We have engineered transgenic C. elegans that express the human b-amyloid peptide, which appears to play a central role in the pathology of Alzheimer's disease. As described previously, these worms, which express the peptide under the control of the
unc-54 promoter, are unhealthy, have larval vacuoles, and exhibit a progressive paralysis. To better understand the molecular mechanism of b peptide toxicity, we would like to determine which, if any, of these phenotypes are specific to the expression of b peptide, and which are non-specific effects resulting from overexpression of a foreign peptide. To address this question we have sought to develop non-toxic b peptide variants, which should 'unmask' the non-specific phenotypes. We have specifically attempted to design b peptide variants that are incapable of aggregating into amyloid fibers, which have previously been reported to be directly correlated with b peptide toxicity. Single amino acid changes in the b(1-42) peptide have provided us with two interesting variants that both fail to produce amyloid aggregates: L17P and M35C. Intriguingly, multiple L17P transgenic lines all show the progressive paralysis observed in strains expressing wild-type b peptide. All M35C lines, however, are significantly healthier than the strains expressing wild-type b peptide. While the L17P variant may allow us to differentiate between b peptide's ability to form aggregates and its toxicity, the M35C variant is currently being used to differentiate between specific and non-specific toxicity resulting from b peptide expression. We have also isolated a second non-aggregating version of the b peptide, the single-chain b dimer. Like the M35C variant, this version of the b peptide does not aggregate and animals that express it are healthier than those expressing the wild-type b peptide. Quantitative immunoblot analyses on these strains demonstrate that the differences we see in aggregation of the variants are due to qualitative differences in the peptides and not quantitative differences in their levels of expression. These results also suggest that the progressive paralysis phenotype is specifically due to b peptide toxicity. We are currently using the non- or less-toxic b peptide variants in conjunction with the wild-type b peptide to look for molecular correlates of b peptide toxicity. We have preliminary results from immunoblots of Hsp16 expression that suggest that this heat shock protein is upregulated in response to b amyloid expression, and that this upregulation is significantly reduced in lines expressing the non-aggregating b dimer.