CXCR3 antagonist VUF10085 binds to an intrahelical site distinct from that of the broad spectrum antagonist TAK-779.
Bottom Line: Validation of modelling was carried out by site-directed mutagenesis of CXCR3, followed by assays of cell surface expression, ligand binding and receptor activation.Contrary to our hypothesis, mutation of Asp-112(2:63) had no observable effects on TAK-779 activity, but clearly decreased the antagonist potency of VUF 10085.Likewise, mutations of Phe-131(3.32) , Ile-279(6.59) and Tyr-308(7.43) were well tolerated and were critical for the antagonist activity of VUF 10085 but not for that of TAK-779.
Affiliation: Leukocyte Biology Section, NHLI Division, Faculty of Medicine, Imperial College, London, UK.Show MeSH
Related in: MedlinePlus
Mentions: The three-dimensional model of the seven helical TM bundle of human CXCR3 was predicted using the ab initio method MembStruk (Vaidehi et al., 2002; Hall et al., 2009). The extra and intracellular loops were added using the method, Modeller. VUF 10085 was built using the LigPrep module from the Schrodinger Glide suite (Schrodinger Inc.). Multiple ligand conformations were generated for the compound and docked using Glide XP (Schrodinger Inc., Portland, OR, USA). Subsequently, a short energy minimization was performed on each docked pose and the binding energy of this optimized pose was calculated. The binding energy was calculated as BE (binding energy) = PE (ligand in fixed protein) − PE (ligand in solvation); where BE is the binding energy and PE is the potential energy. The compound poses were then sorted by binding energy and the top 20 conformations inspected visually to maximize the interactions with residues that are known to interact with ligands in chemokine receptors (Vaidehi et al., 2009). During the course of this work, the crystal structure of CXCR4 bound to a low MW antagonist was published (Wu et al., 2010). Therefore, we also generated a homology model of CXCR3 based on CXCR4 crystal structure as template (pdb ID:3ODU) using the program MODELLER (http://salilab.org/modeller/9v7/manual/node8.html). We selected the top 100 models from MODELLER and clustered these models by their root mean squared deviation in coordinates. The 100 models clustered into five clusters and the best energy structure from the cluster was chosen for docking. We then docked the VUF 10085 antagonist to this model using GOLD (http://www.ccdc.cam.ac.uk/SupportandResources/Support/pages/SupportSolution.aspx?supportsolutionid=110) flexible side chain docking to allow for protein flexibility. The side chains of the residues Y60, W109, D112, F131, F135, H202, Y271 and Y308 were treated as flexible using the built in rotamer library. A distance constraint was placed between D112 and the pyridine nitrogen of the VUF 10085 compound. The best docked pose was selected based on the experimental data in this paper. The final model was used in generating Figure 7A–C.
Affiliation: Leukocyte Biology Section, NHLI Division, Faculty of Medicine, Imperial College, London, UK.