Amyloid Beta Aggregation in the Presence of Naturally-Derived Inhibitors

Abstract

Alzheimer's Disease (AD) is a neurodegenerative disease that causes dementia, nervous system degradation, and death. Currently there are no therapeutic agents available for the treatment of AD despite great efforts by the research community. Many small phenolic compounds have been shown in experiment to be effective at preventing the formation of Aβ fibrils. In this work we use discontinuous molecular dynamics (DMD) simulations to learn how four naturally-occurring phenolic compounds, resveratrol, vanillin, curcumin, and epigallocatechin-3-gallate (EGCG), bind to Aβ(17-36) monomers and affect oligomerization and fibrillation. We use a coarse-grained model/force field for inhibitors which has geometric and energetic parameters that are compatible with the coarse-grained protein model PRIME20. Preliminary results show that the U-shaped protofibril structure formed by Aβ(17-36) is similar to the corresponding part of the Aβ(1-42) fibril model, based on solid state NMR data. The order of peak heights in the peptide-inhibitor radial distribution function shows that the strength of the inhibitor binding affinity is resveratrol > curcumin and EGCG > vanillin. Simulations of 8 Aβ(17-36) peptides aggregating in the presence of 30 inhibitors show that EGCG, resveratrol and curcumin inhibit Aβ(17-36) fibril formation while vanillin only retards the lag phase. Instead of forming an ordered β sheet structure (as occurs in the absence of inhibitors), the peptides remain random coils and form complexes with the inhibitors, binding mainly to the hydrophobic residues near the peptide termini. The relative ability to inhibit Aβ(17-36) fibrillation is EGCG > resveratrol > curcumin > vanillin, consistent with experimental findings on full length Aβ(1-42) aggregation. Our simulations provide molecule-level insights into the mechanisms by which small molecules inhibit Aβ aggregation.