Paralysis and killing of Caenorhabditis elegans by enteropathogenic Escherichia coli requires the bacterial tryptophanase gene

Abstract

Pathogenic Escherichia coli, including enteropatho- genic E. coli (EPEC), enterohaemorrhagic E. coli (EHEC), enteroinvasive E. coli (EIEC) and enterotoxi- genic E. coli (ETEC) are major causes of food and water-borne disease. We have developed a genetically tractable model of pathogenic E. coli virulence based on our observation that these bacteria paralyse and kill the nematode Caenorhabditis elegans. Paralysis and killing of C. elegans by EPEC did not require direct contact, suggesting that a secreted toxin medi- ates the effect. Virulence against C. elegans required tryptophan and bacterial tryptophanase, the enzyme catalysing the production of indole and other mole- cules from tryptophan. Thus, lack of tryptophan in growth media or deletion of tryptophanase gene failed to paralyse or kill C. elegans. While known tryptophan metabolites failed to complement an EPEC tryptopha- nase mutant when presented extracellularly, comple- mentation was achieved with the enzyme itself expressed either within the pathogen or within a coc- ultured K12 strains. Thus, an unknown metabolite of tryptophanase, derived from EPEC or from commen- sal non-pathogenic strains, appears to directly or indirectly regulate toxin production within EPEC. EPEC strains containing mutations in the locus of enterocyte effacement (LEE), a pathogenicity island required for virulence in humans, also displayed attenuated capacity to paralyse and kill nematodes. Furthermore, tryptophanase activity was required for full activation of the LEE1 promoter, and for efficient formation of actin-filled membranous protrusions (attaching and effacing lesions) that form on the sur- face of mammalian epithelial cells following attach- ment and which depends on LEE genes. Finally, several C. elegans genes, including hif-1 and egl-9, rendered C. elegans less susceptible to EPEC when mutated, suggesting their involvement in mediating toxin effects. Other genes including sek-1, mek-1, mev-1, pgp-1,3 and vhl-1, rendered C. elegans more susceptible to EPEC effects when mutated, suggest- ing their involvement in protecting the worms. More- over we have found that C. elegans genes controlling lifespan (daf-2, age-1 and daf-16), also mediate sus- ceptibility to EPEC. Together, these data suggest that this C. elegans/EPEC system will be valuable in elu- cidating novel factors relevant to human disease that regulate virulence in the pathogen or susceptibility to infection in the host.