Rapid diagnosis of mycobacterial infections and quantitation of Mycobacterium tuberculosis load by two real-time calibrated PCR assays

Abstract

Sensitive and specific techniques to detect and identify Mycobacterium tuberculosis directly in clinical specimens are important for the diagnosis and management of patients with tuberculosis (TB). We developed two real-time PCR assays, based on the IS6110 multicopy element and on the senX3-regX3 intergenic region, which provide a rapid method for the diagnosis of mycobacterial infections. The sensitivity and specificity of both assays were established by using purified DNA from 71 clinical isolates and 121 clinical samples collected from 83 patients, 20 of whom were affected by TB. Both assays are accurate, sensitive, and specific, showing a complementary pattern of Mycobacterium recognition: broader for the IS6110-based assay and restricted to the M. tuberculosis complex for the senX3-regX3-based assay. Moreover, the addition of a synthetic DNA calibrator prior to DNA extraction allowed us to measure the efficiency of DNA recovery and to control for the presence of PCR inhibitors. The mycobacterial burden of the clinical samples, as assessed by direct microscopy, correlates with the M. tuberculosis DNA load measured by the senX3-regX3-based assay. In addition, reduced levels of M. tuberculosis DNA load are present in those patients subjected to successful therapy, suggesting a potential use of this assay for monitoring treatment efficacy. Therefore, these assays represent a fully controlled high-throughput system for the evaluation of mycobacterial burden in clinical specimens.

FIG. 1.

Schematic representation indicating spacing, positions, and orientations of primers and the probe of mycobacterial IS6110 and senX3-regX3 IR. (A) Schematic drawing of IS6110 showing locations of sequences amplified by published PCR protocols and the international RFLP probe. The bars above the drawing represent sequences determined previously by Kent et al. (13), Desjardin et al. (3), Eisenach et al. (6), Hellyer et al. (11), and van Embden et al. (33). (B) PCR amplification strategy to identify mycobacteria belonging to the M. tuberculosis complex strains. The TaqregT2 primer is specific for the 53-bp MIRU present in one copy of M. tuberculosis strains but absent in BCG.

FIG. 2.

Comparison between reference curves of the two real-time TaqMan PCR systems. The reference curves were obtained by plotting the C_t values (on the y axis) against the plasmid copy number (IS6110, open squares; senX3-regX3 IR, open triangles). In all of the resulting equations [IS6110, y = 38.85 to 3.422 log (x); senX3-regX3 IR, y = 39.05 to 3.398 log (x)], all of the slope coefficients and the intercept values were similar and no significant differences were found by covariance analysis.

FIG. 3.

Distribution of M. tuberculosis (MTB) DNA loads by senX3-regX3 IR TaqMan assay. PCR-positive samples were classified into three groups according to the results of microscopy and bacterial culture (group 1, diamonds; group 2, triangles; group 3, circles), and the M. tuberculosis load was reported before (filled symbols) and after (empty symbols) the correction due to calibration. Bars indicate the median DNA load copy numbers in each group. The median M. tuberculosis loads measured in the presence and absence of calibration were, respectively, 46,000 (range, 760 to 10,000,000) and 10,400 (range, 500 to 9,000,000) for samples in group 1, 2,750 (range, 600 to 430,000) and 822 (range, 160 to 129,000) for samples in group 2, and 454 (range, 220 to 4,150) and 184 (range, 10 to 1,530) for samples in group 3. Statistically significant differences have been reported only for calibrated values. +, positive; −, negative.