The functional protein state is a dynamic one, and this behaviour should be accurately reported by our biophysical toolset. Hydrogen-exchange mass spectrometry (HX-MS) is a powerful means of probing changes in the conformational ensemble of interacting proteins, and helps shape our understanding of structure and function relationships. This dissertation describes the development of novel tools for HX-MS, geared towards interacting with biologically relevant systems that are prohibitively large and complex for current structural approaches.
First, we introduced a set of standards to correct for dispensing during sample workup. This improved both systematic and random error, and increased the statistical power in differential experiments. Next, we scaled HX-compatible digestion strategies to determine how the analysis of complexes is limited. Digestion with traditional proteases was efficient, with modest coverage of a > 500 kDa sample. Our results suggested that the remaining limiting factors in the analysis of larger systems were related to chromatographic performance. We demonstrated the potential of specific prolyl-endoproteases to mitigate sample complexity. However, we discovered that peptide mapping was inadequate in all proteolytic approaches, and should be resolved if complete HX-MS datasets are desired. A proteomics-inspired nanoHX-MS system was next described. Resolution was improved by eliminating post-column band-broadening with our in-source configuration and a 50-fold improvement in sensitivity was achieved. We then investigated the validity of using overlapping peptides to increase structural resolution. Induction of secondary structure, and charge effects upon interaction with the chromatographic stationary phase perturbed exchange behaviour. Therefore, the fundamental assumption that residue exchange rates are independent of their parent peptide is invalid. These effects must be accounted for to obtain accurate modelling of site-resolved exchange with any high-resolution strategy. Finally, a multivariate strategy was tailored for large scale HX-MS screens. HX-MS readings were complemented by functional data (IC50) and used to characterize a panel of 18 compounds against Eg5, a mitotic kinesin. Canonical mechanisms were confirmed and roughly classified based on inhibitory strength. A modified binding mode and novel allosteric mechanism was discovered for Terpendole E, an inhibitor with activity in clinically relevant resistant Eg5 mutants.