The role of solvent in regulating protein condensation

Abstract

Protein self-assembly gives rise to concentrated liquid droplets, gels, aggregates and crystals which are increasingly implicated in cellular organisation and disease. In particular, membraneless organelles (MLOs), which form through liquid-liquid phase separation (LLPS) of intrinsically disordered protein (IDP) and nucleic acid complexes, have emerged as a key feature of cellular organisation. Despite their apparent ubiquity, the mechanisms underlying the regulation of MLO formation remains poorly understood. LLPS is driven by weak, attractive and short-ranged interprotein interactions, which includes contributions from protein-protein, protein-solvent and solvent-solvent interactions. MLOs often form in response to cellular stress such as changes in temperature, pH and ionic strength which suggests that these parameters help regulate LLPS by modulating the interprotein interactions. In this thesis, we explore how changes in the physiochemical environment tunes the interprotein interactions, with a particular focus on proteinsolvent and solvent-solvent interactions. These are often overlooked in the literature but are important for understanding the regulation of LLPS within a biological context. α-elastin is an IDP, which undergoes LLPS. We show that the LLPS of α-elastin, which exhibits a lower critical solution temperature (LCST), is sensitive to relatively small and physiologically relevant changes in the ionic strength and salt type. We reveal the existence of two distinct regimes, in which LLPS is first suppressed and then enhanced with increasing ionic strength. We have characterised the net interprotein interactions using light scattering and demonstrate that the remarkable sensitivity of the LLPS phase boundary to ionic strength is driven by the interplay between net weak protein-protein interactions and the thermodynamics of the solvent which we explain using the framework of hydrophobic hydration. It has been shown that BSA undergoes condensation driven by the binding of trivalent cations to the protein surface. This effect appears to be salt specific, yet the underlying drivers of this specificity or the nature of the condensed states formed are not well characterised. We study the condensation of BSA in the presence of three trivalent salts LaCl3, HoCl3 and YCl3. Using a combination of LLPS phase boundary measurements and microscopy, we show that the type of condensed state exhibited by BSA (LLPS, spinodal idecomposition or aggregation) is sensitive to the salt type and concentration and that the interprotein interaction strength is correlated with the energetic terms describing ion hydration in the bulk. We extend our study to incorporate lysozyme to assess the generality of the protein-trivalent cation interactions. Wealso assess the effectiveness of the hydrophobic hydration framework to explain how kosmotropic and chaotropic agents, namely alcohol and urea respectively, modulate LLPS for α-elastin. We found that ethanol enhanced LLPS in a ternary protein-solvent-ethanol system whereas in the quaternary system of protein-solvent-ethanol-salt, it suppressed LLPS. This work demonstrates that the interprotein interactions which drive LLPS are sensitive to the hydration of colsolvent and cosolutes, which can drive changes in the properties of water in the bulk and at the interface. Finally, we also investigate how the hydration of aliphatic alcohols (ethanol, methanol and isopropanol) influence the structural stability of BSA and lysozyme. We found that, despite the proteins exhibiting different structural stabilities in response to chemical or thermal denaturation, in alcohol solutions the protein unfolding is remarkably well correlated with the thermodynamics of the solvent.

Date of Award	20 Jan 2026
Original language	English
Awarding Institution	University of Bristol
Supervisor	Jennifer McManus (Supervisor)