Researcher(s)
- Dylan Stare, Chemical Engineering, University of Delaware
Faculty Mentor(s)
- Kenneth Crane-Moscowitz, Chemical and Biomolecular Engineering, University of Delaware
- Abraham Lenhoff, Chemical and Biomolecular Engineering, University of Delaware
- Eric Furst, Chemical and Biomolecular Engineering, University of Delaware
Abstract
Bundlemers are a class of computationally designed peptide sequences which spontaneously self-assemble in aqueous solutions into tetra-helical coiled coil “nanoparticles”. This self-assembly is governed by the positioning of hydrophobic residues within the peptide chain. Provided they remain unperturbed, the remaining amino acids can be readily substituted with any other hydrophilic residue. Different bundlemer sequences have different solution phase behavior; some stay well dispersed in solution while others form soluble aggregates. A bundlemer with a higher surface charge will tend to remain dispersed while one with a lower surface charge may form soluble aggregates. Previous work on protein docking computation neglected electrostatics in favor of hard-sphere attraction models. This code was repurposed to find the effect of electrostatics, using existing structure for reading molecule data and performing computation. Pairwise calculations to determine the screened coulombic potential were employed along 6 well-defined surface paths. With one molecule centered at the origin, the location of the second molecule was described through a combination of spherical coordinates (⍴, φ, θ) and Tait-Bryan angles (α, β, γ) with the rotation order XYZ chosen arbitrarily. Three of the surface paths were defined as rotations in either phi, theta, or alpha. The other three rotations linked two angles together such that the second molecule was always oriented towards the center of the first. These combinations were (-β, θ), (α, φ), and (φ, γ). The other angles were held at constant values, but not necessarily 0. Across the six paths tested, MCP4 featured favorable electrostatic interactions (denoted as E/kT < 0) for several specific orientations near the major axis of the particle, whereas the MCP8 series did not. The systems developed to quantify electrostatics may be further used to find more favorable configurations with deeper energy wells, through a combination of monte-carlo methods and more well-defined paths.