Design, synthesis, molecular modeling and neuroprotective effects of a new framework of cholinesterase inhibitors for Alzheimer’s disease

Abstract

In search of a novel class of compounds against Alzheimer’s disease (AD), a new series of 7-chloro-aminoquinoline derivatives containing methylene spacers of different sizes between the 7-chloro-4-aminoquinoline nucleus and imino methyl substituted phenolic rings, and also their reduced analogues, were designed, synthesized and evaluated as neuroprotective agents for AD in vitro. In spite of the multifaceted feature of AD, cholinesterases continue to be powerful and substantial targets, as their inhibition increases both the level and duration of the acetylcholine neurotransmitter action. The compounds presented inhibitory activity in the micromolar range against acetylcholinesterase (AChE) (imines and amines) and butyrylcholineterase (BChE) (amines). The SAR study revealed that elongation of the imine side chain improved AChE activity, whereas the reduction of these compounds to amines was crucial for higher activity and indispensable for BChE inhibition. The most promising selective inhibitors were not cytotoxic and did not stimulate pro-inflammatory activity in glial cells. Kinetic and molecular modeling studies indicated that they also show mixed-type inhibition for both enzymes, behaving as dual-site inhibitors, which can interact with both the peripheral anionic site and the catalytic anionic site of AChE. They could therefore restore cholinergic transmission and also may inhibit the aggregation of Aβ promoted by AChE. Additionally, one compound showed promising anti-inflammatory activity by reducing the microglial release of NO• at a concentration that is equivalent to the IC₅₀ against BChE (30.32 ± 0.18 µM) and 15-fold greater than the IC₅₀ against AChE (1.97 ± 0.20 µM).

Communicated by Ramaswamy H. Sarma

Keywords:

• Acetylcholinesterase
• butyrylcholinesterase
• Alzheimer’s disease
• 7-chloro-4-aminoquinoline derivatives
• neuroprotection
• neuroinflammation

Acknowledgments

We are grateful to the Molecular Spectroscopy (http://www.uff.br/lame/) and NMR (http://www.laremn.uff.br) Multiuser Laboratories at Universidade Federal Fluminense. We also thank the free software developers (Linux, Gromacs, VMD, Qtiplot, Xmgrace, etc.) who made our molecular modeling possible. T.C.C.F wish to thank the Military Institute of Engineering and the Federal University of Lavras for the software infrastructure. This work was also supported by the excellence project UHK.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

The authors gratefully acknowledge the Brazilian Federal agencies Conselho Nacional de Desenvolvimento Científico e Tecnológico—CNPq (grant numbers 306136/2011-2—M.D.V. and 308225/2018-0—T.C.C.F.; fellowship—R.F.A), Coordenação de Aperfeiçoamento de Pessoal de Nível Superior—CAPES (fellowship—V.S.Z. and visiting professor fellowship—J.A.L.), National Institutes for Science and Technology (INCT-NT grant number 465346/2014-6—F.R.S.L.) and the Rio de Janeiro Research Foundation (FAPERJ, grant numbers 26/201.352/2014—M.D.V., 02/202.961/2017—T.C.C.F.).