Genome scale analysis of the pathogenesis of Staphylococcus aureus bovine mastitis

Staphylococcus aureus is one of the main causes of bovine mastitis, causing a major economic burden to the global dairy industry and a substantial impact on animal welfare. Some bovine S. aureus clones represent an emerging zoonotic threat with spillover events between cows and humans already documented. Antibiotic treatment is often unsuccessful due to the bacteria surviving intracellularly, and currently available vaccines have a low efficacy. It is therefore of critical importance to identify novel therapeutic targets for the control of S. aureus bovine mastitis. In order to control S. aureus mastitis infections, the bacteria must be targeted prior to the establishment of chronic infection. This project therefore focuses on the growth of S. aureus in milk, which is a major constituent of the bovine host environment encountered by the bacteria upon initial entry to the mammary gland. Previous work in our research group examining the host-adaptive evolution of S. aureus has identified the ability of most bovine isolates to clot milk and form a milk associated biofilm; phenotypes that distinguish them from human isolates. Importantly, we have discovered that clotting is associated with enhanced growth in milk, therefore providing a fitness advantage in the dairy niche, and the biofilm is believed to contribute to the establishment and maintenance of chronic infections. To enable the examination of the molecular basis for these adaptive phenotypes, transposon (Tn) mutagenesis, was employed to produce mutant libraries consisting of thousands of individual Tn insertion mutants. Two pooled Tn mutant libraries in clinical S. aureus strains representing the most successful bovine lineages CC97 and CC151 were constructed. The library in the CC151 bovine reference strain RF122t has also been separated into an arrayed library, representing a powerful resource for future studies. Screening of the arrayed RF122t library in milk identified 77 mutants with an altered phenotype during growth in milk, including loss of the clotting phenotype or altered levels of biofilm production. This set of mutants were sequenced to identify the site of the Tn insertion and further characterised to examine the role of the affected gene. This screening revealed the essentiality of the purine biosynthesis pathway to milk clotting, growth in milk and milk-associated biofilm formation with 46 intergenic and intragenic Tn insertion mutants in the pathway identified to be deficient in these phenotypes. The purine biosynthesis pathway is linked to the Krebs cycle and amino acid synthesis, and mutants in lysine and aspartate synthesis also have delayed or loss of milk clotting and reduced milk growth. The assays also identified the S. aureus secreted proteases, aureolysin and a novel cysteine protease, in addition to the protease expression regulators Agr, SarR, YjbH, MgrA and RsbU to be crucial for milk clotting, milk growth and the milk associated biofilm. In addition to the investigation of S. aureus growth in milk, data sets from a previous genome wide association study (GWAS) that detected both genes and single nucleotide polymorphisms (SNPs) that are enriched in bovine S. aureus strains compared to non-bovine S. aureus strains, have been investigated using protein modelling to identify the predicted structural and functional impact of the identified SNPs. Some of the most notable SNPs were predicted to alter putative interaction sites of proteins involved in protein degradation, transcription and iron import. One of the bovine enriched genes identified by the GWAS was located on the staphylococcal pathogenicity island, SaPIbov3 and encoded the major facilitator superfamily (MFS) protein. The mfs gene was cloned into a plasmid expression vector and tested in multiple bovine strains to investigate whether it could alter the clotting phenotype, which it did not. However, a gene deletion mutant in a CC97 strain did form milk clots that were harder to disrupt than those formed by the wild-type. We initially hypothesised that the MFS protein may be involved in glucose transport, however, lactose utilisation studies in multiple strains did not show any difference with the presence or absence of the mfs gene so the function of this protein remains to be elucidated. In summary, this thesis describes investigations into bovine host-adaptive phenotypes by employing the first Tn mutant libraries for bovine S. aureus which has identified key genes and pathways involved in host adaptation. These pathways may represent novel targets for the control of intramammary infection.