The functional aspects of proteins are largely dictated by highly selective protein- protein and protein-ligand interactions, even in situations of high structural homology, where electrostatic factors are the major contributors to selectivity. The vibrational Stark effect (VSE) allows us to measure electrostatic fields in complex environments, such as proteins, by the introduction of a vibrational chromophore whose vibrational absorption energy is linearly sensitive to changes in the local electrostatic field. The works presented here seek to computationally quantify electrostatic fields measured via VSE, with the eventual goal of being able to quantitatively predict electrostatic fields, and therefore Stark shifts, for any given protein-interaction. This is done using extensive molecular dynamics in the Amber03 and AMOEBA force fields to generate large ensembles the GTPase Rap1a docked to RalGDS and [superscript p]²¹Ras docked to RalGDS. We discuss how side chain orientations contribute to the differential binding of different mutations of Rap1a binding to RalGDS, where it was found that a hydrogen-bonding pocket is disrupted by the mutation of position 31 from lysine to glutamic acid. We then show that multi-dimensional umbrella sampling of the probe orientations yields a wider range of accessible structures, increasing the quality of the ensembles generated. A large variety of methods for calculating electrostatic fields are presented, with Poisson- Boltzmann electrostatics yielding the most consistent, reliable results. Finally, we explore using AMOEBA for both ensemble-generation as well as the electrostatic description of atoms for field calculations, where early results suggest that the electrostatic field due to the induce dipole moment of the probe is responsible for predicting qualitatively correct Stark shifts.
| Identifer | oai:union.ndltd.org:UTEXAS/oai:repositories.lib.utexas.edu:2152/31346 |
| Date | 17 September 2015 |
| Creators | Ritchie, Andrew William |
| Contributors | Webb, Lauren J. |
| Source Sets | University of Texas |
| Language | English |
| Detected Language | English |
| Type | Thesis, text |
| Format | application/pdf |
Page generated in 0.0752 seconds