Characterization of zoonotic bacteria from dairy cattle in the era of genomics

Abstract Text:

Dairy farms are well-documented reservoirs for zoonotic pathogens. Salmonella spp., Listeria monocytogenes, Shiga-toxigenic Escherichia coli and Campylobacter spp. are often excreted in the feces of cows and it is common for infected cows to show no signs of illness and not be recognized as sources of human health risks. Historically, comparisons of bacterial isolates from animals and humans were made using molecular genotyping tools, such as pulsed-field gel electrophoresis and rep-PCR, or targeted sequencing techniques, like multilocus sequence typing. The discriminatory power of these tools has been exploited for strain differentiation and epidemiology tracing, the most widespread example being PulseNet. Decreased costs have made whole genome sequencing (WGS) a viable means of comparing the genomes of large numbers of bacterial isolates. Here we describe several examples where we have used comparative genomics and metagenomics to describe relationships between dairy-associated isolates and isolates from other sources and elucidate the ecology of pathogens in dairy farm environments. Genome analysis of 118 Salmonella enterica serotype Kentucky (S. Kentucky) isolates from dairy, poultry, and humans identified some of the poultry and bovine isolates as sequence type (ST) 152, but there was a phylogenetic distinction between the poultry and bovine isolates. The human isolates were primarily distantly related ST198 strains. Three of the dairy isolates were ST198, suggesting dairy animals are a potential reservoir of this human pathogen. We further compared the fecal microbial communities of S. Kentucky-shedding and non-shedding cows to search for potential shifts in community composition associated with S. Kentucky carriage. No significant differences between the two groups were observed suggesting S. Kentucky is a transient commensal gut inhabitant. For L. monocytogenes we sequenced the genomes of 128 isolates from dairy cows and bulk tank milk and compared them to human-associated L. monocytogenesisolates. Phylogenetic inference revealed a high level of diversity among the isolated strains. Multiple sequence and virulence types were identified, including at least four virulence types known to be responsible for large outbreaks. Based on a whole genome phylogenetic analysis, several isolates were closely related to human clinical isolates, such as the strain isolated in the 2011 cantaloupe outbreak that was responsible for multiple deaths. Decreased cost and increased access to WGS is radically changing the understanding of the ecology of bacterial populations. With WGS data much more subtle changes are readily accessible compared with historic methods of distinguishing strains.

Keywords: Salmonella Listeria dairy