Similarities and differences in molecular epidemiology of third-generation cephalosporin-resistant Escherichia coli carried by dogs living in urban and nearby rural settings and associated behavioural risk factors

Abstract

SynopsisO_ST_ABSObjectivesC_ST_ABSOur aims were to compare faecal third-generation cephalosporin-resistant (3GC-R) Escherichia coli isolates from dogs living in a city and in a rural area ~30 km away; to compare isolates from dogs, cattle, and humans in these regions; to determine risk factors associated with 3GC-R E. coli carriage in these two cohorts of dogs. Methods600 dogs were included, with faecal samples processed to recover 3GC-R E. coli using 2 mg/L cefotaxime. WGS was by Illumina; risk factor analyses were multivariable linear regression using the results of an owner-completed survey. Results3GC-R E. coli were excreted by 20/303 rural and 31/297 urban dogs. Dog/human sharing was evident for the dominant canine 3GC-R sequence type, ST963(blaCMY-2). Cattle/dog sharing was evident for CTX-M-14 and CTX-M-32-producing E. coli from rural dogs, including sharing of plasmid pMOO-32, which is common on cattle farms in the area. Feeding raw meat was associated with carrying 3GC-R E. coli in rural dogs, but not in urban dogs, where swimming in rivers was a weak risk factor. ConclusionsGiven clear zoonotic potential for resistant canine E. coli, our work suggests interventions that may reduce this threat. In rural dogs, carriage of 3GC-R E. coli, particularly CTX-M producers, was phylogenetically associated with interaction with local cattle and epidemiologically associated with feeding raw meat. In urban dogs, sources of 3GC-R E. coli appear to be more varied and include environments such as rivers.