Search

Bacillus cereus is an opportunistic pathogen that can cause emetic or diarrheal foodborne illness. Previous studies have identified multiple pathogenic B. cereus strains and characterized a variety of virulence factors. Here, we demonstrate that the virulence and lethality of B. cereus for mammalian cells and host animals involve the interaction of B. cereus flagellin proteins and the host-cell-surface-localized glycosphingolipid Gb3 (CD77, Galα1-4Galβ1-4Glcβ1-Cer). We initially found that B. cereus infection was less lethal for Gb3-deficiencient A4galt−/- mice than for wild-type mice. Subsequent experiments established that some factor other than secreted toxins must account of the observed differential lethality: Gb3-deficiencient A4galt−/- mice were equally susceptible to secreted-virulence-factor-mediated death as WT mice, and we observed no differences in the bacterial loads of spleens or livers of mice treated with B. cereus strain vs. mice infected with a mutant variant of incapable of producing many secreted toxins. A screen for host-interacting B. cereus cell wall components identified the well-known flagellin protein, and both flagellin knockout strain assays and Gb3 inhibitor studies confirmed that flagellin does interact with Gb3 in a manner that affects B. cereus infection of host cells. Finally, we show that treatment with polyclonal antibody against flagellin can protect mice against B. cereus infection. Thus, beyond demonstrating a previously unappreciated interaction between a bacterial motor protein and a mammalian cell wall glycosphingolipid, our study will provide useful information for the development of therapies to treat infection of B. cereus.