Responses to host amyloids and curli amyloid fibrils of Escherichia coli and Salmonella enterica serotype Typhimurium are mediated through Toll-like receptor (TLR) 2. Here we show that TLR2 alone was not sufficient for mediating responses to curli. Instead, transfection experiments with human cervical cancer (HeLa) cells and antibody-mediated inhibition of TLR signaling in human macrophage-like (THP-1) cells suggested that TLR2 interacts with TLR1 to recognize curli amyloid fibrils. TLR1/TLR2 also serves as a receptor for tri-acylated lipoproteins, which are produced by E. coli and other Gram-negative bacteria. Despite the presence of multiple TLR1/TLR2 ligands on intact bacterial cells, an inability to produce curli amyloid fibrils markedly reduced the ability of E. coli to induce TLR2-dependent responses in vitro and in vivo. Collectively, our data suggest that curli amyloid fibrils from enterobacterial biofilms significantly contribute to TLR1/TLR2-mediated host responses against intact bacterial cells.

SUMMARY Amyloid contributes to incapacitating disorders, such as Alzheimer’s disease, by eliciting local inflammation in the brain. However, little is known about how amyloid deposition elicits inflammation. Salmonella enterica serotype Typhimurium produces amyloid-like curli fibrils composed of the protein CsgA. We show that curli fibrils contributed to Nos2 expression in a mouse sepsis model by stimulating Toll-like receptor (TLR) 2. The TLR2 agonist activity of CsgA was markedly reduced by an amino acid substitution (N122A) that also lowered its amyloidogenicity. Synthetic peptides corresponding to residues 111-151 of CsgA or β-amyloid 1-42 from plaques of Alzheimer’s disease stimulated Nos2 production in macrophages and microglia cells through a TLR2-dependent mechanism. This activity was abrogated when a N122A substitution was introduced into the synthetic CsgA peptide. The induction of TLR2-mediated responses by bacterial and eukaryotic amyloids may provide a novel linchpin of pathogenesis, with implications for innate immunity and the immunopathogenesis of Alzheimer’s disease.

Salmonella enterica serotype Typhi (S. Typhi) causes an estimated 21 million annual cases of typhoid fever, a severe systemic infection resulting in 200,000 to 600,000 fatalities per year (5, 37). After ingestion, S. Typhi invades the intestinal mucosa, but symptoms develop only after an average incubation period of 2 weeks (23). The relatively long incubation period of typhoid fever suggests that S. Typhi can evade or suppress detection by the innate immune system during the initial stages of infection (26,28,33). However, the virulence mechanisms that enable S. Typhi to evade components of the innate immune system early after infection have long remained elusive.One arm of the innate immune system involved in detection of invasive microbes is the complement system (9, 38). Complement deposition on the bacterial cell surface and opsonophagocytosis can be prevented by capsular polysaccharides of invasive Gramnegative pathogens, including Neisseria meningitidis, Klebsiella pneumoniae, and Escherichia coli isolates associated with extraintestinal infections (1,15,16,35). S. Typhi produces the virulence (Vi) capsular polysaccharide (8), which is encoded by the viaB locus (17). The viaB locus is a 14-kb DNA region containing genes required for the regulation (tviA), the biosynthesis (tviBCDE), and the export (vexABCDE) of the Vi capsular polysaccharide (36). In S. Typhi, the role of the Vi capsular polysaccharide in reducing complement deposition has not been convincingly demonstrated. The viaB locus of S. Typhi is located on a 134-kb DNA region, termed Salmonella pathogenicity island 7 (SPI-7) (24). SPI-7 is genetically unstable and can be lost upon laboratory passage (3,22). Clinical S. Typhi isolates expressing the Vi capsular polysaccharide tend to bind less complement on their surface in vitro than clinical isolates lacking capsule expression (20). While this report concludes that the Vi capsular polysaccharide inhibits opsonophagocytosis, the evidence is not conclusive, because it is based on comparison of nonisogenic, clinical S. Typhi isolates. Genetic differences between these clinical isolates were not defined but likely included the entire SPI-7 region. Additionally, the in vivo relevance of phenotypes attributed to the Vi capsular polysaccharide remains to be established using animal models.With the exception of higher primates, vertebrate hosts are resistant to infection with the human-adapted S. Typhi, which has prevented the use of animal models to investigate the in vivo relevance of results obtained using tissue culture. While mice orally inoculated with S. Typhi are not suited to study the development of typhoid fever, we reasoned that this animal model could be used for studying isolated steps during infection, provided that the relevant interactions were not species specific. Here we investigated the role of the Vi capsular polysaccharide on complement-mediated phagocytosis and its impact on bacterial clearance during an infection. MATERIALS AND METHODSBacterial strains and culture conditions. Vi...