Molecular simulations of the fluctuating conformational dynamics of intrinsically disordered proteins

Abstract

Intrinsically disordered proteins (IDPs) do not possess well-defined three-dimensional structures in solution under physiological conditions. We develop all-atom, united-atom, and coarse-grained Langevin dynamics simulations for the IDP α-synuclein that include geometric, attractive hydrophobic, and screened electrostatic interactions and are calibrated to the inter-residue separations measured in recent single-molecule fluorescence energy transfer (smFRET) experiments. We find that α-synuclein is disordered, with conformational statistics that are intermediate between random walk and collapsed globule behavior. An advantage of calibrated molecular simulations over constraint methods is that physical forces act on all residues, not only on residue pairs that are monitored experimentally, and these simulations can be used to study oligomerization and aggregation of multiple α-synuclein proteins that may precede amyloid formation.

Calibrated Langevin-dynamics simulations of intrinsically disordered proteins

“…As in our previous studies [38], we employed a purely repulsive Weeks-Chandler-Andersen (WCA) potential, the attractive part of the Lennard-Jones potential, and screened Coulomb potential to model the steric, hydrophobic, and electrostatic interactions, respectively:…”

“…In a recent manuscript [38], we introduced a physical model to describe the fluctuating conformational dynamics of IDPs. The motivation for the new computational model for IDPs stems in part from the fact that commonly used molecular mechanics force fields, such as Amber [36] and CHARMM [3], can bias the simulation results toward folded behavior since they have been calibrated using x-ray crystal structures of folded proteins [26].…”

Structural Ensembles of Intrinsically Disordered Proteins Depend Strongly on Force Field: A Comparison to Experiment

“…A variety of simulation methods (MD, Monte Carlo, metadynamics, replica exchange) and different levels of representation (coarse-grained, implicit solvent, all-atom with explicit water) have been used to obtain IDP ensembles. 7,[21][22][23][24][25][26][27][28][29] There are two main challenges encountered in de novo simulations.…”

Understanding the dynamics of monomeric, dimeric, and tetrameric α‐synuclein structures in water

“…Although it is known that αS is intrinsically disordered in solution , most of the published molecular dynamics studies address the micelle‐bound structure of αS with well‐defined α‐helical secondary structures. Complementing MD studies, Monte‐Carlo simulations and Langevin dynamics modeling of intrinsically disordered αS chains have also been reported. Such models are potentially very efficient; however, they require a calibration or constraining using experimentally determined structural data.…”

“…Complementing MD studies, Monte‐Carlo simulations and Langevin dynamics modeling of intrinsically disordered αS chains have also been reported. Such models are potentially very efficient; however, they require a calibration or constraining using experimentally determined structural data. Parameterization of coarse‐grained models toward ultimate predictive reliability remains in the pipeline.…”