Dual-Probe Assay for Rapid Detection of Drug-Resistant Mycobacterium tuberculosis by Real-Time PCR

Bacterial strains.To confirm the specificity of the real-time PCR system, we used DNAs extracted from M. tuberculosis H37Rv (ATCC 25618), Mycobacterium bovis (Ravenel), M. bovis BCG (Tokyo), Mycobacterium africanum (ATCC 25420), Mycobacterium microti (TC 77), Mycobacterium avium (ATCC 15769), Mycobacterium intracellulare (ATCC 13950), Mycobacterium kansasii (ATCC 12478), Mycobacterium marinum (ATCC 927), Mycobacterium simiae (ATCC 25275), Mycobacterium asiaticum (ATCC 25276), Mycobacterium xenopi (ATCC 19250), Mycobacterium scrofulaceum (ATCC 19981), Mycobacterium gordonae (ATCC 14470), Mycobacterium malmoense (ATCC 29571), Mycobacterium shimoidei (ATCC 27962), Mycobacterium nonchromogenicum (ATCC 19530), Mycobacterium fortuitum (ATCC 6841), Mycobacterium abscessus (ATCC 19977), Mycobacterium tokaiense (ATCC 27282), Mycobacterium austroafricanum (ATCC 33464), Mycobacterium pulveris (ATCC 35154), Mycobacterium smegmatis (ATCC 14468), and Mycobacterium leprae (Thai 53). M. leprae was a kind gift from M. Matsuoka, Leprosy Research Center, National Institute of Infectious Diseases, Tokyo, Japan. In addition to mycobacteria, DNAs from Klebsiella pneumoniae (clinical isolate), Pseudomonas aeruginosa (ATCC 27853), and Staphylococcus aureus (ATCC 29213) were used.

Drug-resistant clinical isolates and DNA isolation.Using the proportion method with Ogawa egg medium, drug resistance was defined as growth of at least 1% of the number of colonies that grew on drug-free medium at critical concentrations of the drugs (i.e., 40 mg of RIF per liter, 0.2 mg of INH per liter, and 2 mg of EMB per liter) (35). The present study examined 45 laboratory strains of RIF-resistant M. tuberculosis (resistant to RIF alone, 7 strains; resistant to RIF and INH, 12 strains; resistant to RIF and EMB, 9 strains; resistant to RIF, INH, and EMB, 17 strains). Genomic DNAs of mycobacteria were isolated from bacteria grown on Ogawa medium by combined chloroform extraction and mechanical disruption (16). DNAs from K. pneumoniae, P. aeruginosa, and S. aureus were extracted with a QIAamp DNA Mini kit (Qiagen Inc., Valencia, Calif.). DNA concentration was estimated by UV absorbance at 260 nm.

Clinical samples and DNA preparation.We examined 27 clinical samples of sputum from patients with pulmonary TB. Drug susceptibility testing was performed as described in a previous study (35). A concentrated smear was prepared from specimens that were decontaminated by using the N-acetyl-l-cysteine-NaOH protocol (17). The basic fuchsin (Ziehl-Neelsen) staining procedure was employed. The DNAs were extracted by using the AMPLICOR respiratory specimen preparation kit (Roche Diagnostic Systems, Inc., Branchburg, N.J.) according to the manufacturer's instructions and were confirmed as M. tuberculosis complex with the COBAS AMPLICOR M. tuberculosis detection kit (Roche Diagnostic Systems, Inc.) (25).

PCR amplification and DNA sequencing.DNA sequences of rpoB (GenBank accession no. L27989 ), embB (U68480 ), and katG (X68081 ) were used for primer design. PCR was performed with each primer set (Table 1) as described previously (19). Both strands of PCR products of the rpoB and the embB genes were sequenced by using respective primers that were used in PCR. For DNA sequencing of the katG mutation sites, alternative sequencing primers were designed, because the amplicon (1,771 bp) was much longer than other genes (Table 1). Direct sequencing of PCR products was performed with a CEQ2000 automate sequencer (Beckman Coulter, Inc., Fullerton, Calif.) and a DTCS quick-start master mix kit (Beckman Coulter, Inc.).

Preparation of TaqMan MGB probes.The TaqMan MGB probes were designed to hybridize with wild-type DNA by using the Primer Express program (Applied Biosystems, Foster City, Calif.). The MGB probes were synthesized by Applied Biosystems. The primers and probes used in the present study are shown in Table 1. Four of eight probes were labeled with 6-carboxyfluorescein (FAM) (emission wavelength, 518 nm) and the remaining four probes were labeled with VIC (emission wavelength, 552 nm), because these two dyes emit luminescence of different wavelengths and can be distinguished individually in one tube.

Real-time PCR and nested PCR.The real-time PCR mixture was prepared in a final volume of 25 μl with 12.5 μl of Universal PCR Master Mix (Applied Biosystems), 25 pmol of primer, the optimal concentration of FAM- and VIC-labeled TaqMan MGB probes (Table 1), and 5 ng of purified DNA or 1 μl of extracted DNA. Real-time PCR analysis was performed with an ABI PRISM 7700 instrument (Applied Biosystems). The conditions of PCR amplification were 50°C for 2 min, 95°C for 10 min, and then 40 cycles of 95°C for 15 s and 60°C for 1 min. Data were analyzed with Sequence Detector software (Applied Biosystems). Fluorescence of hybridized probes was expressed as ΔRn (normalized reporter signal). The number of amplification cycles required for emission of a certain luminescence intensity by each probe (ΔRn = 0.2) reflected the amount of DNA in the sample. This cycle number was called the threshold cycle (Ct). Therefore, the presence of a mutation would result in an increase in Ct.

Nested PCR was performed when signals were undetected in real-time PCR. The PCR was carried out in a 25-μl reaction volume. The reaction mixture contained 1× PCR buffer, 1 U of GC-rich enzyme mix (Roche Diagnostics Corp., Indianapolis, Ind.), a 200 μM concentration of each of the four deoxynucleoside triphosphates, a 1 μM concentration of three kinds of primer sets (Table 1), 1 M GC-rich solution, 1.5 mM MgCl₂, and 10 μl of template DNA. The primer pairs for the rpoB, katG, and embB genes amplified 534-, 464-, and 408-bp fragments, respectively. The PCR conditions were as follows: initial denaturation at 94°C for 5 min and then 40 cycles of 94°C for 30 s, 60°C for 30 s, and 72°C for 30 s. After the first PCR, the amplified product was diluted 100-fold with sterilized water. One microliter of this solution was used for the real-time PCR analysis as described above.

IPC for detecting PCR inhibitors.For analysis of sputum samples from TB patients, an internal process control (IPC) was performed to detect PCR inhibitors PCR simultaneously. As IPC primers, each rpoB primer was added to the 5′ end of the corresponding lambda phage primer (Table 1). The identical sequence of the lambda phage IPC-R probe, which was 5′ labeled with FAM and quenched with 6-carboxytetramethylrhodamine (TAMRA) at the 3′ end, was used for analysis as described previously (13).

Electrophoresis.Amplification products were separated on 3% agarose gels with 1× Tris-acetate-EDTA buffer for 45 min at 100 V.

Real-time PCR.For rapid detection of mutations in the rpoB, katG, and embB genes involving M. tuberculosis resistance to RIF, INH, and EMB, respectively, three primer pairs and eight MGB probes were designed (Fig. 1 and Table 1). Five probes (rpo510/514, rpo514/520, rpo520/524, rpo524/529, and rpo529/533) were used for detection of mutations in the hot spot of rpoB (81 bp between codons 507 and 533 [equivalent to Escherichia coli numbering system {31}]). One probe (TB control probe) was designed outside the hot spot in rpoB as a control for determining the amount of DNA and for identifying M. tuberculosis. Polymorphisms in the 81-bp region of rpoB could be analyzed by using three tubes. One probe each for embB (codon 306) and katG (codon 315) was labeled with FAM and VIC, respectively. These probes were mixed with their four corresponding primers in one tube. Four tubes (three tubes for control and RIF resistance and one tube for INH and EMB resistance) were employed in the assay. Luminescence of all eight probes was detectable by real-time PCR with genomic DNA extracted from M. tuberculosis H37Rv as the template (Fig. 2A). When the DNA had mutations in rpoB at position 516 and in katG at position 315, the corresponding probes showed no luminescence signal (Fig. 2B). Typical results are shown in Fig. 2C and D. These results indicate that the probes used in this study can identify mutations of the target genes. Similar results were obtained by using autoclaved supernatants of M. tuberculosis suspensions instead of purified DNAs (data not shown).

• Open in new tab
• Download powerpoint

FIG. 1.

Design of TaqMan MGB probes for detection of mutations in the rpoB, embB, and katG genes. MGB probes were labeled with two different dyes (FAM or VIC). The DNA sample and PCR primers for rpoB amplification were mixed with each set of probes in three different tubes: (i) TB control and rpo520/524, (ii) rpo510/514 and rpo514/520, and (iii) rpo524/529 and rpo529/533 (A). The emb306 and kat315 probes and embB and katG PCR primers were mixed and reacted in another tube (B). The sequences of primers and probes are listed in Table 1.

• Open in new tab
• Download powerpoint

FIG. 2.

Analysis of DNAs from mycobacteria with eight TaqMan MGB probes by real-time PCR. The templates were genomic DNAs extracted from M. tuberculosis H37Rv (A) as a control, certain mutants of M. tuberculosis (B, rpoB and katG; C, rpoB and embB; and D, rpoB), and M. avium (E). The x axis shows cycle numbers of PCR, and the y axis represents the normalized reporter signal (ΔRn). A horizontal line indicates the threshold (ΔRn = 0.2). The Ct is expressed as the number of cycles to reach the threshold.

Specificity and sensitivity.No luminescence was found when M. avium DNA (up to 50 ng) was analyzed in this system (Fig. 2E). In addition, luminescence was not detected even when more than 50 ng of DNA from other mycobacteria, such as M. intracellulare, M. kansasii, M. marinum, M. simiae, M. asiaticum, M. xenopi, M. scrofulaceum, M. gordonae, M. malmoense, M. shimoidei, M. nonchromogenicum, M. fortuitum, M. abscessus, M. tokaiense, M. austroafricanum, M. pulveris, M. smegmatis, and M. leprae, was used as the template. The presence of sufficient amounts of DNA in the system was confirmed by alternative PCR with mycobacterial 16S rRNA gene primers (6) or dnaJ gene primers (29) (data not shown). As expected, we detected no luminescence with DNAs prepared from K. pneumoniae, P. aeruginosa, and S. aureus. By contrast, the other members of M. tuberculosis complex, such as M. bovis, M. bovis BCG, M. africanum, and M. microti, exhibited similar luminescence in the real-time PCR system (data not shown). Therefore, the specificity of this system for the M. tuberculosis complex was sufficiently high.

By using purified genomic DNA from M. tuberculosis H37Rv, the sensitivity of this system was determined. In direct real-time PCR with the rpoB primer set and TB control probe, 100 fg of genomic DNA could be detected. Therefore, the mutation was detectable efficiently in the presence of 250 fg (Ct = 37) of TB genomic DNA. The detection limit of the real-time PCR was 10 fg of DNA when combined with nested PCR (data not shown).

Assessment of real-time PCR.We extracted and sequenced genomic DNAs from 45 laboratory strains of RIF-resistant M. tuberculosis. They were classified into 20 groups based on their genotypes by nucleotide sequencing of M. tuberculosis DNAs (Table 2). The luminescence intensity of TaqMan MGB probes in the real-time PCR amplification system was expressed as Ct. The Ct was higher when mutations were present in the genes. Using 45 RIF-resistant strains, we measured the Ct derived from an internal TB control probe bound outside the hot spots of the rpoB gene and ΔCt, which expressed the difference between the control and each MGB probe (Table 2). It was found that the ΔCts of probes hybridizing with the region without mutations were low (from −2.44 to 1.61) and that the ΔCt was higher (≥7) when mutations existed in the target DNA that should hybridize with the MGB probe. These results suggest that a ΔCt of more than 7 was associated with a mutation in the nucleotide sequence. No strains that had mutations within the codons 510 to 514 of the rpoB gene were found. To check to the specificity of the rpo510/514 probe, the rpoB gene was cloned and mutations were constructed at codons 511 (CTG→CCG) and 513 (CAA→CTA). Real-time PCR analysis with each construct as the template showed that the luminescence that was derived from probe rpo510/514 disappeared completely (data not shown).

Clinical application of real-time PCR.DNAs extracted from sputa of patients with pulmonary TB were examined to assess to possibility of rapid detection of drug-resistant M. tuberculosis in clinical specimens by real-time PCR. Twenty-five clinical samples were culture positive, and two samples were culture negative. Nine samples were smear positive, and 18 samples were negative (Table 3). Real-time PCR was used to analyze sputum samples for detection of mutations. Sample C5 had a mutation at codon 315 in the katG gene, and sample C11 possessed mutations at codon 531 in the rpoB gene, codon 315 in the katG gene, and codon 306 in the embB gene. These findings were in agreement with the results obtained by DNA sequencing. The drug susceptibility phenotypes of clinical isolates assessed by conventional culture methods were consistent with the genotypes (Table 3). Thus, our real-time PCR system can detect mutations even when clinical samples, such as sputa, are used.

A total of 16 of 27 sputum samples (59.3%) showed strong luminescence in real-time PCR. By contrast, 11 samples showed no luminescence even after 40 cycles of PCR amplification. This was probably due to low concentrations, because IPC was positive for all 11 samples. Consequently, we performed site-specific nested PCR when the amount of DNA was small. The rpoB, katG, and embB genes were amplified by PCR with their corresponding primer sets (Fig. 3). By optimizing the PCR conditions with six primers, three fragments that contained the target sites at a similar concentration were amplified (Fig. 3, lane 4). The targets (rpoB, katG, and embB) were amplified by nested PCR with one tube. The targets were then analyzed by using real-time PCR. Nested PCR products could be analyzed for the presence of mutations associated with drug resistance, although DNA was undetectable in a single-step real-time PCR in these 11 samples. When nested PCR was used before real-time PCR, analysis of all 27 samples could be performed appropriately (Table 4). However, in analysis of clinical samples, the Ct was more variable, resulting in a larger ΔCt (from −3.40 to 3.15) than for purified DNAs extracted from laboratory strains (from −2.44 to 1.61).

• Open in new tab
• Download powerpoint

FIG. 3.

DNA analysis of amplicons by simplex and multiplex PCRs. DNA from sputum sample C1 was amplified by using primers for the rpoB, embB, and katG genes. The product sizes of their amplicons were 534, 408, and 464 bp, respectively (lanes 1 to 3, respectively). Three kinds of primer sets for the rpoB, embB, and katG genes were mixed, and multiplex PCR was performed (lane 4). The M lane was a DNA molecular size marker which contained the 100-bp ladder.