Integrated pharmacophore-based virtual screening and molecular modeling approaches for the identification of sigma-2 receptor antagonists as novel therapeutics against Alzheimer’s disease

Alzheimer's disease (AD) is a progressive and irreversible neurodegenerative disorder with a conundrum pathophysiology. Disruption of cholesterol homeostasis in AD is well known. Recently, sigma-2 ligands have been reported for their involvement in abnormal lipid metabolism associated with AD that may lead to disturbance in amyloid-β (Aβ) production, over activity of β- and γ- secretase enzymes, and neuroinflammation. Therefore, targeting sigma-2 receptor inhibition is a plausible mechanism to combat AD. Computational tools aid in screening substantial chemical libraries to unveil potential leads against the desired protein target in less time and cost-effectively. In the present study, five chemical databases (Molport, Mcule, Zinc, ChEMBL, and Enamine) were screened against a pharmacophore model, followed by removing duplicates. The obtained 12,811 hits were subjected to PAINS (Pan assay interference compounds), BRENK (Brenk's rule-based filters), Lipinski's rule of five, structural diversity, and BBB (blood-brain barrier) permeability filters followed by comprehensive molecular docking studies. Further, the top fifteen hits obtained were evaluated based on their predicted pharmacokinetic and toxicity profiles. The binding free energy (ΔG) calculations were carried out for the selected hits by performing an MMPBSA (Molecular mechanics Poisson-Boltzmann surface area) assay followed by regressive MD simulation studies. Finally, <b>ZINC00184959872</b>, <b>ZINC001704299697,</b> and <b>MolPort-046-745-161</b> were obtained as potential virtual leads against the specific sigma-2 receptor with ΔG values of -34.09, -30.93 and -28.03 kcal/mol, respectively, satisfying all the significant parameters undertaken for the study along with optimum pharmacokinetic properties, minimal toxicity, acceptable RMSD value, and stable protein-ligand complex trajectory throughout the MD simulation run (200 ns).