Mechanical Properties of Surfactant Bilayer Membranes from Atomistic and Coarse-Grained Molecular Dynamics Simulations

Abstract

We use simulations to predict the stability and mechanical properties of two amphiphilic bilayer membranes. We carry out atomistic MD simulations and investigate whether it is possible to use an existing coarse-grained (CG) surfactant model to map the membrane properties. We find that certain membranes can be represented well by the CG model, whereas others cannot. Atomistic MD simulations of the erucate membrane yield a headgroup area per surfactant a(0) of 0.26 nm(2), an elastic modulus K(A) of 1.7 N/m, and a bending rigidity kappa of 5 k(B)T. We find that the CG model, with the right choice for the size and potential well depth of the head, correctly reproduces a(0), kappa, as well as the fluctuation spectrum over the whole range of q values. Atomistic MD simulations of EHAC, on the other hand, suggest that this membrane is unstable. This is indicated by the fact that kappa is of the order of k(B)T, which means that the interface is extremely flexible and diffuse, and K(A) is close to zero, which means that the surface tension is zero. We argue that the CG model can be used if the headgroups are uncharged, dipolar, or effectively dipolar due to headgroup charge screening induced by counterion condensation.

Biological and synthetic membranes: What can be learned from a coarse-grained description?

“…Often this approximation is utilized to extract the interface tension and bending rigidity from computer simulations of small membrane patches [ 353,137,111,354,174,17,182,355]. To this end, one determines the local position of the membrane, h(x, y), from the simulation.…”

“…The large scale fluctuations are controlled by the surface tension, γ, and obey |h(k)| 2 ∼ k −2 . This method has successfully been employed to extract the tension and bending rigidity from simulations of interfaces in polymer blends [ 159,356,357], polymer-solvent interfaces [ 317] liquid crystals [ 358,359], and lipid bilayers [ 353,137,111,354,174,17,182,355]. To do so, one numerically determines the position, h, of the interface, or bilayer, in sub-columns centered around (x i , y j ) and with lateral size ∆ (block analysis [ 360,356]).…”

“…While developed earlier for the elucidation of the statics and dynamics of melts and solutions of very long and uniform polymers [ 4,5,6,7,8,9,10,11], coarse-grained models seem particularly well suited to the study of biological constituents [ 12,13,14,15,16,17], such as the heteropolymers DNA and RNA, as well as relatively shortchain lipids which comprise all biological membranes.…”

A novel method for measuring the bending rigidity of model lipid membranes by simulating tethers

“…However, by far the most common approach in simulations is flicker spectroscopy, both for atomistic simulations [16][17][18] as well as for various coarse-grained methods. [18][19][20][21][22][23][24] Only recently den Otter and Briels have proposed a method by which constraining forces are applied to actively deform the membrane, 25 and Farago and Pincus have proposed a scheme based on the change in free energy of deforming the bilayer. 26 Unfortunately, both active methods involve significant technical and conceptual sophistication.…”

Estimation of the bending rigidity and spontaneous curvature of fluid membranes in simulations

“…Theoretically, the fluctuation spectrum is derived by the perturbations from planar [1,11,12], spherical [13][14][15][16], and cylindrical [17][18][19] membranes. In simulations, the fluctuation spectrum of planar membranes is widely used to measure κ [12,[20][21][22][23][24][25]. The fluctuations of tubular membranes have not been simulated as yet, whereas those * noguchi@issp.u-tokyo.ac.jp of quasispherical vesicles are calculated in Refs.…”