Investigating the aggregation of Alzheimer’s disease-associated proteins in S. cerevisiae

Abstract

Alzheimer’s disease (AD) is the most common type of dementia and is associated with roughly 500,000 new cases each year (2). AD is associated with the aggregation of two proteins in the brain, A-beta peptide (Aβ) and microtubule associated protein tau (MAPT). Aβ and MAPT are capable of adopting a cross-β fibrous protein structure, which can be reproduced and spread via nucleated polymerization and are termed amyloids. Despite such a broad biological impact of amyloids and prions, the mechanism of their initial formation in vivo remains a mystery. In this thesis, I will investigate proteins associated with AD, and the properties of these proteins that control their aggregation. Recent research has indicated that the U1 small nuclear ribonuclear protein 70 (U1-70k) can form detergent-insoluble aggregates in a manner specific to Alzheimer’s disease. U1-70k is strongly correlated with Aβ and tau, both proteins known to play a highly important role in the Alzheimer’s disease cascade and plaque formation. It has been shown that misfolded forms of U1-70k can sequester natively folded U1-70k proteins and cause them to form insoluble aggregates, a characteristic of amyloids (8). The mechanism behind this conversion remains elusive, however. Our research focuses on determining which domains and combinations of domains of the U1-70k protein are necessary for aggregation, and we also examine its interactions with Aβ. Through plasmid construction, expression, and observation under fluorescence microscopy (FM), we demonstrate that the N(1-99) domain alone cannot induce aggregation, but the C(182-437) domain, combined N and M domains, and M(100-181) domain are capable of inducing aggregation. Further SDD-AGE and Western blot analyses indicate that the aggregates formed by the C(182-437) domain are detergent-insoluble, while those formed by the N and M domains as well as the M(100-181) domain alone are detergent-soluble. This leads us to hypothesize that the aggregates formed by the M domain are reversible stress granules. Furthermore, the N and M domains also co-aggregate with Aβ, though the C(182-437) domain does not. We also examine tau’s suitability as a model in yeast for protein interactions and find that its aggregation is transformant-specific and cannot be cured by Hsp104, a heat shock protein found in yeast cells. We find that wild-type repeat domains of tau, the 244-372 amino acid region, aggregates are detergent-soluble.