Abstract
The erm(41) gene confers inducible macrolide resistance in Mycobacterium abscessus subsp. abscessus, calling into question the usefulness of macrolides for treating M. abscessus subsp. abscessus infections. With an extended incubation (14 days), isolates with MICs of ≥8 μg/ml are considered macrolide resistant by current CLSI guidelines. Our goals were to determine the incidence of macrolide susceptibility in U.S. isolates, the validity of currently accepted MIC breakpoints, and the erm(41) sequences associated with susceptibility. Of 349 isolates (excluding those with 23S rRNA gene mutations), 85 (24%) had clarithromycin MICs of ≤8 μg/ml. Sequencing of the erm(41) genes from these isolates, as well as from isolates with MICs of ≥16 μg/ml, including ATCC 19977T, revealed 10 sequevars. The sequence in ATCC 19977T was designated sequevar (type) 1; most macrolide-resistant isolates were of this type. Seven sequevars contained isolates with MICs of >16 μg/ml. The T28C substitution in erm(41), previously associated with macrolide susceptibility, was identified in 62 isolates (18%) comprising three sequevars, with MICs of ≤2 (80%), 4 (10%), and 8 (10%) μg/ml. No other nucleotide substitution was associated with macrolide susceptibility. We recommend that clarithromycin susceptibility breakpoints for M. abscessus subsp. abscessus be changed from ≤2 to ≤4 μg/ml and that isolates with an MIC of 8 μg/ml have repeat MIC testing or erm sequencing performed. Our studies suggest that macrolides are useful for treating approximately 20% of U.S. isolates of M. abscessus subsp. abscessus. Sequencing of the erm gene of M. abscessus subsp. abscessus will predict inducible macrolide susceptibility.