Structure prediction of protein-solid surface interactions reveals a molecular recognition motif of statherin for hydroxyapatite

A molecular description of protein-surface interactions could open new avenues in bionanotechnology and provide a deeper understanding of in vivo phase boundary biophysics. However, current experimental techniques can provide only inferential or incomplete information about the protein-surface interface. We present a novel computational method for modeling the interactions of proteins with solid surfaces using comprehensive sampling and an atomistic description. The approach relies on an all-atom Monte Carlo plus-minimization search algorithm that rapidly and simultaneously optimizes rigid-body and side-chain conformations. We apply the method to the statherin-hydroxyapatite system, an evolved protein-surface interaction that is likely to have one or a few specific structural solutions. The algorithm converges on a set of low energy, entropically favorable structures that are consistent with previous experimental results, namely protein-surface intermolecular distances acquired by solid-state NMR. The simulations isolate particular residues as being primary contributors to the adsorption free energy (hydrogen bonding, van der Waals, and electrostatic energies), in agreement with previous mutagenesis, deletion, and single amino acid experiments. We also report the discovery of a molecular recognition motif where the N-terminal α-helix of statherin places all four of its basic residues to match the periodicity of open phosphate triad clusters across the [001] monoclinic face of the hydroxyapatite surface. Results suggest new experiments that could further elucidate the structural features of this important biological system.