Structure–activity relationship studies of benzyl-, phenethyl-, and pyridyl-substituted tetrahydroacridin-9-amines as multitargeting agents to treat Alzheimer's disease

Article abstract of DOI:10.1111/cbdd.12800

A library of substituted tetrahydroacridin-9-amine derivatives were designed, synthesized, and evaluated as dual cholinesterase and amyloid aggregation inhibitors. Compound 8e (N-(3,4-dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine) was identified as a potent inhibitor of butyrylcholinesterase (BuChE IC₅₀?=?20?nm; AChE IC₅₀?=?2.2?μm) and was able to inhibit amyloid aggregation (40% inhibition at 25?μm). Compounds 9e (6-chloro-N-(3,4-dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine, AChE IC₅₀?=?0.8?μm; BuChE IC₅₀?=?1.4?μm; Aβ-aggregation inhibition?=?75.7% inhibition at 25?μm) and 11b (6-chloro-N-(3,4-dimethoxyphenethyl)-1,2,3,4-tetrahydroacridin-9-amine, AChE IC₅₀?=?0.6?μm; BuChE IC₅₀?=?1.9?μm; Aβ-aggregation inhibition?=?85.9% inhibition at 25?μm) were identified as the best compounds with dual cholinesterase and amyloid aggregation inhibition. The picolylamine-substituted compound 12c (6-chloro-N-(pyridin-2-ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine) was the most potent AChE inhibitor (IC₅₀?=?90?nm). These investigations demonstrate the utility of 3,4-dimethoxyphenyl substituent as a novel pharmacophore possessing dual cholinesterase inhibition and anti-Aβ-aggregation properties that can be used in the design and development of small molecules with multitargeting ability to treat Alzheimer's disease.

Full text of DOI:10.1111/cbdd.12800

Received Date : 01-Apr-2016
Accepted Date : 03-Jun-2016
Article type : Research Article
*Corresponding author: Tel: +1 519 888 4567 ext: 21317; e-mail: praopera@uwaterloo.ca
a School of Pharmacy, Health Sciences Campus, University of Waterloo, 200 University Avenue
West, Waterloo, Ontario, Canada N2L 3G1
Structure-Activity Relationship Studies of Benzyl, Phenethyl and Pyridyl Substituted
Tetrahydroacridin-9-amines as Multi-targeting Agents to treat Alzheimer’s Disease
Wesseem Osman 1, Tarek Mohamed 1, 2, Victor Munsing Sit 1, Maryam S. Vasefi 1, Michael A.
Beazely 1, Praveen P. N. Rao 1, *
1School of Pharmacy, Health Sciences Campus, University of Waterloo, 200 University Avenue
West, Waterloo, Ontario, Canada N2L 3G1
2 Department of Chemistry, University of Waterloo, 200 University Avenue West, Waterloo,
Ontario, Canada N2L 3G1
______________________________________________________________________________
Abstract
A library of substituted tetrahydroacridin-9-amine derivatives were designed, synthesized and
evaluated as dual cholinesterase and amyloid aggregation inhibitors. Compound 8e (N-(3,4-
dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine), was identified as a potent inhibitor of
butyrylcholinesterase (BuChE IC50 = 20 nM; AChE IC50 = 2.2 µM) and was able to inhibit
amyloid aggregation (40% inhibition at 25 µM). Compound 9e (6-chloro-N-(3,4-
This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process which may lead to
differences between this version and the Version of Record. Please cite this article as an
'Accepted Article', doi: 10.1111/cbdd.12800
This article is protected by copyright. All rights reserved.

dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine, AChE IC50 = 0.8 µM; BuChE IC50 = 1.4
µM; Aβ-aggregation inhibition = 75.7% inhibition at 25 μM) and 11b (6-chloro-N-(3,4-
dimethoxyphenethyl)-1,2,3,4-tetrahydroacridin-9-amine, AChE IC50 = 0.6 µM; BuChE IC50 =
1.9 µM; Aβ-aggregation inhibition = 85.9% inhibition at 25 μM) were identified as the best
compounds with dual cholinesterase and amyloid aggregation inhibiton. The picolylamine
substituted compound 12c (6-chloro-N-(pyridin-2-ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine),
was the most potent AChE inhibitor (IC50 = 90 nM). These investigations demonstrates the
utility of 3,4-dimethoxyphenyl substituent as a novel pharmacophore possessing dual
cholinesterase inhibition and anti-Aβ-aggregation properties that can be used in the design and
development of small molecules with multi-targeting ability to treat Alzheimer’s disease.
______________________________________________________________________________
Key words: Acetylcholinesterase, butyrylcholinesterase, beta-amyloid, Alzheimer’s disease,
tetrahydroacridin-9-amines, molecular docking
Introduction
The loss of cholinergic neurons is a characteristic hallmark of Alzheimer's disease (AD) (1-3).
The resulting deficiency in the neurotransmitter, acetylcholine (ACh), leads to poor cell
signalling, especially in the hippocampus and nucleus basalis where there is a high density of
cholinergic neurons. Consequently, there is a decline in cognition and memory leading to
dysfunction. This forms the basis of the cholinergic theory of AD. In this regard, the
cholinesterase (ChE) enzymes, acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE)
are involved in the degradation of ACh in the central nervous system (CNS). They have been a
This article is protected by copyright. All rights reserved.

major pharmacological target, resulting in the development of a number of cholinesterase
inhibitors (ChEIs, Fig 1, 1–3) that are currently the mainstay of AD pharmacotherapy (4-7).
Accumulation of amyloid-beta (Aβ) in the CNS is hypothesized by the amyloid cascade pathway
which shows that misprocessing of amyloid precursor protein (APP), can promote the formation
of Aβ-peptide fragments, that tends to accumulate, resulting in the formation of various
neurotoxic Aβ-aggregates. From a therapeutic perspective, targeting Aβ accumulation is also a
major focus in developing novel AD therapies (8-11). Unfortunately, the current
pharmacotherapy options for AD, based on ChEIs, provide only symptomatic relief and are not
effective over the long term. Consequently, new approaches toward AD treatment are aimed at
developing hybrid molecules that can target multiple pathological routes of AD (12-14).
Tacrine (Fig 1, 4), is a potent dual ChEI and has been used as a template in the design of
novel anti-AD hybrid small molecules (15-24). A number of chemical modifications have been
carried out with tacrine to afford agents with multi-targeting ability which has been elegantly
summarized in a recent review (25). In a previous study (26), we have shown that a 3,4-
dimethoxybenzyl substituent attached to a pyrimidine ring was able to prevent Aβ-aggregation.
Accordingly, in the current study, we used the tetrahydroacridin-9-amine scaffold with chemical
modifications at the 9-amino-position to design potential multi-targeting agents. A library of
1,2,3,4-tetrahydroacridin-9-amine derivatives, with various C-6, C-7 and C-9 substituents were
synthesized and evaluated to assess their ability to (i) inhibit human AChE and BuChE enzymes
and (ii) prevent self-induced Aβ-aggregation, by investigating the aggregation kinetics profile.
Our investigations show that these novel tetrahydroacridin-9-amine derivatives, exhibit potent
ChE inhibition along with anti-Aβ-aggregation properties.
This article is protected by copyright. All rights reserved.

Materials and Methods
General
All the chemicals, reagents and solvents were obtained from commercial vendors (Sigma-
Aldrich, Acros Organics and Alfa Aesar, USA) with ≥ 95% purity, and were used without further
purification. The tetrahydroacridin-9-amine derivatives 7a-c, 8a, 8d, 9a and 10a are known
compounds which were reported earlier (27–31). 1H NMR spectra were obtained and analyzed
using a 300 MHz spectrometer (Bruker® Avance, Department of Chemistry, University of
Waterloo). 1H NMR chemical shifts are in ppm relative to deuterated chloroform (CDCl3).
Coupling constants (J values) were recorded in hertz (Hz) and the following abbreviations were
used to represent multiplets of NMR signals: s = singlet, d = doublet, t = triplet, m = multiplet, br
= broad. Thin layer chromatography (TLC) was performed using Merck 60F254 selica gel plates
and visualized at UV 254 nm. Test compounds were purified by colum chromatography using 60
µM 230-400 mesh silica gel. Melting points were obtained via Fisher-Johns apparatus and were
uncorrected. High resolution mass spectra (HRMS) were recorded on a Thermo Scientific Q
ExactiveTM mass spectrometer with an ESI source, Department of Chemistry, University of
Waterloo. Compound purity was assessed by HPLC (Agilent 1100 series) using a Zobrax Eclipse
XDB-C8 column (4.6 x 150 mm) using 0.1% TFA in water/acetonitrile (30:70) as the eluent
system with a flow rate of 1-1.5 mL (UV detection 254 nm). All the tested compounds were ≥
95% pure.
This article is protected by copyright. All rights reserved.

Chemistry
Synthesis of aminobenzoic (anthranilic) acid derivatives (6a-c)
These intermediates were prepared by employing two different routes as below.
Method A: 1.67 g (10 mmol) of 5a (X = H) was dissolved in ethanol (EtOH) (25 mL) under
argon at 0 °C in a 100 mL round-bottom flask (RBF). Pd/C (0.001 mol, 10% wt) was added to
the reaction mixture followed by drop-wise addition of hydrazine hydrate (1.28 g, 40 mmol). The
reaction mixture was refluxed for 1.5 h at 80 °C then cooled to r.t. Pd/C was filtered off via
cotton plug and EtOH was evaporated in vacuo. Upon drying, the product, 6a, was obtained as a
light yellow/brown solid (80% yield, Scheme 1).
Method B: 2.01 g (10 mmol) of 5b or 5c (X = Cl), were dissolved in EtOH (35 mL) at 0 °C
under argon in a 100 mL RBF. Then treated with tin(II) chloride dihydrate (5.64 g, 25 mmol) for
60 min. at r.t. EtOH was evaporated in vacuo. The dried product was re-dissolved in EtOAc (10
mL). The EtOAc solution was washed with brine solution (6 x 20 mL), dried with MgSO4,
filtered off and EtOAc was evaporated in vacuo to afford the light-yellow solid products, 6b or
6c (23 - 27% yield, Scheme 1).
2-Amino-benzoic acid (6a)
The product was obtained by reducing 5a with hydrazine hydrate and Pd/C. Yield 80%. mp: 144-
146 °C. 1H NMR (300 MHz, CDCl3): δ 7.90 (d, J = 8.0 Hz, 1H), δ 7.31-7.26 (m, 1H), δ 6.67-
6.63 (m, 2H).
This article is protected by copyright. All rights reserved.

Amino-4-chlorobenzoic acid (6b)
The product is obtained by reducing 5b with tin(II) chloride (23%). mp: 205-207 °C. 1H NMR
(300 MHz, CDCl3): δ 7.80 (d, J = 8.5 Hz, 1H), δ 6.66 (d, J = 2.0 Hz, 1H), δ 6.63 - 6.59 (dd, J =
8.5 Hz, 2.0 Hz, 1H).
2-Amino-5-chlorobenzoic acid (6c)
The product is obtained by reducing 5c with tin(II) chloride (27%). mp: 196-198 °C. 1H NMR
(300 MHz, CDCl3): δ 7.86 (d, J = 2.5 Hz,1H), δ 7.25-7.22 (m, 1H), δ 6.62 (d, J = 9.0 Hz, 1H).
General method to synthesize 9-chloro, 6,9- and 7,9- dichloro-1,2,3,4-tetrahydroacridine
derivatives (7a-c)
Compounds 7a-c were synthesized by adding 6a, 6b, or 6c (10 mmol) and cyclohexanone (0.98
g, 10 mmol) in a dry 100 mL RBF at kept at 0 °C in ice bath. Phosphorous oxychloride (POCl3)
(9.3 mL, 100 mmol) was added drop-wise, and the solution was allowed to mix for 5 min.
followed by the addition of 3Å molecular sieves. The reaction mixture was refluxed for 2.5 h at
120 °C and then cooled for 20 min. The solution was poured onto ice and then 80 mL of water,
kept at 0 °C, was added. A precipitate was formed and filtered off. The filtrate was neutralized
with sodium carbonate (Na2CO3) to precipitate out any remaining product. The precipitate was
collected and dissolved in dichloromethane (DCM, 50 mL). The solution was washed five times
with saturated NaHCO3 solution (20 mL). The organic layer was collected, dried with MgSO4
and the solvent was evaporated in vacuo. The crude product (7a-c) was purified by flash
chromatography using DCM as the eluent. The collected fractions were dried under reduced
pressure to afford yellow solid (51-82% yield, Scheme 2).
This article is protected by copyright. All rights reserved.

9-Chloro-1,2,3,4-tetrahydroacridine (7a)
The product was obtained by coupling 6a with cyclohexanone in the presence of POCl3. Yield
51%. mp: 67-69 °C 1H NMR (300 MHz, CDCl3): δ 8.15 (d, J = 8.5 Hz 1H), δ 7.96 (d, J = 8.5
Hz, 1H), δ 7.66 (t, J = 7.0 Hz, 1H), δ 7.61 (t, J = 7.0 Hz, 1H), δ 3.22-3.11 (m, 2H), δ 3.05 - 2.99
(m, 2H), δ 1.97-1.88 (m, 4H).
6,9-Dichloro-1,2,3,4-tetrahydroacridine (7b)
The product was obtained by coupling 6b with cyclohexanone in the presence of POCl3. Yield
82%. mp: 83-85 °C. 1H NMR (300 MHz, CDCl3): δ 8.94 (m, 1H), δ 7.99 (m, 1H), δ 7.47 (dd, J =
9.0 Hz, 2.0 Hz, 1H), δ 3.11-3.09 (m. 2H), δ 3.01-2.99 (m, 2H), δ 1.97-1.88 (m, 4H).
7,9-Dichloro-1,2,3,4-tetrahydroacridine (7c)
The product was obtained by coupling 6c with cyclohexanone in the presence of POCl3. Yield
75%. mp: 79-81°C. 1H NMR (300 MHz, CDCl3): δ 8.11 (d, J = 2.0 Hz, 1H), δ 7.88 (d, J = 9.0
Hz, 1H), δ 7.56 (dd, J = 9.0 Hz, 2.0 Hz,1H), δ 3.08-3.06 (m, 2H), δ 3.00-2.87 (m, 2H), δ 1.96-
1.88 (m, 4H).
General method to synthesize substituted-N-benzyl, phenethyl and pyridyl-substituted-1,2,3,4-
tetrahydroacridine-9-amine derivatives (8a-e, 9a-e, 10a-e, 11a-c and 12a-f)
Method A: Coupling of the various benzylamines (benzylamine, 2- or 3- or 4-
methoxybenzylamines and 3,4-dimethoxybenzylamines) and picolylamines (2- or 3-
This article is protected by copyright. All rights reserved.

picolylamines) to 7a-c was carried out by adding, sodium iodide (0.089 g, 0.6 mmol) to 7a, 7b or
7c (2 mmol each) kept in a 25 mL RBF and was dissolved in n-butanol (5 mL). Benzylamine or
picolylamines (5 mmol each) was added and the reaction mixture was refluxed at 150 ºC for 4.5
h, cooled, and solvent was removed in vacuo. It was re-dissolved in DCM (20 mL) and washed
three times with concentrated brine solution (15 mL). The organic layer was dried with MgSO4,
and the solvent was evaporated in vacuo. The remaining product was purified via flash
chromatography with EtOAc as the eluent twice or thrice to afford compounds 8a-e, 9a-e, 10a-e,
11a-c and 12a-f.
Method B: In another variation, compounds 8a-e, 9a-e, and 10a-e, were synthesized by coupling
7a-c (2 mmol each) with various benzylamines (5 mmol each) in the presence of NaI (0.089 g,
0.6 mmol) and by refluxing at 180 °C for 3 h using phenol (5 mL) as a solvent, instead of n-
butanol. The cooled reaction mixture was dissolved in DCM (20 mL) and washed three times
with concentrated brine solution (15 mL). The organic layer was dried with MgSO4, and the
solvent was evaporated in vacuo. The remaining product was purified via flash chromatography
with EtOAc as the eluent twice or thrice to afford compounds 8a-e, 9a-e and 10a-e. Both
methods A and B, provided yields ranging from 12-47% (Scheme 3 and 4). Analytical data for
the tacrine derivatives 8a-e, 9a-e, 10a-e, 11a-c and 12a-f, is given below.
N-Benzyl-1,2,3,4-tetrahydroacridin-9-amine (8a)
The product was obtained by coupling (7a) with benzylamine. Yield 45%. mp:111-113 ºC. 1H
NMR (300 MHz, CDCl3): δ 7.90 (t, J = 8.0 Hz, 2H), δ 7.49 (t, J = 8.0 Hz, 1H), δ 7.35-7.25 (m,
6H), δ 4.56 (d, J = 6.0 Hz, 2H), δ 4.17 (br s, 1H), δ 3.02 (t, J = 6.0 Hz, 2H), δ 2.57 (t, J = 6.0 Hz,
This article is protected by copyright. All rights reserved.

2H), δ 1.86-1.78 (m, 4H). HRMS (ESI) m/z calcd for C15H14N4 [M + H]+ 289.1705, found:
289.1697.
N-(2-Methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (8b)
The product was obtained as an oil after coupling (7a) with 2-methoxybenzylamine. Yield 40%.
1H NMR (300 MHz, CDCl3): δ 8.35 (d, J = 8.5 Hz, 1H), δ 8.02 (d, J = 8.5 Hz, 1H), δ 7.51 (t, J =
8.0 Hz, 1H), δ 7.33-7.22 (m, 3H), δ 6.89-6.78 (m, 2H), δ 4.88 (s, 2H), 3.80 (br s, 1H), δ 3.78 (s,
3H), δ 3.18-3.08 (m, 2H), δ 2.61-2.56 (m, 2H), 1.87-1.76 (m, 4H). HRMS (ESI) m/z calcd for
C21H23N2O [M + H]+ 319.1810, found: 319.1805.
N-(3-Methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (8c)
The product was obtained as an oil after coupling (7a) with 3-methoxybenzylamine. Yield 32%.
1H NMR (300 MHz, CDCl3): δ 7.90 (t, J = 8.0 Hz, 2H), δ 7.52 (t, J = 8.0 Hz, 1H), δ 7.35-7.23
(m, 2H), δ 6.92-6.81 (m, 3H), δ 4.56 (d, J = 6.0 Hz, 2H), δ 4.21-4.16 (m, 1H), δ 3.75 (s, 3H), δ
3.04 (t, J = 6.0 Hz, 2H), δ 2.61 (t, J = 6.0 Hz, 2H), δ 1.92-1.81 (m, 4H). HRMS (ESI) m/z calcd
for C21H23N2O [M + H]+ 319.1810, found: 319.1803.
N-(4-Methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (8d)
The product was obtained as an oil after coupling (7a) with 4-methoxybenzylamine. Yield 30%.
1H NMR (300 MHz, CDCl3): δ 7.89 (t, J = 8.0 Hz, 2H), δ 7.51 (t, J = 8.0 Hz, 1H), δ 7.32 (t, J =
8.0 Hz, 1H), δ 7.24 (d, J = 8.0 Hz, 2H), δ 6.85 (d, J = 8.0 Hz, 2H), δ 4.54 (d, J = 6.0 Hz, 2H), δ
4.20-4.12 (m, 1H), δ 3.79 (s, 3H), δ 3.03 (t, J = 6.0 Hz, 2H), δ 2.58 (t, J = 6.0 Hz, 2H), δ 1.95-
1.84 (m, 4H). HRMS (ESI) m/z calcd for C21H23N2O [M + H]+ 319.1810, found: 319.1805.
This article is protected by copyright. All rights reserved.

N-(3,4-Dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (8e)
The product was obtained as an oil after coupling (7a) with 3,4-dimethoxybenzylamine (41%).
1H NMR (300 MHz, CDCl3): δ 7.94 (d, J = 8.5 Hz, 1H), δ 7.90 (d, J = 8.5 Hz, 1H), δ 7.52 (t, J =
8.0 Hz, 1H), δ 7.30 (t, J = 8.0 Hz, 1H), δ 6.81 (q, J = 8.0 Hz, 2H), δ 6.76 (s, 1H), δ 4.55-4.52 (m,
2H), δ 4.13 (br s, 1H), δ 3.86 (s, 3H), δ 3.78 (s, 3H), δ 3.02 (t, J = 6.0 Hz, 2H), δ 2.60 (t, J = 6.0
Hz, 2H), δ 1.86-1.80 (m, 4H). HRMS (ESI) m/z calcd for C22H25N2O2 [M + H]+ 349.1916,
found: 349.1909.
N-Benzyl-6-chloro-1,2,3,4-tetrahydroacridin-9-amine (9a)
The product was obtained as an oil after coupling (7b) with benzylamine. Yield 41%. 1H NMR
(300 MHz, CDCl3): δ 8.01-7.94 (m, 1H), δ 7.89 (d, J = 9.0 Hz, 1H), δ 7.38-7.29 (m, 6H), δ 4.63
(s, 2H), δ 4.27 (br s, 1H), δ 3.09-3.00 (m, 2H), 2.60-2.54 (m, 2H), 1.91-1.79 (m, 4H). HRMS
(ESI) m/z calcd for C20H20ClN2 [M + H]+ 323.1315, found: 323.1309.
6-Chloro-N-(2-methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (9b)
The product was obtained as an oil after coupling (7b) with 2-methoxybenzylamine. Yield 25%.
1H NMR (300 MHz, CDCl3): δ 7.91 (d, J = 8.0 Hz, 2H), δ 7.27 (s, 1H), δ 7.15 (d, J = 7.0 Hz,
1H), δ 6.87-6.77 (m, 3H), δ 4.56 (s, 2H), δ 3.81 (br s, 1H), δ 3.80 (s, 3H), δ 3.02-3.00 (m, 2H), δ
2.60-2.58 (m, 2H), δ 1.85-1.82 (m, 4H). HRMS (ESI) m/z calcd for C21H22N2O [M + H]+
353.1421, found: 353.1414.
This article is protected by copyright. All rights reserved.

6-Chloro-N-(3-methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (9c)
The product was obtained as an oil after coupling (7b) with 3-methoxybenzylamine. Yield 15%.
1H NMR (300 MHz, CDCl3): δ 7.92 (s, 1H), δ 7.87 (d, J = 9.0 Hz, 1H), δ 7.30-7.26 (m, 2H), δ
6.90-6.84 (m, 3H), δ 4.57 (s, 2H), δ 4.19 (br s, 1H), δ 3.77 (s, 3H), δ 3.01 (t, J = 6.0 Hz, 2H), δ
2.57 (t, J = 6.0 Hz, 2H), δ 1.87-1.85 (m, 4H). HRMS (ESI) m/z calcd for C21H22N2O [M + H]+
353.1421, found: 353.1415.
6-Chloro-N-(4-methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (9d)
The product was obtained as an oil after coupling (7b) with 4-methoxybenzylamine. Yield 14%.
1H NMR (300 MHz, CDCl3): δ 8.05 (d, J = 9.0 Hz, 1H), 7.95-7.80 (m, 2H), 7.43 (dd, J = 9.0 Hz,
2.0 Hz, 1H), δ 7.27-7.19 (m, 2H), δ 6.85 (d, J = 8.0 Hz, 1H), δ 4.53-4.51 (m, 2H), δ 4.29 (br s,
1H), δ 3.80 (s, 3H), δ 3.10-2.97 (m, 2H), δ 2.58-2.54 (m 2H),1.89-1.83 (m, 4H). HRMS (ESI)
m/z calcd for C21H22N2O [M + H]+ 353.1421, found: 353.1414.
6-Chloro-N-(3,4-dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (9e)
The product was obtained by coupling (7b) with 3,4-dimethoxybenzylamine. Yield 35%. mp:
131-133 ºC. 1H NMR (300 MHz, CDCl3): δ 7.90-7.87 (m, 2H), δ 7.27-7.24 (m, 1H), δ 6.87-6.81
(m, 2H), δ 6.75 (s, 1H), δ 4.53 (s, 2H), δ 4.13 (br s, 1H), δ 3.87 (s, 3H), δ 3.79 (s, 3H), δ 3.00 (t,
J = 6.0 Hz, 2H), δ 2.55 (t, J = 6.0 Hz, 2H), δ 1.87-1.75 (m, 4H). HRMS (ESI) m/z calcd for
C22H24ClN2O2 [M + H]+ 383.1526, found: 383.1519.
This article is protected by copyright. All rights reserved.

N-Benzyl-7-chloro-1,2,3,4-tetrahydroacridin-9-amine (10a)
The product was obtained as an oil after coupling (7c) with benzylamine. Yield 36%.1H NMR
(300 MHz, CDCl3): δ 8.54 (d, J = 9.0 Hz, 1H), δ 8.17 (s, 1H), δ 7.59 (dd, J = 9.0 Hz, 2.0 Hz,
1H), δ 7.44-7.25 (m, 5H), δ 5.91(br s, 1H), δ 5.06 (s, 2H), δ 3.29 (t, J = 6.0 Hz, 2H), δ 2.59 (t, J
= 6.0 Hz, 2H), δ 1.90-1.87 (m, 4H). HRMS (ESI) m/z calcd C20H20ClN2 [M + H]+ 323.1315,
found: 323.1310.
7-Chloro-N-(2-methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (10b)
The product was obtained as an oil after coupling (7c) with 2-methoxybenzylamine. Yield 12%.
1H NMR (300 MHz, CDCl3): δ 8.00 (d, J = 2.0 Hz, 1H), δ 7.79 (d, J = 9.0 Hz, 1H), δ 7.42 (dd, J
= 9.0 Hz, 2.0 Hz, 1H), δ 7.20 (t, J = 8.5 Hz, 1H), δ 7.10 (d, J = 8.5 Hz, 1H), δ 6.81 (t, J = 8.5 Hz,
2H), δ 4.46 (s, 2H), δ 4.38 (br s, 1H), δ 3.81 (s, 3H), δ 2.97 (t, J = 6.0 Hz, 3H), δ 2.58 (t, J = 6.0
Hz, 2H), δ 1.85-1.78 (m, 4H). HRMS (ESI) m/z calcd for C21H22N2O [M + H]+ 353.1421, found:
353.1415.
7-Chloro-N-(3-methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (10c)
The product was obtained as an oil after coupling (7c) with 3-methoxybenzylamine. Yield 17%.
1H NMR (300 MHz, CDCl3): δ 7.94-7.93 (m, 1H), δ 7.81 (d, J = 9.0 Hz, 1H), δ 7.43 (dd, J = 9.0
Hz, 2.0 Hz, 1H), δ 7.27-7.22 (m, 1H), δ 6.88-6.81 (m, 3H), 4.51 (d, J = 6.0 Hz, 2H), δ 4.11 (br s,
1H), δ 3.74 (s, 3H), δ 2.99 (t, J = 6.0 Hz, 2H), δ 2.55 (t, J = 6.0 Hz, 2H), δ 1.84-1.81 (m, 4H).
HRMS (ESI) m/z calcd for C21H22N2O [M + H]+ 353.1421, found: 353.1415.
This article is protected by copyright. All rights reserved.

7-Chloro-N-(4-methoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (10d)
The product was obtained as an oil after coupling (7c) with 4-methoxybenzylamine. Yield 20%.
1H NMR (300 MHz, CDCl3): δ 7.96 (s, 1H), δ 7.83 (d, J = 9.0 Hz, 1H), δ 7.50-7.46 (m, 1H), δ
7.19 (d, J = 8.0 Hz, 2H), δ 6.86 (d, J = 8.0 Hz, 2H), δ 4.51 (s, 2H), δ 4.03 (br s, 1H), δ 3.80 (s,
3H), δ 3.00 (t, J = 6.0 Hz, 2H), δ 2.54 (t, J = 6.0 Hz, 2H), 1.85-1.82 (m, 4H). HRMS (ESI) m/z
calcd for C21H22N2O [M + H]+ 353.1421, found: 353.1414.
7-Chloro-N-(3,4-dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine (10e)
The product was obtained after coupling (7c) with 3,4-dimethoxybenzylamine. Yield 20%. mp:
113-115 ºC.1H NMR (300 MHz, CDCl3): δ 7.96 (d, J = 2.0 Hz, 1H), δ 7.82 (d, J = 9.0 Hz, 1H), δ
7.46 (dd, J = 9.0 Hz, 2.0 Hz, 1H), δ 6.87-6.81 (m, 2H), δ 6.75 (s, 1H), δ 4.50 (d, J = 6.0 Hz, 2H),
δ 4.09 (br s, 1H), δ 3.87 (s, 3H), δ 3.80 (s, 3H), δ 2.99 (t, J = 6.0 Hz, 2H), δ 2.56 (t, J = 6.0 Hz,
2H), δ 1.87-1.83 (m, 4H). HRMS (ESI) m/z calcd for C22H24ClN2O2 [M + H]+ 383.1526, found:
383.1518.
N-(3,4-Dimethoxyphenethyl)-1,2,3,4-tetrahydroacridin-9-amine (11a)
The product was obtained as an oil after coupling 7a with 3,4-dimethoxyphenethylamine. Yield
31%. 1H NMR (300 MHz, CDCl3): δ 7.91-7.54 (m, 2H), δ 7.49 (t, J = 9.0 Hz, 2.0 Hz, 1H), δ
7.24 (dd, J = 9.0 Hz, 2.0 Hz, 1H), δ 6.82-6.72 (m, 2H), δ 6.64 (s, 1H), δ 4.50 (d, J = 6.0 Hz, 2H),
δ 4.05 (br s, 1H), δ 3.77 (s, 3H), δ 3.74 (s, 3H), 3.72-3.61 (m, 2H), δ 3.00 (t, J = 6.0 Hz, 2H), δ
2.84 (t, J = 6.0 Hz, 2H), δ 2.45 (t, J = 6.0 Hz, 2H), δ 1.90-1.82 (m, 4H). MS (ESI) m/z calcd for
C23H27N2O2 [M + H]+ 363.20, found: 363.18.
This article is protected by copyright. All rights reserved.

6-Chloro-N-(3,4-dimethoxyphenethyl)-1,2,3,4-tetrahydroacridin-9-amine (11b)
The product was obtained as a solid after coupling 7b with 3,4-dimethoxyphenethylamine. Yield
25%. mp: 87-89 ºC.1H NMR (300 MHz, CDCl3): δ 7.87 (d, J = 2.0 Hz, 1H), δ 7.74 (d, J = 9.0
Hz, 1H), δ 7.20 (d, J = 2.0 Hz, 1H), δ 6.82-6.6.72 (m, 2H), δ 6.63-6.60 (m, 1H), δ 4.02 (br s,
1H), δ 3.82 (s, 3H), δ 3.78 (s, 3H), δ 3.76-3.73 (m, 2H), δ 2.96 (t, J = 6.0 Hz, 2H), δ 2.84 (t, J =
6.0 Hz, 2H), δ 2.42 (t, J = 6.0 Hz, 2H), δ 1.85-1.81 (m, 4H). MS (ESI) m/z calcd for
C23H26ClN2O2 [M + H]+ 397.16, found: 397.14.
7-Chloro-N-(3,4-dimethoxyphenethyl)-1,2,3,4-tetrahydroacridin-9-amine (11c)
The product was obtained as a solid after coupling 7c with 3,4-dimethoxyphenethylamine. Yield
27%. mp: 96-99 ºC.1H NMR (300 MHz, CDCl3): δ 7.82 (d, J = 9.0 Hz, 1H), δ 7.76 (d, J = 2.0
Hz, 1H), δ 7.43 (dd, J = 9.0 Hz, 2.0 Hz, 1H), δ 6.84-6.73 (m, 2H), δ 6.70-6.67 (m, 1H), δ 3.84 (s,
3H), δ 3.83-3.81 (br s, 1H), δ 3.80 (s, 3H), δ 3.73-3.70 (m, 2H), δ 3.02-2.98 (m, 2H), δ 2.86 (t, J
= 6.0 Hz, 2H), δ 2.43 (t, J = 6.0 Hz, 2H), δ 1.89-1.82 (m, 4H). HRMS (ESI) m/z calcd for
C23H26ClN2O2 [M + H]+ 397.16, found: 397.15.
N-(Pyridin-2-ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine (12a)
The product was obtained by coupling (7a) with 2-picolylamine. Yield 35%. mp:103-105 ºC 1H
NMR (300 MHz, CDCl3): δ 8.64 (d, J = 5.0 Hz, 1H), δ 8.09 (d, J = 8.0 Hz, 1H), δ 7.89 (d, J =
8.0 Hz, 1H), δ 7.60 (td, J = 8.0 Hz, 2.0 Hz, 1H), δ 7.51 (td, J = 8.0 Hz, 2.0 Hz, 1H), δ 7.38-7.33
(m, 1H), δ 7.17-7.21 (m, 2H), 5.86 (br s, 1H), δ 4.76 (d, J = 5.0 Hz, 2H), δ 3.10-3.00 (m, 2H), δ
2.90-2.80 (m, 2H), δ 1.92-1.87 (m, 4H). HRMS (ESI) m/z calcd for C19H20N3 [M + H]+
290.1657, found: 290.1650.
This article is protected by copyright. All rights reserved.

N-(Pyridin-3-ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine (12b)
The product was obtained as an oil after coupling (7a) with 3-picolylamine. Yield 40%. 1H NMR
(300 MHz, CDCl3): δ 8.60 (d, J = 2.0 Hz, 1H), δ 8.53 (dd, J = 5.0 Hz, 1.5 Hz, 1H), δ 7.98-7.84
(m, 2H), δ 7.62-7.53 (m, 2H), δ 7.32 (t, J = 8.0 Hz, 1H), δ 7.27-7.23 (m, 1H), δ 4.57 (d, J = 6.0
Hz, 2H), δ 4.07-4.11 (m, 1H), δ 3.03 (t, J = 6.0 Hz, 2H), δ 2.61 (t, J = 6.0 Hz, 2H), δ 1.89-1.83
(m, 4H). HRMS (ESI) m/z calcd for C19H20N3 [M + H]+ 290.1657, found: 290.1650.
6-Chloro-N-(pyridin-2-ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine (12c)
The product was obtained by coupling (7b) with 2-picolylamine. Yield 36%. mp:117-119 ºC. 1H
NMR (300 MHz, CDCl3): δ 8.65 (d, J = 5.0 Hz, 1H), δ 8.08-7.99 (m, 2H), δ 7.66 (td, J = 8.0 Hz,
1.6 Hz, 1H), δ 7.29 (dd, J = 9.0 Hz, 2.0 Hz, 1H), δ 7.28-7.23 (m, 2H), δ 5.99 (br s, 1H), δ 4.86
(d, J = 6.0 Hz, 2H), δ 3.08-3.05 (m, 2H), δ 2.82-2.79 (m, 2H), δ 1.92-1.90 (m, 4H). HRMS (ESI)
m/z calcd for C19H19ClN3 [M + H]+ 324.1268, found: 324.1261.
6-Chloro-N-(pyridin-3-ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine (12d)
The product was obtained by coupling (7b) with 3-picolylamine. Yield 26%. mp: 87-89 ºC 1H
NMR (300 MHz, CDCl3): δ 8.59 (d, J = 2.0 Hz, 1H), δ 8.55 (dd, J = 5.0 Hz, 1 Hz, 1H), δ 7.93 (d,
J = 2.0 Hz, 1H), δ 7.84 (d, J = 9.0 Hz, 1H), δ 7.58 (d, J = 8.0 Hz, 1H), δ 7.26 (dd, J = 9.0 Hz, 2.0
Hz, 2H), δ 4.60 (d, J = 6.0 Hz, 2H), δ 4.25 (br s, 1H), δ 3.01 (t, J = 6.0 Hz, 2H), δ 2.57 (t, J = 6.0
Hz, 2H), δ 1.91-1.89 (m, 4H). HRMS (ESI) m/z calcd for C19H19ClN3 [M + H]+ 324.1268, found:
324.1261.
This article is protected by copyright. All rights reserved.

7-Chloro-N-(pyridin-2-ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine (12e)
The product was obtained as an oil after coupling (7c) with 2-picolylamine. Yield 47%. 1H
NMR (300 MHz, CDCl3): δ 8.66-8.63 (m, 1H), δ 8.10 (d, J = 2.0 Hz, 1H), δ 7.83 (d, J = 8.5 Hz,
1H), δ 7.68-7.62 (m, 1H), δ 7.46 (dd, J = 9.0 Hz, 2.0 Hz, 1H), δ 7.24-7.19 (m, 2H), δ 5.93 (br s,
1H), δ 4.79-4.73 (m, 2H), δ 3.10-2.99 (m, 2H), δ 2.85-2.75 (m, 2H), δ 1.93-1.81 (m, 4H). HRMS
(ESI) m/z calcd for C19H19ClN3 [M + H]+ 324.1268, found: 324.1261.
7-Chloro-N-(pyridin-3-ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine (12f)
The product was obtained by coupling (7c) with 3-picolylamine. Yield 27%. mp:126-128 ºC. 1H
NMR (300 MHz, CDCl3): δ 8.59 (d, J = 2.0 Hz, 1H), δ 8.55 (dd, J = 5.0 Hz, 1.5 Hz, 1H), δ 7.93-
7.90 (m, 2H), δ 7.59 (d, J = 8.0 Hz, 1H), δ 7.48 (dd, J = 9.0 Hz, 2.0 Hz, 1H), δ 7.28-7.23 (m,
1H), δ 4.60 (d, J = 6.0 Hz, 2H), δ 4.24 (br s, 1H), δ 3.03 (t, J = 6.0 Hz, 2H), δ 2.57 (t, J = 6.0 Hz,
2H), δ 1.87-1.84 (m, 4H). HRMS (ESI) m/z calcd for C19H19ClN3 [M + H]+ 324.1268, found:
324.1261.
Biological assays
Cholinesterase inhibition assay
Test compound (8a-e, 9a-e, 10a-e, 11a-c and 12a-f) were evaluated for ChEI activity by
following Ellman's method (32). In this method, test compounds compete with substrates
acetylthiocholine iodide (ATChI) and S-butyrylthiocholine iodide (BuTChI) for human AChE
(Sigma-Aldrich, St. Louis, MO) and human BuChE (Sigma-Aldrich, St. Louis, MO),
respectively. The concentration at which test compounds produce 50% inhibition of ChE activity
This article is protected by copyright. All rights reserved.

(IC50 values) were evaluated and were compared with ChE inhibition data for reference
compounds tacrine, donepezil and galantamine. Test compounds were prepared in DMSO
(maximum concentration used 1% v/v) and 10 µL each (0.001-25 µM final concentration
range), was incubated for 5 minutes at room temperature with 160 µL of 1.5 mM DTNB, 50 µL
0.22 U/mL AChE (in 50 mm Tris.HCl, pH 8.0, 0.1% w/v bovine serum albumin, BSA) or 50 µL
0.12 U/mL BuChE (in 50 mm Tris.HCl, pH 8.0, 0.1% w/v bovine serum albumin, BSA). After
the incubation period, 30 µL of ATChI (15 mM, prepared in ultra pure water) or BuTChI (15
mM, prepared in ultra pure water) were added to 96-well plates. Test compound inhibition of
ChE was measured via UV absorbance (412 nm wavelength) using a microplate reader
(SpectraMax M5) at various time intervals (t = 0, 1, 2, 3, 4 and 5 min). Appropriate controls,
without the test compounds and ChE were included. The inhibitory concentration (IC50 values)
was calculated from the concentration–inhibition dose response curve on a logarithmic scale
based on two independent experiments (n = 3).
Aβ-aggregation inhibition assay
The anti-Aβ aggregation activity of test compounds (8e, 9e, 10e and 11a-c), were evaluated
using the ThT-based fluorescence assay (33–35). The Aβ1-42 hexafluoro-2-propanol (HFIP)
(rPeptide, Geogia, USA) stock solution was prepared by dissolving in 1% NH4OH solution, to a
1mg/mL stock solution, followed by dilution in phosphate buffer (pH 8.0) to 500 μM. Stock
solutions of test compounds were prepared in DMSO, diluted in phosphate buffer (pH 8.0), and
were sonicated for 30 min. The final DMSO concentration per each well was 1% v/v or lower.
This article is protected by copyright. All rights reserved.

The ThT fluorescent dye stock solution (15 μM) was prepared in 50 mM glycine buffer (pH 8.5).
The aggregation kinetics assay was carried out using a Corning® 384-well flat, clear bottom
black plates. Each well contains 44 μl of ThT, 20 μl of phosphate buffer (pH 8.0), 8 μl of test
compounds in different concentrations (1, 5, 10 and 25 μM, final concentration) and 8 μl of Aβ1-
42 (5 μM final concentration). The plate was incubated at 37 °C with a plate cover under shaking
and fluorescence was measured every 5 min. using a SpectraMax M5 multimode plate reader
(excitation = 440 nm and emission = 490 nm) over a period of 14 h. Appropriate control
experiments that contain Aβ1-42 and test compound alone were evaluated as well. The known Aβ
aggregation inhibitor orange G (Sigma-Aldrich, St. Louis, MO) was used as a reference
compound. The percentage inhibition was calculated using the equation 100% control – [(IFi –
IFo)] where 100% control indicates no inhibitor, IFi and IFo are the fluorescence intensities in
the presence and absence of ThT. The results were expressed as percentage inhibition of two
separate experiments with triplicate measurements
Molecular docking studies
The modeling experiments were carried out using Discovery Studio (DS), Structure-Based-
Design version 4.0, software program from BIOVIA/Accerlys, San Diego, USA. Tacrine
derivatives 8e, 9e, 10e and 12c were constructed using the small molecules module in DS. The x-
ray coordinates of human AChE (pdb id:4EY7) and BuChE (pdb id:2XQJ) were obtained from
RCSB protein data bank. The enzymes were prepared for docking using the macromolecules
module in DS. A 15 Å binding site sphere was defined for both AChE and BuChE, using bound
This article is protected by copyright. All rights reserved.

ligands, which were deleted after. Docking simulation was carried out using the CDOCKER
algorithm which included simulated annealing with 2000 heating steps, 700k target temperature
and 5000 cooling steps, with a 300K cooling target temperature to obtain 10 docked poses. The
entire simulation was carried out using CHARMm force field. The docked poses were evaluated
using CDOCKER energy, CDOCKER interaction energy in kcal/mol and by considering the
type and number of polar and nonpolar contacts (35, 36). The coordinates for the Aβ-dimer
model was obtained from RCSB protein data bank (pdb id: 2LMN), whereas the octamer
assembly of Aβ-steric zipper was built using the known x-ray crystal structure (pdb id:3OVJ).
The docking simulation was carried out using the CDOCKER algorithm as reported previously
(37, 38) by defining 20 Å (Aβ-dimer) and 15 Å (Aβ-steric zipper) sphere respectively.
Cell Viability Assay
Test compounds 8e, 9e and 12a–d were assessed for their ability to maintain the viability of SH-
SY5Y neuroblastoma cells (35, 39). Tacrine was used a reference standard. Viable cells produce
mitochondrial dehydrogenase which reduce 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl
tetrazolium bromide (MTT) into a formazan intermediate, while non-viable cells will be unable
to generate formazan which can be monitored by measuring UV absorption at 570 nm. The
percentage cells, which remain viable after being incubated with test compounds is directly
related to the concentration of MTT reduced to formazan, as compared to control groups. SH-
SY5Y cells were plated at a density of 4 x 105 cells/mL in growth media consisting of 1:1
Dulbecco's modified eagle media (DMEM) and Ham's F12 with the addition of glutamate (2.5
mM) and 10% fetal bovine serum (FBS). Cells were incubated for 24 h at 37 ºC with test
compound solutions (25 μM each) with no more than 2% DMSO. The MTT solution was added
to test wells, which made up 10% of well volume, followed by 3 h incubation at 37 ºC in 5%
This article is protected by copyright. All rights reserved.

CO2. SH-SY5Y cells were then solubilized, with an MTT reagent solution (10% Triton X-100
and 0.1 N HCl in anhydrous isopropanol) making up 50% of the final well volume. Wells were
read at 570 nm absorbance wavelength with the final percent viability being determined via
percent reduction of MTT into formazan compared to control wells (untreated with test
compound).
Results and Discussion
Chemistry
The tetrahydroacridin-9-amine derivatives 8a-e, 9a-e, 10a-e, 11a-c and 12a-f were synthesized
as shown in Scheme 1-4. The initial synthesis of the 2-aminobenzoic acid intermediate
(anthranilic acids 6a-c, Scheme 1) was accomplished by reducing the corresponding 2-
nitrobenzoic acid (5a-c), to amines in the presence of hydrazine over palladium-carbon (Pd/C)
catalyst in ethanol (40). Although these conditions provided good yield (80%), it led to
dechlorination of 4-and 5-chloro-2-nitrobenzoic acids (5a and 5b) respectively. In order to
prevent the dechlorination, an alternative method was employed where tin(II) chloride (SnCl2)
was used as a reducing agent, which afforded the desired 2-aminobenzoic acids (6b and 6c,
Scheme 1) in lower yields (23–27%) (41). In the next step, the tricyclic, 9-chloro-
tetrahydroacridine precursor (7a-c, Scheme 1), was obtained by coupling the 2-aminobenzoic
acid (6a-c) with cyclohexanone in the presence of phosphorous oxychloride (POCl3) (18). Yields
ranged from moderate to good range (50–80%). The tetrahydroacridine (THA) derivatives 8a-e,
were synthesized by carrying out a nucleophilic substitution reaction, where 7a (X = H), was
coupled with various substituted-benzylamines (benzylamine, 2- or 3- or 4-
methoxybenzylamines and 3,4-dimethoxybenzylamines respectively, Scheme 2) in the presence
of sodium iodide (NaI) as a catalyst and under reflux at 150 °C, using n-butanol or phenol as a
This article is protected by copyright. All rights reserved.

solvent, to afford 8a-e in 32–45% yield (Scheme 2) (42). Similar conditions were employed to
synthesize target compounds 9a-e, 10a-e and 11a-c (12–41%, Schemes 2 and 3). In addition, the
C-9 N-pyridinylmethyl-substituted tacrine derivatives (12a-f) were synthesized by coupling 7a-c
with picolylamines as shown in Scheme 4 (26–47% yield).
Cholinesterase inhibition
The synthesized tetrahydroacridin-9-amine derivatives 8a-e, 9a-e, 10a-e, 11a-c and 12a-f were
evaluated for their inhibition activity toward human AChE and BuChE using Ellman’s assay
(Table 1) (26, 32). The structure-activity relationship (SAR) data shows, these tetrahydroacridin-
9-amine derivatives exhibited dual inhibition of both AChE and BuChE with some derivatives
exhibiting potent inhibition in the nanomolar range. The C-9 position was substituted with
methoxy-benzylamines to develop bivalent inhibitors of AChE. In the N-benzyl-1,2,3,4-
tetrahydroacridine-9-amine series of compounds (8a-e), the presence of a methoxy group (R1, R2
and R3) attached to the C-9 N-benzylamine substituent, either at ortho-, meta- or para-position,
enhanced AChE inhibitory potency, with either a 4-or a 3,4-dimethoxy group providing the best
AChE inhibition (8d, AChE IC50 = 2.2 μM; 8e, AChE IC50 = 2.2 μM). These tetrahydroacridin-
9-amine derivatives were less potent inhibitors of AChE compared to tacrine (AChE IC50 = 0.16
μM). Interestingly, compounds 8a-e, exhibited potent inhibition of BuChE (IC50 = 0.02–0.4 μM
range, Table 1). In this regard, the compound 8e, with a 3,4-dimethoxybenzylamine substituent
was identified as the most potent and selective inhibitor of BuChE (IC50 = 0.02 μM, SI = 110,
Table 1) and was 2-fold more potent compared to the reference agent tacrine (BuChE IC50 = 0.04
μM).
This article is protected by copyright. All rights reserved.

In further SAR optimization, C-6 or C-7 positions were substituted with a chlorine group (9a-e
and 10a-e, X = Cl, Table 1). Generally, the presence of a chlorine substituent either at C-6 or C-7
position, enhanced AChE selectivity compared to unsubstituted derivatives 8a-e (Table 1). In the
C-6 chloro-substituted series, compounds 9c-e, exhibited almost similar AChE inhibition activity
profile (AChE IC50 = 0.6-0.8 μM range, Table 1) and were about 3-4 fold more potent compared
to the reference agent galantamine (AChE IC50 = 2.6 μM). The C-7 chloro-substituted
compounds 10a-e, exhibited dual inhibition of AChE and BuChE, with the 3,4-
dimethoxybenzylamine substituted compound 10e, identified as the most potent AChE inhibitor
(AChE IC50 = 1.3 μM, Table 1). Among these series of C-6 and C-7 substituted compounds, 9e
(6-chloro-N-(3,4-dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine) was identified as the best
compound with dual AChE and BuChE inhibition profile (AChE IC50 = 0.8 μM; BuChE IC50 =
1.4 μM). Encouraged by the potent and dual AChE/BuChE inhibition exhibited by
tetrahydroacridin-9-amine derivatives 8e and 9e, possessing a C-9 3,4-dimethoxybenzylamine
substituent, we varied the carbon chain length at the C-9 position by preparing the
phenethylamine derivatives 11a-c (Table 1). In this regard, the C-6 unsubstituted compound 11a,
exhibited potent and highly selective BuChE inhibition (BuChE IC50 = 0.08 μM; AChE IC50 =
4.5 µM; SI = 562, Table 1). Substitution of the chlorine at either C-6 or C-7 was detrimental to
BuChE inhibition (compounds 11b and 11c) and 11b was identified as the most potent dual
AChE and BuChE inhibitor (AChE IC50 = 0.6 μM; BuChE IC50 = 1.9 μM) in the phenethylamine
series of compounds.
In other C-9 modifications, the effect of replacing the phenyl ring with either a 2- or 3-pyridyl
substituent was explored (12a-f, Table 1). In this series, compounds exhibited moderate to potent
inhibition of both AChE (IC50 = 0.09–6.7 μM range) and BuChE (IC50 = 0.07– >25 μM range).
This article is protected by copyright. All rights reserved.

Compound 12c with a 2-picolylamine and a C-6 chlorine substituent, was the most potent AChE
inhibitor discovered in this entire compound library (AChE IC50 = 0.09 μM, Table 1). It was
about 1.7 fold more potent compared to tacrine (AChE IC50 = 0.16 μM), whereas the C-6
unsubstituted compound 12a, possessing a C-9 2-picolylamine substituent, was the most potent
BuChE inhibitor (IC50 = 0.07 μM, Table 1) with high selectivity toward BuChE (SI = 957, Table
1). In contrast, the presence of a C-6 chlorine substituent promoted AChE selectivity with
compound 12d exhibiting about 63-fold selectivity over AChE relative to BuChE inhibition
(AChE IC50 = 0.2 μM; BuChE IC50 = 12.7 μM, Table 1). Among this picolylamine series, 12c
was identified as the best compound with dual AChE/BuChE inhibition profile (AChE IC50 =
0.09; BuChE IC50 = 1.6 μM, Table 1). It should be noted that the presence of a 2-pyridyl
substituent, closer to the central pyridine-4-amine of the THA template, has the potential to
chelate metal ions due to the presence of 1,4-diamine system, as reported by Ramamoorthy and
coworkers (43).
The binding interactions of the most potent, dual AChE and BuChE inhibitors 8e, 9e and 12c,
were investigated by conducting molecular docking experiments (35). The binding mode of 8e
(N-(3,4-dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine) within the human AChE enzyme
shows that the tricyclic THA ring was oriented closer to the catalytic site and was stacked
against Trp86 (cation binding site, Figure 2a). The pyridine and phenyl aromatic rings of THA
underwent π-π interactions with indole ring of Trp86 (distance < 5 Å) and the non-aromatic
cyclohexene, was in close proximity to Trp86, His447 and Ser203 (distance < 5 Å). The C-9 3,4-
dimethoxybenzyl substituent was oriented perpendicular to the plane of tetrahydroacridine and
was closer to the peripheral anionic site (PAS) as per our hypothesis, in the vicinity of Tyr72,
Asp74, Tyr124 and Trp286. The benzyl aromatic ring underwent π-π interactions with aromatic
This article is protected by copyright. All rights reserved.

rings of Tyr341 and Tyr124 (distance < 5.5 Å). The 3- and 4-OMe groups, underwent van der
Waal’s interactions with Trp286 and Val294 (distance < 5 Å). These results show that compound
8e, acts as a bivalent inhibitor of AChE due to its ability to interact with both the catalytic site
and PAS (44, 45). Its binding in human BuChE shows that unlike its binding to AChE, the THA
ring was oriented closer to the catalytic site (Figure 2b) with the central pyridine forming a
hydrogen bond with Ser198 (distance < 3.5 Å) and binds in a U-shaped conformation, which was
not seen in AChE. The THA ring was in van der Waal’s contact with amino acids Gly115,
Gly116 and Gly117 (distance < 5 Å). Furthermore, the THA phenyl ring underwent T-shaped π-
π interactions with Phe329 (distance < 5.5 Å) whereas the cyclohexene underwent π-alkyl
interactions with Trp82 (cationic subsite). The C-9 benzylamine NH forms a hydrogen bond with
Thr120 and the 3,4-dimethoxybenzyl substituent was in a hydrophobic pocket made up of Ser79,
Tyrp82, Phe329, Tyr332 and Trp430 (distance < 5 Å). These observations account for the potent
inhibition exhibited by 8e (BuChE IC50 = 0.02 µM).
The binding interactions of the C-6 chlorinated compound 9e, in the AChE active site shows
that, the tetrahydroacridine ring was in van der Waal’s contact with Trp86 and underwent a
number of π-π and π-alkyl interactions (distance < 5 Å) as seen with 8e (Figure 2c). In addition,
the C-6 chlorine substituent, underwent hydrophobic interactions with Pro88 (distance < 5.2 Å).
The C-9 3,4-dimethoxybenzyl substituent was oriented toward the PAS and was in a nonpolar
region consisting of Tyr72, Tyr124, Trp286, Phe295, Phe297 and Tyr341. The benzyl aromatic
ring underwent π-π interactions, with Tyr341 (distance < 4 Å), whereas the 3- and 4-OMe
substituents were in close proximity to Tyr72 and Val294 (Figure 2c). These observations
support the ability of 9e (AChE IC50 = 0.8 µM) to act as a bivalent inhibitor. Its binding mode in
BuChE shows that the tetrahydroacridine ring was closer to the catalytic site and the glycine
This article is protected by copyright. All rights reserved.

pocket (Figure 2d). The central pyridine ring formed a hydrogen bond with Ser198 (distance <
2.9 Å). Interestingly, the C-6 chlorophenyl group of tetrahydroacridine, was oriented in a
hydrophobic pocket consisting of Trp231, Leu286, Val288 and Phe329 (distance < 5 Å) where it
underwent π-halogen and π-π interactions. The non-aromatic cyclohexene ring of
tetrahydroacridine, underwent π-alkyl interactions with Trp82 (distance < 5Å). The C-9 3,4-
dimethoxybenzyl substituent was oriented away from the entrance and was in a nonpolar pocket
consisting of Ala328, Trp430, Met437 and Tyr440 (Figure 2d).
The binding interactions of the most potent AChE inhibitor (12c, IC50 = 90 nM) possessing a 2-
picolylamine substituent at C-9 (6-chloro-N-(pyridin-2-ylmethyl)-1,2,3,4-tetrahydroacridin-9-
amine), shows that in the AChE active site, the tetrahydroacridine ring was closer to a
hydrophobic area made up of Trp86, Tyr337, Phe338, Tyr341 and His447 (distance < 5 Å,
Figure 3a). Interestingly, the tetrahydroacridine phenyl ring with a C-6 chloro-substituent,
underwent a number of π-π and π-halogen interactions with Phe297, Phe338 and Tyr341
(distance < 5 Å) which seems to enhance its binding affinity toward AChE. Furthermore, the C-9
2-picolylamine substituent was not oriented toward the PAS. Instead, it was closer to Trp86, the
cationic subsite where it underwent π-π interactions (distance < 5 Å, Figure 3a). These
investigations show that a 3,4-dimethoxybenzyl PAS-binding pharmacophore (26) is necessary
to exhibit bivalent binding. The binding of 12c in BuChE shows an extended, linear
conformation which was not seen with the C-9 3,4-dimethoxybenzylamine substituted tacrine
derivatives (Figure 3b). The tetrahydroacridine ring was stacked against Trp82 (distance < Å).
The C-6 chlorophenyl substituent of tetrahydroacridine underwent π-π and π-halogen
interactions with Trp82 (distance < 5 Å), whereas the cyclohexene ring underwent π-alkyl
This article is protected by copyright. All rights reserved.

interactions with Trp82, Trp430 and van der Waal’s interactions with Ala328 (distance < 5 Å).
The C-9 picolylamine was in van der Waal’s contact with Tyr332 (distance < 5 Å, Figure 3b).
Self-induced Aβ1-42 aggregation inhibition
The Aβ-aggregation activity profile of tetrahydroacridin-9-amine derivatives 8e, 9e, 10e and
11a-c, possessing a 3,4-dimethoxybenzylamine substituent, was investigated by thioflavin T
(ThT) based fluorescence spectroscopy (33–35). The time-dependent Aβ1-42 aggregation kinetics
in the presence and absence of various concentrations of test compounds (1, 5, 10 and 25 µM)
was measured by monitoring ThT fluorescence (excitation wavelength = 440 nm; emission
wavelength = 490 nm) over a 14 h incubation period and were compared with the known
inhibitor orange G (Figure 4). All the compounds tested, exhibited a concentration dependent
decline in fluorescence intensities as a function of time. The C-6 unsubstituted tetrahydroacridin-
9-amine derivative 8e exhibited 40% inhibition and was less potent compared to orange G
(65.6% inhibition, Figure 4). In compound 11a, the addition of a methylene group at C-9, led to
moderate enhancement in Aβ-aggregation inhibition (47.2% inhibition). Significantly, the
presence of chlorine substituent at either C-6 or C-7 position led to drastic increase in Aβ-
aggregation inhibition, with compounds 9e, 10e, 11b and 11c exhibiting 75.7–87.5% inhibition
range (Figure 4). The aggregation kinetics profile of the most potent inhibitor, compound 10e
(87.5% inhibition), is shown in Figure 5. These studies show that, in the absence of 10e, a
characteristic Aβ-aggregation curve was observed, with a short lag phase, steep growth phase
and a saturation phase. At 1 µM, compound 10e, did not affect the aggregation profile. However,
at a concentration of 5 µM, 10e exhibited weak inhibition (11% at 14 h), with greater inhibition
This article is protected by copyright. All rights reserved.

seen at 10 µM (30.4% inhibition). At 25 µM, 10e exhibited significant inhibition (87.5%, Figure
5) and was superior to orange G (65.6% inhibition at 25 µM, Figure 4).
In order to understand the binding interactions of 10e, with Aβ-aggregates, we conducted
molecular docking of 10e, with full length Aβ-dimer and Aβ-steric zipper assembly consisting of
the 16KLVFFA21 hexapeptide segment (37, 38). In the Aβ-dimer model, at the C-terminal, 10e
was closer to 17LVFFAED23 segment while at the N-terminal, it was interacting with 32IGL34
(Figure 6). The cyclohexyl ring of tetrahydroacridine, underwent π-alkyl and π-π interactions
with leucine and phenylalanine respectively (distance < 5.4 Å, Figure 6). The central pyridine
ring formed T-shaped π-π interaction with a phenylalanine (distance < 5.3 Å) and the phenyl
ring with a C-7 chloro-substituent was oriented toward the N-terminal, where a halogen-alkyl
interaction was seen with the side chains of leucine and isoleucine respectively (distance < 5 Å).
The aromatic ring of C-9, 3,4-dimethoxylbenzylamine substituent, underwent T-shaped π-π
interactions with a phenylalanine (distance < 5.5 Å, Figure 6). These studies suggest that 10e can
stabilize the dimer assembly, and prevent further aggregation. In the steric-zipper
16KLVFFA21-hexapeptide assembly, 10e was oriented in a linear, perpendicular fashion, to the
fiber axis (Figure 7). The THA ring underwent a number of nonpolar contacts with the amino
acid side chains that line the steric zipper interface. The cyclohexyl ring underwent π-alkyl and
π-π interactions, with leucine, valine and phenylalanine respectively (distance < 5.3 Å) whereas
the central pyridine was involved in π-alkyl and π-π interactions on either side, with amino acid
side chains of phenylalanine and valine respectively (distance < 5 Å, Figure 7). The C-7
chlorophenyl ring was in a nonpolar region consisting of phenylalanines on either side (distance
< 5.3 Å). The C-9 3,4-dimethoxybenzylamine substituent underwent π-π interactions with
phenylalanine whereas the 3- and 4-OMe substituents underwent hydrogen bonding interactions
This article is protected by copyright. All rights reserved.

with lysine amino acids on either side of the steric zipper interface (distance < 3.1 Å) as shown
in Figure 7. These studies suggest that 10e has the potential to bind various Aβ-aggregates
including dimers, oligomers and other higher-order aggregates.
Cytotoxicity in SH-SY5Y neuroblastoma cells
The cell viability of some potent ChE inhibitors 8e, 9e and 12a–d, was evaluated using the 3-
(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium (MTT) bromide assay (Figure 8) (26, 39).
The SH-SY5Y cell viability in the presence of 25 µM of test compounds range from 58.9–100%.
The C-9 3,4-dimethoxybenzylamine substituted compounds 8e and 9e, exhibited good cell
viability (93.8 and 81.8% respectively), compared to tacrine (89.6% cell viability, Figure 8). In
the C-9 picolylamine series of compounds, 12a was toxic with cell viability of 58.9% whereas
other compounds tested in this series (12b–d) exhibited good cell viability (75.4–100%). These
compounds had ClogP values in the range of 4.02–5.99 and compounds in the less lipophilic
picolylamine series (except 12a) exhibited better cell viability profile.
Conclusions
Our investigations on novel tetrahydroacridin-9-amine derivatives 8a-e, 9a-e, 10a-e, 11a-c and
12a-f, have shown that, C-9 modification provides multi-targeting ability, with dual
cholinesterase and amyloid aggregation inhibition. From this library, compound 8e (N-(3,4-
dimethoxybenzyl)-1,2,3,4-tetrahydroacridin-9-amine), was identified as the most potent BuChE
inhibitor with nanomolar inhibition (IC50 = 20 nM), whereas 12c (6-chloro-N-(pyridin-2-
ylmethyl)-1,2,3,4-tetrahydroacridin-9-amine) was identified as the most potent AChE inhibitor
(IC50 = 90 nM). Significantly, compound 9e (6-chloro-N-(3,4-dimethoxybenzyl)-1,2,3,4-
tetrahydroacridin-9-amine) was identified as the best compound with multi-targeting ability
This article is protected by copyright. All rights reserved.

(AChE IC50 = 0.8 µM; BuChE IC50 = 1.4 µM; Aβ1-42 inhibition at 25 µM = 75.7%; cell viability
= 81.8%). These SAR studies provide valuable insight in the application of medicinal chemistry
principles to design and develop multi-targeting agents, as potential disease-modifying therapies
to treat AD.
Acknowledgments
The authors would like to thank Katherine Tran, Faculty of Science, University of Waterloo for
assistance in compound synthesis and purification, Ontario Graduate Scholarship, Rx&D
Graduate Scholarship (WO), Ontario Mental Health Foundation for a PhD studentship (TM),
NSERC-Discovery Grant to MB (RGPIN 371384-2013), to PR (RGPIN 03830-2014), Canada
Foundation for Innovation (CFI-JELF) Leaders Fund (PR), Ontario Research Fund (ORF) and
ERA (PR), Ministry of Research and Innovation, Government of Ontario, Canada for financial
support.
Conflict of Interest
All authors have no conflict of interest to declare.
References
1. Schliebs R., Arendt T. (2011) Cholinergic system in aging and neuronal degeneration. Behav
Brain Res 221: 555–563.
2. Mufson EJ, Counts SE, Perez SE, Ginsberg SD. (2008) Cholinergic system during the
progression of Alzheimer’s disease: therapeutic implications. Expert Rev Neurother 8:1703–
1718.
3. Wolfe, MS. (2002) Therapeutic strategies for Alzheimer’s disease. Nat Rev Drug Discov 1:
859–866.
This article is protected by copyright. All rights reserved.

4. Massoulie J, Bon S. (1982). The molecular forms of cholinesterase and acetylcholinesterase in
vertebrates. Annu Rev Neurosci 5:57–106.
5. Soreq H, Seidman S. (2001) Acetylcholinesterase-new roles for an old actor. Nat Rev
Neurosci 2: 294–302.
6. Darvesh S, Hopkins DA, Geula C. (2003) Neurobiology of butyrylcholinesterase. Nat Rev
Neurosci 4: 131–138.
7. Birks J. (2006) Cholinesterase inhibitors for Alzheimer’s disease. Cochrane Database Syst
Rev 1, CD005593.
8. Selkoe DJ. (2001). Alzheimer’s disease results from the cerebral accumulation and
cytotoxicity of amyloid beta-protein. J Alzheimers Dis 3:75–80.
9. Hardy J, Selkoe DJ. (2002) The amyloid hypothesis of Alzheimer’s disease: progress and
problems on the road to therapeutics. Science 297:353-356.
10. Karran E, Mercken M, De Strooper B. (2011) The amyloid cascade hypothesis for
Alzheimer’s disease: an appraisal for the development of therapeutics. Nat Rev Drug Discov 10:
698–712.
11. Mucke L, Selkoe DJ. (2012) Neurotoxicity of amyloid β-protein: synaptic and network
dysfunction. Cold Spring Harb Perspect Med 2:a006338.
12. Cavalli A, Bolognesi ML, Minarini A, Rosini M, Tumiatti V, Recanatini M, Melchiorre C.
(2008) Multi-target-directed ligands to combat neurodegenerative diseases. J Med Chem 51:
347–372.
13. Zheng H, Fridkin M, Youdim M. (2014) From single target to multitarget/network
therapeutics in Alzheimer’s therapy. Pharmaceuticals 7: 113–135.
This article is protected by copyright. All rights reserved.