Abstract
Amyotrophic lateral sclerosis (ALS) and Parkinson’s disease (PD) are neurodegenerative diseases associated with the build-up of misfolded protein aggregates. The treatments available for these protein-misfolding disorders are palliative, and no treatments address their underlying symptoms. The yeast protein Hsp104 acts as a disaggregase which can process substrates that need to be refolded/remodeled. While its activity against yeast substrates is robust, Hsp104 can only weakly solubilize certain aggregates and amyloid in human cells; however, it can be engineered to enhance its activity. Introduction of single missense mutations have been shown to potentiate Hsp104, allowing it to solubilize aggregates of proteins including TDP-43, FUS, which are implicated in ALS, as well as α-synuclein which is implicated in PD. While first-generation Hsp104 variants suppress misfolding in certain systems, they can lead to toxicity because they lack substrate specificity. A recent discovery shows that potentiated variants harboring mutations in nucleotide binding domain 1 (NBD1) are less likely to elicit off-target effects. Here, I describe our efforts to identify variants that have reduced off-target effects, by preferentially targeting specific substrates. We have constructed a saturation mutagenesis library of Hsp104 NBD1 variants, which we couple with a yeast-based selection strategy and next-generation sequencing, allowing us to assay the effects of all possible missense mutations. By screening in this way, we can assay domain-wide trends against the three different disease substrates. This comprehensive screen has revealed substrate-specific trends and has allowed us to generate new hypotheses about the basis for substrate specificity, while also giving us new insight into the structural basis for potentiation. Our study has uncovered four variants of particular interest. We find that they counter toxicity of TDP-43, FUS, and a-synuclein with a diminished temperature sensitive phenotype. Interestingly, these four variants have also shown reduced activity in thermotolerance assays. This suggests that the Hsp104 variants are more specifically targeting our proteins of interest as compared to the broader yeast proteome. Discovering novel Hsp104 variants with desirable phenotypes will guide the engineering of more precisely tuned protein disaggregases, which may hold promise as new therapeutic agents for neurodegenerative disorders. Our studies reveal insights into the basis for Hsp104 potentiation, providing us with a deeper understanding of the effects of potentiation on Hsp104 structure and how Hsp104 interacts with and targets substrates.
Committee Chair
Timothy Wencewicz
Committee Members
Alireza Ghanbarpour; Jennifer Heemstra; Meredith Jackrel; Yusuke Okuno
Degree
Doctor of Philosophy (PhD)
Author's Department
Chemistry
Document Type
Dissertation
Date of Award
4-27-2026
Language
English (en)
DOI
https://doi.org/10.7936/cg5n-f257
Author's ORCID
https://orcid.org/0000-0001-8933-6827
Recommended Citation
Miller, Karlie Roxanne, "Deep Mutational Scan of Hsp104 NBD1 Uncovers Substrate Selective Variants" (2026). Arts & Sciences Graduate Student Theses and Dissertations. 3735.
The definitive version is available at https://doi.org/10.7936/cg5n-f257