Abstract
Phage-encoded Shiga toxin (Stx) released by Shiga toxin-producing E. coli (STEC) can kill multiple eukaryotic bacterial predators, including Acanthamoeba castellanii, Tetrahymena thermophila and Caenorhabditis elegans. However, the impact of Stx type, Stx amount, and the serogroup of the E. coli on the effectiveness of this exotoxin are poorly understood. These factors impact the severity of Stx-mediated disease in humans and therefore, by studying their role in modulating predator–prey interactions, we may gain insight into how these virulence factors evolved to contribute to human pathogenicity. Herein, we investigated the effects of these factors on predator killing by measuring the efficiency with which five different hemolytic uremic syndrome (HUS)-causing STEC strains consume and/or kill A. castellanii and C. elegans. These strains express various combinations of Stx types and amounts and O-antigens. We found that variations in Stx types and amounts significantly affect the ability of a given bacterial strain to kill predator A. castellanii and C. elegans, with higher Stx1 titers (HUSEC 31 vs. 19) and the presence of Stx2 alone (HUSEC 20) correlating with significantly greater predator killing. These attributes also affect STEC pathogenicity in humans, suggesting that ecological selective pressures for anti-predator defense inadvertently drive the evolution of strains with higher virulence potential in humans.