Whelen, A. Christian

Detection of mutations in the rpoB gene of Mycobacterium tuberculosis can be used as an accurate predictor of rifampin resistance in the majority of strains tested. Simple but highly accurate screening methods must be developed for the detection of these mutations. Either DNA sequence analysis or single-strand conformation polymorphism (SSCP) screening can be used to detect rpoB mutations, but these techniques either are expensive or yield results that may prove difficult to interpret when used in a clinical setting. This report describes the use of dideoxy fingerprinting (ddF) as a postamplification screening method to identify rifampin-resistant genotypes. The ddF protocol was performed on the amplified rpoB fragment with no preparatory steps, thus making ddF practical for laboratories equipped for polyacrylamide gel electrophoresis. When compared with the results of SSCP analysis, ddF results were more easily interpreted and contained more sequence-dependent information that facilitated differentiation of functionally significant and silent mutations. The ddF method was used for genotypic determination of rifampin susceptibility of 20 multidrug-resistant strains of M. tuberculosis. The results of this analysis were concordant with DNA sequence analysis and conventional clinical laboratory methods.

Recent analysis of the gene encoding the beta subunit of Mycobacterium tuberculosis RNA polymerase (rpoB) has demonstrated a small region that harbors the mutations most frequently associated with rifampin resistance. Earlier reports have described a high degree of sequence conservation of rpoB among mycobacteria other than M. tuberculosis and other GC-rich bacteria that can lead to false-positive amplification when applied directly to clinical specimens. We developed reagents for PCR amplification that are based on signature nucleotides discovered by comparative sequence analysis of the rpoB genes of organisms phylogenetically related to M. tuberculosis. The specificities of the reagents were challenged with 20 isolates of multiple-drug-resistant M. tuberculosis and more than 20 species of mycobacteria other than M. tuberculosis and other GC-rich organisms. A single-tube heminested PCR protocol was devised to obtain sensitivity equal to those of an IS6110-based PCR assay and culture in spiked sputum experiments. The assay correctly identified 21 of 24 (87.5%) culture-positive specimens, 13 of which were acid-fast smear-negative, in a panel of 51 clinical specimens. Three specimens that were false-positive initially were negative upon repeat testing when the assay was modified to eliminate the potential for aerosol carryover of the first-round amplification product during the open-tube addition of the second set of reaction reagents. This assay is the most sensitive and specific test to date for the direct detection of M. tuberculosis rpoB in clinical specimens. This rapid PCR-based assay can be used for the simultaneous identification of M. tuberculosis and its rifampin susceptibility genotype.

Clinical microbiology is in the midst of a new era. Methodology that is based on nucleic acid detection has slowly appeared in the diagnostic laboratory, and is revolutionizing our ability to assist physicians in the diagnosis and management of patients suffering from infectious diseases. Much like the introduction of immunoassays built around hybridoma technology in the 1980s, considerable doubt and promise exist hand in hand in the 1990s. Conventional testing that is technically straight forward, informative, and timely will always be a part of clinical microbiology; however, considerable room for improvement exists with organisms/diseases for which laboratory methods are limited. Nucleic acid methodology will have its greatest and long-awaited impact in this arena.