Free Energy Calculations for the Peripheral Binding of Proteins/ Peptides to an Anionic Membrane. 1. Implicit Membrane Models

摘要

The binding of peptides and proteins to the surface of complex lipid membranes is important in many biological processes such as cell signaling and membrane remodeling. Computational studies can aid experiments by identifying physical interactions and structural motifs that determine the binding affinity and specificity. However, previous studies focused on either qualitative behaviors of protein/ membrane interactions or the binding affinity of small peptides. Motivated by this observation, we set out to develop computational protocols for bimolecular binding to charged membranes that are applicable to both peptides and large proteins. In this work, we explore a method based on an implicit membrane/solvent model (generalized Born with a simple switching in combination with the Gouy-Chapman-Stem model for a charged interface), which we expect to lead to useful results when the binding does not implicate significant membrane deformation and local demixing of lipids. We show that the binding free energy can be efficiently computed following a thermodynamic cycle similar to protein-ligand binding calculations, especially when a Bennett acceptance ratio based protocol is used to consider both the membrane bound and solution conformational ensembles. Test calculations on a series of peptides show that our computational approach leads to binding affinities in encouraging agreement with experimental data, including for the challenging example of the bringing of flexible MARCKS-ED peptides to membranes. The calculations highlight that for a membrane with a significant fraction of anionic lipids, it is essential to include the effect of ion adsorption using the Stern model, which significantly modifies the effective surface charge. This implicit membrane model based computational protocol helps lay the groundwork for more systematic analysis of protein/peptide binding to membranes of complex shape and composition.