Methods and Informatics to Analyze Intact Protein Sequence and Structure by Ion Mobility-Mass Spectrometry

Methods for rapid assessment of intact protein sequence and structure are increasingly important to understanding biology and treating disease. Mass spectrometry and ion mobility have emerged as effective tools for assessing protein sequence and structure, but face significant technical challenges in evaluating intact proteins and converting data from gas-phase ions to solution-relevant information. Protein complexes preserved intact for analysis by mass spectrometry typically generate low quality fragmentation, in part due to their relatively low charge. In Chapter 2, we develop a chemical modification method that affixes stable, intrinsically-charged reagents to proteins and demonstrate improvements in sequence coverage for protein complexes. In Chapter 3, we use the reagents developed in Chapter 2 to alter the competition between fragmentation pathways of intact proteins and demonstrate the immense capacity of protein ions to accommodate excess charge through charge solvation. We show that fragmentation can be directed to the site of fixed charge, a novel observation for intact, multiply-charged protein ions, and that mobile proton-mediated fragmentation can be restored by capping carboxylic acids, blocking the charge-remote pathway and resulting in improved sequence coverage. The conclusions reached in Chapters 2 and 3 depended on the development of software tools to analyze the complex data generated during fragmentation of intact proteins on an IM-MS platform. Ion mobility provides an additional dimension of separation when coupled to MS, enabling resolution of overlapping fragment ion signals that could not be resolved by MS alone. To utilize this additional information, we adapted tools being developed for bottom-up IM-MS proteomics, IMTBX and Grppr, to analyze intact protein fragmentation by IM-MS. IM-MS also has the potential to provide structural insight into intact proteins and complexes, including through the use of collision-induced unfolding (CIU). Similar to analysis of protein fragmentation by IM-MS, CIU experiments generate complex datasets requiring informatics to recover useful information about protein structure. In Chapter 5, we develop a software suite to provide automated annotation of gas-phase stability shifts and statistical classification of CIU data to support the development of high-throughput screening methods using CIU. We also utilize Gaussian fitting to distinguish and remove chemical noise from protein signal in membrane protein CIU to enable stability shift and classification analyses in these challenging datasets. In Chapter 6, we develop a framework to include all observed charge states in CIU analyses rather than a single charge state. Adducts associated with intact proteins can be lost as charged species, reducing the charge of the protein and contaminating nearby charge states through “charge stripping.” We characterize charge stripping of intact monoclonal antibodies and develop an algorithm to predict and remove the contaminating signals. The software tools for CIU data developed in Chapter 5 annotate only a single charge state of CIU data at a time. In Chapter 7, we expand the classification method of Chapter 5 to incorporate data from all charge states, or any other perturbation that alters the CIU pathway, into a single “multi-state” classifier. We are able to generate robust multi-state classifiers to distinguish kinase inhibitor binding modes and a highly similar innovator/biosimilar biotherapeutic pair, which had each proven challenging to differentiate using a single charge state. In total, the work presented in this thesis describes improved methods for analysis of protein complexes by IM-MS, including derivatization for improved sequence analysis and informatics for both sequencing and structural analyses.PHDChemistryUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttps://deepblue.lib.umich.edu/bitstream/2027.42/151656/1/dpolasky_1.pd