Evaluation of Genotype MTBC Assay for Differentiation of Clinical Mycobacterium tuberculosis Complex Isolates

Abstract

A commercially available DNA strip assay was evaluated for the ability to differentiate Mycobacterium tuberculosis complex species. M. bovis subsp. bovis, M. bovis subsp. caprae, M. bovis BCG, M. africanum subtype I, and M. microti were unequivocally identified. M. tuberculosis, M. canetti, and M. africanum subtype II showed a unique hybridization pattern.

The Mycobacterium tuberculosis complex (MTBC) comprises the closely related species M. tuberculosis, M. bovis, M. bovis BCG, M. africanum, M. microti, and M. canetti which cause tuberculosis in humans and animals (23, 24, 25). Despite their close genetic relatedness, as demonstrated, e.g., by DNA-DNA hybridization or by sequencing of the 16S rRNA gene (4, 5, 6, 10, 13, 14, 24), the members of the MTBC differ in host range and pathogenicity (25). The natural hosts of M. tuberculosis and M. africanum are humans, whereas M. bovis can cause disease in a wide range of domestic or wild animals like cattle or goats, as well as in humans (25). According to their biochemical characteristics, two major subgroups of M. africanum, corresponding to their geographic origin in West (subtype I) or East (subtype II) Africa, have been described. Numerical analyses of biochemical characteristics have revealed that M. africanum subtype I is more closely related to M. bovis, whereas subtype II more closely resembles M. tuberculosis (3). Among M. bovis isolates, two subspecies have been described: M. bovis subsp. bovis (resistant to pyrazinamide) and M. bovis subsp. caprae (susceptible to pyrazinamide [16]). The attenuated bacillus Calmette-Guérin (BCG) vaccine strain is derived from a virulent M. bovis strain and has been used as a vaccine against tuberculosis and increasingly as cancer immunotherapy (1). M. microti has been reported to infect small rodents like voles and more recently also humans (15, 24, 25). M. canetti has been described as a novel pathogenic taxon of the MTBC, and rare cases have been reported in patients living mainly in Africa (23).

Rapid identification of MTBC isolates can be easily performed with commercially available gene probes or PCR methods targeting, e.g., the identical 16S rRNA gene and internal transcribed spacer sequences or a number of specific repetitive elements like the insertion sequence IS6110 or the direct repeat locus (11, 19). In contrast, routine differentiation within the MTBC so far is based on a number of phenotypic characteristics and biochemical tests like nitrate reduction or niacin accumulation (25). These tests need sufficient bacterial growth, are time consuming, do not allow unambiguous species identification in every case, and may not be performed routinely in every laboratory.

To overcome these problems, several DNA-based techniques have been established. Spoligotyping and other molecular methods have been demonstrated to be useful tools for rapid species differentiation (8, 11, 13, 18, 20, 21, 24). A PCR test identifying the RD1 deletion was found to be useful for the identification of M. bovis BCG (22). By our own work, we established a PCR-restriction fragment length polymorphism (RFLP) assay that is based on DNA single-nucleotide polymorphisms in the gyrB gene and allows rapid differentiation of M. bovis subsp. bovis, M. bovis subsp. caprae, and M. microti, as well as clear identification of M. africanum subtype I strains (14). M. tuberculosis, M. africanum subtype II, and M. canetti, however, displayed identical gyrB DNA sequences and were indistinguishable in this analysis. In a very recent report, Parsons and coworkers (17) described a rapid and simple approach to the differentiation of members of the MTBC by a PCR-based genomic deletion assay. By a series of multiplex PCR assays, MTBC isolates can be differentiated; M. africanum subtype II, however, cannot be distinguished from M. tuberculosis by this technique either.

Although the gyrB PCR-RFLP and the PCR-based genomic deletion assays provide very useful tools for the discrimination of members of the MTBC, these techniques may be more suited for specialized molecular laboratories rather than for routine diagnostic laboratories. In the latter case, a rapid, easy-to-perform, commercially available technique, such as gene probes for identification of mycobacterial species, may be more suitable and reach broader acceptance in laboratories.

In this work, we evaluated a new commercially available DNA strip assay (Genotype MTBC; Hain Lifescience GmbH, Nehren, Germany) intended for the differentiation of members of the MTBC and identification of M. bovis BCG. This assay is based on gyrB DNA sequence polymorphisms and the RD1 deletion of M. bovis BCG. Specific oligonucleotides targeting these polymorphisms are immobilized on membrane strips. Amplicons derived from a multiplex PCR react with these probes during hybridization. A well-characterized collection of MTBC isolates was analyzed in a blinded fashion, and the results obtained were compared with those obtained by classical biochemical and phenotypic tests, as well as with other molecular tests. Moreover, clinical isolates of M. tuberculosis, M. bovis, M. microti, and M. africanum were included.

Strains analyzed.

For the first series, a set of 30 MTBC strains comprising the type strains M. tuberculosis H37 ATCC 27294, M. bovis ATCC 19210, M. bovis BCG ATCC 27289, and M. africanum ATCC 25420 and 26 well-characterized isolates of M. tuberculosis (n = 4), M. bovis subsp. bovis (n = 6), M. bovis subsp. caprae (n = 3), M. bovis BCG (n = 3), M. africanum subtype I (n = 3), M. africanum subtype II (n = 3), M. canetti (n = 1), and M. microti (n = 3) was analyzed. Identification of these isolates to the species level was based on a detailed analysis of their phenotypic and genetic characteristics, including analysis of the gyrB gene and RD1 deletion assay for M. bovis BCG strains. Additionally, a collection of 82 clinical isolates differentiated by standard laboratory procedures (9) and three different M. bovis BCG vaccine strains (Connaught, RIVM, and Tice) were included.

All isolates were identified as members of the MTBC with gene probes (AccuProbe; GenProbe, San Diego, Calif.).

Biochemical tests and susceptibility testing.

Biochemical analyses for differentiation included colony morphology, nitrate reduction on modified Dubos broth, a niacin accumulation test (INH test strips; Difco, Detroit, Mich.), and growth in the presence of thiophen-2-carboxylic acid hydrazide (1 μg/ml). Growth characteristics on Lebek medium and on bromcresol purple medium were determined as described previously (12). Drug susceptibility was determined by the proportion method on Löwenstein-Jensen medium (in accordance with the German DIN guidelines) and/or the modified proportion method in BACTEC 460TB (Becton Dickinson and Company, Cockeysville, Md.).

DNA procedures.

For isolation of DNA, 1 loopful of bacteria was suspended in distilled water (400 μl), subjected to sonication for 15 min, and boiled for 20 min in a water bath. Three to 5 μl of this suspension was directly used for PCR amplification. Spoligotyping of strains was performed as described by Kamerbeek et al. (7), PCR-RFLP analysis of the gyrB gene was performed as described previously (14), and PCR-based identification of M. bovis BCG RD1 was performed as described by Talbot et al. (22).

Genotype MTBC assay.

The Genotype MTBC assay was performed as recommended by the manufacturer. Briefly, for amplification, 35 μl of a primer nucleotide mixture (provided with the kit), amplification buffer containing 2.5 mM MgCl₂ and 1.25 U of HotStarTaq polymerase (Qiagen, Hilden, Germany), and 5 μl of DNA in a final volume of 50 μl were used. The amplification protocol consisted of 15 min of denaturation at 95°C, followed by 10 cycles comprising 30 s at 95°C and 120 s at 58°C, an additional 20 cycles comprising 25 s at 95°C, 40 s at 53°C, and 40 s at 70°C, and a final extension at 70°C for 8 min. Hybridization and detection were performed with an automated washing and shaking device (Profiblot; Tecan, Maennedorf, Switzerland). The program was started after mixing 20 μl (in the first series) or 10 μl (in the second series) of the amplification products with 20 μl of denaturing reagent (provided with the kit) for 5 min in separate troughs of a plastic well. After it was placed in the device, 1 ml of prewarmed hybridization buffer was automatically added, followed by a stop to put the membrane strips into each trough. The hybridization procedure was performed at 49°C for 0.5 h and followed by two washing steps. For colorimetric detection of hybridized amplicons, streptavidin-conjugated alkaline phosphatase and the appropriate substrate were added. After a final washing, strips were air dried and fixed on paper. Each strip contains 13 probes; the specificity of each probe is listed in Fig. 1. Amplification and hybridization controls verified the test procedures; an MTBC control confirmed the identification as MTBC. A template sheet showing the positions of the lines and the interpretation table, both provided with the kit, were used for interpretation of the test results.

FIG. 1.

Results representative of all of the patterns obtained with the Genotype MTBC assay. The positions of the oligonucleotides and the marker line are shown on the left. The specificity and targeted genes of the lines are as follows: 1, conjugate control; 2, amplification control (23S rRNA); 3, MTBC specific (23S rRNA); 4 to 12, discriminative for the MTBC species (gyrB); 13, M. bovis BCG (RD1); M, marker line for correct orientation of the strip. Patterns of the strips shown: 1, M. tuberculosis or M. africanum subtype II or M. canetti. 2, M. africanum subtype I; 3, M. microti; 4, M. bovis subsp. bovis; 5, M. bovis BCG; 6, M. bovis subsp. caprae.

All assays were performed without knowledge of the previous differentiation.

In this study, we evaluated the new commercially available Genotype MTBC assay for species differentiation within the MTBC, which is based on the principle of reverse hybridization. Fragments of the 23S rRNA gene, of the gyrB gene, and of the RD1 region are amplified simultaneously in a multiplex PCR with biotinylated primers. The targeted fragments of gyrB and RD1 comprise the species-specific sequences, whereas the amplified fragments of the 23S rRNA gene comprise sequences (i) covering G+C-rich gram-positive bacteria (amplification control) and (ii) specific for all members of the MTBC. Amplified fragments are coincubated with membrane strips that are coated with 13 oligonucleotides targeting the complementary DNA sequences. The location and specificity of each oligonucleotide are shown in Fig. 1. The reverse hybridization is followed by biotin-streptavidin-mediated detection of hybridized fragment. Six different patterns can be obtained (Fig. 1).

In the first part of the study, the Genotype MTBC assay was performed with type strains of M. tuberculosis H37, M. bovis, M. bovis BCG, and M. africanum and 26 MTBC isolates (M. tuberculosis [n = 4], M. bovis subsp. bovis [n = 6], M. bovis subsp. caprae [n = 3], M. bovis BCG [n = 3], M. africanum subtype I [n = 3], M. africanum subtype II [n = 3], M. canetti [n = 1], and M. microti [n = 3]) that have been characterized in detail previously. Species identification was based on classical phenotypic and genetic analyses, including PCR-RFLP of the gyrB gene and an RD1 deletion assay for BCG. The Genotype MTBC assay results for these strains gave 100% agreement with the previous differentiation, confirming the concordance between genetic data and hybridization patterns. As expected, M. tuberculosis, M. canetti, and M. africanum subtype II showed characteristic but identical patterns in the Genotype MTBC assay because these species have identical gyrB gene sequences.

In the second part of this study, three different BCG vaccine strains and 82 clinical or veterinary isolates were included (Table 1). All of the strains were differentiated by conventional biochemical and/or genetic analyses. Interpretation of the Genotype MTBC hybridization patterns was performed on the basis of the description included in the test. The hybridization patterns were all unequivocal and could easily be allocated to species (Fig. 1). Results were thereafter compared with the classical differentiation results. The data are summarized in Table 1.

TABLE 1.

Results obtained with clinical and veterinary MTBC isolates

Genotype MTBC pattern	No. of isolates	Differentiation	Comment(s)
M. tuberculosis	31	29 M. tuberculosis	20 M. tuberculosis Beijing type strains, 9 M. tuberculosis non-Beijing type strains; resistant and susceptible strains
		2 M. africanum subtype II	African patients
M. africanum I	4	4 M. africanum subtype I	African patients
M. bovis subsp. bovis	17	17 M. bovis subsp. bovis	Patient isolates
M. bovis subsp. caprae	17	17 M. bovis subsp. caprae	14 patient isolates, 3 cattle isolates
BCG	7	7 M. bovis BCG	3 vaccine strains (Connaught, RIVM, Tice); 4 M. bovis BCG patient isolates
M. microti	9	9 M. microti	5 patient, 3 veterinary isolates, 1 isolate no data; 1 subtype llama, 8 subtype vole

Of the 31 isolates identified as M. tuberculosis on the basis of the Genotype MTBC pattern, 29 strains (93.5%) have previously been identified as classical M. tuberculosis. The patterns were identical, irrespective of the susceptibility of the strain or of whether it belongs to the Beijing family. The remaining two strains were differentiated as M. africanum subtype II by conventional methods.

Definite differentiation could be obtained for the four M. africanum subtype I strains, which resemble M. bovis rather than M. tuberculosis, e.g., in susceptibility to thiophen-2-carboxylic acid hydrazide. Within the cluster of M. bovis, the subspecies bovis (n = 17) and caprae (n = 17) and the attenuated BCG strains (n = 7) could be clearly identified. Furthermore, both subtypes (llama and vole) of M. microti strains (n = 9) showed the characteristic M. microti pattern, as suggested for this species.

These results demonstrate that, with the Genotype MTBC assay, with the exception of M. tuberculosis, M. africanum subtype II, and M. canetti, all of the MTBC species can be unambiguously identified. With these species, identical hybridization patterns were obtained. However, the close relationship of M. tuberculosis, M. africanum subtype II, and M. canetti was also confirmed by analysis of variable regions within the genome of MTBC species (2). So far, no unequivocal genetic marker has been found with which to differentiate M. africanum subtype II (the East African strains) from M. tuberculosis (2, 13). The designation of these strains as M. africanum probably should be reconsidered. M. canetti is a smooth variant of M. tuberculosis and has been isolated mainly from African patients (24). A specific genomic deletion in M. canetti was reported that may be used for amplification-based differentiation (2). However, the prevalence of M. africanum subtype II and M. canetti in non-African populations is extremely low; thus, this may not pose an aggravating mistake when this analysis is performed on strains deriving from non-African populations.

Conventional differentiation of members of the MTBC relies on a combination of the results of several tests based on the growth characteristics and biochemical properties of the strains. Most assays require sufficient culture material, usually grown on solid slants for 2 to 3 weeks; special media are necessary; and sometimes borderline results are even obtained, rendering definite differentiation more difficult, particularly in cases of MTBC species other than M. tuberculosis. The Genotype MTBC assay offers a new promising technique with which to overcome several of these obstacles. Since this assay is an amplification-based technique, differentiation can be performed with no need for extensive culture growth and thus can readily be done with liquid cultures from primary isolations with no need for further cultivation on solid media. Evaluation of the sensitivity of this test with liquid media from primary isolations is under way in a multicenter study. The time required to perform the assay is 4 to 5 h, thus reducing the time required to obtain diagnostic results. Compared to other genetic techniques, this assay does not require special technical skill, such as, e.g., that required for a line blotter for spoligotyping, or work with potentially carcinogenic substances like ethidium bromide, thus enabling even routine diagnostic laboratories to perform this test. It can easily be included in routine work flows. Furthermore, the amplification and hybridization conditions are identical to those of the Genotype Mycobacteria assay (Hain Lifescience GmbH) for identification of several mycobacterial species, allowing parallel performance of both assays in one working operation.

In conclusion, the Genotype MTBC assay allows differentiation of MTBC species with an easy-to-perform reverse hybridization assay. It has the potential to replace sophisticated techniques required for differentiation of members of the MTBC available mainly in reference laboratories.