It has been established that the function of membrane proteins is affected by the surrounding lipid-bilayer. For instance, it has been shown experimentally that cholesterol affects GPCR function, and recent crystal structures have revealed co-crystalized cholesterol molecules on the protein surface. However, the underlying mechanism of cholesterol-aided modulation is unknown and it is not clear how realistic the cholesterol binding sites are, or whether there exist additional binding sites not observed in the crystal structures. We have been simulating the adenosine A2a and adrenergic B2 receptors in both the active and inactive conformational states. To reach appropriate time scales, we have utilized coarse-grained simulations with the MARTINI force field. For each protein and membrane composition, we have run five repeats of 50 �s, obtaining good statistics. Using these simulations, we are able to determine the binding affinity of cholesterol and other lipids to the receptors. These reveal that most of the GPCR surface is covered with cholesterol or lipid molecules that bind with a weak affinity. It is therefore clear that the experimentally observed cholesterol site is not unique. Furthermore, we are able to characterize the kinetics of the bound molecules. Most of the interactions with the receptor are short-lived with average lifetimes of a few nanoseconds. However, a few binding sites show lifetimes up to the sub-microsecond time scale. Comparing the simulations of the active and inactive states provides information about the role of cholesterol in GPCR modulation. However, the MARTINI simulations are inherently limited because we have to restrain the protein to its crystal structure, thereby removing all possibility of conformational change. Dual-resolution simulations remove this limitation so that we can more accurately describe cholesterol�GPCR interactions. Initial simulations of small-molecule permeability and membrane peptides will be used as illustrations of this powerful technique.

Simulation of membrane proteins using multi-scale models

248th National Meeting (2014)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.