The importance of solvation in the context of non-covalent biomolecular interactions is already widely acknowledged within the drug design community. In this context, water is recognized as an active contributor to protein-ligand binding thermodynamics. However, the implementation of solvent thermodynamics into routine structure-based drug discovery pipelines remains partly unresolved. Oftentimes, the thermodynamic parameters of structurally resolved water molecules cannot be clearly defined and can become ambiguous within the derivation of rational enthalpy-entropy-based SAR. Here, we present novel strategies and solvent free energy functionals which can accurately predict the enthalpic and entropic contributions for protein-bound water molecules. Also, solvation contributions are partitioned to either the protein-bound state or the pre-binding state of the ligand molecule. In addition, the solvent functionals easily allow for partitioning of the solvent thermodynamics into contributions from substructures/fragments of the ligand molecules. This enables a new route towards design based on customized solvation thermodynamics within fragment-based drug discovery. The parameters of the functionals are derived from MD simulations/GIST analysis and optimized with experimental thermodynamic data (ITC) by using exhaustive Monte Carlo sampling within the parameter space of the functionals. As a test system, we applied our methodology to a set of over 80 Trypsin and Thrombin systems, each with known thermodynamic binding profiles obtained from ITC experiments. With our approach, we were able to rationalize enthalpy-entropy contributions of solvent molecules for the investigated systems and also match them with experimental binding profiles. Furthermore, we propose new ligand molecules with designed solvent thermodynamics, based on the properties calculated from the fragment-wise partitioning of solvent thermodynamics. The methodology is integrated into a python library, as well as a command line program and will be available to the scientific community at the GitHub page of the presenting author (https://github.com/wutobias).

New semi-empirical solvent functionals for structure-based drug discovery: Routes to customized solvent thermodynamics

2019 Spring National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.