Computational Investigations of Water and Its Role in Biological Systems

Jeremy Palmer

Water is one of the most abundant and important substances on earth, playing a central role in processes ranging from global climate regulation to biological function. Despite its ubiquitous nature and centuries of scientific investigation, there are still major unanswered questions regarding water’s unusual physical properties and how these properties influence biological processes essential to life as we know it.  In my talk, I will discuss recent developments we have made using advanced computational techniques to investigate water’s interactions with biological systems and the low-temperature thermodynamic implications of its experimentally observed behavior.

The first part of my presentation will focus on exploring how proteins are affected by dehydration and subsequent rehydration. Understanding this process is essential to developing rational formulation strategies for biological therapeutics, which are commonly packaged as freeze-dried, or lyophilized, powders. Lyophilization is widely used in the pharmaceutical industry as a means to prevent aggregation and degradation process that may occur in aqueous solution. While lyophilization can stabilize many biologicals, some exhibit diminished therapeutic activity when reconstituted with water due to irreversible structural changes that occur during the freeze-drying process.  I will demonstrate a new computational approach we have developed to study water sorption on protein crystals and powders, and show that this technique allows us, for the first time, to make rigorous contact with experimental measurements designed to probe the protein hydration process, including sorption isotherms, calorimetric heats of sorption, and data from X-ray diffraction. I will also show that we are able to go beyond the current capabilities of experimental techniques and investigate how changes in hydration level can give rise to microscopic mechanical stresses that may alter the structure and biological activity of some proteins.

In the second part, I will use state-of-the-art computational techniques to explore the thermodynamic implications, specifically at deeply supercooled conditions, of water’s well-known anomalies, such as its negative thermal expansion, and its increased compressibility and heat capacity upon cooling. I will also address the need for a unified viewpoint of aqueous thermodynamics in order to understand water’s role as a biological solvent.  One thermodynamically consistent interpretation of experimental observations posits that water becomes highly compressible when cooled below its freezing point due to the presence of a second critical point associated with a first-order phase transition between two distinct metastable liquid phases. Since the region of the phase diagram where this hypothetical critical point would occur is below the homogenous nucleation temperature of bulk water, obtaining direct experimental evidence to falsify the second critical point hypothesis has so far proved to be a significant challenge. Although this intriguing scenario ultimately requires experimental validation, I will present compelling results obtained from advanced free energy analysis techniques that show that an atomistic water model phase-separates into two liquid polymorphs at sufficiently low temperatures.  With these techniques, we are able to precisely locate points of coexistence for the two liquids, as well as their limits of stability, and we provide bounds on the location of the second critical point. Moreover, I will present numerical calculations that demonstrate, for the first time, the coexistence of two metastable (liquid) phases, as well as the formation of a stable ice phase, at the same temperature and pressure. This unusual behavior may have broader implications in liquid state theory and thermodynamics.