Implicit Solvent Models and the Energy Landscape for Aggregation of the Amyloidogenic KFFE Peptide.

摘要: This study compares the performance of four implicit solvent models in describing peptide aggregation. The are effective energy function-1 (EEF1) and three generalized Born (GB) models: one following original implementation Still (GB1), analytical continuum electrostatics (ACE) potential, GB with "simple switching" (GBSW). For each model first step aggregation, namely dimerization, is investigated for KFFE peptide, which shortest peptides known to form amyloid fibrils vitro. Using basin-hopping global optimization replica exchange molecular dynamics simulations, we conclude that considered, EEF1 potential provides most reliable description formation precursors. It produces results closest experimental findings a partial β-strand conformation solution along exhibiting antiparallel structure. ACE GB1 potentials also show significant β-propensity but fail produce stable dimers. GBSW on other hand, supports very dimer structure, turn rather than β conformation.