Glucokinase (GK) is an enzyme that catalyzes the ATP-dependent phosphorylation of glucose to form glucose-6-phosphate, and it is a tightly regulated checkpoint in glucose homeostasis. The monomeric enzyme possesses a highly exotic kinetic profile, with a sigmoidal dependence on glucose, which has been the source of vigorous investigation and debate in the last several decades. This unique regulatory behavior can be thought of as a remarkable glucose sensor, which may result in hyperglycemia when it is not active enough and hypoglycemia when it is too active. This interdisciplinary study, which draws on small angle X-ray scattering (SAXS) integrated with atomistic molecular dynamics simulations and experimental glucose binding thermodynamics, I reveal the critical regulation of the glucose sensor is due to a solvent controlled switch. Moreover, this solvent controlled switch manifests a regulatory mechanism of GK; a specific local conformational change that leads to an enzyme structure that has a much more favorable solvation energy than most of the protein ensemble. These findings have direct implications for the design of small molecule GK activators as anti- diabetes therapeutics as well as for understanding how proteins can be designed to have built-in regulatory functions via solvation energy dynamics.

This examines the interplay of structure and local electrostatic fields in protein-protein and protein-solvent interactions. The partial charges of the protein amino acids and the polarization of the surrounding solvent create a complex system of electrostatic fields at protein-protein and protein-solvent interfaces. An approach incorporating vibrational Stark effect (VSE) spectroscopy, dissociation constant measurements, and molecular dynamics (MD) simulations was used to investigate the electrostatic interactions in these interfaces. Proteins p21Ras (Ras) and Rap1A (Rap) have nearly identical amino acid sequences and structures along the effector-binding region but bind with different affinities to Ral guanine nucleotide dissociation stimulator (RalGDS). A charge reversion mutation at position 31 alters the binding affinity of Ras and Rap with RalGDS from 0.1 [mu]M and 1 [mu]M, to 1 [mu]M and 0.5 [mu]M, respectively. A spectral probe was placed at various locations along the binding interface on the surface of RalGDS as it was docked with Ras and Rap single (position 30 or 31) and double mutants (both positions). By comparing the probes' absorption energies with the respective wild-type (WT) analogs, VSE spectroscopy was able to measure molecular-level electrostatic events across the protein-protein interface. MD simulations provided a basis for deconvoluting the structural and electrostatic changes observed by the probes. The mutation at position 31 was found to be responsible for both structural and electrostatic changes compared to the WT analogs. Furthermore, previous identification of positions N27 and N29 on RalGDS as "hot spots" that help discriminate between structurally similar GTPases was supported. The RalGDS probe-containing variants and three model compounds were placed in aqueous solvents with varying dielectric constants to measure changes in absorption energy. We investigated the ability of the Onsager solvent model to describe the solvent induced changes in absorption energy, while MD simulations were employed to determine the location and solvation of the probes at the protein-solvent interface. The solvent accessible-surface area, a measure of hydration, was determined to correlate well with the change in magnitude of the probe's absorption energy and the displaced solvent by the probe. / text