Metagenomic analysis of UK retail foods finds limited evidence for associations between food production method and antimicrobial resistance gene burden.

Abstract

Food is produced by a range of methods including extensive (organic and free range), intensive (conventional) and wild-caught production systems. Antimicrobial use varies between different food production systems, which may affect the microbial populations as well as the prevalence and diversity of antimicrobial resistance genes (ARGs) found on food at retail. In this study, shotgun metagenomics was used to investigate the microbial and ARG composition of 25 pork, 33 beef, 33 lamb, 60 chicken, 31 salmon and 41 leafy green samples collected in Norfolk, England, and labelled as extensive, wild caught or intensive. Food microbiomes consisted predominantly of spoilage-associated organisms including Pseudomonas, Lactococcus and Psychrobacter. Significant differences in bacterial diversity were found between intensive and extensive systems on chicken, and 22 differentially abundant genera were identified between production systems across beef, chicken and salmon. Genes conferring resistance to tetracyclines and beta-lactams comprised the majority of the food resistome across all commodities. Across most measures used to compare food resistomes between production methods, no significant differences were detected, except on chicken and salmon where differences in beta-diversity between production methods were detected, albeit with low effect sizes. Overall, these results suggest that differently produced foods, at least when tested at retail and in this region, may present a similar risk of antimicrobial resistance across the commodities investigated within this study. However, specific associations were identified with the microbial composition across chicken, beef and salmon, suggesting that production method may drive some variation in the microbial population structure on food products. Additional work at the farm or food processing levels is required to identify the drivers of these differences between production systems.