ИСТИНА

Tankyrase enzymes (TNKS), members of the poly(ADP-ribose) polymerase (PARP) superfamily, control the Wnt pathway and represent a promising target in the search for potential anticancer agents that can also be used in synergic combinations with drugs targeting other pathways. Using the molecular docking and machine learning-based virtual screening techniques along with the physico-chemical and ADMET profile prediction as well as molecular dynamics simulations, we have identified a number of candidate compounds in a subset of the ZINC database. Out of 36 compounds biologically evaluated in vitro for their inhibition of the TNKS2 enzyme using immunochemical assay, 13 compounds belonging to three different chemotypes have shown various levels of inhibitory activity. For two compounds, the IC50 values below 10 nM and 30 nM were obtained. In retrospective analysis of the results, it was found that, although quite useful in virtual screening, the relatively simple scores based on molecular docking (even with target-specific machine learning-based scoring functions) or MM-PBSA methods proved unsuitable [1] for predicting the effect of structural modification or for accurate ranking of the compounds based on their binding energies. On the other hand, the molecular dynamics simulations and Free Energy Perturbation (FEP) calculations allowed us to further decipher the structure-activity relationships. In matching molecular pairs or networks of congeneric compounds, the Relative Free Energy Perturbation (RFEP) technique enables efficient activity ranking. These approaches can be applied at the subsequent lead optimization stages.