Potentiated Hsp104 variants suppress the toxicity of most overexpressed dosage-sensitive yeast genes

Maintenance of optimal gene expression levels is critical for cell viability and homeostasis. However, misregulation of gene expression can and regularly occur. One type of detrimental misregulation involves overexpression of a single gene that can cause organismal death is dosage sensitivity, which is often due to increased concentration of the protein encoded by the gene. Deleterious increases in the expression of specific proteins are associated with various neurodegenerative diseases such as Parkinson’s and Alzheimer’s Diseases as well as other cellular maladies including various cancers and Down Syndrome. In yeast, it has been estimated that ~20% of genes are toxic when overexpressed. The physicochemical properties and function of a protein seem to dictate whether it will be toxic upon overexpression. However, the mechanism by which individual proteins become toxic when overexpressed is typically unclear, which complicates the development of agents that counter toxicity of diverse dosage-sensitive genes. The overarching goal of this thesis was to rationally engineer a ‘buffer’ that universally mitigates the toxicity of dosage-sensitive genes. To meet this goal, we turned to Hsp104, a hexameric, ring-shaped AAA+ ATPase and protein-remodeling factor found in yeast, which protects yeast from toxicity associated with aggregated and misfolded proteins induced by chemical, heat, or age-related stress. An engineered variant of Hsp104, Hsp104 A503S, displayed potentiated activity and suppressed proteotoxicity of various neurodegenerative disease proteins, including TDP-43, FUS, and α-synuclein in yeast, whereas wild-type Hsp104 was ineffective. Inspired by this striking activity, we determined whether Hsp104A503S could combat the toxicity of diverse yeast dosage-sensitive genes. Surprisingly, Hsp104 A503S suppressed the toxicity of nearly 98% of dosage-sensitive genes tested, whereas wild-type Hsp104 rescued none. Expression of Hsp70- or Hsp90-class chaperones also failed to suppress toxicity of the majority of dosage-sensitive genes. To achieve this broad rescue of dosage-sensitive genes, Hsp104 A503S required critical tyrosines in pore-loops that engage substrate during protein remodeling and translocation across the central channel of Hsp104. Moreover, ATPase activity at NBD1 or NBD2 was required for Hsp104 A503S to alleviate toxicity of dosage-sensitive genes. Rescue of toxicity by Hsp104A503S was not typically due to decreases in toxic protein expression or disaggregation of amyloid. In addition, neither autophagy nor proteasome activity was required for Hsp104A503S to rescue the toxicity of dosage-sensitive genes. Rather, Hsp104A503S effectively prevented the formation of labile, SDS-soluble aggregates, which correlated with alleviation of toxicity. With null mutants, we established that the intrinsic function of several dosage-sensitive kinases and phosphatases was crucial for overexpression toxicity. In vitro functional assays with Ppz1 (a dosage-sensitive protein phosphatase), indicated the phosphatase activity was reduced by Hsp104A503S and not by Hsp104. Lastly, we demonstrated that Hsp104A503S suppressed the toxicity of the potent oncogenic kinase, v-Src, in yeast, decreasing protein levels and kinase activity in yeast. Thus, we suggest that in addition to preventing formation of labile, SDS-soluble aggregates Hsp104A503S can also suppress dosage sensitivity by directly unfolding or otherwise deactivating toxic protein such as Ppz1 and v-Src. These studies establish that potentiated protein-remodeling factors like Hsp104A503S can serve as a powerful buffer that mitigates the toxicity of nearly all dosage-sensitive yeast genes