The design and synthesis of potential acetylcholinesterase inhibitors and NMDA antagonists as neuroprotective agents

Abstract

The envisaged project included the in silico assessment and synthesis of potential acetylcholinesterase inhibitors, structurally related to donepezil, and of N-methyl D-aspartate antagonists related to an active compound that was identified by the Liotta group. These inhibitors were hypothesised to act as neuroprotective agents targeting the symptomatic effects of Alzheimer’s disease. Three scaffolds were initially targeted. Scaffolds A and B, were based upon previous in-house work related to tricyclic thiazoles. These scaffolds would allow for the inhibition of acetyl cholinesterase at the catalytic anionic site (scaffold A), and at the catalytic anionic site and peripheral anionic site (scaffold B). The third, scaffold C, was envisaged to act as a potential dual neuroprotective agent inhibiting acetylcholinesterase and acting as an N-methyl D-aspartate antagonist. Computational analysis for scaffolds A to C made predictions for each scaffold’s ligand molecular docking scores (mimicking the ligands docked into the active site of acetylcholinesterase and providing an output as a negative score). The more negative the better the affinity the ligands have for acetylcholinesterase. The most potent inhibitors (docking scores < -13.000) included fifteen ligands out of the fifty-seven ligands under study. Further assessment of CYP2C9 inhibition suggested that only one of the compounds would not be a CYP2C9 inhibitor. A Qikprop analysis, predicting the drug-likeness of the ligands under study, was also conducted. All the ligands obeyed the molecular weight, QPlogBB, number of H-donor or H-acceptor ranges for orally available drugs. However, all the ligands disobeyed the QPlogHERG rule which indicates that there is a significant difference in the dose response between acetylcholinesterase and hERG inhibition. Three ligands disobeyed the QPlogPo/w rule, while the total polar surface area rule was only broken by a single ligand. As a result, the potential drug-like molecules found in scaffolds A to C was limited to twelve ligands. Finally, SwissADME bioavailability analysis was also performed predicting that seventeen of the studied ligands would be bioavailable. Comparing both SwissADME and the Schrodinger analysis for drug-likeness a total of six ligands were identified for synthesis. Ultimately, four ligands were targeted for synthesis, however, due to synthetic challenges these could not be accessed and a fourth, scaffold D, based upon an isoxazole core was identified. Scaffold D underwent a lower intensity computational analysis which included molecular docking, Qikprop analysis and CYP2C9 inhibition of fifty ligands which included twenty five α,β-unsaturated ketones and their respective cyclised isoxazole forms. The in silico analysis identified four ligands that were considered as drug-like with a docking score within 1 log unit of donepezil. The Qikprop analysis indicated that all the ligands obeyed the molecular weight, QPlogPo/w, QPlogBB, H-donor, H-acceptor and total polar surface area rules for drug-likeness and only four ligands did not violate the QPlogHERG rule. The CYP2C9 inhibition study suggested that relative to donepezil, none of the ligands should cause adverse drug reactions. Several cyclic and acyclic ketones were reacted with aldehydes affording a range of α,β-unsaturated ketones. A total of fourteen ligands were successfully synthesised in modest to good yields, however, we were unsuccessful in converting these to their related isoxazoles. Biological activity studies between the successfully synthesised α,β-unsaturated ketones and acetylcholinesterase will be conducted in the future using the appropriate acetylcholinesterase electric eel enzyme in an enzymatic assay. This analysis will determine the inhibitory concentration of 50 % of the ligand (IC50). Cytotoxicity studies will also be conducted and compared to donepezil which has an IC50 value of 0.05 ± 0.06 μM. The ligands falling below those values could then be considered as lead compounds in future studies.