Binned into log-sized bins, ranging from more than 1000-fold selectivity in either direction to 1. doi:10.1371/journal.pone.0049910.gModel RefinementAs shown previously, adapting the orthosteric sites of GPCR homology models to known ligands improves pose fidelity and hit rates [20]. Thus, for optimization of model O, binding site purchase L-DOPS residues ?within a 6 A radius around the equivalent position of 1 (the ligand in 3EML) were iteratively refined with CHARMM [21] and MODELLER. The residues selected for optimization were also compared to mutagenesis studies of the A1AR in recognition of agonists and antagonists [22,23]. Residues that caused major changes in binding affinity (up to 100-fold decrease) after alanine substitution were checked 18334597 against the selection of residues within 6 ?A of the ligand. In all cases, the residues that contributed to a loss of binding affinity after alanine substitution were included in the selection. For the part of the refinement using CHARMM, the CHARMm22 force field (Accelrys, Inc.) was used, and harmonic ?restraints with a force constant of 50 kcal/mol?A2 and a minimum ?at 2.4 A were assigned to the hydrogen bonds formed between the respective ligand and Asn2546.55, the key recognition residue in the A1AR binding pocket. A known ligand of the A1AR (4-[(3benzyl-5-phenyl-triazolo[4,5-e]pyrimidin-7-yl)amino]cyclohexan1-ol; 5, [24]) was placed manually in the binding site (to ensure correct orientation, i.e. maintenance of the two hydrogen bonds with Asn2546.55) and force-field minimized while keeping the adjacent residues fixed. The optimized ligand pose was then included in the following re-modeling step with MODELLER. This procedure of force-field minimizing the ligand and remodeling with MODELLER was repeated until the atomic positions of the active site residues and the ligand converged. To check for bias introduced by the optimization with the reference triazolopyrimidine derivative 5, a second AR antagonist (1-(8-butyl-2furan-2-yl-8H-pyrazolo[4,3-e] [1,2,4]triazolo[1,5-c]pyrimidin-5yl)-3-(4-nitro-phenyl)-urea, 6 [25]) was manually placed in the binding site, again making making sure that the hydrogen bonds with Asn2546.55 are formed, and minimized with PLOP [26,27].Residues whose interaction with the ligand had unfavorable force field energy values (Ala662.61, Ile692.64, Phe171, Leu2506.51, and Ile2747.39) were sidechain-optimized followed by minimization together with the ligand. Both ligands had been part of a set of 3276 A1AR binders extracted from the WOMBAT database [28]. They were selected for the refinement process because they docked in poses interacting with Asn2546.55 and ranked highly when docked to model O. The final refined structure, termed model A, was used in the first docking round (see below and Fig. 1A). Using the ligand data acquired in round one, the orthosteric binding site of the A1AR was optimized a second time. In this round of refinement (resulting in model B; Fig. 1B), residues were chosen based on their deviation from the corresponding residues in the A2AAR structure. In particular, extracellular loop 3 (ECL3; residues Phe2596.60 to Cys2637.28) and adjacent residues in helix 7 (up to Ser2677.32) were rebuilt to maintain the salt bridge between His2647.29 and Glu172. Moreover, the “toggle switch” Trp2476.48 and the adjacent His2516.52, which showed large deviations of up to 140u in their x1 Nazartinib angles, were manually flipped and then minimized. No restraints were applied during loop r.Binned into log-sized bins, ranging from more than 1000-fold selectivity in either direction to 1. doi:10.1371/journal.pone.0049910.gModel RefinementAs shown previously, adapting the orthosteric sites of GPCR homology models to known ligands improves pose fidelity and hit rates [20]. Thus, for optimization of model O, binding site residues ?within a 6 A radius around the equivalent position of 1 (the ligand in 3EML) were iteratively refined with CHARMM [21] and MODELLER. The residues selected for optimization were also compared to mutagenesis studies of the A1AR in recognition of agonists and antagonists [22,23]. Residues that caused major changes in binding affinity (up to 100-fold decrease) after alanine substitution were checked 18334597 against the selection of residues within 6 ?A of the ligand. In all cases, the residues that contributed to a loss of binding affinity after alanine substitution were included in the selection. For the part of the refinement using CHARMM, the CHARMm22 force field (Accelrys, Inc.) was used, and harmonic ?restraints with a force constant of 50 kcal/mol?A2 and a minimum ?at 2.4 A were assigned to the hydrogen bonds formed between the respective ligand and Asn2546.55, the key recognition residue in the A1AR binding pocket. A known ligand of the A1AR (4-[(3benzyl-5-phenyl-triazolo[4,5-e]pyrimidin-7-yl)amino]cyclohexan1-ol; 5, [24]) was placed manually in the binding site (to ensure correct orientation, i.e. maintenance of the two hydrogen bonds with Asn2546.55) and force-field minimized while keeping the adjacent residues fixed. The optimized ligand pose was then included in the following re-modeling step with MODELLER. This procedure of force-field minimizing the ligand and remodeling with MODELLER was repeated until the atomic positions of the active site residues and the ligand converged. To check for bias introduced by the optimization with the reference triazolopyrimidine derivative 5, a second AR antagonist (1-(8-butyl-2furan-2-yl-8H-pyrazolo[4,3-e] [1,2,4]triazolo[1,5-c]pyrimidin-5yl)-3-(4-nitro-phenyl)-urea, 6 [25]) was manually placed in the binding site, again making making sure that the hydrogen bonds with Asn2546.55 are formed, and minimized with PLOP [26,27].Residues whose interaction with the ligand had unfavorable force field energy values (Ala662.61, Ile692.64, Phe171, Leu2506.51, and Ile2747.39) were sidechain-optimized followed by minimization together with the ligand. Both ligands had been part of a set of 3276 A1AR binders extracted from the WOMBAT database [28]. They were selected for the refinement process because they docked in poses interacting with Asn2546.55 and ranked highly when docked to model O. The final refined structure, termed model A, was used in the first docking round (see below and Fig. 1A). Using the ligand data acquired in round one, the orthosteric binding site of the A1AR was optimized a second time. In this round of refinement (resulting in model B; Fig. 1B), residues were chosen based on their deviation from the corresponding residues in the A2AAR structure. In particular, extracellular loop 3 (ECL3; residues Phe2596.60 to Cys2637.28) and adjacent residues in helix 7 (up to Ser2677.32) were rebuilt to maintain the salt bridge between His2647.29 and Glu172. Moreover, the “toggle switch” Trp2476.48 and the adjacent His2516.52, which showed large deviations of up to 140u in their x1 angles, were manually flipped and then minimized. No restraints were applied during loop r.
