Simulating Bilayers of Nonionic Surfactants with the GROMOS-Compatible 2016H66 Force Field

Polyoxyethylene glycol alkyl ether amphiphiles (C i E j ) are important nonionic surfactants, often used for biophysical and membrane protein studies. In this work, we extensively test the GROMOS-compatible 2016H66 force field in molecular dynamics simulations involving the lamellar phase of a series of C i E j surfactants, namely C 12 E 2, C 12 E 3, C 12 E 4, C 12 E 5, and C 14 E 4 . The simulations reproduce qualitatively well the monitored structural properties and their experimental trends along the surfactant series, although some discrepancies remain, in particular in terms of the area per surfactant, the equilibrium phase of C 12 E 5, and the order parameters of C 12 E 3, C 12 E 4, and C 12 E 5 . The polar head of the C i E j surfactants is highly hydrated, almost like a single polyethyleneoxide (PEO) molecule at full hydration, resulting in very compact conformations. Within the bilayer, all C i E j surfactants flip-flop spontaneously within tens of nanoseconds. Water-permeation is facilitated, and the bending rigidity is 4 to 5 times lower than that of typical phospholipid bilayers. In line with another recent theoretical study, the simulations show that the lamellar phase of C i E j contains large hydrophilic pores. These pores should be abundant in order to reproduce the comparatively low NMR order parameters. We show that their contour length is directly correlated to the order parameters, and we estimate that they should occupy approximately 7–10% of the total membrane area. Due to their highly dynamic nature (rapid flip-flops, high water permeability, observed pore formation), C i E j surfactant bilayers are found to represent surprisingly challenging systems in terms of modeling. Given this difficulty, the results presented here show that the 2016H66 parameters, optimized independently considering pure-liquid as well as polar and nonpolar solvation properties of small organic molecules, represent a good starting point for simulating these systems.