Relationship between antibiotic susceptibility and genotype in Mycobacterium abscessus clinical isolates

Abstract

This study aimed to determine the antibiotic susceptibility and resistance related genotypes of Mycobacterium abscessus. One hundred sixty-two clinical isolates were collected. Genomic data were obtained by whole genome sequencing. Single nucleotide polymorphism (SNP) analysis was conducted using the NCBI GenBank database and BLAST algorithm. The following genes were of interest: erm(41), rrl and rrs. Erm(41) was further divided into 3 sequevars: erm(41)C28, erm(41)T28, and M type [erm(41) with deletions in nucleotides 64 and 65, or 159 through 432]. Antibiotic susceptibility was assessed at 3 days (early reading time, ERT) and 14 days (late reading time, LRT) after clarithromycin (CLA) treatment. Three patterns of CLA resistance were observed. (1) Fifty-five (acquired resistance) isolates [45 erm(41)T28, 1 erm(41)C28 and 9 M type] exhibited MIC ≥8 mg/L at ERT; among these isolates, 10 had an rrl 2058/2059 mutation. (2) Sixty-two subsp. abscessus and 2 subsp. massiliense (induced resistance) isolates exhibited MIC ≤4 mg/L at ERT, but ≥8 mg/L at LRT. (3) Forty-three (sensitive and intermediate) isolates [14 erm(41)C28, 1 erm(41)T28, and 28 M type] exhibited MIC ≤4 mg/L at both ERT and LRT. No rrs 1408 mutation or other meaningful SNP was found in 3 amikacin-resistant isolates. No correlation was found between rrl, erm(41) or rrs and susceptibility to the 8 other antibiotics tested. The rrl and erm(41) genotypes could predict the CLA resistance of M. abscessus clinical isolates. China has a large number of CLA-resistant M. abscessus isolates with erm(41)T28 sequevar. Treatment of M. abscessus infections should be based upon a comprehensive consideration of factors that include genotype and geographic location.