Using deep mutational scanning to study protein function and disease

Abstract

Hsp90 is a crucial molecular chaperone that regulates proteostasis by facilitating the refolding and activation of various signaling proteins. Its importance in protein folding and activation has made it a potential target for the treatment of cancer and neurodegenerative diseases. However, due to the diverse sequence space of Hsp90's clients, it has been challenging to characterize the determinants driving Hsp90 function and client recognition. Recent advances in DNA sequencing technology have allowed for multiplexed assays capable of studying thousands of variants and cells in a single experiment. In this thesis, I review the deep mutational scanning technique and its most recent advancements in protein science. Deep mutational scanning can further our understanding of protein biochemistry by generating comprehensive maps of effects of mutations on protein function. To demonstrate its utility, I use deep mutational scanning to investigate the molecular determinants of Hsp90's recognition of Src kinase, a model proto-oncogene. Through this work, I identify novel structural hotspots on the catalytic domain that drive Src's dependence on Hsp90. This study proposes a new model of Hsp90 client recognition, which may serve as a framework for a unified model of Hsp90 chaperoning of client kinases. Overall, this research showcases the potential of deep mutational scanning in furthering our understanding of protein biochemistry and the mechanisms underlying Hsp90 client recognition.