REGULATION AND EXPRESSION OF AMYLOID CURLI IN SALMONELLA INFECTION

Abstract

Biofilms were first observed by Antonie van Leeuwenhoek in the mid-17th century from samples of his own oral and fecal microbiota. Leeuwenhoek was limited by the technology of the times; much of his research was spent characterizing what could be seen with a rudimentary microscope. Today, knowledge of biofilms in vitro has grown substantially with modern technology; however, knowledge of biofilms in vivo is still lacking. By forming a biofilm, the bacterium increases its persistence and chances of survival outside of the host as it awaits transmission to a new host. Most of the biomass of the enteric biofilm is a thick extracellular matrix composed of curli, cellulose, BapA, and extracellular DNA. Curli are functional amyloids with a rich β-sheet structure strikingly similar to pathological and immunomodulatory human amyloids such as β-amyloid, α-synuclein, and serum amyloid A. It is often debated if enteric bacteria express curli and form biofilms in the gut during infection due to the contrasting conditions between life outside versus inside the host. However, there is indirect evidence in support of curli expression in both humans and mice during infection. Furthermore, bacterial infections are important environmental triggers of autoimmunity and can contribute to autoimmune disease onset and severity. Here, we investigated the synthesis of curli in the gastrointestinal tract during Salmonella infection and how host-derived metabolites, like nitrate, are important regulators of biofilm expression in vivo. S. Typhimurium uses signals from the host environment to enhance its pathogen fitness during infection. Together, these studies provide a link between biofilm associated infections and autoimmune responses in the host. In these studies, we sought to show curli expression in the gastrointestinal tract of mice during S. Typhimurium infection. We determined that csgD was expressed and that S. Typhimurium produced curli during acute infections in susceptible mouse strains, both with and without streptomycin-pretreatment, and during chronic infections in genetically resistant mouse strains (NRAMP-positive). Oral infection with S. Typhimurium triggered an autoimmune response in mice, which correlated with in vivo synthesis of curli. The production of curli was associated with an increase in anti-dsDNA autoantibodies and joint inflammation in infected mice. This result demonstrated that a biofilm component expressed in the intestinal tract can trigger subsequent autoimmune responses within the host. Reactive arthritis (ReA) is a painful form of inflammatory arthritis that can develop following gastrointestinal infections. The mechanisms of how enteric pathogens induce ReA are not known; however, curli fibrils are highly conserved in enteric bacteria and could represent a link between enteric infections and ReA development. Oral exposure to purified curli alone was not enough to trigger an autoimmune response as curli needed to cross the epithelial barrier and become systemic to negatively impact the host. Overall, these data show a leaky gut induced by invasive pathogens like S. Typhimurium or dysbiosis common in autoimmune disease, allows curli-producing bacteria and curli complexes to translocate to sterile tissues and contribute to known complications of enteric infections and exacerbation of autoimmune diseases. Enteric bacteria like S. Typhimurium and Escherichia coli (E. coli), form biofilms during the environmental phase of their life cycle to protect the bacteria from various environmental insults. The in vitro conditions necessary for curli expression in the laboratory are well-known, but these conditions are not reproduced in the host leading many to theorize in vivo expression of curli is not possible; however, our research showed in vivo expression of curli in acute and chronic infection models of gastroenteritis and Typhoid fever in mice. This led us to our next question: what could be triggering curli expression in vivo? We began by considering the metabolites and byproducts produced within the inflammatory gut environment. A study by Smith et al. showed a mutation in anaerobic respiration, specifically those in the NarQ-NarP sensor response, decreased curli expression, suggesting involvement of electron transport chains in curli production. narQ is one of four genes responsible for regulating anaerobic respiratory gene expression in response to nitrate and nitrite. Furthermore, recent studies involving Burkholderia pseudomallei and uropathogenic E. coli revealed an association between nitrate reduction and biofilm dispersal and biofilm extracellular matrix (ECM) biosynthesis in vitro, respectively. Therefore, we investigated whether nitrate could serve as a signal for biofilm regulation during S. Typhimurium infection. We began our studies by investigating the effects of nitrate on S. Typhimurium biofilms in vitro and in vivo. We found that biofilms exposed to nitrate lost their integrity and instigated the dispersal of biofilm-associated bacteria. Furthermore, nitrate exposure facilitated a reduction in csgA gene expression. There was less CsgA protein expression and production in biofilms treated with nitrate than biofilms grown in the absence of nitrate. In contrast, flagellar motility was increased in the presence of nitrate. We discovered that the addition of nitrate to S. Typhimurium cultured under biofilm-forming conditions led to a dramatic decrease in cyclic di-GMP (c-di-GMP) levels. c-di-GMP inversely regulates biofilm formation and motility; high levels are associated with attachment to surfaces, ECM production, and biofilm formation and repression of motility, whereas low levels of c-di-GMP are associated with increased motility and virulence. Decreased levels of c-di-GMP led to decreased production of biofilm components and increased motility. Together, these data showed increased concentrations of nitrate influenced an intracellular molecular switch in c-di-GMP production, and, consequently, caused the bacterial community to shift from a sessile, biofilm former to a more motile, virulent state.