Different Environments for a Realistic Simulation of GPCRs‐Application to the M₂ Muscarinic Receptor

Abstract

A model of the human M(2) muscarinic receptor was taken as an example for a class A G-protein coupled receptor to explore the influence of different environments in a molecular dynamics simulation (MDS) on the protein structure. The most commonly used environment is the vacuum, although it is very unnatural for a transmembrane protein. As an alternative a membrane-like system, consisting of a lipophilic central layer and two aqueous flanking layers, was tested. The most realistic system that can be applied is a phospholipid bilayer with a surrounding physiological sodium chloride solution. From all systems good protein structures were received, nevertheless clear differences between the systems were detected in the structural comparison of the models. Subsequently it was analyzed whether the observed structural differences influence ligand binding. For this purpose the antagonist (S)-scopolamine was docked into the binding cavity, which is well known by many reported single and multiple point mutations. As expected from the observed structural variations triggered by the type of environment employed in MDS, also differences in the binding mode of (S)-scopolamine were detected, all contacts, however, which are known to be important were found.

Update 1 of: Computational Modeling Approaches to Structure–Function Analysis of G Protein-Coupled Receptors

“…Despite the difficulties in using the available high resolution structures as templates for modeling the homologous GPCRs, we think that, at the moment, comparative modeling remains preferable to the most effective ab initio approaches, and it has, indeed, been widely used since the year 2000 ,,,− ,,,,,,,,,,− (see also references in the red updates through the text).…”

“…A significant number of cases also exist in which no energy refinement has been done before and after the ligand–receptor docking. Overall, the receptor systems were subjected to short MD simulation time periods, which only in a few cases exceeded 500 ps (i.e., refs ,,,,,,,,,,,,,,− ,,,,,, among the articles on the free and/or ligand-bound receptor forms reviewed in this article). In some cases, the lengths of MD simulations were exceedingly short (i.e., 250–500 ps unrestrained production phases) for the molecular systems under study (i.e., all-atoms TM-receptor model in explicit methane ,,, or lipid/water ,, ).…”

Computational Modeling Approaches to Structure−Function Analysis of G Protein-Coupled Receptors

“…Despite the difficulties in using the crystal structure of rhodopsin as a template for modeling the homologous GPCRs, we think that, at the moment, comparative modeling remains preferable to the most effective ab initio approaches, and it has, indeed, been widely used since the year 2000. ,,,− ,,,,,,,− ,− …”