Vibrational solvatochromism of nitrile infrared probes: beyond the vibrational Stark dipole approach.

Systematic probing of local environments around biopolymers is important for understanding their functions. Therefore, there has been growing interest in in situ measurements of molecular granularity and heterogeneity through the systematic analysis of vibrational frequency shifts of carbonyl and nitrile infrared probes by vibrational Stark dipole theory. However, here we show that the nitrile vibrational frequency shift induced by its interaction with the surrounding molecules cannot be solely described by electric field-based theory because of the exchange-repulsion and dispersion interaction contributions. Considering a variety of molecular environments ranging from bulk solutions to protein environments, we explore the distinct scenarios of solute-environment contacts and their traces in vibrational frequency shifts. We believe that the present work could provide a set of clues that could be potentially used to design a rigorous theoretical model linking vibrational solvatochromism and molecular topology in complex heterogeneous environments.

Understanding and Manipulating Electrostatic Fields at the Protein-Protein Interface Using Vibrational Spectroscopy and Continuum Electrostatics Calculations.

Biological function emerges in large part from the interactions of biomacromolecules in the complex and dynamic environment of the living cell. For this reason, macromolecular interactions in biological systems are now a major focus of interest throughout the biochemical and biophysical communities. The affinity and specificity of macromolecular interactions are the result of both structural and electrostatic factors. Significant advances have been made in characterizing structural features of stable protein-protein interfaces through the techniques of modern structural biology, but much less is understood about how electrostatic factors promote and stabilize specific functional macromolecular interactions over all possible choices presented to a given molecule in a crowded environment. In this Feature Article, we describe how vibrational Stark effect (VSE) spectroscopy is being applied to measure electrostatic fields at protein-protein interfaces, focusing on measurements of guanosine triphosphate (GTP)-binding proteins of the Ras superfamily binding with structurally related but functionally distinct downstream effector proteins. In VSE spectroscopy, spectral shifts of a probe oscillator's energy are related directly to that probe's local electrostatic environment. By performing this experiment repeatedly throughout a protein-protein interface, an experimental map of measured electrostatic fields generated at that interface is determined. These data can be used to rationalize selective binding of similarly structured proteins in both in vitro and in vivo environments. Furthermore, these data can be used to compare to computational predictions of electrostatic fields to explore the level of simulation detail that is necessary to accurately predict our experimental findings.

Optimizing electrostatic field calculations with the Adaptive Poisson-Boltzmann Solver to predict electric fields at protein-protein interfaces II: explicit near-probe and hydrogen-bonding water molecules.

We have examined the effects of including explicit, near-probe solvent molecules in a continuum electrostatics strategy using the linear Poisson-Boltzmann equation with the Adaptive Poisson-Boltzmann Solver (APBS) to calculate electric fields at the midpoint of a nitrile bond both at the surface of a monomeric protein and when docked at a protein-protein interface. Results were compared to experimental vibrational absorption energy measurements of the nitrile oscillator. We examined three methods for selecting explicit water molecules: (1) all water molecules within 5 Å of the nitrile nitrogen; (2) the water molecule closest to the nitrile nitrogen; and (3) any single water molecule hydrogen-bonding to the nitrile. The correlation between absolute field strengths with experimental absorption energies were calculated and it was observed that method 1 was only an improvement for the monomer calculations, while methods 2 and 3 were not significantly different from the purely implicit solvent calculations for all protein systems examined. Upon taking the difference in calculated electrostatic fields and comparing to the difference in absorption frequencies, we typically observed an increase in experimental correlation for all methods, with method 1 showing the largest gain, likely due to the improved absolute monomer correlations using that method. These results suggest that, unlike with quantum mechanical methods, when calculating absolute fields using entirely classical models, implicit solvent is typically sufficient and additional work to identify hydrogen-bonding or nearest waters does not significantly impact the results. Although we observed that a sphere of solvent near the field of interest improved results for relative field calculations, it should not be consider a panacea for all situations.

Optimizing electrostatic field calculations with the adaptive Poisson-Boltzmann Solver to predict electric fields at protein-protein interfaces. I. Sampling and focusing.

Continuum electrostatics methods are commonly used to calculate electrostatic potentials in proteins and at protein-protein interfaces to aid many types of biophysical studies. Despite their ubiquity throughout the biophysical literature, these calculations are difficult to test against experimental data to determine their accuracy and validity. To address this, we have calculated the Boltzmann-weighted electrostatic field at the midpoint of a nitrile bond placed at a variety of locations on the surface of the protein RalGDS, both in its monomeric form as well as when docked to four different constructs of the protein Rap, and compared the computation results to vibrational absorption energy measurements of the nitrile oscillator. This was done by generating a statistical ensemble of protein structures using enhanced molecular dynamics sampling with the Amber03 force field, followed by solving the linear Poisson-Boltzmann equation for each structure using the Applied Poisson-Boltzmann Solver (APBS) software package. Using a two-stage focusing strategy, we examined numerous second stage box dimensions, grid point densities, box locations, and compared the numerical result to the result obtained from the sum of the numeric reaction field and the analytic Coulomb field. It was found that the reaction field method yielded higher correlation with experiment for the absolute calculation of fields, while the numeric solutions yielded higher correlation with experiment for the relative field calculations. Finer grid spacing typically improved the calculation, although this effect was less pronounced in the reaction field method. These sorts of calculations were also very sensitive to the box location, particularly for the numeric calculations of absolute fields using a 10(3) Å(3) box.

Role of electrostatics in differential binding of RalGDS to Rap mutations E30D and K31E investigated by vibrational spectroscopy of thiocyanate probes.

The human protein Rap1A (Rap) is a member of the Ras superfamily of GTPases that binds to the downstream effector Ral guanine nucleotide dissociation stimulator (RalGDS). Although Ras and Rap have nearly identical amino acid sequences and structures along the effector binding surface, the charge reversal mutation Rap K31E has previously been shown to increase the dissociation constant of Rap-RalGDS docking to values similar to that of Ras-RalGDS docking. This indicates that the difference in charge at position 31 could provide a mechanism for Ral to distinguish two structurally similar but functionally distinct GTPases, which would be of vital importance for appropriate biological function. In this report, vibrational Stark effect spectroscopy, dissociation constant measurements, and molecular dynamics simulations were used to investigate the role that electrostatic field differences caused by the charge reversal mutation Rap K31E play in determining the binding specificity of RalGDS to Rap versus Ras. To do this, six variants of RalGDS carrying a thiocyanate electrostatic probe were docked with three Rap mutants, E30D, K31E, and E30D/K31E. The change in absorption energy of the thiocyanate probe caused by RalGDS docking to these Rap variants was then compared to that observed with wild-type Ras. Three trends emerged: the expected reversion behavior, an additive behavior of the two single mutations, and cancelation of the effects of each single mutation in the double mutant. These observations are explained with a physical model of the position of the thiocyanate probe with respect to the mutated residue based on molecular dynamics simulations.