Evaluation of a direct colorimetric assay for rapid detection of rifampicin resistant Mycobacterium tuberculosis

Background: Susceptibility test is a useful tool for selecting and modifying appropriate treatment for tuberculosis. Objective: To evaluate a direct colorimetric assay based on 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyl tetrazolium bromide (MTT) for a rapid detection of rifampicin resistance. Methods: Sputum was inoculated directly into 7H9 broth supplemented with PANTA (antibiotic mixture) and oleic acid albumin dextrose catalase (OADC) to which MTT was added after 1-3 weeks. Bacterial growth is indicated by the formation of a colored formazan product, which can be measured with a spectrophotometer. The assay was read on the first, second or third week of incubation depending on when a measurable formazan was observed in the drug free control tube. The results were compared against the standard indirect modified proportion method of susceptibility testing on 7H10 media. Single sputum samples from 78 re-treatment cases were tested. Results: The MTT assay was found to be 90.5% sensitive and 100% specific. In more than two-thirds of the samples, results were already available by the end of week two. Conclusion: The direct colorimetric assay is a simple rapid test for rifampicin susceptibility testing. It significantly shortens the time required to obtain a drug susceptibility test result and could be useful to screen for MDR-TB. [Ethiop.J.Health Dev. 2005;19(1):51-54] Introduction Early detection of drug resistance is one of the essential steps in the management of tuberculosis. The effectiveness of a standard anti-TB treatment regimen correlates well with the in-vitro drug susceptibility pattern of the infecting tubercle bacilli. The results of drug susceptibility tests help select a proper treatment regimen or modify treatment regimen for a better management of patients and for surveillance and timely control of the spread of drugresistant TB in the community. The drug susceptibility testing of M. tuberculosis can be done either indirectly using cultured mycobacterial colonies isolated from clinical specimens or directly on the clinical samples (1). Conventional susceptibility testing is done after bacterial isolation using agar-based or egg-based media and takes seven to twelve weeks. It is time consuming and laborious. The direct susceptibility test with conventional methods takes three to four weeks. The time required for the susceptibility test is significantly reduced when relatively new methods like a standard radiometric method (BACTEC) are used. In the BACTEC method, the direct and indirect tests need an average time of 11 and 18 days respectively (2). However, BACTEC is very expensive for routine use in resource-poor settings. We had previously developed an affordable and reliable colorimetric assay using a dye called 3-[4, 5dimethylthiazol-2-yl]-2, 5diphenyl tetrazolium bromide (MTT) for a rapid detection of rifampicin resistance in M. tuberculosis isolates using the BACTEC method as a reference (3). A colorimetric assay using MTT was first introduced by Mossman as a quantitative measure of mammalian cell survival and proliferation (4). MTT is a yellow tetrazolium salt, which is converted into a blue formazan by dehydrogenases of a live cell. The assay is based on the principle that the amount of formazan produced is directly proportional to the number of live cells (4). The formazan can then be measured by a spectrophotometer after solubilization (5). We previously reported on the indirect MTT assay in which susceptibility testing is done on a mycobacterial isolate. The MTT-based antimycobacterial drug susceptibility test requires only three to seven days (3,6). However, the time required to obtain a primary isolate to perform an indirect susceptibility test is relatively longer and may limit the wider application of new rapid methods. In this study, we evaluated a direct MTT assay for rapid detection of rifampicin-resistant M. tuberculosis in sputum samples collected under program conditions. Rapid assay for rifampicin resistance 52 ______________________________________________________________________________________ Ethiop.J.Health Dev. 2005;19(1) Methods Specimens Sputum samples were collected from 100 consecutive, consenting re-treatment cases (patients who became smear positive again after a previous treatment for tuberculosis) visiting St Peter TB Specialized Hospital in Addis Ababa between December 2001 and October 2002. Sputum specimens were collected in screw-capped clean plastic sputum cups and kept at 4 C until transported within four hours of collection to the TB laboratory at the Armauer Hansen Research Institute (AHRI) where they were processed immediately. Culture The sputum was digested and decontaminated using Petroff’s method (7). The pellets were resuspended in 2.5 ml PBS. An aliquot of 100μl of the sample was then cultured into two tubes containing Lowenstein Jensen (LJ) media for primary isolation. The tubes were incubated at 37C and examined weekly for bacterial growth until the eighth week. An aliquot of 500μl from the remaining resuspended sample was inoculated into each of four tubes containing 3ml Middlebrook 7H9 broth supplemented with Oleic acid Albumin Dextrose Catalase (OADC) enrichment (Becton-Dickinson) and antibiotics mixture (PANTA) (Becton-Dickinson). PANTA is a standard antibiotic cocktail that contains Polymyxin B, Amphotericin B, Nalidixic acid, Trimethoprim and Azlocillin. One of the four tubes contained rifampicin at 2 mg/l final concentration (Sigma) while the rest were drug free controls. They were then incubated at 37C and the MTT assay was done on the day after a visible formazan product was observed in the control medium. However, when no formazan was seen on the first and second week of incubation, the assay would be performed in the third week. MTT assay The MTT assay was done as described previously (8). Briefly, 300μl of 5 mg/ml MTT solution was added into each tube. The tubes were vortexed and incubated for 4 hrs at 37C. The formazan produced was dissolved with solubilization buffer (20% sodium dodecyl sulfate) in a 50% aqueous solution of dimethyl formamide Optical density (OD570) was then measured at 570nm (Novaspec II photometer, Pharmacia Biotech Ltd, UK), using a tube containing 7H9 broth, PBS, MTT and solubilization buffer as a reference. Relative optical density unit (RODU) values were calculated by dividing the OD of the drug containing tubes with the OD of drug free control. The correlation of the MTT assay at OD570 in 7H9 broth to colony forming units on standard 7H10 agar media was determined in initial experiments. Based on the distinctive difference between positive and negative samples observed, appropriate cut-off values were determined. No CFU was observed for broth cultures with < 0.1 OD570. Therefore, the MTT results were defined as interpretable if OD of a control tube was ≥ 0.1. A strain was defined as rifampicin-susceptible when the relative optical density unit (RODU) value was below 0.2 and resistant when it was above 0.5. RODU between 0.2 and 0.5 was considered indicative of borderline resistance. Among the four tubes, two control tubes were used to check formazan formation on the first and second weeks. The MTT assay on the rifampicin containing tube and one control was performed a day after MTT assay on a control tube showed visible formazan formation. When the MTT assay on control tubes did not show formazan formation in the first two weeks, the remaining tubes were used on the third week. A test for bacterial contamination was performed for each sample inoculated into 7H9 broth by inoculation of a loopful of the corresponding broth on nutrient agar and incubating it at 37C for 24 hours before performing the MTT assay on the particular tube. Mycobacterial isolates were identified as M. tuberculosis using standard biochemical tests (9). Results Of 100 processed samples, 78 gave interpretable results (Table 1). The rest were excluded because of contamination of media (n=4), lack of growth in drugfree control medium of MTT assay (OD 570 <0.1) (n=7) and lack of growth on LJ medium (n=11). From the 11 isolates that were culture negative five grew on 7H9 broth and had MTT results. Among the five isolates, four were susceptible to rifampicin and one was resistant. Three of the culture negative isolates came out with uninterpretable MTT result. Among the seven isolates that showed uninterpretable MTT result only two did not show growth on LJ while the others had growth and most were MDR. A relative value, RODU was employed to control for possible strain differences in the ability to reduce MTT (8) and define resistance and susceptibility. The mean (+ standard deviation, SD) RODU value of resistant strains was 1.08±0.5 whereas the mean (+ SD) RODU value of susceptible strains was 0.04±0.06. Except for one susceptible isolate with an RODU of 0.3, all others had an RODU value < 0.2. This result matched with the RODU values for resistant and susceptible strains in the standardized indirect MTT assay for rifampicin reported earlier (3,6,8). Rapid assay for rifampicin resistance 53 ______________________________________________________________________________________ Ethiop.J.Health Dev. 2005;19(1) Table 1: Comparison of the sensitivity test results using direct MTT assay and the standard proportional method Standard proportional method MTT assay Susceptible Intermediate Resistant Total Susceptible 56 0 2 58 Intermediate 1 0 0 1 Resistant 0 0 19 19