Preferential Binding Parameter Research Articles

Molecular mechanism of the common and opposing cosolvent effects of fluorinated alcohol and urea on a coiled coil protein.

Alcohols and urea are widely used as effective protein denaturants. Among monohydric alcohols, 2,2,2-trifluoroethanol (TFE) has large cosolvent effects as a helix stabilizer in proteins. In contrast, urea efficiently denatures ordered native structures, including helices, into coils. These opposing cosolvent effects of TFE and urea are well known, even though both preferentially bind to proteins; however, the underlying molecular mechanism remains controversial. Cosolvent-dependent relative stability between native and denatured states is rigorously related to the difference in preferential binding parameters (PBPs) between these states. In this study, GCN4-p1 with two-stranded coiled coil helices was employed as a model protein, and molecular dynamics simulations for the helix dimer and isolated coil were conducted in aqueous solutions with 2 M TFE and urea. As 2 M cosolvent aqueous solutions did not exhibit clustering of cosolvent molecules, we were able to directly investigate the molecular origin of the excess PBP without considering the enhancement effect of PBPs arising from the concentration fluctuations. The calculated excess PBPs of TFE for the helices and those of urea for the coils were consistent with experimentally observed stabilization of helix by TFE and that of coil by urea. The former was caused by electrostatic interactions between TFE and side chains of the helices, while the latter was attributed to both electrostatic and dispersion interactions between urea and the main chains. Unexpectedly, reverse-micelle-like orientations of TFE molecules strengthened the electrostatic interactions between TFE and the side chains, resulting in strengthening of TFE solvation.

Molecular dynamics simulations of K+–Cl− ion pair in polar mixtures of acetone and water: Preferential solvation and structural studies

Molecular dynamics simulation studies have been performed for a series of polar mixtures of acetone and water containing a K+–Cl− ion pair. We have obtained the potentials of mean force (PMFs) for 15 mixtures spanning the whole composition range from 0% to 100% acetone. The PMFs indicate the presence of a stable contact ion pair (CIP), a solvent assisted ion pair (SAIP) and a solvent separated ion pair (SSIP) in all the compositions with acetone mole fraction (xacetone)≤0.50. However, for xacetone>0.50 and xacetone≤0.90, PMFs suggest the presence of only a stable CIP and SAIP. For xacetone>0.90, PMFs show the existence of only a stable CIP. While the stability of CIPs in mixtures increases with xacetone, the reverse trend is observed for SAIPs. The determination of thermodynamic properties suggests that entropy favours the CIPs and SAIPs in all the mixtures. The analysis of radial distribution functions using Kirkwood–Buff (KB) integrals explains the preferential solvation of the ion pair. We observed that in all the mixtures, K+–Cl− ion pair is preferentially solvated by water. Running coordination numbers of solvent molecules around the ion pair and preferential binding parameters also support the above observations. The dynamical aspects have been explored by calculating self-diffusion constants and hydrogen bond dynamics.

Various Contributions to the Osmotic Second Virial Coefficient in Protein−Water−Cosolvent Solutions

An analysis of the cosolvent concentration dependence of the osmotic second virial coefficient (OSVC) in water-protein-cosolvent mixtures is developed. The Kirkwood-Buff fluctuation theory for ternary mixtures is used as the main theoretical tool. On its basis, the OSVC is expressed in terms of the thermodynamic properties of infinitely dilute (with respect to the protein) water-protein-cosolvent mixtures. These properties can be divided into two groups: (1) those of infinitely dilute protein solutions (such as the partial molar volume of a protein at infinite dilution and the derivatives of the protein activity coefficient with respect to the protein and water molar fractions) and (2) those of the protein-free water-cosolvent mixture (such as its concentrations, the isothermal compressibility, the partial molar volumes, and the derivative of the water activity coefficient with respect to the water molar fraction). Expressions are derived for the OSVC of ideal mixtures and for a mixture in which only the binary mixed solvent is ideal. The latter expression contains three contributions: (1) one due to the protein-solvent interactions B2(p-s), which is connected to the preferential binding parameter, (2) another one due to protein/protein interactions (B2(p-p)), and (3) a third one representing an ideal mixture contribution (B2(id)). The cosolvent composition dependencies of these three contributions were examined for several water-protein-cosolvent mixtures using experimental data regarding the OSVC and the preferential binding parameter. For the water-lysozyme-arginine mixture, it was found that OSVC exhibits the behavior of an ideal mixture and that B2(id) provides the main contribution to the OSVC. For the other mixtures considered (water-Hm MalDH-NaCl, water-Hm MalDH-(NH4)2SO4, and water-lysozyme-NaCl mixtures), it was found that the contribution of the protein-solvent interactions B2(p-s) is responsible for the composition dependence of the OSVC on the cosolvent concentration, whereas the two remaining contributions (B2(p-p)) and B2(id)) are almost composition independent.

Solubility and local structure around a dilute solute molecule in an aqueous solvent: From gases to biomolecules

A new approach regarding the solubility of various sparingly soluble solutes, such as proteins, drugs and gases, and the local structure around a solute molecule is presented. This approach is based on an expression for the activity coefficient derived through the Kirkwood–Buff fluctuation theory of solutions. First, an expression for the solubility of proteins in aqueous solutions in terms of the preferential binding parameter is derived and criteria for salting-out or salting-in by various cosolvents obtained. Second, the methodology developed for the solubility of proteins in water + cosolvent mixtures is extended to the solubility of sparingly soluble gases in the same kinds of solvents. The derived equation was successfully applied to the experimental data regarding the solubilities of oxygen, carbon dioxide and methane in water + sodium chloride. In addition, the excesses (or deficits) of water and sodium chloride molecules in the vicinity of a gas molecule have been calculated to conclude that the infinitely dilute solute gas molecules are preferentially hydrated.

Local Composition in the Vicinity of a Protein Molecule in an Aqueous Mixed Solvent

This paper is focused on the composition of a cosolvent in the vicinity of a protein surface (local composition) and its dependence on various factors. First, the Kirkwood-Buff theory of solution is used to obtain analytical expressions that connect the excess or deficit number of cosolvent and water molecules in the vicinity of a protein surface with experimentally measurable quantities such as the bulk concentration of the mixed solvent, the preferential binding parameter, and the molar volumes of water and cosolvent. Using these expressions, relations between the preferential binding parameter (at a molal concentration scale) and the above excesses (or deficits) are established. In addition, the obtained expressions are used to examine the effect of the nonideality of the water + cosolvent mixtures and of the molar volume of the cosolvent on the excess (or deficit) number of cosolvent molecules in the vicinity of the protein surface. It is shown that at least for the mixed solvents considered (water + urea and water + glucose) the nonideality of the mixed solvent is not an important factor in the local compositions around a protein molecule and that the main contribution is provided by the nonidealities of the protein-water and protein-cosolvent mixtures. Special attention is paid to urea as cosolvent, because urea is one of only a few compounds with a concentration at the protein surface larger than its concentration in the bulk. The composition dependence of the excess of urea around a protein molecule is calculated for the water + lysozyme + urea mixture at pH = 7.0 and 2.0. At pH = 7.0, the excess of urea becomes almost composition independent at high urea concentrations. Such independence could be explained by assuming that urea totally replaces water in some areas of the protein surface, whereas on the remaining areas of the protein surface both water and urea are present with concentration comparable to those in the bulk. The Schellman exchange model was used to relate the preferential binding parameter in water + lysozyme + urea mixtures to the urea concentration.

Stability of proteins in the presence of carbohydrates; experiments and modeling using scaled particle theory

The effects of sucrose and fructose on the free energy of unfolding, Δ G N → D , and on the change in hydrodynamic radius, R H, upon unfolding were measured for RNase A and α-lactalbumin. Recently we analyzed the results for RNase A and showed that the effects of the carbohydrates on the protein's thermal stability can be accurately accounted for by scaled particle theory (SPT), and are thus largely entropic in nature. In this paper we extend this analysis to α-lactalbumin and demonstrate the generality of this finding. We also investigate the relationship between SPT and the thermodynamic formalism of preferential interactions. The preferential binding parameters calculated using SPT are in excellent agreement with experimentally measured values available in the literature. This agreement is expected to hold as long as enthalpic interactions between the cosolute and the protein are not important, as appears to be the case here. Finally we use the experimental data and SPT to calculate the change in the number of sugar molecules excluded from the protein surface during unfolding from knowledge of the preferential binding parameter for the native and denatured state of the protein.

Preferential hydration and solubility of proteins in aqueous solutions of polyethylene glycol

This paper is focused on the local composition around a protein molecule in aqueous mixtures containing polyethylene glycol (PEG) and the solubility of proteins in water + PEG mixed solvents. Experimental data from literature regarding the preferential binding parameter were used to calculate the excesses (or deficits) of water and PEG in the vicinity of β-lactoglobulin, bovine serum albumin, lysozyme, chymotrypsinogen and ribonuclease A. It was concluded that the protein molecule is preferentially hydrated in all cases (for all proteins and PEGs investigated). The excesses of water and deficits of PEG in the vicinity of a protein molecule could be explained by a steric exclusion mechanism, i.e. the large difference in the sizes of water and PEG molecules. The solubility of different proteins in water + PEG mixed solvent was expressed in terms of the preferential binding parameter. The slope of the logarithm of protein (lysozyme, β-lactoglobulin and bovine serum albumin) solubility versus the PEG concentration could be predicted on the basis of experimental data regarding the preferential binding parameter. For all the cases considered (various proteins, various PEGs molecular weights and various pHs), our theory predicted that PEG acts as a salting-out agent, conclusion in full agreement with experimental observations. The predicted slopes were compared with experimental values and while in some cases good agreement was found, in other cases the agreement was less satisfactory. Because the established equation is a rigorous thermodynamic one, the disagreement might occur because the experimental results used for the solubility and/or the preferential binding parameter do not correspond to thermodynamic equilibrium.

Relationship between preferential interaction of a protein in an aqueous mixed solvent and its solubility

The present paper is devoted to the derivation of a relation between the preferential solvation of a protein in a binary aqueous solution and its solubility. The preferential binding parameter, which is a measure of the preferential solvation (or preferential hydration) is expressed in terms of the derivative of the protein activity coefficient with respect to the water mole fraction, the partial molar volume of protein at infinite dilution and some characteristics of the protein-free mixed solvent. This expression is used as the starting point in the derivation of a relationship between the preferential binding parameter and the solubility of a protein in a binary aqueous solution. The obtained expression is used in two different ways: (1) to produce a simple criterion for the salting-in or salting-out by various cosolvents on the protein solubility in water, (2) to derive equations which predict the solubility of a protein in a binary aqueous solution in terms of the preferential binding parameter. The solubilities of lysozyme in aqueous sodium chloride solutions (pH = 4.5 and 7.0), in aqueous sodium acetate (pH = 8.3) and in aqueous magnesium chloride (pH = 4.1) solutions are predicted in terms of the preferential binding parameter without any adjustable parameter. The results are compared with experiment, and for aqueous sodium chloride mixtures the agreement is excellent, for aqueous sodium acetate and magnesium chloride mixtures the agreement is only satisfactory.

A protein molecule in an aqueous mixed solvent: Fluctuation theory outlook

In the present paper a procedure to calculate the properties of proteins in aqueous mixed solvents, particularly the excesses of the constituents of the mixed solvent near the protein molecule and the preferential binding parameters, is suggested. Expressions for the Kirkwood-Buff integrals in ternary mixtures and for the preferential binding parameter were derived and used to calculate various properties of infinitely dilute proteins in aqueous mixed solvents. The derived expressions and experimental information regarding the partial molar volumes and the preferential binding parameters were used to calculate the excesses (deficits) of water and cosolvent (in comparison with the bulk concentrations of protein-free mixed solvent) in the vicinity of ribonuclease A, ribonuclease T1, and lysozyme molecules. The calculations showed that water was in excess in the vicinity of ribonuclease A for water/glycerol and water/trehalose mixtures, and the cosolvent urea was in excess in the vicinity of ribonuclease T1 and lysozyme. The derivative of the activity coefficient of the protein with respect to the mole fraction of water was also calculated. This derivative was negative for the water/glycerol and water/trehalose mixed solvents and positive for the water/urea mixture. The mixture of lysozyme in the water/urea solvent is of particular interest, because the lysozyme at pH 7.0 is in its native state up to 9.3M urea, while at pH 2.0 it is denaturated between 2.5 and 5M and higher concentrations of urea. Our results demonstrated a striking similarity in the hydration of lysozyme at both pHs. It is worthwhile to note that the excesses of urea were only weakly composition dependent on both cases.

Unilamellar DMPC Vesicles in Aqueous Glycerol: Preferential Interactions and Thermochemistry

Glycerol is accumulated in response to environmental stresses in a diverse range of organisms. Understanding of favorable in vivo effects of this solute requires insight into its interactions with biological macromolecules, and one access to this information is the quantification of so-called preferential interactions in glycerol-biopolymer solutions. For model membrane systems, preferential interactions have been discussed, but not directly measured. Hence, we have applied a new differential vapor pressure equipment to quantify the isoosmotic preferential binding parameter, Γ μ1 , for systems of unilamellar vesicles of DMPC in aqueous glycerol. It is found that Γ μ1 decreases linearly with the glycerol concentration with a slope of −0.14 ± 0.014 per molal. This implies that glycerol is preferentially excluded from the membrane-solvent interface. Calorimetric investigations of the same systems showed that the glycerol-DMPC interactions are weakly endothermic, and the temperature of the main phase transition increases slightly (0.16°C per molal) with the glycerol concentration. The results are discussed with respect to a molecular picture which takes into account both the partitioning of glycerol into the membrane and the preferential exclusion from the hydration layer, and it is concluded that the latter effect contributes about four times stronger than the former to the net interaction.

Link between protein-solvent and weak protein-protein interactions gives insight into halophilic adaptation.

Malate dehydrogenase (Hm MalDH) from the extreme halophile Haloarcula marismortui is a very acidic protein with extensive ion binding properties. It is a good model for the study of solvation-solubility relationships. We measured the small-angle neutron or X-ray scattering profiles of folded and stable Hm MalDH at various protein concentrations and derived the second virial coefficients A(2). In NaCl, CsCl, KF, KCl, and NaCH(3)CO(2), A(2) values are positive, indicating globally repulsive protein-protein interactions. Below 1 M MgCl(2) and MgSO(4) or above 2 M (NH(4))(2)SO(4), A(2) rapidly decreases. From structure factor modeling with DLVO (Derjaguin, Landau, Verwey, and Overbeek)-like potentials, an effective diameter of 80-82 A is found for the protein particle in solution, compatible with its structural dimensions; the effective charge of the particle is undefined because of the high salt concentration. The strong variations of the protein-protein interaction are correlated to an attractive potential whose depth evolves with the salinity but in an opposite way in Mg salts and (NH(4))(2)SO(4). A repulsive Donnan term, corresponding to counterion dissociation, and an attractive term related to previously measured preferential salt binding parameters are discussed from well-established thermodynamics considerations and qualitatively account for the behavior of the protein-protein interactions in the various solutions. Because a solvation shell with a composition different from bulk induces protein-protein attraction, molecular adaptation to high salt would be directed to allow protein-salt interactions in order to avoid water or salt enrichment at the surface of the protein and thus preserve its solubility.

Solvent interactions of halophilic malate dehydrogenase.

Malate dehydrogenase from the extreme halophilic Haloarcula marismortui (Hm MalDH) is an acidic protein that is unstable below molar salt concentrations. The solvated folded protein was studied by small-angle neutron scattering in solvents containing salt: NaCl, NaCH(3)CO(2), KF, NH(4)Cl, NH(4)CH(3)CO(2), (NH(4))(2)SO(4), MgCl(2), and MgSO(4). It was found that the global solvent interactions depend mainly on the nature of the cation. Complementary mass density measurements in MgCl(2), NaCl, NaCH(3)CO(2), and (NH(4))(2)SO(4) allowed determining the partial molal volumes of the protein, which were found to increase slightly with the salt, and the preferential salt binding parameters for each solvent condition. These are strongly dependent on the cation type and salt concentration. Hm MalDH can be modeled as an invariant particle binding 4100 water molecules in MgCl(2) and 2000 +/- 200 in NaCl, NaCH(3)CO(2), or (NH(4))(2)SO(4). The number of salt molecules associated to the particle decreases from about 85 to 0 in the order MgCl(2) > NaCl = NaCH(3)CO(2) > (NH(4))(2)SO(4). Alternatively, we considered exchangeable sites for water and salt with the effects of solvent nonideality. It does not change the description of the solvent interactions. Solvent anions act on Hm MalDH stability through a limited number of strong binding sites, as those seen at the interfaces of Hm MalDH by crystallography. Cations would act through some strong and numerous weak binding sites defined on the folded protein, in possible addition to nonspecific hydration effects.

Thermodynamic relationships between protein-solvent and protein-protein interactions.

How the solvent modulates the weak inter-particle interactions in solution and affects macromolecule solubility is not yet understood. Well-established thermodynamic relationships link second virial coefficient and preferential solute binding parameter. We present the meaning of these thermodynamic parameters and the way to measure them. When a solvation shell has a composition different from the bulk solvent, a negative contribution is found in the second virial coefficient corresponding to an effective attraction between the macromolecules in solution. A quantitative evaluation using simple models of solvated particles in solution suggests that solvation could induce, at high or low concentration of a small molecule solute, attractive inter-particle interactions corresponding to favorable crystallization conditions.

Binding of small alcohols to a lipid bilayer membrane: does the partitioning coefficient express the net affinity?

The total vapor pressures at 26°C of binary (water–alcohol) and ternary (water–alcohol–vesicle) systems were measured for six short chain alcohols. The vesicles were unilamellar dipalmitoyl phosphatidylcholine (DMPC). The data was used to evaluate the effect of vesicles on the chemical potential of alcohols expressed as the preferential binding parameter of the alcohol–lipid interaction, Γ 23. This quantity is a thermodynamic (model-free) measure of the net strength of membrane–alcohol interactions. For the smaller investigated alcohols (methanol, ethanol and 1-propanol) Γ 23 was negative. This is indicative of so-called preferential hydration, a condition where the affinity of the membrane for water is higher than the affinity for the alcohol. For the longer alcohols (1-butanol, 1-pentanol, 1-hexanol) Γ 23 was positive and increasing with increasing chain length. This demonstrates preferential binding, i.e. enrichment of alcohol in the membrane and a concomitant depletion of the solute in the aqueous bulk. The measured values of Γ 23 were compared to the number of alcohol–membrane contacts specified by partitioning coefficients from the literature. It was found that for the small alcohols the number of alcohol–membrane contacts is much larger than the number of preferentially bound solutes. This discrepancy, which is theoretically expected in cases of very weak binding, becomes less pronounced with increasing alcohol chain length, and when the partitioning coefficient exceeds approximately 3 on the molal scale (10 2 in mole fraction units) it vanishes. Based on this, relationships between structural and thermodynamic interpretations of membrane partitioning are discussed.