Specificity of IS6110-Based DNA Fingerprinting and Diagnostic Techniques for Mycobacterium tuberculosis Complex

Abstract

Restriction fragment length polymorphism and hybridization of DNA extracted from Mycobacterium tuberculosis, nontuberculous mycobacteria, and nonmycobacterial species with a probe derived from IS6110 confirmed that IS6110 was specific to M. tuberculosis complex. In addition, DNA amplification with IS6110-specific primers yielded a 181-bp fragment only in DNA from M. tuberculosis complex isolates.

The discovery of polymorphic DNA sequences in Mycobacterium tuberculosis and the subsequent use of methods based on the insertion sequence IS6110 both for amplification and for typing of isolates of M. tuberculosis complex have facilitated early diagnosis of tuberculosis and differentiation of strains, respectively. However, the specificity of these methods for M. tuberculosis has recently been questioned.

Kent et al. (13) reported that strains of nontuberculous mycobacteria (NTM) were positive by PCR amplification of a sequence homologous to a central 181-bp fragment of IS6110. In a subsequent study, Gillespie and colleagues (7) emphasized that the observed homology was demonstrated by Southern blotting. They also described the presence in PCR products of nonmycobacterial species of discrete bands that hybridized with a probe derived from the 181-bp fragment. They suggested that the central sequence found to be homologous in the NTM isolates overlaps that of the internationally recognized probe, IS6110 (19), for the standard restriction fragment length polymorphism (RFLP) method. In a subsequent study by Hellyer et al. (9) no DNA amplification of any of the 27 NTM isolates analyzed, 26 of which had been used in the study by Kent et al. (13), was observed. Similar results were obtained by Mulcahy et al. (15).

McHugh et al. (14) described the presence of IS6110 elements in multiple copies in 14 strains of NTM that hybridized with a probe derived from the 181-bp fragment of IS6110 at 50°C. They did not, however, clarify the exact conditions under which hybridization was performed, including details of the hybridizing solution as well as the time taken for both hybridization and posthybridization washes. For instance, protocols incorporating formamide use hybridization temperatures of 42°C (8), while those that do not use higher hybridization temperatures of 60 to 68°C (1, 12). In both techniques, appropriate posthybridization stringency washes at temperatures ranging between 65 and 68°C are usually used. In this study, we investigated the specificity of IS6110-based assays for identification and typing of M. tuberculosis complex.

The M. tuberculosis reference strain Mt. 14323, clinical isolates of NTM (M. avium complex, M. xenopi, M. kansasii, M. gordonae, M. fortuitum, M. chelonae, M. malmoense, and M. marinum), and nonmycobacterial species (Nocardia sp. and Escherichia coli NCTC 10418) were analyzed. Genomic DNA was extracted by using procedures described previously (20). The concentration of the extracted DNA was determined by measuring absorbance at 260 nm. DNA from a Mycobacterium bovis BCG strain (Glaxo, Greenford, United Kingdom) was included to serve as a positive control for PCR amplification. Twenty-five nanograms of each DNA sample was amplified in a two-tube nested protocol with IS6110 specific primers; the outer primers were Tb 294 (5′-GGACAACGCCGAATTGCGAAGGGC-3′) and Tb 850 (5′-TAGGCGTCGGTGACAAAGGCCACG-3′), and the inner primers were Tb 505 (5′-ACGACCACATCAACC-3′) and Tb 670 (5′-AGTTTGGTCATCAGCC-3′) (13, 14, 21). All primers were used at concentration of 0.5 pmol/μl. Using the outer primers, reactions were amplified for 30 cycles (denaturation at 93°C for 20 s, annealing at 65°C for 30 s, and extension at 72°C for 1 min). One microliter of the product of the reaction with the outer primers was amplified by the inner primers for 30 cycles (denaturation at 93°C for 20 s, annealing at 48°C for 30 s, and extension at 72°C for 30 s). PCR products were analyzed by agarose gel electrophoresis.

One microgram of DNA from each isolate was digested in duplicate with PvuII. One duplicate was hybridized with the INS1/INS2-derived IS6110 probe (19). The other duplicate was hybridized with a digoxigenin-labeled 312-bp 16S RNA fragment of M. tuberculosis H37Rv (nucleotides 248 to 560) (2) which is reported to be conserved in all prokaryotes (5, 18). Hybridization was carried out at 42°C in a solution containing 50% formamide, 6× SSC (1× SSC is 0.15 M NaCl plus 0.015 M sodium citrate), 0.1% (wt/vol) N-lauroylsarcosine, 0.02% (wt/vol) sodium dodecyl sulfate, and 5% (wt/vol) skimmed milk powder, followed by two posthybridization washes of 2× SSC, 0.1% (wt/vol) sodium dodecyl sulfate, and 0.02% (wt/vol) N-lauroylsarcosine at 65% for 15 min each. The digoxigenin-labeled DNA probe was detected with alkaline phosphatase-conjugated anti-digoxigenin antibody (Boehringer, Mannheim, Germany).

Hybridization with the IS6110 probe was seen with DNA from M. tuberculosis but not with DNA from NTM or nonmycobacterial species (Fig. 1). Even loading of DNA in all lanes was confirmed visually by ethidium bromide staining and from the comparable hybridization signals from all lanes when hybridization was performed on the duplicate membrane with the conserved 16S rRNA gene fragment (data not shown). This indicated that appropriate amounts of DNA were used for IS6110 hybridization analysis.

FIG. 1.

Isolates of mycobacterial and nonmycobacterial species hybridized with the international INS1/INS2 IS6110 probe. Lane(s): M, Lambda HindIII molecular weight (size) marker; 1 and 12, M. tuberculosis Mt. 14323; 2, M. avium complex; 3, M. xenopi; 4, M. kansasii; 5, M. gordonae; 6, M. fortuitum; 7, M. chelonae; 8, M. malmoense; 9, Nocardia sp.; 10, M. marinum; and 11, E. coli NCTC 10418.

Amplified PCR products of the expected size (181 bp) were seen in DNA extracted from M. tuberculosis complex isolates but not in DNA from NTM or from nonmycobacterial isolates (Fig. 2). Some amplified products were seen in M. avium complex, M. xenopi and E. coli isolates, but these were of the wrong size. Low levels of nonspecific priming can often cause the generation of nonspecific products in a sensitive nested PCR format such as this (10, 21). All seven negative controls were negative for amplified products.

FIG. 2.

Amplified products of various mycobacterial and nonmycobacterial species obtained by using IS6110 two-tube nested PCR. Lane(s): M, 100-bp ladder (Promega, Madison, Wis.); 1, M. avium complex; 2, M. xenopi; 3, M. kansasii; 4, M. gordonae; 5, M. fortuitum; 6, M. chelonae; 7, M. malmoense; 8, Nocardia sp.; 9, M. marinum; 10, E. coli NCTC 10418; 12 and 13, M. tuberculosis Mt. 14323; 15 through 21, negative controls; 23 and 24, BCG strain (Glaxo) positive controls (amplification of 0.05 and 0.5 pg of BCG DNA, respectively); 11, 14 and 22, empty. Arrows indicate the position of a 181-bp IS6110 fragment.

The results obtained in this study support the recent observations by Hellyer et al. (11). Although some homology between members of the IS3 family of insertion sequences exists, this should not cause a problem in PCR or RFLP assays performed under appropriate conditions. RFLP requires carefully controlled hybridization conditions and high-stringency washes to minimize low-homology binding. In the study by McHugh et al. (14) bacteriophage lambda molecular weight markers had undefined bands on their Southern blots, demonstrating binding which presumably was due to nonspecific hybridization. Concurrently analyzed M. tuberculosis controls which might help to differentiate nonspecific binding are not shown.

Although Gillespie et al. (7) dismissed the likelihood of PCR contamination, it remains possible that their product became contaminated during the DNA extraction procedure. The apparent ‘reduced specificity’ shown in some studies (3, 6, 16, 17) could be explained by contamination and carryover of PCR products in samples. Moreover, in a French study (4), cited by McHugh et al. (14), in which the false PCR-positive rate was 7%, the authors did not attribute this to DNA from NTM. Cross contamination with M. tuberculosis DNA or amplicons was an equally likely explanation.

Results obtained in this study confirm that the IS6110-based assays used both for DNA fingerprinting of M. tuberculosis complex and for diagnosis of tuberculosis disease are specific. We highlight the importance of maintaining appropriate stringency during hybridization and washing when performing RFLP assays and of excluding contamination in PCR analysis.