(464a) Computation of Amino Acid-Mineral Surface Thermodynamics and Interaction Modes via Importance Sampling

Understanding interactions between biomolecules and minerals is essential to the development of biomimetics and the treatment of biomineralization diseases. Yet few force fields are validated for the interactions of biomolecules at mineral interfaces, and there are few rigorous benchmarks of the effectiveness of the conformational sampling in this context. Toward those ends, we used new detailed-balance Monte Carlo methods in the Rosetta molecular modeling suite to generate equilibrium canonical ensembles of the twenty naturally occurring amino acids interacting with stable, flat surfaces of calcite, hydroxyapatite, calcium oxalate monohydrate, and brushite. We tested the convergence of the algorithms and found that all converge energetically. In several cases of strong adsorption, there are multiple structural modes of interaction with similar binding energies. Adsorption free energies range from –40 to 0 kcal/mol depending on the surface and amino acid. Negatively charged amino acids consistently adsorb most strongly, with the electrostatic term dominating the interactions. Hydroxyapatite binds amino acids tightest, and calcite the weakest. We compare results to published molecular dynamics studies and experimental adsorption and single molecule force spectroscopy studies. Our findings have implications for studies of protein-mineral interactions and design of biomolecules to control crystal growth.