Antimicrobial resistance genotypes and phenotypes of Campylobacter coli isolated from different sources over a 16-year period in Brazil

Abstract

This study aimed to identify antimicrobial resistance (AMR) genotypes in 63 C. coli strains isolated from humans (12), animals (21), the environment (20) and food (10) in Brazil using whole genome sequencing tools (WGS) and compare them with the results obtained by antimicrobial susceptibility testing against (AST) some important antimicrobials of clinical use. The phenotypic resistance profiles were determined by minimal inhibitory concentrations and disk diffusion technique. The prediction of the resistance genes was performed using ABRicate v.0.8 and the Resistance Gene Identifier software (RGI) of the CARD. The percentage of C. coli strains phenotypically resistant to antimicrobials were: ampicillin 44.4%, doxycycline 20.6%, tetracycline 20.6%, ciprofloxacin 12.7%, nalidixic acid 12.7%, streptomycin 6.3%, erythromycin 4.8% and gentamicin 1.6%. The genes blaOXA-605 / blaOXA-61,tet(O), cmeB, aadE-Cc, aph (3 ') - IIIa, sat4 and aad9 were detected in 54%, 22.2%, 9.5%,6.3%, 1.6%, 1.6% and 1.6% strains, respectively. Mutations T86I in the QRDR region of gyrA were detected in 8 (12.7%) strains. The correlation between AST and WGS was 100%, 92.9%, 82.4%, 80%, for quinolones, tetracycline, β-lactam and aminoglycoside classes. The rates of C. coli strains resistant to β- lactams and quinolones may represent a public health concern. The partial correlation between AST and WGS shows that improvement in antibiotic resistance databases may be required to minimize this discrepancy observed in some antimicrobial classes and to become an acceptable tool to both clinical microbiologists and regulatory agencies.