exploring variation within salmonella dublin using sequencing and multi-model phenotyping to define markers of virulence and zoonosis

Salmonella enterica is an economically important bacterial pathogen that compromises food safety and One Health globally. Cattle are a major reservoir of Salmonella including S.Dublin, which causes invasive disease in cattle. S.Dublin spreads via the bloodstream to internal organs causing severe and often fatal disease in calves and abortions in pregnant cows. Adult cattle can carry S.Dublin asymptomatically and shed S.Dublin in faeces and milk leading to its spread within herds and into the foodchain. Over 25% of herds in the UK are infected with S.Dublin and it is the predominant serovar isolated from cattle. S.Dublin is a zoonotic pathogen. The consumption of raw milk and raw milk products and direct contact with infected animals are major risk factors for human disease. S.Dublin causes invasive disease in humans. It is at least 30% more invasive than Salmonella serovars that cause gastroenteritis; over 75% of S.Dublin patients need hospitalisation compared to the 27% infected with other serovars and the fatality rate is 4-15% compared to 0.3% for other serovars. The bacterial factors responsible for this invasive behaviour in cattle and humans are poorly defined. S.Dublin is an emerging public health concern. It is difficult to eliminate from cattle herds. There are an increasing number of incidents related to human consumption of raw milk in the UK. Outbreaks due to raw milk and cheese have been reported in Europe. Multidrug resistance is being increasingly reported in cattle and human isolates globally, and there are currently no effective vaccines against S.Dublin. We will use cutting-edge genome sequence analyses combined with observations of virulence in multiple in vitro and in vivo models to define genetic markers of virulence and zoonosis. Guided by publicly available short-read sequence data for S.Dublin, we will assemble a collection of genetically diverse isolates from cattle and humans and use established assays to screen for virulence in bovine and human epithelial and immune cells. We will perform long-read sequencing to characterise additional variation in S.Dublin genomes like large-scale rearrangement of chromosomal fragments and to detect the presence and features of extrachromosomal DNA molecules like plasmids with antimicrobial resistance genes. We will identify isolates with high- and low-virulence profiles using correlation and genome-wide association analyses to define genetic markers of overall in vitro virulence and markers specific to increased virulence in human cells. We will confirm the role of defined markers in vivo using our established ethical 3Rs sequencing-based method in cattle, which are an excellent model for human salmonellosis, thereby maximising the benefits of our research to animal and human health. Our preliminary genomic analysis has already recognised core-genome diversity beyond the predominant genomic cluster, sequence type 10, and variable distribution of acquired virulence factors like the S.Typhi Vi capsular polysaccharide. We also identified large-scale chromosomal rearrangements in S.Dublin genomes using long-read sequencing and found that isolates that vary in sequence type or genome structure exhibit differential virulence in vitro in immune cells. This research will advance our fundamental knowledge of S.Dublin virulence by linking genotype to phenotype. Defining genetic markers to predict high-risk isolates will provide valuable information to animal and public health stakeholders that aligns with long-term BBSRC research priorities to improve surveillance and control programmes; prevent zoonoses; improve food safety; ensure safe and sustainable agriculture; and identify targets to develop effective vaccines and interventions and drive innovation.