Over-Oxidation of DJ-1 Leads to Loss of Function and Structural Integrity at Physiological Temperatures

Human DJ-1 is a ubiquitously expressed homodimer that is believed to play a role in several cellular processes, most notably protection against oxidative stress. However, a highly conserved cysteine residue, Cys106, can be oxidized to a sulfinate (-SO2-) form and over-oxidized to a sulfonate (-SO3-) form. This over-oxidized form has been shown to cause a loss of structural integrity at physiological temperatures. Furthermore, DJ-1 and its oxidative states may play significant roles in multiple disease pathways, such as Parkinson’s Disease, cancer, and emphysema. Characterizing the structural, dynamic and functional differences of DJ-1’s oxidative states may be vital in understanding the protein’s role in these diseases. Over the last decades, nuclear magnetic resonance (NMR) has emerged as one of the most powerful analytical tools for the characterization of biomolecular structure and dynamics. This dissertation describes the use of NMR to investigate the unique differences in dynamics between the oxidative states of DJ-1 at and below physiological temperatures. By implementing techniques such as chemical exchange saturation transfer (CEST), the global and per residue dynamics of DJ-1 at physiological temperature showed significant differences between the stability of the reduced and oxidized forms of DJ-1 and the over-oxidized form. Furthermore, drastic differences were observed between all three oxidative states at low temperatures, indicating that temperature plays a significant role in the protein’s stability. A computational method, molecular dynamics simulations, were performed to further validate the results seen in the experimental NMR data. Furthermore, DJ-1 has been identified as a potential therapeutic protein target for drug discovery as it pertains to Parkinson’s Disease. To determine potential chemical leads that interact with DJ-1, virtual screening and chemometric analysis were performed on large, drug-like compound libraries. Compounds with high binding affinity based on the virtual screen were experimentally tested to determine if the compounds were true binders to be used in further drug discovery analysis