TL;DR
This study uses molecular simulations to disentangle the roles of packing, hydrophobicity, and hydrogen bonding in structuring water around a protein helix. They found that water-water hydrogen bonding, which underlies hydrophobicity, is the dominant force, pushing water away from hydrophobic regions and creating a smoother apparent surface than predicted by packing alone.
What Problem Does This Paper Address?
Understanding how proteins interact with water is fundamental to biochemistry and molecular biology. Proteins fold into complex 3D structures, and their surfaces are a mix of polar (hydrophilic) and nonpolar (hydrophobic) regions. The question is: what governs the structure of water molecules at this protein-water interface? Traditionally, protein structure has been explained by tight packing of amino acid side chains and hydrogen bonding within the protein itself. However, the behavior of water, with its own strong hydrogen-bonding network, at the protein surface is more complex. This paper aims to specifically quantify the relative importance of three key factors in organizing this interfacial water: 1) the physical "packing" of water molecules against the protein's van der Waals surface, 2) the "hydrophobicity" effect driven by water-water hydrogen bonding, and 3) direct "protein-water hydrogen bonding."
Methods
The researchers employed molecular dynamics simulations, specifically Monte Carlo simulations, to model a polyalanine alpha-helix surrounded by water molecules. To isolate the contributions of different interactions, they ran four distinct simulations:
- Standard Simulation: A "normal" simulation with both the protein helix and water molecules fully charged, representing a realistic scenario.
- Uncharged Helix Simulation: The charges on the protein helix were "switched off" to remove the direct protein-water hydrogen bonding contribution, allowing the assessment of packing and hydrophobicity's role.
- Fluid "Argon" Simulation: All charges on the water molecules were switched off, effectively turning off water-water hydrogen bonding. This simulation used Lennard-Jones parameters for the water molecules, mimicking a fluid like argon, to study the effect of packing alone.
- Liquid "Argon" Simulation: Similar to the fluid "argon" simulation, but with parameters adjusted to represent a liquid state, further emphasizing the role of packing.
They analyzed the distribution and orientation of water molecules around the helix, as well as the energies involved (water-water, water-protein, and total solvent energy).
Datasets
Not discussed in the paper. The study relies on in silico simulations of a model system, not experimental datasets.
Key Insights and Ideas
- Water-Water Hydrogen Bonding Dominates: The most striking finding is that the strong tendency of water molecules to hydrogen bond with each other (hydrophobicity) is the primary driver of water structure around the helix, more so than direct protein-water hydrogen bonds or simple packing.
- "Smoother" Apparent Surface: Hydrophobic regions of the protein appear "smoother" to water molecules than they do to a non-hydrogen-bonding fluid like argon. This is because water molecules tend to "move back" from hydrophobic surfaces to maintain their own hydrogen bond network, rather than fitting into every crevice.
- Packing vs. Hydrogen Bonding Trade-off: While packing is important, it's often constrained by the need for water to maintain its hydrogen bond network. Water molecules can pack tightly into hydrophilic regions where they can form hydrogen bonds with the protein, but they avoid hydrophobic crevices to preserve water-water hydrogen bonds.
- Implications for Molecular Surfaces: The results suggest that traditional molecular surface definitions (like Richards-Connolly surfaces), which treat water as a simple packing sphere, might overestimate the accessibility of hydrophobic regions to water.
Key Results
- Distribution Changes: Switching off helix charges (removing protein-water H-bonds) caused water molecules to "move back" from the helix, particularly from carbonyl oxygens. Switching off water charges (removing water-water H-bonds) led to a dramatic restructuring, with molecules packing much more closely and filling crevices.
- Energy Analysis: Water-water interactions were significantly more attractive in the standard and uncharged-helix simulations compared to the "argon" simulations. Conversely, water-protein interactions were more favorable around hydrophilic regions.
- "Argon" vs. Water: The "argon" simulations showed much higher radial distribution peaks, indicating denser packing, and a more pronounced filling of crevices compared to water simulations.
- Hydrophobicity Correlation: A synthesized function (ln ζ) that combined oxygen distribution and total energy showed a strong correlation with hydrophobicity, being lowest around hydrophobic carbons and higher around hydrophilic oxygens.
Limitations and Future Work
- Model System: The study used a simplified polyalanine helix, not a complex, multi-domain protein. The results may not directly translate to all protein surfaces.
- Water Model: The TIP3P water model was used, and while effective for many purposes, it's not polarizable. The authors acknowledge that more advanced models (e.g., with distributed multipoles) could provide more detailed electrostatic insights.
- Energetic Interpretation: While the simulations provide valuable insights into energy contributions, the authors note that the interpretation of entropic effects is challenging and requires further work.
- Future Work: The authors suggest further investigation into the utility of the synthesized function (ln ζ) for characterizing entropy and hydrophobicity and the use of more sophisticated water models. They also hint at ongoing work to better assess the utility of their developed function.
Summary updated