The Influence of Cosolvent Interactions Upon Interfaces and Macromolecules
Restricted (Penn State Only)
- Author:
- Mandalaparthy, Varun
- Graduate Program:
- Chemistry
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- April 14, 2022
- Committee Members:
- Paul Cremer, Major Field Member
William Noid, Chair & Dissertation Advisor
Edward O'Brien, Major Field Member
Xiantao Li, Outside Unit & Field Member
Philip Bevilacqua, Program Head/Chair - Keywords:
- protein stability
polymers
proteins
osmolytes
thermodynamics
statistical mechanics
lattice models
protein folding
urea
tmao
simulations
theory
solute interactions
molecular dynamics
monte carlo - Abstract:
- Proteins are biological macromolecules that serve a wide variety of vital functions that often rely upon the protein conformation or conformational ensemble. Perturbations to this conformation (or to the relevant conformational ensemble) can have severe ramifications for biological survival. The thermodynamic stability of protein conformational equilibria is influenced not only by external factors, such as temperature, pressure, and pH, but also by the composition of the surrounding solution. In particular, direct and indirect interactions with cosolutes can significantly perturb protein equilibria. Proteins work in complex multicomponent solutions with many interacting cosolutes. It is often assumed that these cosolutes are somehow ``neutral'' towards the protein or that they exert additive effects upon protein stability. However, recent studies question the general validity and applicability of this assumption. In particular, experiments performed by the Cremer laboratory indicate that osmolyte mixtures can exert dramatic nonadditive effects upon the conformational transitions of thermoresponsive polymers and elastin-like polypeptides. In this dissertation, I investigate the origin of these nonadditive effects. In particular, I relate these nonadditive effects to interactions between cosolutes and investigate the conditions for which nonadditivity is observed. In the first parts of this dissertation, I develop a dilute solution theory to investigate the influence of neutral (i.e., non-ionic) cosolutes upon the properties of liquid-vapor interfaces. These relatively simple interfaces have been commonly employed as an ``ideal hydrophobe'' in order to understand the influence of solution conditions upon the hydrophobic forces that stabilize many folded proteins. Moreover, the Gibbs-Duhem equation implies a fundamental relationship between the influence of solution conditions upon interfacial properties and the conformational equilibria of dilute proteins. Specifically, the free energy of the macromolecular transition corresponds to the surface tension of the liquid-vapor interface, while the preferential interaction of solutes for the macromolecular conformations correspond to the surface excess. This dilute solution theory directly relates interactions between cosolutes to the surface tension and the excess of cosolutes adsorbed at the interface. In contrast to the vast majority of prior theories, we do not adopt the Bragg-Williams random mixing approximation that is invoked in, e.g., regular solution theory or Flory-Huggins theory. Rather, we employ a perturbative approach that is exact to the lowest order that treats cosolute interactions. We initially adopt a simple lattice model to validate the accuracy of this theory and to qualitatively explore its predictions. We then extend the formalism for much more realistic off-lattice models, i.e., classical particle-based models of molecules with both short-ranged dispersive and long-ranged electrostatic interactions that are commonly employed in molecular dynamics simulations of osmolytes in aqueous solutions. These studies suggest that cosolutes may be classified into three distinct categories- surfactants, depletants, and surface-neutral molecules- based upon their affinity for the air/water interface. Molecules in different categories reflect cosolute interactions in qualitatively different manners. Our theory identifies common regimes in which mixed cosolutes can exert additive influences upon interfacial properties. This additivity can stem either from a lack of interactions between the cosolutes or from a cancellation between the interactions. This theory also predicts that, in many cases, the surface properties of dilute solutions are dominated by additive effects that reflect the intrinsic preferences of cosolutes. In such cases, nonadditive effects are either not observed at all or they are not observed until relatively high concentrations. However, for surface-neutral molecules, the nonadditive effects due to cosolute interactions can be observed even at low concentrations. Experimental measurements suggest that, while betaine is a depletant, TMAO, urea, and proline are all examples of surface-neutral molecules. Our theory reveals an important distinction between the intrinsic and effective preferences of cosolutes for the interface. In particular, this theory predicts that cosolute-cosolute interactions can convert intrinsic surfactants to weak depletants and, conversely, convert intrinsic depletants to weak surfactants. In some cases, these interactions can qualitatively reverse the influence of solutes upon interfacial properties and even flip the sign of the solute surface excess. Our numerical simulations validate this predicted transition for both lattice and off-lattice models. Moreover, for the class of neutral solutes considered in this work, the dilute solution theory appears valid out to surprisingly high concentrations, i.e., out to concentrations of several molar. In the second part of this dissertation, I employ this dilute solution theory to interpret the experimental results from the Cremer laboratory that motivated this dissertation. In this case, I consider the influence of cosolute-cosolute interactions upon the conformational transition of a single macromolecule. This theory can accurately model the influence of cosolvent interactions upon the measured transition temperature. In particular, the theory explains the surprising observation that stabilizers (e.g., TMAO) can significantly reduce the efficacy of destabilizers (i.e., urea).