Mass spectrometry (MS) has emerged as a powerful tool for epitope mapping, protein-ligand interaction, protein-protein interaction, aggregation, and effect of solution environment on protein conformation because they provide high-throughput data with relatively high structural resolution. Two popular MS-based approaches are hydrogen deuterium exchange-mass spectrometry (HDX-MS) and fast photochemical oxidation of proteins (FPOP), which complement classical biophysical and biochemical techniques in achieving higher structural resolution. The research presented in this dissertation is focused on the application of mass spectrometry-based footprinting techniques in characterizing the biophysical properties of Part I: pH-dependent conformation change of diphtheria toxin T domain (Chapters 2-4)); Part II: Ca2+ binding proteins and the role of Ca2+ regulation (Chapters 5-6); and Part III: protein-protein interaction including epitope mapping of IL-23 (Chapter 7) and Marburg virus protein VP24 (Chapter 8). Chapter 1 serves as an introduction to mass spectrometry instrumentation and standard LC-MS workflow. Two mass spectrometry based-footprinting techniques are introduced: (1) hydrogen deuterium exchange (HDX), and (2) fast photochemical oxidation of proteins (FPOP). Part I focuses on the development of pH-dependent HDX-MS for the conformation study of diphtheria toxin T domain. In Chapter 2, we describe the use pH-dependent HDX to study the pH-dependent conformation change of wild-type diphtheria toxin T domain monomer along its translocation pathway. In Chapter 3, we study the pH-dependent dissociation and reformation of T domain dimer. In Chapter 4, we apply the same method to a T domain mutant H223Q to further investigate the role of key histidine residues in triggering the conformation change. Part II focuses on the application of HDX mass spectrometry for the study of calcium binding proteins. Chapter 5 describes the Ca2+-binding property of ACaM and its Ca2+-regulated interaction with myosin VI. In chapter 6, HDX is also applied to an EF-hand Ca2+ binding protein, DREAM, for the study of its Ca2+ binding sites and stoichiometry. Part III of the dissertation focuses on the development and application of MS-based footprinting methods to investigate protein-protein interaction. Chapter 7 describes the methodology of fast photochemical oxidation of proteins (FPOP) for epitope mapping of IL-23 interacting a therapeutic antibody from Bristol-Myers Squibb. Chapter 8 discusses the use of HDX, FPOP, and NEM chemical labeling for the study of Marburg virus protein VP24 and its interaction with the host protein Keap1 Kelch domain. These seven studies on characterization of protein conformation dynamics, Ca2+ binding protein, and protein-protein interaction show the successful application of mass spectrometry in the structural study of large biomolecules.

Mass spectrometry (MS)-based protein footprinting characterizes protein structure and protein-ligand interactions by interrogating protein solvent-accessible surfaces by using chemical reagents as probes. The method is highly applicable to protein or protein-ligand complexes that are difficult to study by conventional means such as X-ray crystallography and nuclear magnetic resonance. In this dissertation, we describe the development and application of MS-based protein footprinting from three perspectives, including I) protein aggregation and amyloid formation (Chapter 2-3), II) protein-ligand interactions (Chapter 4-5), and III) in-cellulo structures and dynamic motion of membrane proteins (Chapter 6). Fast Photochemical Oxidation of Proteins (FPOP) is the main methodology implemented in the work presented in this dissertation. Chapter 1 provides an overview of FPOP and discusses its fundamentals as well as its important applications in both academic research and biotechnology drug development. In Part I, Chapter 2 covers the early method development of FPOP for monitoring amyloid beta (A[beta]) aggregation. In this work, we demonstrated the high sensitivity and spatial resolution of the method in probing the solvent accessibility of A[beta] at global, sub-regional, and some amino-acid residue levels as a function of its aggregation, and revealed A[beta] species at various oligomeric states identified by their characteristic modification levels. In Chapter 3, we extended the application of the platform to assess the effect of a putative polyphenol inhibitor on amyloid formation and to provide insights into the mechanism of action of the inhibitor in remodeling A[beta] aggregation pathways. In Part II, we evaluated different protein footprinting techniques, including FPOP, hydrogen-deuterium exchange (HDX), and carboxyl group footprinting, for probing protein-ligand (drug candidates) interaction in the context of a therapeutic development. Chapter 4 focused on protein-protein interaction by investigating the epitope of IL-6 receptor for two adnectins that have similar apparent biophysical properties. In Chapter 5, we probed the hydrophobic binding cavity of bromodomain protein for a small molecule inhibitor. This study serves as an example of interrogating protein-small molecule interactions. The two studies in Part II demonstrate the unique capabilities and limitations of protein footprinting methods in protein structural characterization. In Part III, we pushed the boundary of MS-based protein footprinting by applying the method to footprint live cells and investigate the dynamic structures/motion of membrane-transport proteins in their native cellular environment. We employed protein engineering, suspension cell expression and isotopic-encoded carboxyl group footprinting to identify salt bridges in two proteins, GLUT1 and GLUT5, that control their alternating access motions for substrate translocation. With functional analysis and mutagenesis, live-cell footprinting provides new insights into the transport mechanism of proteins in the major facilitator superfamily. The five studies in the dissertation demonstrate the powerful capability of MS-based protein footprinting in protein structural biology and biophysics research. The method also holds great potential in studying more complicated biological systems and solving demanding problems related to protein structure and properties.

Converting gene-sequence information into functional information about a protein is a major challenge of post-genomic biology. Proteins have a variety of functions from serving as catalysts to acting as structural components; all these functions are closely related to protein structure. The first step to understand protein function is often a structural study of that protein. Two major approaches, NMR spectroscopy and X-ray crystallography, can provide an atomic-level, 3D structural model of a protein. The applications of these high resolution approaches, however, are limited by protein size, conformational flexibility, and aggregation propensity. To obtain complementary structural information about proteins, a variety of approaches from traditional structural biology (e.g., circular dichroism, fluorescence spectroscopy) to new advances (e.g., computational prediction, protein footprinting) are required. Mass spectrometry (MS) has become an important tool for studying protein structure, dynamics, interactions, and function. In particular, detailed characterization of protein-ligand interactions is now possible, a critical step toward understanding biological function. Mass spectrometric analysis of protein structure can take two approaches. First, protein-ligand interactions can be probed by chemical labeling followed by MS analysis to determine the resulting mass shift (extent of labeling) and the location of the labeling. This approach in a titration format gives protein-ligand affinities. The labeling takes place in solution, where biochemistry occurs, and can be under physiological conditions, whereas the mass spectrometer is used for analysis typically by bottom-up proteomic strategies. In the other approach, protein assemblies can also be transferred directly into the gas phase and interrogated by MS to afford structural insights. One can view this is a top-down approach. The measurements refer to a gas-phase species, and that raises the question of whether the outcomes of the measurements have relevance to the structure and properties of proteins in solution or in a living system. Although there are differences in experimental format, results, and sensitivity between the two approaches of MS-based protein structural analysis, the similarity of those approaches must not be overlooked. All MS-based structural analyses rely heavily on the identification of peptides, purified protein species, or protein complexes. This analysis has been accelerated by the developments of MS instrumentation and methodology in protein analysis; the structural information provided by MS-based analysis is greatly facilitated by having a structural model of the protein. The integrated results from MS approaches, traditional structural biology approaches (e.g., NMR and X-ray), and computational modeling give more complete structural information of proteins than that from any one of the approaches alone. In the first part of thesis, we focus on the development and application of chemical-labeling methods (protein footprinting) in studies of protein conformation. In the second part, a combined top-down approach of native ESI and electron-capture dissociation (ECD) in FTICR MSis presented for structural studies of protein assemblies in the gas phase.

Integral mass spectrometry (MS) has emerged as an important tool for protein structural characterization. It readouts are a broad range of structural information, including stoichiometry, interactions, conformations and conformation change, and dynamics. Protein footprinting is a pivotal component in the intergral MS toolkit.My dissertation centers around the development and application of protein footprinting to characterize protein structure. It is divided into seven chapters. Chapter 1 serves as the introduction for integral mass spectrometry in structural proteomic. In Chapter 2, we extended the fast-photochemical oxidation of proteins (FPOP) platform by adding the trifluoromethyl radical (·CF3) as a new reagent. We discovered that ·CF3 footprint proteins in a complementary way as hydroxide radicals. The ·CF3 radical has exceptional reactivity, modifying 18 amino acids out of 20. Further studies demonstrate that it can report the conformational change between holo-myoglobin and apo-myoglobin and can define the topology of the VKOR membrane protein. This work bridges trifluoromethylation chemistry in materials and medicinal chemistry to that in structural biology. In Chapter 3, collaborated with Dr. Mark Chance's laboratory in Case Western Reserve University (CWRU) to apply ·CF3 chemistry on the synchrotron platform, which is the first platform that uses ·OH for protein footprinting. Synchrotron radiolysis generates ·CF3 in water media by ionizing water molecules to give ·OH. The ·CF3 shows complementary chemical reactivity with canonical ·OH labeling yet results in higher reactivity coverage. The ·CF3 reagent is the second footprinting reagent enabled by synchrotron since 1999. This work serves as a proof-of-concept to demonstrate that synchrotron platform is adaptable to other novel chemistries that can increase footprinting coverage. Further, taking advantage of X-ray irradiation, we achieved direct protein trifluoromethylation in the absence of metal catalysis or peroxide for the first time, with the synchrotron platform. In Chapter 4, we devloped a laser-mediated radical method for integral membrane protein (IMP) footprinting. Classical footprinting methods use hydrophilic reagents to label protein surfaces. IN so doing, we generate structural information by measuring the solvent accessibility of the backbone or side chains in aqueous media. Owing to the amphipathic nature of IMP, this new approach exploits the highly hydrophobic nature of perfluoroalkyl iodine together with tip sonication to ensure efficient labeling of a transmembrane domain (TM). The chemistry yields 100% reactivity coverage for tyrosine, and complete IMP labeling in a fast fashion. The resulting protein modification, which is resistant to hydrolysis, compatible with proteolysis, and amenable of tandem mass analysis, is appropriate for footprinting by bottom-up analysis. (Collaboration with Dr. Weikai Li from Wash U Medical School). In Chapter 5, we investigated an array of digestion conditions by using different combination of protease and additives to optimize the coverage of IMP digestion. IMPs are under-represented in conventional bottom-up proteomic analysis that generally favors soluble, abundant and easy-to-digest proteins. The new protrocol of IMP digestion significantly decreases our workload for sample preparation, allows us to avoid common contaminants that impair LC-MS, and generally yields >90% sequence coverage by generating peptides suitable for structural proteomic studies. Further, the deep analysis enable us to identify a "sweet spot" in the digestion protocol that may provide guidance to choose a suitable protease in structural proteomics in future. In Chapter 6, apart from methodology development, we used hydrogen/deuterium exchange mass spectrometry (HDX-MS) to characterize the binding interface for Mxra8-immune complex and Mxra8-chikungunya virus protein complex. The cell adhesion molecule Mxra8 is identified as a receptor for multiple arthritogenic alphaviruses such as chikungunya virus. We identified putative binding sites for eight anti-Mxra8 monoclonal antibodies (mAbs). HDX-MS enables us to classify the novel mAbs, predict their competing binding interface with chikungunya virus, and provide a molecular level explanation for the observation that mAbs can block the Chikungunya virus infection. From the HDX kinetic curves, we also observe that the mAbs have higher affinity than do Chikungunya virus proteins when binding with Mxa8. Finally, the HDX data help to assign the orientation of Mxra8 on the Cryo-EM structure of Chikungunya virus complex (Collaboration with Dr. Daved H. Fremont and Dr. Michael s Diamond, Wash U School of Medicine). In Chapter 7, we provide a conclusion for my dissertation. We will discuss challenges and opportunities for protein footprinting, and its role in the expanding toolkit of structural proteomics.

Mass spectrometry (MS) is an essential tool to study proteins whose structures are of great importance in biological systems. The primary structures of proteins can be determined by the powerful sequencing capabilities of MS. The recent advancements in instrumentation and methodology have made MS increasingly valuable in probing secondary, tertiary and quaternary structures, as well as binding strength, interfaces and in solution dynamics of proteins and protein complexes. Various protein footprinting techniques, including hydrogen-deuterium exchange (HDX) and fast photochemical oxidation of proteins (FPOP), encode structural information onto the protein molecule in different forms of modifications, and then MS is utilized to interpret the mass shifts resulted from modifications and extract the structural information. Protein footprinting coupled with bottom-up proteomics, which utilizes front-end LC separation and tandem mass spectrometry, has gained a solid ground in protein biophysics. On the other hand, opportunities emerge as native MS, ion-mobility separation, gas-phase activation and fragmentation techniques allow new approaches to be developed. In the first part of this dissertation, we describe epitope mapping of three malaria antigens (Plasmodium vivax Duffy binding protein in Chapter 2, Plasmodium vivax and falciparum cell-traversal protein for ookinetes and sporozoites in Chapter 6) and one flavivirus antigen (West Nile virus envelope protein domain III (DIII) in Chapter 4) by HDX in combination with bottom-up MS. We also report epitope mapping of DIII antigen by FPOP (Chapter 5). Challenged by highly disulfide-linked antigens, sample complexity and discontinuous epitopes with only a few residues each, we implemented immunoprecipitation, non-canonical quenching and digestion protocols to achieve complete sequence coverage and map the epitopes with high confidence and spatial resolution. In the second part (Chapter 3), we describe the usage of native MS and ion mobility to characterize antigen-antibody complexes formed by the Duffy binding protein antigen with various antibodies targeting different epitopes. The last part (chapter 7 and 8) describes the development an on-line HDX, native-spray platform in conjunction with top-down MS. The strategy is validated by determining the amide hydrogen exchange rates of a model peptide at the residue level. With evidence for adequate mixing efficiency, high sequence coverage, low hydrogen scrambling and capable data analysis, we applied the platform to study solution-phase amyloid beta 1-40 monomer structure by continuous-labeling and monitoring exchange kinetics and to probe the dimerization interfaces of human insulin by pulse-labeling experiment. These seven studies demonstrate the applications of the mature bottom-up and promising top-down MS on characterizing protein conformation and protein-protein interactions.

Most cellular processes occur through changes in protein conformation, interactions, and/or ligand binding. Molecular understanding of these events would ideally be achieved by probing them quantitatively and at high structural resolution in living cells, but this is not possible with existing structural techniques. Emerging technologies for high-throughput protein footprinting provide a possible approach in which a protein's per-residue solvent exposure reports on its local conformation, but the goals of comprehensive coverage of all amino acids, quantitative accuracy, and sufficient sensitivity remain challenging, even for measurements made on purified protein. I provide an overview of footprinting approaches in Chapter 1, presenting the possibilities enabled by the technology and describing the substantial hurdles that have stood in the way of it being a viable technology for in-cell analysis of protein conformation. In Chapter 2, I present an in-cell footprinting technique that can quantitatively monitor solvent accessibility at virtually all of a protein's residues using our cysTRAQ label, enabling sensitive detection of footprinted peptides by mass spectrometry. CysTRAQ footprinting exploits three technical innovations, the development and benchmarking of which are described in detail. The first innovation is the development of a shotgun footprinting approach that enables comprehensive coverage of a protein of interest. Using commercially-available arrayed oligonucleotides, we generate hundreds of single cysteine mutants in an expression plasmid in one day and pool the resulting cysteine libraries for in-cell labeling. Second, we designed mass-tagged cysteine alkylating agents that are cheap, compact, and cell-permeable. These cysTRAQ reagents enable precise kinetic footprinting analysis, employing a novel strategy for LC-MS quantification of modified cysteine probes in tandem mass spectra. Finally, we devised a strategy for > 300-fold purification of cysTRAQ-labeled peptides from the vast excess of unmodified peptides, improving both signal to background in isotope ratio data and amino acid coverage in target proteins. This enrichment is mediated by the cysTRAQ reagent's compact hydroxamate moiety, which we present as a novel affinity tag enabling purification by immobilized ytterbium affinity chromatography. Combining these technical advances, we demonstrate the use of cysTRAQ footprinting on a bacterial ribose-binding protein (RBP) in live cells. We describe two experimental formats suited to different types of biological questions: one in which cysTRAQ footprinting encodes the absolute solvent exposure at a given site on the protein and one in which it encodes the relative solvent exposure differences between two conformational states. We compared RBP's footprints in three different environments, finding that labeling rates on purified protein correlated well with rates in the E. coli periplasm (its native environment) and in the cytoplasm. These rates spanned more than three orders of magnitude, providing data on both exposed and buried regions of the protein. Importantly, periplasmic data from both the absolute and the relative cysTRAQ encoding schemes pinpointed RBP's dynamic ligand-binding interface--the residues with ribose-dependent changes in solvent exposure are located in the mouth and hinge regions of RBP previously implicated in ribose binding. This high-throughput approach provides quantitative measures of conformation at single amino acid resolution at more than 50% of the residues in the protein. We also demonstrate how footprinting can guide the design of real-time fluorescent sensors of conformational change. Finally, Chapter 3 describes some of the difficult questions that can be addressed using this novel approach to in-cell measurements of protein structure and reactivity.