A Coarse-Grain Model for Efficient Simulation of Self-Assembling Amyloidogenic Peptide Systems*

Prior experiments with amyloid forming peptides have demonstrated that short peptides exhibit distinct self-assembly behavior when constrained to a surface resulting in rapid formation of amyloid fibrils. Molecular dynamics simulations of such systems can provide key insights into this self-assembly process. However, the size and time scales associated with such systems is too large for atomistic approaches to study self-assembly. Thus, we are developing an implicit-solvent coarse-grained model of amyloidogenic peptides for the investigation of their surface mediated self-assembly in these systems using MD. The model consists of three CG beads per residue where the central bead has either hydrophobic or hydrophilic character depending on the amino acid it represents. The two side-chain beads form a dipole which replicates that found in an all atom amide backbone. The simulations of initially randomly oriented peptides demonstrate rapid self-assembly into fibrillar structures upon adsorption to a surface. Additionally, the simulations show that fibrillar structures are favored at lower peptide concentrations while surface covering films are favored at high peptide concentrations which is in agreement with experimental evidence.