Prevalence and distribution of extended-spectrum β-lactamase and AmpC-producing Escherichia coli in two New Zealand dairy farm environments

Abstract

Antimicrobial resistance (AMR) is a global threat to human and animal health, with the misuse and overuse of antimicrobials being suggested as the main driver of resistance. In a global context, New Zealand (NZ) is a relatively low user of antimicrobials in animal production. However, the role antimicrobial usage on pasture-based dairy farms, such as those in NZ, plays in driving the spread of AMR within the dairy farm environment remains equivocal. Culture-based methods were used to determine the prevalence and distribution of extended-spectrum β-lactamase (ESBL)- and AmpC-producing Escherichia coli from farm environmental samples collected over a 15-month period from two NZ dairy farms with contrasting management practices. Whole genome sequencing was utilised to understand the genomic epidemiology and antimicrobial resistance gene repertoire of a subset of third-generation cephalosporin resistant E. coli isolated in this study. There was a low sample level prevalence of ESBL-producing E. coli (faeces 1.7%; farm dairy effluent, 6.7% from Dairy 4 and none from Dairy 1) but AmpC-producing E. coli were more frequently isolated across both farms (faeces 3.3% and 8.3%; farm dairy effluent 38.4%, 6.7% from Dairy 1 and Dairy 4, respectively). ESBL- and AmpC-producing E. coli were isolated from faeces and farm dairy effluent in spring and summer, during months with varying levels of antimicrobial use, but no ESBL- or AmpC-producing E. coli were isolated from bulk tank milk or soil from recently grazed paddocks. Hybrid assemblies using short- and long-read sequence data from a subset of ESBL- and AmpC-producing E. coli enabled the assembly and annotation of nine plasmids from six E. coli, including one plasmid co-harbouring 12 antimicrobial resistance genes. ESBL-producing E. coli were infrequently identified from faeces and farm dairy effluent on the two NZ dairy farms, suggesting they are present at a low prevalence on these farms. Plasmids harbouring several antimicrobial resistance genes were identified, and bacteria carrying such plasmids are a concern for both animal and public health. AMR is a burden for human, animal and environmental health and requires a holistic "One Health" approach to address.

Keywords: AmpC; ESBL; Escherichia coli; antimicrobial resistance; dairy; genomics; third-generation cephalosporin resistance.

FIGURE 1

Resistance profiles of E. coli (n = 52) isolated across 14 farm samples. The number of isolates per sample is indicated in the brackets. Isolates with intermediate resistance to cefotaxime (n = 15) were grouped as resistant in the Upset plot. CPD, cefpodoxime (10 μg); FOX, cefoxitin (30 μg); STR, streptomycin (10 μg); CTX, cefotaxime (30 μg); TET, tetracycline (30 μg); CIP, ciprofloxacin (5 μg).

FIGURE 2

Maximum-likelihood tree of core genome single nucleotide polymorphism (SNP) analysis of ESBL and AmpC-producing E. coli (n = 12). E. coli CP014316 (ST131, bla_CTX–M–15 positive) was used as the reference and nodes are coloured by sequence type, as indicated in the figure legend. The scale bar indicates the proportion of the core genome alignment over which core SNPs have been calculated. The number of SNPs between isolates in the same cluster is indicated on the figure.

FIGURE 3

Hierarchical cluster tree constructed using Jaccard distances for the presence or absence data from 37 virulence genes identified using Virulence Finder. The tree was edited using the Interactive Tree of Life webserver. Isolate metadata is included for ESBL or AmpC phenotype, farm and sequence type, as indicated by the colour keys. The virulence genes are grouped and annotated by general function.

FIGURE 4

(A) Collection dates of samples positive for ESBL- and/or AmpC-producing E. coli over the 15-month study period. The phenotype is represented by shape as indicated in the figure legend. (B) Antimicrobial use per month (mg active ingredient per population correction unit) on Dairy 1 and Dairy 4 during the 15-month study period.