Computational Studies on Phosphodiesterase-5 Inhibitors to Design Novel Lead Compounds for the Treatment of Erectile Dysfunction

2D and 3D Quantitative Structural Activity Relationship studies using Molecular Field Analysis (MFA) and Receptor surface Analysis (RSA) methods along with pharmacophore hypothesis using Catalyst version 4.7 were performed on a series of Phosphodiesterase 5 (PDE-5) inhibitors. The best equations with training set consisting 41 molecules, produced r2 value of 0.788 and r2cv value of 0.618 in 2D-model and r2 value of 0.844 and r2cv value of 0.810 in MFA-model and r2 value of 0.853 & r2cv of 0.799 in the RSA-model. Pharmacophore models were generated using 20 molecules as training set. The best quantitative harmacophore model consists of one hydrogen bond acceptor, one hydrophobic aliphatic and two ring aromatic features. We have constructed a large set of 75 test compounds, and conformational studies were done as described earlier. The estimated activities were scored using hypothesis 1 as the pharmacophore. Out of 25 highly active compounds (<50nM), 15 were accurately predicted as highly active and the remaining were all predicted as moderately active. Out of the 33 moderately active compounds (50-1000nM), 4 were predicted as inactive and one was predicted highly active. Out of the 16 inactive compounds (>1000nM), 8 were predicted to be inactive and 8 were predicted to be moderately active.