Chapter 10 Atomistic and Mean Field Simulations of Lateral Organization in Membranes

In this Chapter we have described modeling of lipid bilayers starting from the atomistic level, where molecular dynamics or Monte Carlo are the method capable of providing details of atomistic phenomena at this level. Molecular dynamics, must be applied with caution to successfully model a hydrated lipid bilayer, due to complex and competing interatomic interactions. Individual atoms may experience large forces from chemical bonds, dihedral forces, steric forces, electrostatic forces, and van der Waals attractions. While individual forces may be quite large, there are often cancellations so that the subtle details of the energy landscape are highly complex and non-intutitve. For example, small differences in each dihedral along a single lipid hydrocarbon chain, due to an interplay between various terms in Eq. (1) will propagate to large differences in chain conformations. In general, the development of MD force fields for lipid bilayers is a process that cannot be overlooked. Depending on whether hydrogen atoms are to be included or excluded, the force fields will be quite different. In general, both types of force fields are of the highest quality when tested against experimental smaller "model compounds" in independent simulations before being applied to a lipid simulation. In this vein, the model for molecular water is an important consideration. We described in detail the determination of force field parameters in the United Atom Approximation in this Chapter. While atomistic simulations are of great interest and provide many new insights, they are limited by the relatively slow rate of diffusion of lipids in a bilayer to describing the bilayer over time scales where individual molecules do not move very far laterally. However, atomistic simulations provide clues for larger scale thermodynamic properties of bilayers in the local interactions. We described efforts to use input from MD simulations to construct models based on self consistent mean field theory that can describe larger scale lateral organizational, development, over time and across the membrane. Self Consistent Mean Field Theory, based on models with parameters calculated from atomistic simulations has, as we illustrated in this chapter, great promise for understanding the properties of lipid bilayers of complex composition. The long range goal of this type of modeling is to begin to understand properties of membranes of biological composition.