Characteristics of antimicrobial resistance in Escherichia coli isolated from retail meat products in North Carolina

Abstract

Background Escherichia coli is commonly used as an indicator for antimicrobial resistance (AMR) in food, animal, environment, and human surveillance systems. Our study aimed to characterize AMR in E. coli isolated from retail meat purchased from grocery stores in North Carolina, USA as part of the National Antimicrobial Resistance Monitoring System (NARMS). Materials and methods Retail chicken (breast, n = 96; giblets, n = 24), turkey (n = 96), and pork (n = 96) products were purchased monthly from different counties in North Carolina during 2022. Label claims on packages regarding antibiotic use were recorded at collection. E. coli was isolated from meat samples using culture-based methods and isolates were characterized for antimicrobial resistance using whole genome sequencing. Multi-locus sequence typing, phylogroups, and a single nucleotide polymorphism (SNP)-based maximum-likelihood phylogenic tree was generated. Data were analyzed statistically to determine differences between antibiotic use claims and meat type. Results Of 312 retail meat samples, 138 (44.2%) were positive for E. coli, with turkey (78/138; 56.5%) demonstrating the highest prevalence. Prevalence was lower in chicken (41/138; 29.7%) and pork (19/138;13.8%). Quality sequence data was available from 84.8% (117/138) of the E. coli isolates, which included 72 (61.5%) from turkey, 27 (23.1%) from chicken breast, and 18 (15.4%) from pork. Genes associated with AMR were detected in 77.8% (91/117) of the isolates and 35.9% (42/117) were defined as multidrug resistant (MDR: being resistant to ≥3 distinct classes of antimicrobials). Commonly observed AMR genes included tetB (35%), tetA (24.8%), aph(3’’)-lb (24.8%), and blaTEM-1 (20.5%), the majority of which originated from turkey isolates. Antibiotics use claims had no statistical effect on MDR E. coli isolates from the different meat types (X2 = 2.21, p = 0.33). MDR was observed in isolates from meat products with labels indicating “no claims” (n = 29; 69%), “no antibiotics ever” (n = 9; 21.4%), and “organic” (n = 4; 9.5%). Thirty-four different replicon types were observed. AMR genes were carried on plasmids in 17 E. coli isolates, of which 15 (88.2%) were from turkey and two (11.8%) from chicken. Known sequence types (STs) were described for 81 E. coli isolates, with ST117 (8.5%), ST297 (5.1%), and ST58 (3.4%) being the most prevalent across retail meat types. The most prevalent phylogroups were B1 (29.1%) and A (28.2%). Five clonal patterns were detected among isolates. Conclusions E. coli prevalence and the presence of AMR and MDR were highest in turkey retail meat. The lack of an association between MDR E. coli in retail meat and antibiotic use claim, including those with no indication of antimicrobial use, suggests that additional research is required to understand the origin of resistance. The presence of ST117, an emerging human pathogen, warrants further surveillance. The isolates were distinctly diverse suggesting an instability in population dynamics.