Application of a Membrane Protein Structure Prediction Web Service GPCRM to a Gastric Inhibitory Polypeptide Receptor Model

A novel versatile tool named GPCRM has been developed. It targets structure prediction of a distinct protein family of G protein-coupled receptors (GPCRs). In principle, GPCRM builds a GPCR model using a MODELLER-based homology modeling procedure. In addition, that commonly used procedure was improved by using comparison of sequence profiles, multiple template structures and the extensive loop refinement in Rosetta. We applied our method to predict a three dimensional structure of a gastric inhibitory polypeptide receptor (GIPR) from the secretin-like class B of human GPCRs. The GIPR model was also tested in an ensemble docking study in which we investigated plausible interactions of four potential antagonists with that receptor. Out of those four ligands we suggested ChEMBL_1933363 as the most potent antagonist of GIPR based on the Glide docking results.