Computational modelling of potent β-secretase (BACE1) inhibitors towards Alzheimer's disease treatment.

Abstract Researchers have identified the β-amyloid precursor protein cleaving enzyme 1 (BACE1) in the multifactorial pathway of Alzheimer's disease (AD) as a drug target. The design and development of molecules to inhibit BACE1 as a potential cure for AD thus remained significant. Herein, we simulated two potent BACE1 inhibitors (AM-6494 and CNP-520) to understand their binding affinity at the atomistic level. AM-6494 is a newly reported potent BACE1 inhibitor with an IC50 value of 0.4 nM in vivo and now picked for preclinical considerations. Umibecestat (CNP-520), which was discontinued at human trials lately, was considered to enable a reasonable evaluation of our results. Using density functional theory (DFT) and Our Own N-layered Integrated molecular Orbital and Molecular Mechanics (ONIOM), we achieved the aim of this investigation. These computational approaches enabled the prediction of the electronic properties of AM-6494 and CNP-520 plus their binding energies when complexed with BACE1. For AM-6494 and CNP-520 interaction with protonated BACE1, the ONIOM calculation gave binding free energy of −62.849 and −33.463 kcal/mol, respectively. In the unprotonated model, we observed binding free energy of −59.758 kcal/mol in AM-6494. Taken together thermochemistry of the process and molecular interaction plot, AM-6494 is more favourable than CNP-520 towards the inhibition of BACE1. The protonated model gave slightly better binding energy than the unprotonated form. However, both models could sufficiently describe ligand binding to BACE1 at the atomistic level. Understanding the detailed molecular interaction of these inhibitors could serve as a basis for pharmacophore exploration towards improved inhibitor design.