Genes associated with fitness and disease severity in the pan-genome of mastitis-associated Escherichia coli

Bovine mastitis caused by Escherichia coli may manifest as subclinical through severe acute disease and can be transient or persistent in nature. Little is known about bacterial factors that impact clinical outcomes or allow some strains to outcompete others in the mammary gland (MG) environment. Mastitis-associated E. coli (MAEC) may have distinctive characteristics which may contribute to the varied nature of the disease. In this study, we sequenced the genomes of 96 MAEC strains isolated from cattle with clinical mastitis (CM). We utilized clinical severity data to perform genome-wide association studies to identify accessory genes associated with strains isolated from mild or severe CM, or with high or low competitive fitness during in vivo competition assays. Genes associated with pathogenic or commensal strains isolated from bovine and avian sources were also identified. A type-2 secretion system (T2SS) and a chitinase (ChiA) exported by this system were strongly associated with pathogenic isolates compared with commensal strains. Strains carrying these genes also had higher competitive fitness during experimental intramammary infections. Deletion of chiA from MAEC isolates decreased their adherence to cultured bovine mammary epithelial cells, suggesting that the increased fitness associated with strains possessing this gene may be due to better attachment in the MG. Importance Bovine mastitis caused by MAEC compromises animal health and inflicts substantial product losses in dairy farming. Given their high levels of intraspecies genetic variability, virulence factors of commonly used MAEC model strains may not be relevant to all members of this group. Here we analyzed clinical data as well as fitness (quantified in a mouse MG model) of diverse MAEC isolates to identify accessory genes that contribute to infection. We demonstrated a novel role for chitinase in promoting attachment to mammary epithelial cells. Reverse genetic approaches can be applied to the collection of strains and their complete genome sequences that we have presented here. Overall, these results provide a much richer understanding of MAEC and suggest bacterial processes that may underlie the clinical diversity associated with mastitis and their adaptation to this unique environment. ### Competing Interest Statement The authors have declared no competing interest.