Comparison of Two Nonradioactive, Single-Strand Conformation Polymorphism Electrophoretic Methods for Identification of rpoB Mutations in Rifampin- Resistant Isolates of Mycobacterium tuberculosis.

Background: Diverse mutations in an 81 bp region of the rpoB gene are found in approximately 95% of rifampin-resistant (RIFr) Mycobacterium tuberculosis isolates. Various methods to detect these mutations have been evaluated for their usefulness as rapid screens for rifampin resistance. Methods and Results: Two nonradioactive variations of single-strand conformation polymorphism (SSCP) electrophoresis were optimized and evaluated for their ability to distinguish nine rpoB mutations present in a collection of 51 RIFr M. tuberculosis isolates. One of the methods used polymerase chain reaction products (128 bp) encompassing the 81 bp region of the rpoB gene, which were denatured in the presence of methyl mercury hydroxide, subjected to polyacrylamide gel electrophoresis (PAGE) and detected by staining with ethidium bromide. For the second method, fluorogenically labeled primers were used to generate products that were electrophoresed in an ABI Model 310 Genetic Analyzer equipped with a 3% GeneScan Polymer Column (Applied Biosystems Inc; Foster City, CA). Mobility shifts for all nine mutations were clearly discernible from the wild-type pattern by methods when tested in blind analyses. When an additional 30 isolates were tested by both SSCP methods in a blinded fashion, correlations with RIF susceptibility testing were complete for susceptible and homogeneously resistant isolates. Among three isolates with heterogeneously resistant populations, however, two were correctly identified by fluorescent SSCP compared with one by the PAGE SSCP method. Subpopulations of the His 526-->Tyr rpoB mutant, which is frequently encountered among RIFr strains, could be detected using templates prepared from mixtures of broth cultures with a susceptible strain. Conclusions: SSCP electrophoresis is useful for rapid screening for RIF resistance in susceptible and fully resistant isolates of M. tuberculosis. However, conventional susceptibility testing is still necessary for two reasons: (1) <100% of RIF strains have mutations in the 81 bp hotspot rpoB genomic region, and (2) SSCP may not offer sufficient sensitivity to detect clinically important emergent mutant subpopulations, especially those present as <10% of the total population in a sample. Whereas PAGE SSCP is less costly than fluorescent SSCP, the latter method is somewhat easier to perform and generates quantitative data.