Molecular Dynamics Simulations of Phosphatidylcholine Membranes: A Comparative Force Field Study

Abstract

Molecular dynamics simulations provide a route to studying the dynamics of lipid bilayers at atomistic or near atomistic resolution. Over the past 10 years or so, molecular dynamics simulations have become an established part of the biophysicist's tool kit for the study of model biological membranes. As simulation time scales move from tens to hundreds of nanoseconds and beyond, it is timely to re-evaluate the accuracy of simulation models. We describe a comparative analysis of five freely available force fields that are commonly used to model lipid bilayers. We focus our analysis on 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) and 1-palmitoyl-2-oleoyl-sn-glycero-3-phosphocholine (POPC) bilayers. We show that some bilayer properties have a pronounced force field dependence, while others are less sensitive. In general, we find strengths and weaknesses, with respect to experimental data, in all of the force fields we have studied. We do, however, find some combinations of simulation and force field parameters that should be avoided when simulating DPPC and POPC membranes. We anticipate that the results presented for some of the membrane properties will guide future improvements for several force fields studied in this work.

Phosphatidylcholine reverse micelles on the wrong track in molecular dynamics simulations of phospholipids in an organic solvent

“…However, a recent study shows that CHARMM36 lipid parameters perform well in the case of high-curvature inverse hexagonal phases. 66 Additionally, Piggot et al 4 report CHARMM36 to be the only one of the force fields they examined to reproduce correctly also the C2 carbon atom deuterium order parameters for both DPPC and POPC (palmitoyloleoylphosphatidylcholine) lipids. This means the C2 carbon atom which is facing the head group could be assumed to sample the same conformations as the lipids in the NMR experiments.…”

“…We chose these particular models because the lipid model has been validated also at low hydration with the water model, 37 verified parameters are available for both saturated and unsaturated lipids, 37 and force field consistent parameters for cyclohexane exist. 38 Furthermore, the accuracy of the lipid model compares favourably to other commonly used lipid force fields in a recent review, 1 or is among the best performing in several others 2,4 with specifically the deuterium order parameters most accurately reproduced. 4 In total, the force field is a very good candidate for good performance in describing the PC/cyclohexane system.…”

“…38 Furthermore, the accuracy of the lipid model compares favourably to other commonly used lipid force fields in a recent review, 1 or is among the best performing in several others 2,4 with specifically the deuterium order parameters most accurately reproduced. 4 In total, the force field is a very good candidate for good performance in describing the PC/cyclohexane system. Periodic boundary conditions were applied in all three dimensions and long-range electrostatic interactions were calculated using the PME method, 40 with a fourth-order smoothing spline.…”

“…CHARMM36 force field performs very well in describing all the studied components individually or in aqueous environment, 37,38 and is currently one of the highest regarded lipid force fields, with good accuracy in water environment. [1][2][3][4] However, the FIG. 7.…”

Controlling Carbon-Nanotube—Phospholipid Solubility by Curvature-Dependent Self-Assembly

“…47 Equations of motion were integrated with the leap-frog algorithm using a time step of 2 fs. Similar simulation parameters have been recommended to be used with the CHARMM36 force-field and GROMACS by Piggot et al 45 The stochastic velocity rescaling thermostat of Bussi et al 48 was used with reference temperature of 325 K and a relaxation time constant of 0.5 ps. Water, lipids, and the carbon substrate were coupled separately to the heat bath.…”