Worm Breeder's Gazette 13(1): 66 (October 1, 1993)

These abstracts should not be cited in bibliographies. Material contained herein should be treated as personal communication and should be cited as such only with the consent of the author.

Senile Worms?

Chris Link

Dept. of Biology, University of Denver, Denver, CO 80208

As I described in my poster at the Madison meeting, I am interested in using transgenic C. elegans animals to model amyloid diseases, such as Alzheimer's disease. The basic idea is to engineer worms to express amyloid-forming human proteins, hope these proteins do something to worms, then use classic suppressor genetics to identify biological factors that are involved in amyloid formation or toxicity. (Whether or not this is a dumb idea is unclear.) On the assumption that constitutive expression of these proteins could possibly have a dominant lethal effect (suboptimal for recovering transformants!), my initial constructs (primarily using Andy Fire's expression vectors) used heat shock and her-1 promoters. These promoters were used to drive an artificial signal peptide/ß1-42 minigene (derived from the human ß amyloid precursor protein) or, as a control, wild-type human transthyretin. I was unable to detect any phenotypes in animals transgenic for these constructs.

I have now done the experiments I should have done in the first place: express these proteins with a high-level promoter and see if any phenotypes result. I have recovered two independent transmitting lines with transthyretin driven by the unc-54 promoter/enhancer in Fire vector pDP30 .38.These animals stain strongly and fairly uniformly with anti-transthyretin antisera; some animals show intense coelomocyte staining. My interpretation of these results is that since this construct contains the natural signal peptide, it is being produced and secreted by the body wall muscle cells and scavenged by the coelomocytes. These transgenic animals have no obvious phenotypes. I have recovered 4 transmitting lines transgenic for the corresponding unc-54 /signalpeptide/ß1-42 minigene. At least two of these lines have a discernible phenotype: at 25°, V4 -V3of Rol progeny show a progressive paralysis as adults. All of these lines show muscle-specific deposits of anti-ß amyloid immunoreactivity; I do not have good enough data yet to determine if the sicker lines show more immunoreactivity.

I do not have enough experience with unc-54 constructs to know if the unc-54 /ß1-42results are exciting, or just what one expects when you express high levels of any generic heterologous protein. (Is a bogus phenotype better than no phenotype?) At present I have two candidate integrated lines generated by gamma irradiation; both are significantly sicker than the parental extrachromosomal array line, but neither have been out-crossed.