Prediction of estrogen receptor β ligands potency and selectivity by docking and MM-GBSA scoring methods using three different scaffolds

This study aimed to identify the docking and molecular mechanics-generalized born surface area (MM-GBSA) re-scoring parameters which can correlate the binding affinity and selectivity of the ligands towards oestrogen receptor beta (ER beta). Three different series of ER beta ligands were used as dataset and the compounds were docked against ER beta (protein data bank (PDB) ID: 1QKM) using Glide and ArgusLab. Glide docking showed superior results when compared with ArgusLab. Docked poses were then rescored using Prime-MM-GBSA to calculate free energy binding. Correlations were made between observed activities of ER beta ligands with computationally predicted values from docking, binding energy parameters. ER beta ligands experimental binding affinity/selectivity did not correlate well with Glide and ArgusLab score. Whereas calculated Glide energy (coulomb-van der Waal interaction energy) correlated significantly with binding affinity of ER beta ligands (r(2) = 0.66). MM-GBSA re-scoring showed correlation of r(2) = 0.74 with selectivity of ER beta ligands. These results will aid the discovery of novel ER beta ligands with isoform selectivity.