We use protein engineering and directed evolution to develop specialized molecular machines to reverse the protein misfolding implicated in human disease. In neurodegenerative diseases such as Parkinson’s disease (PD), Alzheimer’s disease (AD), and Huntington’s disease (HD), specific proteins misfold to take on the amyloid conformation. While traditionally viewed as an intractable and highly toxic protein conformation, amyloid is not always toxic. In fact, yeast employ amyloid for beneficial purposes and have evolved pathways to construct and disassemble amyloid. We are interested in reformulating this pathway and applying it to target human disease. The protein controlling this pathway in yeast is Hsp104, and in yeast Hsp104 disaggregates proteins from amyloid fibrils, pre-amyloid oligomers, and disordered aggregates. However, Hsp104 has only limited activity against amyloid fibrils comprised of proteins such as α-synuclein and Aβ (implicated in PD and AD, respectively), as it does not ordinarily encounter these proteins. Thus, we aim to enhance Hsp104 activity and re-engineer Hsp104 substrate specificity. We use yeast as a model system because Hsp104 is present in yeast and powerful genetic tools are available to manipulate yeast. Also, expression of disease-associated substrates in yeast results in the accumulation of aggregated proteins that confer toxicity, which recapitulates their pathologies in human disease and has empowered the identification of genetic risk factors in humans. Using these models, we have developed a series of Hsp104 variants that potently disassemble TDP-43, FUS, and α-synuclein aggregates and amyloids that are implicated in amyotrophic lateral sclerosis (ALS) and PD. Our work is the first example of using protein engineering to restore proteostasis. The variants we have developed suppress the toxicity of these misfolded proteins and also clear preformed aggregates, allowing the substrates to return to their proper localization. Certain variants potently suppress neurodegeneration in a <i>C. elegans</i> PD model and reverse aggregation in mammalian fibroblasts. In addition to their possible therapeutic applications, these variants might be employed as chemical biology probes to study the mechanism of protein-misfolding disorders. Also by better understanding the mechanism by which Hsp104 disassembles amyloid, new strategies to eliminate amyloid might be developed.

Our first generation variants are highly promising, but require further tuning to improve their characteristics. Thus we employ protein engineering and directed evolution to tune the activity of Hsp104 and other chaperones in order to develop protein remodeling factors with desired properties. We predict that just as numerous other proteins have evolved from generalists to specialists over many years of evolution, we can evolve Hsp104 from a generalist to a specialized molecular machine. We are also interested in testing the variants in various model systems including <i>C. elegans, Drosophila,</i> and motor neurons derived from reprogrammed patient fibroblasts. Protein misfolding is implicated in numerous other diseases. Thus we are interested in developing yeast models for these disorders and developing Hsp104 variants to counter the misfolding of these substrates. Finally, we are interested in using pure protein biochemistry and structural biology to better understand Hsp104 structure and mechanism. Amyloid is incredibly stable, thus we seek to understand how Hsp104 disassembles amyloid.

• Lab Website