A series of pyrrolidine derivatives have been used to study the main structural requirements for designing novel Mcl-1 inhibitors. For this purpose, three models CoMSIA, CoMFA and HQSAR were generated using QSAR molecular modeling techniques. The statistical results of the CoMFA (Q² = 0.689; R = 0.999; R²_pred = 0.986), CoMSIA (Q² = 0.614; R² = 0.923; R²_pred = 0.815) and HQSAR (Q²= 0.603; R² = 0.662; R²_pred = 0.743) models showed good stability and predictability. The results of the models were presented as contours and colored fragments indicating the favorable and unfavorable contribution to the inhibitory activity of Mcl-1. Based on the obtained results, four new compounds were designed with more potent predicted pIC₅₀ inhibitory activity. The ADME/Tox results and the pharmacokinetic properties revealed that these four compounds are orally bioavailable and show good permeability. In addition the four compounds showing non-inhibitors of CYP3A4 and CYP2D6 with the exception of Pred03. At the level of toxicity profile, the compounds Pred01, Pred02 and Pred03 showed interesting results and showed no AMES toxicity, no hERG inhibition and no skin sensitization. Molecular docking results were used to uncover the mode of interaction between the ligand and key residues of protein binding site. Molecular docking results were supported by molecular simulation and binding free energy estimation (MMPBSA). These results demonstrate the stability of the analyzed compounds in the target protein binding site during a 100 ns trajectory. Finally, all these results create a strong lead to develop promising new Pyrrolidine-based inhibitors against Mcl-1. Communicated by Ramaswamy H. Sarma

The p53 is a tumor suppressor protein that adjusts cell cycle and growth arrest as well as genes that restore DNA damage and apoptosis. Murine double minute 2 (MDM2) is a main p53 antagonist. We created a novel QSAR model using a series of highly active spiro [pyrrolidin-3,2-oxindoles] that consisted of 29 compounds that were experimentally validated to inhibit the MDM2-p53 interaction. Three optimal models have been developed CoMFA/E+S, CoMSIA/S+H+A and HQSAR have revealed good statistical results, but the CoMSIA mode only which validates all the external validation tests applied successfully. Based on the CoMSIA/S+H+A model was carefully chosen to design four compounds with values ​​of inhibitory activity greater than the highly active compound in the data set. The Newly designed compounds were docked in the target receptor binding site (ID: 4LWU). The newly designed compound Pred 01 showed the highest affinity with a value of -9.4kcal / mol, while compound N°04 which represents the data set and control compound (Nutlin-3) showed binding energies of the order of -8.8 kcal / mol and -8.2kcal / mol, respectively. In addition, the roles of lipinski and veber were estimated, the results obtained demonstrate that the proposed molecules involve good oral bioavailability and an ability to diffuse through different biological barriers. For in-depth study, The Pred01 / receptor, N°04 / receptor and Nutlin-3 / receptor complexes were selected via dynamic simulation analyzes with a simulation time of 100 ns and, also their free binding energy was examined operating the MM-GBSA approach. The molecular docking results obtained accentuate the crucial residues responsible for the ligand / protein interaction, providing insight into the mode of interaction. The MD simulation analysis confirms the conformational stability of the selected complexes during the MD trajectory, and the fluctuations recorded are insignificant. The results of MM-GBSA reveal that the new compound Pred 01 exhibits the lowest free energy, which confirms the result of molecular docking.