New Photo-Crosslinking Mass Spectrometry Approaches for the Study of Intrinsically Disordered Proteins

Abstract

Loss of proteostasis, the process of maintaining protein health and structure, is one of the hallmarks of aging. Loss of proteostasis can lead to the accumulation of aggregates and can lead to Parkinson’s, Alzheimer’s, and cataracts. Intrinsically disordered proteins play key roles in both the prevention of and aggregation in these aging-associated diseases. Small heat shock proteins (sHSPs) are one type of intrinsically disordered protein that help prevent protein aggregation. sHSPs are chaperones that prevent protein aggregation by binding aggregation-prone proteins before they aggregate. The disordered N-terminal region of sHSPs is known to be the chaperone active region. Still, not much is known about its interactions or how it prevents aggregation due to the difficulty in studying the disordered region. In this dissertation, photo-crosslinking mass spectrometry approaches that were designed to study intrinsically disordered proteins are described.Chapter 2 describes an informatic method for identifying residue-level crosslinks from the reagent benzolphenylalanine (BPA). BPA is a non-canonical amino acid that is incorporated site-specifically and reacts with any amino acids when treated with UV light. Because BPA can react with any amino acids, it can be used to identify residue-level crosslinks, but most informatic tools are not designed for residue-level crosslink identification. The informatic workflow in Chapter 2 makes it possible to identify residue-level BPA crosslinks on a large scale and could be applied to other photo-reactive amino acids. Chapter 3 describes a statistical workflow based on bootstrapping to compare qualitative data quantitatively. When applied to BPA crosslinking data, it determined that different sites of BPA incorporation yield significantly different crosslinks and that the crosslinks are not consistent solely with the reactivity of BPA. These findings establish that the crosslinks are probing structural features of the protein. The statistical method in Chapter 3 is highly versatile and could be applied to a wide variety of hypotheses. Chapter 4 describes the use of targeted-data-dependent acquisition to increase the number of identifications. This optimization of the mass spectrometry methods resulted in up to eight times more crosslinked peptide spectral matches from a single dataset. The increase in identifications from work in Chapter 4 increases the feasibility of applying the BPA crosslinking method to different systems, and similar techniques could be used to increase identifications for other crosslinking reagents. Chapter 5 describes the use of stable isotope labeling of amino acids in cell culture (SILAC) to quantify the depletion of peptides from crosslink formation. It describes best practices for sample preparation and data analysis to reduce the noise in the data, making it most feasible to quantify small degrees of change. Overall, the work in this dissertation improves the ease and feasibility of photo-crosslinking mass spectrometry, which has great implications for the study of highly biologically relevant intrinsically disordered proteins.