Mutants resistant to Bacillus thuringiensis Cry toxins (called Bre for Bt toxin resistant) have been isolated in C. elegans. These mutants identify five genes (
bre-1through bre-5), each of which is required for C. eleganssusceptibility to Cry5B, a member of the major family of insecticidal Cry proteins. Two of these genes,
bre-3and bre-5, have been identified, and their predicted functions have been investigated.
bre-5 encodes a putative glycosyltransferase, belonging to a family which incudes
beta-1,3 Gal- and
beta-1,3 GlcNAc transferases. Members of this family can be found in nematodes, insects, and mammals. The predicted product of
bre-3 is remarkably similar (60% identical) to a Drosophila protein of demonstrated importance for insect development, but unknown biochemical activity. BRE-3 shares distant kinship with other beta-glycosyltransferases from bacteria. Given that BRE-3 and BRE-5 resemble glycosyltransferases, we have investigated the possibility that these genes are involved in the biosynthesis of a specific glycoconjugate. To demonstrate genetically that these enzymes act in a common pathway, we have quantified resistance in
bre-3 and
bre-5 loss-of-function single and double mutant strains. Early results from an assay based on growth inhibition indicate that dose response curves for the single mutants, as well as the double mutant, are virtually superimposable, suggesting that these genes control a single pathway. Analysis of genetic mosaics, along with ectopic expression and immunostaining experiments demonstrate that
bre-3and bre-5 are necessary and sufficient in the cells of the gut epithelium to mediate susceptibility, and that their protein products localize to cytoplasmic puncta which may be in the secretory pathway. These results are consistent with a role for BRE-3 and BRE-5 in the synthesis of a glycan structure required for the binding of toxin to the apical surface of gut epithelial cells. The evidence obtained so far for the inability of
bre-3 and
bre-5mutant animals to bind toxin is based on the observation that these mutants are highly resistant to endocytosis of dye-labeled toxin into their gut cells, while normal animals exhibit rapid uptake of toxin under identical conditions. We have observed striking defects in the glycoconjugate profile of
bre-3 and
bre-5 mutants, which suggest an epistatic relationship between the genes, and are currently working to identify the precise substrates of BRE-3 and BRE-5 activity and their relationship to Bt toxin action.