In Vivo Expression of the Bacterial Amyloid Curli

Abstract

Salmonella enterica serotype Typhimurium is a rod-shaped, motile, Gram-negative bacterium that causes gastroenteritis in immunocompetent individuals. S. Typhimurium produces an extracellular protein termed curli, a bacterial amyloid with a cross beta-sheet tertiary structure that is common across all amyloids. Curli formation is critical for biofilm formation by enteric pathogens such as S. Typhimurium and E. coli. Curli expression requires the production of multiple proteins, which are encoded by two operons known as csgBAC and csgDEFG. Curli production can be induced in vitro by low temperature and low osmolarity, which is evident by growth on T-medium plates for 72 hours at 28oC. Earlier studies have shown that curli is expressed in sepsis patients with E. coli, as well as in mice after S. Typhimurium infection. This is evidenced by the production of antibodies to CsgA, the major subunit of curli. Our lab has shown that curli fibers are recognized by the TLR2/TLR1 complex of the innate immune system during infection. Infection with curli expressing bacteria causes elevated levels of proinflammatory cytokines, nitric oxide, and autoantibodies. Nonetheless, the details of curli expression in vivo during bacterial infection remain unknown. The focus of these studies was to elucidate the location where bacteria expresses curli in vivo during infection. Initially, we used S. Typhimurium strains carrying plasmids with csgB and csgD promoter regions fused to the gfp gene to study curli expression in vivo by use of flow cytometry. Unfortunately, we were unable to determine curli expression with this model, due to the diminished fluorescence intensity of GFP under anaerobic conditions in the gastrointestinal tract. As the question of curli expression in vivo was left unanswered, we next used a long-term infection model of S. Typhimurium with the goal of determining seroconversion to curli as well as the location and timing of curli expression. Using CBA/J mice infected with wild-type S. Typhimurium or a curli mutant strain, we were able to identify seroconversion to CsgA in the mice infected with the wild-type strain through ELISA and western blot analysis. We were also able to identify autoantibody production in mice infected with the wild-type strain through ELISA. However, we were unable to determine curli expression in the feces of mice either by western blot or qPCR data. We were also able to identify autoantibody production in mice infected with the wild-type strain through an anti-double stranded DNA ELISA. Preliminary findings lead us to hypothesize that curli expression may occur very early on in infection, and may be expressed inside cells such as macrophages. Overall, our results partially elucidate curli expression in vivo, although more research is needed in order to answer our remaining questions regarding location and timing of expression.