Application of molecular framework-based data-mining method in the search for beta-secretase 1 inhibitors through drug repurposing

Targeting beta-secretase 1, also known as beta-amyloid precursor protein-cleaving enzyme (BACE-1) for the inhibition of amyloid production, has been intensely studied in the last decades in the search for stopping Alzheimer’s disease (AD) progression. The chances of finding a druggable BACE-1 inhibitor may be increased by drug repurposing, as this kind of molecules already fulfil certain requirements needed for further advancement. The study describes the development and application of a data-mining method based on molecular frameworks and descriptor values of tested BACE-1 inhibitors, suitable for filtering large compound databases, in order to find molecules with high potency against this protease. A total of 465 compounds extracted from the literature, tested against BACE-1, were analysed for finding molecular descriptor values and frameworks that ensure a high probability of strong inhibition. Resulting conclusions were used for filtering DrugBank database, containing ∼8700 approved and experimental drugs, obtaining 26 structures characterized by four major Bemis–Murcko frameworks: 2‐[3‐(2‐cyclohexylethyl)cyclohexyl]‐decahydronaphthalene, 3‐(2‐cyclohexylethyl)‐1,1′‐bi(cyclohexane), [5‐(cyclohexylmethyl)‐8‐cyclopentyloctyl]cyclohexane and (3‐cyclohexylcyclopentyl)cyclohexane. The compounds were further studied by molecular docking using the structure of the closed form of the enzyme, which revealed seven compounds already involved in trials targeting BACE-1 inhibition, confirming the method’s specificity. The compounds that afforded the best binding energies were DB06925 (tyrosine-protein kinase inhibitor), DB12285 (Verubecestat) and DB08899 (Enzalutamide). Moreover, docking results indicated several other molecules with high in silico inhibitory potency that can be further studied for developing a potential treatment for AD.Communicated by Ramaswamy H. Sarma