Amyloids are protein aggregates characterized by their insolubility in detergents and ability to form fibrils. They are often associated with various diseases, including neurodegenerative disorders, type 2 diabetes and certain forms of cancer. Amyloids also play important roles in bacteria and different physiological processes in both lower and higher eukaryotes. Together with the laboratory of Prof. Y.O. Chernoff we have developed a comprehensive approach for screening new potentially amyloidogenic proteins. This involves using bioinformatics algorithms to predict protein amyloidogenicity and further verifying using a yeast model. We have created a yeast test system specifically designed to study changes in phenotype in genetically modified Saccharomyces cerevisiae strains [1]. This system involves the production of recombinant amyloidogenic proteins fused with reporter proteins Sup35N or YFP. Using yeast assay, we have investigated 22 human proteins that were predicted to be amyloidogenic by ArchCandy algorithm [2]. Currently, additional in vitro biochemical tests are underway with proteins that have shown the potential to form amyloids in yeast models. There are also plans to evaluate the amyloid-forming ability of specific human proteins in mammalian cell cultures. These various approaches appear to be enhancing our comprehension of the impact of amyloid formation in health and disease.
BACKGROUND: Retrotransposons are mobile genetic elements that replicate via reverse transcription and constitute substantial fractions of eukaryotic genomes; they are also considered evolutionary precursors of retroviruses, and can affect host fitness. Intriguingly, some retrotransposon proteins share motifs with amyloids—fibrous protein aggregates with cross-β architecture that readily self-assemble into polymeric structures and can, in some cases, self-propagate in an infectious manner (prions). AIM: To identify and characterize potential amyloid-forming regions within the integrase of Saccharomyces cerevisiae Ty1 retrotransposon, which mediates integration of transposon copies into the host genome. METHODS: Computational analysis of the Ty1 integrase sequence was performed with the ArchCandy algorithm to identify putative amyloidogenic motifs. To evaluate the amyloidogenic potential of candidate regions, we employed a yeast-based nucleation assay. Aggregation was visualized by expressing Ty1Int(AD)-GFP fusion constructs. Colocalization of Ty1 amyloidogenic domain with full length in yeast cells was evaluated by confocal microscopy. RESULTS: We identified and experimentally validated an amyloidogenic region within the Ty1 integrase, designated Ty1Int(AD). ArchCandy predicted the region with amyloidogenic potential, and these predictions were confirmed in a yeast prion-nucleation assay and by expression of the Ty1Int(AD)-GFP fragment, which formed detergent-resistant aggregates. Confocal microscopy showed co-localization of these aggregates with native Ty1 integrase fused with YFP, indicating recruitment of the full-length protein into inclusions. CONCLUSION: Together, these results identify a previously unrecognized amyloidogenic region within Ty1 integrase, possessing amyloid-like properties, and suggest that aggregation of this domain may regulate retrotransposon activity by altering integrase availability and/or function. This is therefore important for the design, optimization, and biosafety assessment of retrotransposon based vectors and other GMO constructs.