110 PCR. However, the detection of M. kansasii by the devR assay was not considered to be a serious concern as the yield of mycobacteria other than tubercle (MOTT) bacilli from clinically suspected cases is negligible in India (Katoch, 2004). In our own experience too, MOTT bacilli were not isolated while handling over 2000 cultures.
To determine the limit of detection of M. tuberculosis DNA, serial DNA dilutions ranging from 500 ng to 10 fg were added to devR PCR reactions. By the gel assay, up to 100 fg DNA was detected. The fluorimetric and the visual methods were somewhat more sensitive and could detect 25 fg DNA (Fig. 2). When MB 1 and MB 2 were used together, the devR assay was as sensitive as the multicopy IS6110 assay (Fig. 2a). The Z factor was used to evaluate the fluorimetric assays (Zhang et al., 1999); values of 0.65 and 0.7 were obtained for devR and IS6110 assays, respectively, indicating that these assays were of good quality.
|
). In this study, although USP smears had a definite utility, in that they were positive in 85 culture-positive samples versus 24 that were positive by direct smear examination, 35 culture-positive samples were still missed by the USP smear test (Table 2). All these 35 samples were detected by at least three PCR assays, indicating that PCR has a role in the rapid diagnosis of smear-negative TB.
Using culture as the ‘gold’ standard, devR assay sensitivity was 92.5 %, 86.7 % and 90.8 % for the fluorimetric, visual and gel detection formats, respectively. The cutoff was apparently set higher for a positive for the fluorimetric assay than for the visual assay, yet the specificities of the devR fluorimetric and visual assays were equal (92.9 %). Perhaps this highlights that the very small numbers of false-positive PCR results did not permit a significant comparison of the specificities of these two assays. The sensitivity of the IS6110 assay, in particular that of the fluorescence-based formats, was comparable at 91.7 %, 88.3 % and 85 % for the fluorimetric, visual and gel detection formats, respectively. The efficiency of all the assays ranged between 85.8 % and 92.6 % (Table 2).
The various detection formats were compared using the kappa agreement test whereby values between 0.6 and 0.8 indicate good agreement between any two tests (Altman, 1991). The devR visual and fluorimetric assays had kappa values of 0.72 and 0.75, respectively, in comparison to the gel assay. Likewise, the IS6110 visual and fluorimetric assays had kappa values of 0.79 and 0.74 respectively, in comparison to the gel assay. The fluorescence-based assays also had good agreement amongst themselves: a value of 0.77 for both devR and IS6110 assays. On the basis of the kappa agreement test results (P <0.0001), we conclude that end-point detection with fluorescent probes performed well in comparison to the conventional gel assay; in fact the fluorimetric format surpassed the gel assay in its performance.
Among the 120 culture-positive samples that were analysed, 35 samples were missed by one or more PCR assays (false-negative). Of these, 29 samples were either negative or had low bacterial loads as determined by USP smears. PCR false-negative samples included five cases that were missed by all six PCR assays. Regarding false-positive PCR results (7 %), two samples were positive in all six PCR assays, one sample was positive in all three IS6110 assays and the devR gel assay, two samples were positive in the devR gel assay and IS6110 visual assay and one sample was positive in only the IS6110 fluorimetric assay. One culture-positive sample was negative by IS6110 PCR but detected by devR PCR. The absence of IS6110 sequences in this particular M. tuberculosis isolate was established by PCR. The isolation of M. tuberculosis strains from India that lack IS6110 has been reported (Narayanan et al., 2001; Radhakrishnan et al., 2001) and our findings underline the utility of devR PCR in detecting IS6110-negative organisms.
Amongst the beacon-based assays, devR PCR was as sensitive as the multicopy target IS6110 PCR without compromising on specificity. This is ascribed to the use of two beacons in the devR asymmetric assay versus one beacon in the IS6110 assay. Molecular beacons have been used for determining drug resistance among cultured M. tuberculosis isolates (Piatek et al., 1998) and also for M. tuberculosis detection in sputum (Li et al., 2000). The majority of assays for pathogen detection use molecular beacons in automated real-time formats. These technologies would be out of reach in settings where they are most required. In this study, DNA amplification was conveniently carried out in a regular thermal cycler and the products were detected at the end point using molecular beacons in either the visual or fluorimetric formats. The advantages of the closed-tube visual assay are quite obvious as it avoids handling of tubes after PCR and thus minimizes cross-contamination between samples. However, the efficiency of the visual assays was somewhat lower than that of the fluorimetric assays.
The importance of detecting smear-negative TB cannot be overemphasized; up to half of the new TB cases are reportedly smear-negative (Bennedsen et al., 1996), which leads to delays in the institution of therapy, increased transmission from undetected infectious cases and associated increased medical costs. In this study, USP smear microscopy detected an additional 51 % of the culture-positive samples compared to direct smear (Table 2). PCR was able to detect all culture-positive samples that were missed by the USP smear. The results of various combinations of PCR assays were analysed and the highest efficiency (94.6 %) was noted when the devR visual and fluorimetric assays were considered together (Table 2).
In conclusion, we have demonstrated the feasibility of using molecular beacons in simplified formats for TB diagnosis. To the best of our knowledge, this is the first report that uses two molecular beacons in combination with asymmetric PCR to increase the sensitivity of the end-point fluorescence assay. We believe that the assay performance can be further improved by use of dUTP and uracil glycosylase, the use of molecular beacons with a higher fluorescence ratio in the presence of the target sequence and by digital capture of the fluorescence. When used in conjunction with the low-cost USP smear test, the beacon-based assays promise to accurately and rapidly diagnose smear-negative TB. However, with a specificity of 93 % and a false-positive rate of 7 %, PCR results should be interpreted in the context of clinical presentation, especially in low-incidence areas.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  P. Kumar, K. Nath, B. Rath, M. K. Sen, P. Vishalakshi, D. S. Chauhan, V. M. Katoch, S. Singh, S. Tyagi, V. Sreenivas, et al. Visual Format for Detection of Mycobacterium tuberculosis and M. bovis in Clinical Samples Using Molecular Beacons J. Mol. Diagn., September 1, 2009; 11(5): 430 - 438. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Haldar, N. Sharma, V. K. Gupta, and J. S. Tyagi Efficient diagnosis of tuberculous meningitis by detection of Mycobacterium tuberculosis DNA in cerebrospinal fluid filtrates using PCR J. Med. Microbiol., May 1, 2009; 58(5): 616 - 624. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Chakravorty, J. S. Tyagi, A. Cattamanchi, J. L. Davis, W. Worodria, S. Yoo, and L. Huang Performance of Modified Universal Sample Processing Method in a Field Study in Uganda J. Clin. Microbiol., February 1, 2009; 47(2): 517 - 518. [Full Text] [PDF]
|
|
|
|
 |
|
 L. Mosi, H. Williamson, J. R. Wallace, R. W. Merritt, and P. L. C. Small Persistent Association of Mycobacterium ulcerans with West African Predaceous Insects of the Family Belostomatidae Appl. Envir. Microbiol., November 15, 2008; 74(22): 7036 - 7042. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Papaparaskevas, D. P. Houhoula, A. Siatelis, and A. Tsakris Molecular-Beacon-Based Real-Time PCR for Detection and Quantification of Mycobacterium tuberculosis DNA in Clinical Samples J. Clin. Microbiol., September 1, 2008; 46(9): 3177 - 3178. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
