Improving Internal Peptide Dynamics in the Coarse-Grained Martini Model: Application to Amyloid Peptides

Abstract

The assembly of misfolded protein aggregates into amyloid fibrils is at the heart of the development of neurodegenerative disorders such as Alzheimer's disease and prion diseases [1, 2]. Despite its significance, the driving forces behind the aggregation of peptides and protein misfolding are not well understood. To gain molecular insight into the aggregation of amyloid peptides, we carry out computer simulations using the recently developed MARTINI coarse-grained (CG) model [3]. Compared to the more traditional atomistic simulations, CG models offer the possibility of following protein folding events, which typically occur on the millisecond timescale. The current MARTINI model, in particular, is able to reproduce a wide range of lipid properties as well as lipid-protein interactions for rigid proteins. Protein folding and aggregation however often involves significant transitions between secondary structures and hence requires that the proteins be flexible during the simulations. We will present recent advances on our extension of the MARTINI model to more accurately describe the internal flexibility of peptides and small proteins. The model is applied to simulations of amyloid peptides of different lengths in water. Its performance is assessed by comparing the distributions of various structural properties with their counterparts from atomistic simulations.