An energy-based mapping method for identifying the in-plane orientations of polypeptides and other macromolecules at crystalline interfaces

We present an energy-based algorithm, POINTER, which can determine the permissible alignments of a polypeptide (or other macromolecule) with respect to the lattice vectors of an interfacial surface (this alignment is defined by the angle theta). The algorithm represents both the interface and the macromolecule in three dimensions. For each value of theta, incremental moves of the macromolecule occur in the x, y, z direction along the theta orientation, as well as rotation (omega, gamma, zeta) of either the macromolecular chain or the interfacial slab. We utilized a simple forcefield that consists of a dipole-dipole, dipole-charge, or charge-charge electrostatic interaction term and a Lennard-Jones attraction-repulsion term to describe the nonbonding interactions between macromolecular atoms and interfacial atoms. We benchmarked our method by modeling ice- and mineral-interaction polypeptides on various Miller planes of hexagonal ice and inorganic solids, respectively. In addition, we searched phase space for a simpler, nonpolypeptide system: The ice-nucleating C31 alcohol monolayer (comprised of 61 C31 molecules) in contact with the {001} plane of hexagonal ice. Our results indicate that the POINTER simulation method can reproduce the macromolecule orientation observed for each benchmark system. In addition, our simulations point to a number of factors-polypeptide binding site structure, the positioning of hydrophobic residues near the interface, and interface topology-which can influence the adsorption orientation of polypeptides on hexagonal ice and inorganic solids. (C) 2000 American Institute of Physics. [S0021-9606(00)70211-X].