Identification and characterization of bacterial and host epithelial factors altered during Edwardsiella infections

Bacteria in the genus Edwardsiella cause hemorrhagic septicemia in fish hosts, and severe diarrhea in immunocompromised humans. These phenotypes generally require effector proteins secreted through type III and type VI secretion systems. In this thesis, I investigate 2 common bacterial targets, cytoskeleton and intercellular junctions, in epithelial cells to understand the sub-cellular alterations caused by Edwardsiella during these infections. Using two robust epithelial infection models (HeLa and Caco-2 cells) I show that host microtubules are destroyed during the progression of Edwardsiella infections, while the actin and intermediate filaments remain unaltered. My evidence points to the host microtubule severing enzymes as key players in the microtubule disassembly event as katanin A1 and the katanin A1 subunit-like 1 proteins both localize to the microtubule cut-sites. The novelty of this phenotype extends to the bacteria, as this phenotype is independent of both type III and type VI secretion systems. Negative screening using an E. piscicida transposon insertion mutant library identified 15 bacterial genes needed for host microtubule severing event. The thesis concludes with preliminary examination of host intercellular junction alterations where I demonstrate that the tight junction protein claudin-3 is dissociated from cell peripheries in infected cells. In conclusion, I uncovered two steps of Edwardsiella’s infectious process where host structural components are targeted—leading to phenotypes observed during edwardsiellosis.