3D-QSAR and pharmacophore mapping of 1H-indole-2,3-dione adducts inhibiting beta-amyloid aggregation and other major targets related to Alzheimer’s Disease.

Fibrillation of b-amyloid (Ab) is recognized as key neurodegenerative process leading to the Alzheimer’s disease (AD), which is among major causes of illness and death especially in developed countries. Unfortunately, current therapeutic approaches for AD are still limited to the symptomatic treatment of memory impairment by acetylcholinesterase (AChE) inhibitors and memantine as NMDA antagonist, whereas research efforts are focused on targeting Ab aggregation and related neurotoxicity and the discovery of molecules able to inhibit one or more steps of the amyloid cascade1. In this context, our medicinal chemistry research programs have long been focused on a multitarget approach having the anti-amyloidogenic activity as the central feature, and in previously works we investigated the chemical space around the isatin/indolinone molecular core, disclosing potent in- vitro inhibitors of Ab aggregation, acting at different stages of fibrillogenesis2,3. In this study, thirty- six indole derivatives, 3-(2-phenylhydrazono) isatins and structurally related 1H-indole-3- carbaldehyde compounds, were synthesized and assayed as inhibitors of Ab aggregation whit IC50 values from sub- to two-digit micromolar range4. Some of the new compounds were tested also for their inhibitory activity against monoamine oxidases (MAO A/B), showing noteworthy multitarget activity. Moreover, a cell-based assay in tau overexpressing bacterial cells disclosed promising additional activity of some derivatives against tau aggregation. The data of about ninety already reported and newly synthesized indolinone derivatives were used to generate a pharmacophore hypothesis of anti-amyloidogenic activity, discriminating ‘active’ from ‘inactive’ (or poorly active) compounds. An atom-based 3D-QSAR model was also derived for about 80% of ‘active’ compounds. The 3D-QSAR model (encompassing 4 PLS latent variables), featuring predictive statistics both in training (n = 45, q2 = 0.596) and external test set (n = 14, r2ext = 0.695), usefully complemented the pharmacophore model, shedding light on the physicochemical features mainly correlated with the Ab anti-aggregating potency. This in-silico approaches allowed us to furnish more confident predictions for de novo design and/or molecular optimization of new indolinone-based anti-AD small molecule agents.