Session S41: Physics of Proteins: Protein Structure and Interactions

Dynamics and mechanism of ultrafast water-protein interactions

Protein hydration is essential to protein stability, flexibility, dynamics and function. We have used a tryptophan scan with femtosecond spectroscopy to probe global surface water dynamics and characterize the coupled interactions of water and proteins. With extensive temperature-dependent studies, we found that water plays the dominant role to drive relaxation on the picosecond time scales. By measuring both water and protein relaxations, it shows that hydration water drives local protein fluctuations, a clear beta-relaxation, and such results are significant for the understanding of protein dynamics and functions.   

Effect of protein crystal hydration on side chain conformational heterogeneity

The structure of protein crystals is determined in part by water-mediated interactions involving both protein surface-ordered (hydration) and bulk water, and so is sensitive to the relative humidity of the environment. Monoclinic lysozyme provides a remarkable model for studying structural changes induced by dehydration, as it maintains excellent order for relative humidities (r.h.) down to 5{\%}, corresponding to solvent content of 9{\%} by volume, much smaller than the 88{\%} (22{\%} by volume) at which lysozyme loses its enzymatic activity. Although the main chain conformation does not change significantly, the effect of dehydration on side chain conformations has not been systematically studied. High resolution (1.1 to 1.7 A) structural data sets for monoclinic lysozyme at r.h. between 99{\%} and 11{\%} have been analyzed to identify major and minor side chain conformers at each humidity, and to map out how the side chain conformational ensemble evolves with hydration. Modest dehydration produces comparable overall effects to cooling to T$=$100 K, but with conformational changes largely confined to solvent-exposed residues. The largest side chain conformation changes occur at humidities that deplete water within the first two hydration shells.   

Determination of Protein Surface Hydration by Systematic Charge Mutations.

Protein surface hydration is critical to its structural stability, flexibility, dynamics and function. Recent observations of surface solvation on picosecond time scales have evoked debate on the origin of such relatively slow motions, from hydration water or protein charged sidechains, especially with molecular dynamics simulations. Here, we used a unique nuclease with a single tryptophan as a local probe and systematically mutated neighboring three charged residues to differentiate the contributions from hydration water and charged sidechains. By mutations of alternative one and two and all three charged residues, we observed slight increases in the total tryptophan Stokes shifts with less neighboring charged residue(s) and found insensitivity of charged sidechains to the relaxation patterns. The dynamics is correlated with hydration water relaxation with the slowest time in a dense charged environment and the fastest time at a hydrophobic site. On such picosecond time scales, the protein surface motion is restricted. The total Stokes shifts are dominantly from hydration water relaxation and the slow dynamics is from water-driven relaxation, coupled with local protein fluctuations.   

Biomolecular solvation study of proteins in liquid water by a wide range gigahertz-to-terahertz spectroscopy

Solvent dynamics within biomolecular solvation layers play a major role in enzyme activity, but obtaining an accurate and quantitative picture of solvent activity around proteins is challenging. Due to the strong absorption of water in the gigahertz-to-terahertz frequencies, it is challenging to study properties of the solvent dynamics as well as conformational changes protein in water. We have developed a highly sensitive dielectric gigahertz-to-terahertz frequency-domain spectroscopy system for probing the collective dynamics of proteins and solvent. Using this technique, we investigate the complex dielectric response of bovine serum albumin and lysozyme proteins in aqueous environment on a wide frequency range from 0.1 GHz up to 2 THz. We explore the conformation flexibility of proteins and compare the hydration dynamics around proteins to understand the effects of surface-mediated solvent dynamics, relationships among different measures of interfacial solvent dynamics, and protein-mediated solvent dynamics.   

Roles of urea and TMAO on the interaction between extended non-polar peptides

Urea and trimethylamine n-oxide (TMAO) are small molecules known to destabilize and stabilize, respectively, the structure of proteins when added to aqueous solution. To unravel the molecular mechanisms of these cosolvents on protein structure we perform explicit all-atom molecular dynamics simulations of extended poly-alanine and polyleucine dimers. We use an umbrella sampling protocol to compute the potential of mean force (PMF) of dimers at different concentrations of urea and TMAO. We find that the large non-polar side chain of leucine is affected by urea whereas backbone atoms and alanine's side chain are not. Urea is found to occupy positions between leucine's side chains that are not accessible to water. This accounts for extra Lennard-Jones bonds between urea and side chains that favors the unfolded state. These bonds compete with urea-solvent interactions that favor the folded state. The sum of these two energetic terms provide the enthalpic driving force for unfolding. We show here that this enthalpy correlate with the potential of mean force of poly-leucine dimers. Moreover, the framework developed here is general and may be used to provide insights into effects of other small molecules on protein interactions. The effect of the TMAO will be in the presentation.   

Theory Of Salt Effects On Protein Solubility

Salt is one of the major factors that effects protein solubility. Often, at low salt concentration regime, protein solubility increases with the salt concentration(salting in) whereas at high salt concentration regime, solubility decreases with the increase in salt concentration(salting out). There are no quantitative theories to explain salting in and salting out. We have developed a model to describe the salting in and salting out. Our model accounts for the electrostatic Coulomb energy, salt entropy and non-electrostatic interaction between proteins. We analytically solve the linearized Poisson Boltzmann equation modelling the protein charge by a first order multipole expansion. In our model, protein charges are modulated by the anion binding. Consideration of only the zeroth order term in protein charge doesn't help to describe salting in phenomenon because of the repulsive interaction. To capture the salting in behaviour, it requires an attractive electrostatic interaction in low salt regime. Our work shows that at low salt concentration, dipole interaction is the cause for salting in and at high salt concentration a salt-dependent depletion interaction dominates and gives the salting out. Our theoretical result is consistent with the experimental result for Chymosin protein   

Computational and Experimental Study of Neuroglobin and Mutants

\\ Neuroglobin (Ngb) is a hexacoordinated heme protein that is closely related to hemoglobin and myoglobin and normally found in the brain and nervous systems. It is involved in cellular oxygen homeostasis and reversibly binds to oxygen with a higher binding affinity than hemoglobin. To protect the brain tissue from hypoxic or ischemic conditions, Ngb increases oxygen availability. We have previously shown that a mutant form of Ngb reduces nitrite to nitric oxide 50x faster than myoglobin and 500x faster than hemoglobin. It also tightly binds to carbon monoxide (CO) with an association rate that is 500x faster than hemoglobin. To analyze the structure of neuroglobin and the characteristics causing these phenomena, we performed 3 sets of 1 microsecond molecular dynamic (MD) simulations of wild-type oxidized and reduced human Ngb and their C46A, C55A, H64L, and H64Q mutants. We also directly compare our MD simulations with time-resolved absorption spectroscopy. These studies will help identify treatments for diseases involving low nitric oxide availability and carbon monoxide poisoning.   

Analysis of Cavity Volumes in Proteins Using Percolation Theory

Molecular packing is studied in a diverse set of globular proteins in their native state ranging in size from 34 to 839 residues An new algorithm has been developed that builds upon the classic Hoshen-Kopelman algorithm for site percolation combined with a local connection criterion that classifies empty space within a protein as a cavity when large enough to hold a spherical shaped probe of radius, R, otherwise a microvoid. Although microvoid cannot fit an object (e.g. molecule or ion) that is the size of the probe or larger, total microvoid volume is a major contribution to protein volume. Importantly, the cavity and microvoid classification depends on probe radius. As probe size decreases, less microvoid forms in favor of more cavities. As probe size is varied from large to small, many disconnected cavities merge to form a percolating path. For fixed probe size, microvoid, cavity and solvent accessible boundary volume properties reflect conformational fluctuations. These results are visualized on three-dimensional structures. Analysis of the cluster statistics within the framework of percolation theory suggests interconversion between microvoid and cavity pathways regulate the dynamics of solvent penetration during partial unfolding events important to protein function.   

Characterization of DNA-protein interactions using high-throughput sequencing data from pulldown experiments

Methyl-binding domain (MBD) protein pulldown experiments have been a valuable tool in measuring the levels of methylated CpG dinucleotides. Due to the frequent use of this technique, high-throughput sequencing data sets are available that allow a detailed quantitative characterization of the underlying interaction between methylated DNA and MBD proteins. Analyzing such data sets, we first found that two such proteins cannot bind closer to each other than 2 bp, consistent with structural models of the DNA-protein interaction. Second, the large amount of sequencing data allowed us to find rather weak but nevertheless clearly statistically significant sequence preferences for several bases around the required CpG. These results demonstrate that pulldown sequencing is a high-precision tool in characterizing DNA-protein interactions.   

Modeling Adsorption Kinetics (Bio-remediation of Heavy Metal Contaminated Water).

My talk will focus on modeling the kinetics of the adsorption and filtering process using differential equations, stochastic methods, and recursive functions. The models have been developed in support of our interdisciplinary lab group which is conducting research into bio-remediation of heavy metal contaminated water via filtration through biomass such as spent tea leaves. The spent tea leaves are available in large quantities as a result of the industrial production of tea beverages. The heavy metals bond with the surfaces of the tea leaves (adsorption).