Indian Journal of Medical Microbiology Medknow Publications on behalf of Indian Association of Medical Microbiology ISSN: 0255-0857 EISSN: 1998-3646 Vol. 25, Num. 2, 2007, pp. 115-120

Indian Journal of Medical Microbiology, Vol. 25, No. 2, April-June, 2007, pp. 115-120

Original Article

Drug resistance profile of human Mycobacterium avium complex strains from India

D Venugopal, S Kumar, M Isa, M Bose

Department of Microbiology, VP Chest Institute, University of Delhi, New Delhi - 110 007

Correspondence Address: Department of Microbiology, VP Chest Institute, University of Delhi, New Delhi - 110 007, India (email: mridulabose@hotmail.com)

Date of Submission: 19-Mar-2006
Date of Acceptance: 16-Nov-2006

Code Number: mb07034

Abstract

Keywords: Broth microdilution, Mycobacterium avium complex, minimum inhibitory concentration

Infections caused by nontuberculous mycobacteria (NTM ) have gained increasing importance in the recent years. Among NTMs, infections caused by M. avium complex (MAC, M. avium and M. intracellulare ) are widely prevalent. ^[1] MAC causes pulmonary disease indistinguishable from tuberculosis in humans. These are major opportunistic pathogens causing disseminated pulmonary infection among immunocompromized hosts. The therapeutic efficacies of antituberculosis drugs against such infections are still to be defined conclusively.

The main lacuna in studying the therapeutic efficacy of antimicrobial agents or the emergence of drug resistance among MAC organisms is lack of a standard technique to study drug susceptibility of NTM in the laboratory. Mycobacterial susceptibility testing commonly uses the proportion method that was originally developed for M. tuberculosis. This method often yields misleading results when applied for NTM including MAC. A wide variation in the results of assays for the susceptibility to antimicrobial agents has been observed for the MAC isolates. Therefore, testing against a single drug concentration, as done for M. tuberculosis , is not suitable for MAC.

The BACTEC broth susceptibility is a rapid method that involves measurement of the production of radio labelled CO ₂ by mycobacteria growing in broth containing a radio labelled fatty acid as a substrate. ^{[2],[3],[4],[5]} The advantages of this system are that mycobacteria grow more rapidly in broth and the system allows for early detection of growth. Thus, the susceptibility results are obtained early. However, the BACTEC system is expensive and testing of more than one drug at multiple concentrations is not cost effective.

Although a fully quantitative method may be the best way to determine the degree of resistance of a NTM strain, minimum inhibitory concentration (MIC) determination has been found to be suitable for the purpose. A broth microdilution MIC method was described for rapidly growing mycobacteria. ^[6] Wallace et al successfully tried this method for slow growing mycobacteria by using 7H9 broth. ^[7] This is a rapid and simple method that does not require any expensive or dedicated equipment to perform the test or read the results. Such a method should be very useful for any laboratory interested in undertaking drug susceptibility of NTM.

We applied the rapid broth microdilution technique to a collection of MAC strains. These were isolated from sputum of patients suffering from chronic pulmonary mycobacteriosis. A number of antituberculosis drugs at multiple concentrations were used to determine MIC for these isolates.

Materials and Methods

Test cultures

We tested 49 clinical isolates of MAC, isolated from human subjects attending the outpatient department of V. P. Chest Institute, University of Delhi, New Delhi, India. The strains were identified as MAC by biochemical methods ^[8] and molecular markers, viz,DT1/DT6 sequence ^[9] , IS1245 ^[10] , mig gene. ^[11] The cultures were maintained on Lowenstein Jensen (LJ) slants.

Antimicrobial agents and concentrations

The methods followed were according to Wallace et al ^[7] and Telles and Yates. ^[12] The antimicrobial agents tested and the ranges of concentration used are listed in [Table - 1]. All drugs were kept as 1% stock suspensions (w/v) in distilled water, except for rifampicin (RIF) and clofazimine (CLOF), which were dissolved in dimethylformamide and stored at -20°C.

Preparation of Mycobacterial inoculum for broth microdilution

Once there was sufficient growth on the LJ slant, the isolates were subcultured in Middlebrook 7H9 broth (DIFCO) supplemented with ADC enrichment (DIFCO) and incubated at 37°C for seven days. Before use, the cultures were vortexed with glass beads for 20-30 second and then left for three to five seconds to allow for settling of heavy particles. The suspension of organisms was matched to the optical density of no. 1 McFarland standard.

Preparation of broth microdilution plates

Sterile 96-well U bottomed, microtitre plates (Tarsons, India) were used. The stock suspensions of the drugs were prepared in Middlebrook 7H9 medium and 0.2 mL of the highest concentration of each drug [Table - 1] was added to the respective wells of the first row of the plates except the first and the last well. Two-fold serial dilutions of the drugs were prepared directly in the microtitre plates [Table - 2]. The prepared plates were either used immediately or were sealed in plastic bags and stored at -20°C and used within one week of preparation. Before inoculation, the plates were removed from the freezer and brought to room temperature. The plates were inoculated with 5mL of inoculum in each well except the blanks (last well of column 1,11 and 12). The plates were placed in a plastic bag each and sealed. The plastic bags were essential to prevent evaporation in the wells and biologic containment.

Reading

The plates were read by two methods:

1. Visually by comparison with the drug free controls. Readings were taken on day seven and day 14, without disturbing the plates, by looking for macroscopic growth with an indirect light source. The wells were read as + (growth equal to the control), ± (growth reduced), - (no growth).
2. With ELISA plate reader at a wavelength of 620 nm. The percentage reduction was calculated by comparison of the average readings of the control wells (C) containing no drug, with the drug containing wells (D) after the background reading value (from wells containing media but no organisms) (B) has been removed i.e, percentage reduction 100- (D-B) x 100/(C-B).

MIC definition

By visual reading the MIC was defined as the lowest concentration of the drug that exhibited no growth. Plate reader defined it as the lowest dilution giving a 99% reduction in growth.

Drug susceptibility by proportion method

Proportion method is one of the widely accepted and commonly used methods for testing susceptibility to antituberculosis drugs . Quadrant Petri dishes were prepared with Middlebrook 7H10 agar (DIFCO), so as to have different concentration of drug in each quadrant; the fourth quadrant being the drug-free control. The 7H10 agar was overlaid on various drug-incorporated discs. The concentrations of drug/disc are listed in [Table - 3]. Some of the drug discs were purchased commercially (Hi-Media) and others were prepared by using commercially available blank discs. The drugs in chemically pure form were either purchased from Sigma Chemical Co. or were gifts from WoMajor discrepanciesckhardt Pharmaceutical Co, India. Each quadrant was inoculated with 0.1 mL of a bacterial suspension from seven-day-old 7H9 broth culture adjusted to the optical density of no. 1 McFarland standard. The inoculum was spread on the plates by tilting and twirling and incubated at 37°C in the presence of 5% CO ₂ for 21 days. In accordance with the principle of proportion method the organisms were considered "susceptible" to a given concentration if number of colonies in drug containing quadrants were less than 1% of the number of colonies in the drug-free quadrants.

Results

Drug susceptibility of 49 strains of MAC was undertaken using rapid broth microdilution. [Table - 4] depicts the MICs obtained by the broth microdilution method with our strains as against the published results of Heifets. ^[13] [Table - 5] shows the resistance/sensitivity profile of these 49 strains as determined by broth microdilution technique (ELISA reader method). The visual reading and ELISA plate reading were compared to find out the degree of agreement between these two methods of reading the results. The criteria for declaring the strains as resistant, moderately resistant, sensitive, moderately sensitive were according to the guidelines suggested by Heifets et al . ^[13] The results of agar dilution method and broth microdilution method were also compared [Table - 6]. The results have been tabulated as resistant (R), moderately resistant (MR), sensitive (S), moderately sensitive (MS) and inconclusive results (C). The figure depicts the resistance profile of a representative strain of MAC to various antituberculous drugs in a 96 well ELISA plate.

All isolates grew well in the broth medium and produced sharp, easily discernible growth endpoints [Figure - 1]. MIC was generally read when organisms reached good macroscopic growth in control wells after 14 days. More than 40% of the isolates were sensitive to ciprofloxacin (42.86%), roxithromycin (71.43%) and amikacin (55.11%). However, very high degree of resistance was observed for the first line drugs. Only 22.85% sensitivity was seen in the case of both isoniazid and ethambutol. Similarly, rifampicin and clofazimine showed only 28.57% sensitivity [Table - 5].

A comparison of the results obtained in broth (MIC) and in agar (proportion method) for isoniazid (INH), streptomycin (STR), ethambutol (EMB), rifampicin (RIF), kanamycin (KANA), amakacin (AMIKA), clofazimine (CLOF), ciprofloxacin (CIPRO) and roxithromycin (ROXITH) is shown in [Table - 6]. A 100% agreement was noted for kanamycin susceptibility and 83.6% agreement was seen for RIF, AMIKA, CLOF, CIPRO and ROXITH. However, major discrepancy was noted when results were compared for INH, STR and EMB for proportion method and MIC.

Discussion

In India, laboratory diagnosis of NTM has been confined mainly to isolation and identification. A number of workers have isolated and identified MAC as a major group of NTM. ^[14] The increasing significance of NTM in this part of the world has been reflected by the BCG trial held at Chingleput in Madras where 22.6% of the total NTM isolates were identified as MAC. ^[15] Recently, a report has appeared confirming the association of MAC with HIV-AIDS in India. ^[16] Although BACTEC 460 system has been used for assaying drug susceptibility of NTM also; this is an expensive method not accessible to many laboratories. In comparison, broth microdilution is a simple and inexpensive technique that can be standardized by any laboratory willing to initiate drug susceptibility testing for NTM. The present study was therefore an attempt to determine MIC of various antitubercular drugs for the clinical isolates of MAC by broth microdilution method. The MIC results obtained were compared with the results of single concentration drug susceptibility assays in agar using recommended dilutions of antituberculosis drugs. Heifets′ ranges of MICs have been taken as the reference point. The MIC obtained by us was found to be comparable except that, in the case of isoniazid (0.32 mg mL ^-1 ) it was slightly lower while in case of rifamipicin (7.8 mg mL ^-1 ), it was higher than reported [Table - 4].

Major discrepancies between the agar dilution and broth micro dilution method were seen for ciprofloxacin, amikacin and roxithromycin. In each case, the isolates tended to be resistant in agar but susceptible in broth. Similar results have been reported for M. avium when tested against INH, RIF, EMB and STR. ^[7] Hawkins and Gross noted a similar poor correlation for STR when comparing susceptibility measured in broth (BACTEC) with standard agar dilution for M. marinum and M. kansasi (Gross, program abstract 1200, 24 ^th Intersci. Conf. Antimicrob. Agents Chemother, 1984). Discrepancies were also noted when MAC isolates were tested against INH, STR, EMB and RIF. Wallace et al ^[7] also observed the same trends for RIF, EMB and STR. Isolates were usually resistant to the single, fixed drug concentration in agar but were inhibited at MIC well below this concentration in broth. Steadham et al ^[4] also reported similar discrepancies for MAC when they compared the BACTEC broth method with standard agar dilutions. The reasons for these discrepancies are not readily apparent but they are a consistent finding. Heifets and Lindholm-Levy ^[17] found that the MIC determined by the broth dilution method (radiometric) was four to eight times lower than those found by the agar plate method. Royo et al ^[18] studied the correlation between microdilution and macrodilution techniques in MB7H9 broth by testing the MIC of sparfloxacin against M. kansasii and found all the strains to be susceptible by both methods with an MIC of 0.25 μg mL ^-1 . Heifets ^[19] found the broth determined MIC of seven antituberculosis drugs viz, isoniazid, rifampin, ethionamide, streptomycin, amikacin, kanamycin and capreomycin to be lower than the agar determined MIC for MAC. These differences may be caused as suggested by some reports by such factors as higher absorption and degradation of drug in solid medium ^[20] or the longer period of incubation required when using agar medium leading to greater degradation of the drug. ^[21] Furthermore, on solid medium bacteria grew on the surface and a concentration gradient could develop during the incubation and growth of bacteria leading to altered or pseudo resistance. In contrast, in liquid medium, there are more cell to drug contact as the bacterial population is submerged in the drug-containing medium. The third factor is the low bactericidal effect of some drugs against MAC, which may result in secondary growth of the temporarily suppressed bacterial population after the inhibitory concentration of a drug has diminished during the long period of growth, particularly on agar plates.

Although broth microdilution method is not likely to allow direct susceptibility testing because of the problem of contamination, it would allow for easy evaluation of all first- and second-line drugs, as well as levels of resistance of the culture positive isolates. This method has the convenience of testing several drugs in a single plate and at multiple concentrations. Thus a gradient of growth gives a very clear indication of resistance or sensitivity even visually. In contrast, agar dilution is cumbersome in case of a necessity for testing multiple concentrations and generally reports higher MIC than broth microdilution.

The results reported here compared well with the reference reports from Heifets′ laboratory and thus may contribute to the development of a standardized, rapid, inexpensive, feasible and reproducible technique for MAC susceptibility testing in the Indian as well as a global context.

References

Copyright 2007 - Indian Journal of Medical Microbiology

The following images related to this document are available:

Photo images