Discovery of novel 2,5-dihydroxyterephthalamide derivatives as multifunctional agents for the treatment of Alzheimer's disease

Article abstract of DOI:10.1016/j.bmc.2018.11.015

A series of 2,5-dihydroxyterephthalamide derivatives were designed, synthesized and evaluated as multifunctional agents for the treatment of Alzheimer's disease. In vitro assays demonstrated that most of the derivatives exhibited good multifunctional activities. Among them, compound 9d showed the best inhibitory activity against both RatAChE and EeAChE (IC₅₀ = 0.56 μM and 5.12 μM, respectively). Moreover, 9d exhibited excellent inhibitory effects on self-induced Aβ_1–42 aggregation (IC₅₀ = 3.05 μM) and Cu²⁺-induced Aβ_1–42 aggregation (71.7% at 25.0 μM), and displayed significant disaggregation ability to self- and Cu²⁺-induced Aβ_1–42 aggregation fibrils (75.2% and 77.2% at 25.0 μM, respectively). Furthermore, 9d also showed biometal chelating abilities, antioxidant activity, anti-neuroinflammatory activities and appropriate BBB permeability. These multifunctional properties highlight 9d as promising candidate for further studies directed to the development of novel drugs against AD.

Full text of DOI:10.1016/j.bmc.2018.11.015

Accepted Manuscript
Discovery of novel 2,5-dihydroxyterephthalamide derivatives as multifunction-
al agents for the treatment of Alzheimer’s disease
Qing Song, Yan Li, Zhongcheng Cao, Hongyan Liu, Chaoquan Tian, Ziyi Yang,
Xiaoming Qiang, Zhenghuai Tan, Yong Deng
PII:
S0968-0896(18)31761-9
https://doi.org/10.1016/j.bmc.2018.11.015
BMC 14619
DOI:
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
15 October 2018
10 November 2018
12 November 2018
Please cite this article as: Song, Q., Li, Y., Cao, Z., Liu, H., Tian, C., Yang, Z., Qiang, X., Tan, Z., Deng, Y.,
Discovery of novel 2,5-dihydroxyterephthalamide derivatives as multifunctional agents for the treatment of
Alzheimer’s disease, Bioorganic & Medicinal Chemistry (2018), doi: https://doi.org/10.1016/j.bmc.2018.11.015
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Discovery of novel 2,5-dihydroxyterephthalamide derivatives as multifunctional
agents for the treatment of Alzheimer’s disease
a
a
a
a
a
a
Qing Song , Yan Li , Zhongcheng Cao , Hongyan Liu , Chaoquan Tian , Ziyi Yang , Xiaoming
a
b
a,*
Qiang , Zhenghuai Tan , Yong Deng
a
Department of Medicinal Chemistry, Key Laboratory of Drug-Targeting and Drug Delivery System
of the Education Ministry, Sichuan Engineering Laboratory for Plant-Sourced Drug and Sichuan
Research Center for Drug Precision Industrial Technology, West China School of Pharmacy, Sichuan
University, Chengdu, 610041, P. R. China
b
Institute of Traditional Chinese Medicine Pharmacology and Toxicology, Sichuan academy of
Chinese Medicine Sciences, Chengdu, 610041, P. R. China
*
Corresponding Author.
E-mail: dengyong@scu.edu.cn (Yong Deng)
1

Abstract
A series of 2,5-dihydroxyterephthalamide derivatives were designed, synthesized and evaluated as
multifunctional agents for the treatment of Alzheimer’s disease. In vitro assays demonstrated that
most of the derivatives exhibited good multifunctional activities. Among them, compound 9d
showed the best inhibitory activity against both RatAChE and EeAChE (IC = 0.56 μM and 5.12
5
0
μM, respectively). Moreover, 9d exhibited excellent inhibitory effects on self-induced Aβ1-42
2
+
aggregation (IC50 = 3.05 μM) and Cu -induced Aβ
aggregation (71.7% at 25.0 μM), and
1-42
2
+
displayed significant disaggregation ability to self- and Cu -induced Aβ
aggregation fibrils
-42
1
(75.2% and 77.2% at 25.0 μM, respectively). Furthermore, 9d also showed biometal chelating
abilities, antioxidant activity, anti-neuroinflammatory activities and appropriate BBB permeability.
These multifunctional properties highlight 9d as promising candidate for further studies directed to
the development of novel drugs against AD.
Keywords:
Alzheimer’s disease; 2,5-Dihydroxyterephthalamide derivatives; Acetylcholinesterase inhibitors; Aβ
aggregation inhibitors; Antioxidant; Anti-neuroinflammatory agents
2

1
. Introduction
Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by memory
loss and cognitive impairments, which can cause both physical and mental injury even resulting in
1
,2
death. According to the World Alzheimer Report, there were 46.8 million people worldwide
3
suffering from dementia in 2015 and this number will reach 131.5 million in 2050. AD has
4
multifactorial pathophysiological hallmarks. Although the exact etiology of AD remains unknown,
many factors have been described to play definitive roles in its pathogenesis, including degeneration
of cholinergic neurons, β-amyloid (Aβ) deposition, τ-protein aggregation, dyshomeostasis of
5
biometals, oxidative stress, and neuroinflammation.
“Cholinergic hypothesis” of AD demonstrated there was a functional impairment of basal
forebrain cholinergic neurons with the consequence of the decreased levels of acetylcholine (ACh)
6
resulting in learning and memory dysfunctioning. Following this, Acetylcholinesterase (AChE) was
validated as a therapeutic target to increase cholinergic levels, and AChE inhibitors are effective in
temporarily restoring cholinergic function. Recently, AChE was indicated to be involved in Aβ
aggregation-promoting action. Bivalent AChE inhibitors simultaneously blocking both the catalytic
active site (CAS) and particularly the peripheral anionic sites (PAS) might not only alleviate the
cognitive deficit of AD patients by elevating ACh levels, but also act as disease-modifying agents
7
delaying amyloid plaque formation. Butyrylcholinesterase (BuChE) is also the key enzyme that
plays important roles in cholinergic transmission by hydrolyzing the neurotransmitter ACh. However,
serious inhibition of BuChE may contribute to the potential peripheral side effects of ChE inhibitors,
for instance tacrine, the dual AChE and BuChE inhibitor, has shown severe hepatotoxicity and many
8
other adverse effects.
“Amyloid hypothesis” indicates that the abnormal production and accumulation of Aβ forms
9
insoluble amyloid fibrils, which can deposit in senile plaques in the critical regions of brain. There
are two major forms of Aβ peptides. Aβ1-40, the dominant one, is soluble and less neurotoxic. Another
one is Aβ1-42 peptide, which constitutes 5-10% in normal conditions. It is considered to be aberrant in
1
0
AD brains and can result in larger oligomers and fibrils. So counteracting brain damage by
inhibiting the Aβ1-42 aggregation or promoting its depolymerization can be a potent treatment for AD.
Increasing evidence suggests that Aβ plaques directly interfere with the electron transfer chain
3

11
and produce free radicals, resulting in oxidative stress. Due to high reactivity, some of these
superoxide free radicals can react with nitrate, leading to the formation of highly reactive
1
2
oxygenation of the nitrous oxide free radicals. On the other hand, the increase of production of
reactive oxygen species (ROS) and free radicals leads to deleterious effects on the cell components,
1
3
and it’s accepted as central processes in AD pathophysiology. What’s more, the process of cellular
oxidation is linked to the oxidation reduction of certain metals. The changes of ion concentration can
2
+
2+
cause oxidative stress and increase the production of ROS. Cu , Zn and other metals can also
influence the processes of protein aggregation. Amyloid precursor protein (APP) and Aβ are able to
1
4
form complexes with these biometals, which can promote the aggregation of Aβ. According to
these, molecules with antioxidant and metal-chelating abilities may be a good therapy for AD.
With the emphasis on the inflammatory mechanism of AD in recent years, it’s proposed that
1
5
neuroinflammation may be another important pathogenesis of AD. Neuroinflammation of AD is
mainly manifested by activation of microglia cells. Under normal physiological conditions, microglia
1
6
cells are resting. However, in neurodegenerative diseases, microglia cells are constantly activated
and release a variety of inflammatory cytokines leading to chronic neuroinflammation. Studies
showed these inflammatory cytokines could also aggravate oxidative stress in the brain and further
stimulate the generation, aggregation and deposition of Aβ. Other studies have shown that activation
of microglia occurs before neurodegenerative symptoms appear, and may be an early event of
1
7,18
pathological changes in AD.
Thus, anti-neuroinflammatory agents are considered potential
candidates for anti-Alzheimer drugs.
Epidemiological studies suggest that nonsteroidal anti-inflammatory drugs (NSAIDs) can
significantly reduce the incidence of AD. Researches showed NSAIDs could even control the release
of inflammatory factors and oxygen free radicals caused by the deposition of Aβ, and reduce the
19,20
amount and area of Aβ deposition.
Salicylic acid (Figure 1a) is a classic NSAID with a long
history of medicication. Besides the anti-inflammatory activity, salicylic acid has been extensively
2
1,22
studied for the antioxidative and metal-chelating properties.
It even has certain anticholinesterase
2
3
activity (IC = 346.0 μM). These researches show that salicylic acid and its derivatives can be used
5
0
against multiple targets in AD disease, suggesting that they can be used as a precursor drug to design
new anti-AD chemicals.
Due to the complex etiology of AD, the single-target directed drugs that have reached clinical
4

trials have failed. Recently, more researches in AD have been paid attention to the development of
‘multi-target-directed ligands’’ (MTDLs), by which new scaffolds aiming at two or more diseases
‘
targets are designed. Based on previous research of our group, the introduction of the
alkylbenzylamine group to genistein significantly increased the inhibitory activity of AChE, and the
2
4
modified genistein derivatives are multifunctional for AD treatment. In addition, inspired by the
2
5
multitarget anti-AD compound Memoquin (Figure 1b) reported by Melchiorre's group, bialkaline
central modification can significantly enhance the activity of compounds. Therefore, we designed a
series of 2,5-dihydroxyterephthalamide (DHTA) derivatives (Figure 1c) by combining
2
,5-dihydroxyterephthalic acid with appropriate secondary amines using carbon spacers of different
lengths, and evaluated for their multifunctional biological activities about effects on Aβ aggregation
and disaggregation, anti-inflammatory activity, AChE inhibition activity, anti-oxidative and metal
chelating properties.
Figure 1. Design strategy followed for the development of multifunctional DHTA derivatives.
2
. Results and discussion
2
.1 Chemistry
The synthetic pathways of target derivatives were summarized in Scheme 1. Compound 2 was
synthesized from diethyl succinate (1) by Claisen-Dieckmann condensation under the condition of
5

26
sodium ethanol/ethanol. Then, in the presence of N-chlorosuccinimide, compound 3 was obtained
2
7
through aromatization reaction. It was hydrolyzed under alkaline conditions to get the key
2
8
intermediate 4. Compound 4 was condensed with the corresponding primary amine 5-7 in the
presence of 1-(3-dimethylaminopropyl)-3-ethyl carbon imine hydrochloride (EDCI) and
2
9
1
-hydroxybenzotriazole (HOBt) in THF to provide the DHTA derivatives 8-10. All of the target
1
compounds were characterized by H NMR and ESI-MS, and most of them were further
1
3
characterized by C NMR. The purity of all target compounds was determined by high-performance
liquid chromatography (HPLC) analysis to be over 96.5 %.
Scheme 1. Synthesis of DHTA derivatives 8-10. Reagents and conditions: (i) NaOEt, EtOH, reflux,
5
8
h; then 1.0 mol/L H SO , r.t., 5 h; (ii) N-Chlorosuccinimide, HOAc, 80 °C, 5 h; (iii) NaOH, H O,
2 4 2
0 °C, argon atmosphere, 5 h; then conc. HCl; (iv) corresponding amine 5~7, EDCI, HOBt, Et N,
3
THF, r.t., overnight.
2
.2 Biological evaluation
2
.2.1 Evaluation of AChE and BuChE inhibitory activities of DHTA derivatives
The inhibitory activities of all the synthesized DHTA derivatives against AChE and BuChE in
2
4,30
vitro were evaluated according to modified Ellman’s method.
AChE inhibitory activity was
measured from both rat cortex homogenate and Electrophorus electricus, and BuChE inbibitory
activity was measured from rat serum. Donepezil and rivastigmine were used as reference
compounds. The ChE inhibitory results are summarized in Table 1.
Most of the tested compounds showed moderate AChE inhibitory activities with IC50 values
6

ranging from submicromolar to micromolar (0.56-30.80 μM), which were better than that of
rivastigmine. Comparing the activities of compound 4 with DHTA derivatives, it could be found that
the introduction of the alkylbenzylamine group could significantly increase the AChE inhibitory
activities as expected. Further structure-activity relationship analysis showed that the length of
methylene side chain and the type of terminal benzyl amine also had important effects on the activity.
The four methylene length side chain compounds (9a-d, IC = 0.56-7.37 μM) had higher activity
5
0
than the corresponding three and six methylene length compounds (8a-d, IC50 = 0.67-11.30 μM;
0a-d, IC50 = 1.68-10.60 μM). Therefore, a four-methylene linker was beneficial for AChE
1
inhibitory activity. Fixing the side chain length, it could be found that compounds with (2-methoxy
benzyl) ethylamine moieties (8d, 9d and 10d) had higher activity, indicating that (2-methoxy benzyl)
ethylamine moiety is the best substitute group. And the introduction of (4-dimethylamine benzyl)
ethylamine and 2,3,4,5-tetrahydrobenzo[f][1,4]oxazepine heterocyclic moieties (9f and 9g)
significantly decreased the activity. Among the tested derivatives, compound 9d exhibited the most
potent inhibitory activity aginst both RatAChE and EeAChE (IC50 = 0.56 ± 0.022 μM and 5.12 ±
0
2
.16 μM, respectively), which was more excellent than rivastigmine (IC50 = 37.10 ± 1.22 μM and
3.20 ± 0.44 μM, respectively) under our experimental condition. But 9d had relatively poor activity
2
5
than Memoquin (IC = 1.55 ± 0.11 nM) although it possessed almost the same substitute group,
5
0
the reason may lie in the difference of the mother nucleus.
In addition, most DHTA derivatives were inactive or weak on BuChE inhibitory activity. This
result showed that the designed target compounds were potent inhibitors with high selectivity toward
AChE. Compound 8c, 9c and 10c had relatively good BuChE inhibitory activity with IC values of
5
0
1
7.30 ± 0.33 μM, 19.20 ± 0.65 μM and 40.50 ± 0.56 μM, respectively, indicating that (2-methoxy
benzyl) methylamine moiety may be favorable for the contacts with BuChE.
Table 1. In vitro inhibition of AChE and BuChE and Oxygen Radical Absorbance Capacity (ORAC,
Trolox Equivalents) by DHTA derivatives and reference compounds.
a
IC (μM) ± SD
5
0
e
Compd.
n
NR R2
ORAC
1
b
c
d
RatAChE
RatBuChE
EeAChE
3
4
—
—
—
—
> 100
> 100
> 100
0.96 ± 0.02
4.70 ± 0.10
> 100
> 100
> 100
7

8
8
a
3
3
11.30 ± 0.20
7.80 ± 0.19
9.21 ± 0.37
0.67 ± 0.013
5.21 ± 0.18
4.05 ± 0.18
7.37 ± 0.23
0.56 ± 0.023
4.17 ± 0.16
27.40 ± 0.83
30.80 ± 1.11
5.60 ± 0.17
4.60 ± 0.12
10.60 ± 0.56
1.68 ± 0.04
37.10 ± 1.22
> 100
> 100
14.10 ± 0.17 1.21 ± 0.05
26.80 ± 0.75 0.95 ± 0.01
b
8
c
3
17.30 ± 0.33 39.70 ± 0.63 1.37 ± 0.06
8d
9a
9b
3
> 100
> 100
11.40 ± 0.27 1.10 ± 0.04
12.60 ± 0.22 1.52 ± 0.06
14.00 ± 0.33 0.93 ± 0.02
4
4
> 100
9
c
4
19.2 ± 0.65
> 100
5.87 ± 0.11
5.12 ± 0.16
0.64 ± 0.01
0.97 ± 0.05
9d
4
9
e
4
> 100
10.80 ± 0.13 1.16 ± 0.03
47.40 ± 0.17 3.23 ± 0.07
64.80 ± 3.01 1.02 ± 0.01
13.10 ± 0.28 1.01 ± 0.02
22.90 ± 0.70 1.27 ± 0.06
9
f
4
> 100
9
g
4
> 100
10a
6
41.9 ± 1.05
47.2 ± 0.73
40.5 ± 0.56
> 100
10b
6
10c
6
6.11 ± 0.11
6.75 ± 0.17
23.20 ± 0.44
0.62 ± 0.02
1.43 ± 0.03
10d
6
g
Rivastigmine
Donepezil
—
—
—
—
9.87 ± 0.23
n.t.
g
0.015 ± 0.002 20.70 ± 0.36 0.021 ± 0.003
n.t.
a
IC50 values represent the concentration of inhibitor required to decrease enzyme activity by 50% and are the
mean of three independent experiments, each performed in triplicate (SD = standard deviation ).
b
From 5% rat cortex homogenate.
c
From rat serum.
d
From Electrophorus electricus.
e
The mean ± SD of the three independent experiments. The data are expressed as μM of Trolox equiv/μM of
tested compound.
8

f
n.a. = no active. Compounds defined “no active” means that percent inhibition is less than 5.0% at a
concentration of 50 μM in the assay conditions.
g
n.t. = not tested.
2
.2.2 Evaluation of AChE inhibition mechanism
To further understand the inhibitory mechanism of DHTA derivatives, compound 9d, the most
31
potent AChE inhibitor among these compounds, was chosen for a kinetic study using EeAChE.
From the illustration of the reciprocal Lineweaver-Burk plots (Figure 2), it could be seen that as the
concentration of 9d increased, both the slopes (decreased Vmax) and intercepts (higher K ) increased.
m
This pattern indicated a mixed-type inhibitory behavior for 9d as the consequence of binding to both
the CAS and PAS of AChE.
Figure 2. Kinetic study on the mechanism of EeAChE inhibition by compound 9d. In the absence
and presence of 9d, overlapping Lineweaver-Burk reciprocal plots of the initial velocity of AChE at
increasing substrate concentration (0.1-0.4 mM) are shown. Lines were derived from a weighted
least-squares analysis of data points.
2
.2.3 Molecular modeling studies
To explore the possible interaction pattern between DHAT derivatives and Torpedo californica
(Tc)AChE (PDB code: 1EVE), a molecular modeling research was conducted using the docking
32
program, AutoDock 4.2 package with Discovery Studio 2.5. Compound 9d was selected for our
molecular modeling studies after comprehensive consideration of the results of AChE inhibition
assay and kinetic study. As the results shown in Figure 3, compound 9d occupied the entire
9

enzymatic CAS, the mid-gorge sites and the PAS, while simultaneously binding to the catalytic site
and peripheral site. In the TcAChE-9d complex, the phenolic hydroxyl groups and the carbonyl
groups of amide in 9d could form intermolecular hydrogen bond with the carbonyl group of amino
acid residues Asp72 and the hydroxyl group of Try121 respectively (9d-OH• • •O=Asp72; 9d=O• •
•OH-Tyr121). The N-(2-methoxybenzyl) ethylamine moiety interacted with Trp84 on the CAS via
parallel π-π interactions. It also had a potential hydrophobic interaction with Gly117, Gly118,
Gly123, Tyr130, and Glu199. And the amide fragment and the benzene ring at the center of the
molecule could also interact with Phe331 and Tyr334 respectively via a certain parallel π-π
interaction. In addition, the molecular folded conformation of 9d induced a hydrophobic interaction
with residues such as Tyr70, Ser81, Trp279, Leu282, Ser286, Ile287, Phe288, Arg289, Phe290,
Phe330 and His440 in the active sites of enzymes. In conclusion, the results of enzyme kinetics
studies and molecular docking studies are consistent, indicating that compound 9d can
simultaneously binding to the CAS and PAS parts of AChE, which explains the moderate AChE
inhibition activity of DHTA derivatives.
Studies showed that the amino acid sequences and crystal structures were different between
3
3
EeAChE and RatAChE. So there may be different binding site between compounds with
acetylcholinesterase from different sources. To further elucidate why the IC50 value of 9d against
EeAChE (IC50 = 5.12 μM) is much higher than that against RatAChE (IC = 0.56 μM), docking
50
studies of 9b with mouse AChE (mAChE, PDB code: 4B85) was also performed (Figure S2,
Supplementary Material). The results showed, in the mAChE-9d complex, the two phenolic hydroxyl
groups in 9d could respectively form intermolecular hydrogen bond with the carbonyl group of
amino acid residues Tyr124 and the hydroxyl group of Phe295 (9d-OH• • •O=Tyr124; 9d-O• •
•OH-Phe295). The carbonyl groups of amide in 9d could form intermolecular hydrogen bond with
the hydroxyl groups of Arg296 and Phe295 (9d=O• • •OH- Arg296; 9d=O• • •OH- Phe295). In
addition, the N-(2-methoxybenzyl) ethylamine moiety interacted with Trp86 via parallel π-π
interactions, and the salicylic acid nucleus could also interact with Tyr341 via parallel π-π
interactions. Compared with TcAChE-9d complex, there were more intermolecular hydrogen bonds
in the mAChE-9d complex, which may give the reason of the higher activity of 9d against RatAChE
than that against EeAChE.
10

Figure 3. Representation of compound 9d (grey stick) interacting with residues in the binding site of
TcAChE (PDB code: 1EVE), highlighting the protein residues involved in the major interactions with
the inhibitor. Picture was generated using Discovery Studio 2.5.
2
.2.4 Antioxidant activity assay of DHTA derivatives
The antioxidant activities of DHTA derivatives were evaluated by the well-established
3
4
ORAC-FL method (oxygen radical absorbance capacity by fluorescein) and the results are shown
in Table 1. The activities of tested compounds to scavenge radicals were expressed as Trolox (a
water-soluble vitamin E analog) equivalent. And their relative activity at concentration of 5 μM was
compared with the highly potent compound Trolox. All of the compounds exhibited significant
antioxidant activities ranging from 0.62 to 3.23-fold of Trolox. And the parent compounds 3 and 4
were also tested, which had ORAC-FL values of 0.96 and 4.70 Trolox equivalents, respectively,
Most of the derivatives had similar antioxidant activities to compound 3 due to the amidation of
25
carboxyl groups. Compared with the lead compound Memoquin (0.77 Trolox equivalents) , most of
the derivatives had better antioxidant activities than it. There was no significant effect on the activity
of other tertiary amine species and the side chain length. Among them, compound 9f showed the
most potent antioxidant activity of this family with ORAC-FL values of 3.23 Trolox equivalents.
Representative compound 9d showed good antioxidant activity with an ORAC-FL value of 0.97
Trolox equivalent.
11

2
.2.5 Metal-chelating studies of representative compound 9d
The chelating abilities of representative compound 9d toward biometals such as Cu , Zn , Fe
2
+
2+
2+
3
+
35,36
and Al were evaluated by UV-visual spectrometry,
and the results are shown in Figure 4. It
could be seen that, after the addition of CuCl , the UV-visual spectrum of 9d changed significantly.
2
The absorption peak at the original 260 nm shifted to 274 nm with the absorption intensity increasing
significantly, and the absorption peak at 360 nm disappeared, indicating that compound 9d had a
2
+
certain chelating effect on Cu . After the addition of AlCl , the absorption intensity at 260 nm was
3
increased, and the absorption peak at the original 360 nm was moved to 351 nm, indicating the
3
+
interaction between compound 9d with Al . In addition, after adding FeSO to the solution of 9d,
4
the UV-Visual spectrum significantly changed and the absorption intensity increased dramatically,
2
+
indicating 9d and Fe had certain interaction. However, there was no significant change in the
spectrum after ZnCl was added. These results showed that 9d had the ability to interact with most
2
biometals.
2
+
To determine the stoichiometry of 9d-Cu complex, molar ratio method was performed by
preparing the methanol solution of 9d with ascending amounts of CuCl . The UV spectra were used
2
to obtain the absorbance of the 9d complexes and different concentrations of CuCl at 274 nm. As
2
indicated in Figure 4B, the absorbance linearly increased initially and then plateaued. The point for
the straight lines to intersect were determined to be at a mole proportion of 1.90, revealing 1:2
2
+
stoichiometry for 9d-Cu complex.
Figure 4. UV spectra of compound 9d (A) (37.5 μM in methanol) alone or in the presence of CuCl2,
12

2+
ZnCl , FeSO or AlCl (37.5 μM, in methanol). Determination of the stoichiometry of complex-Cu
2
4
3
by using the molar ratio method of titrating the methanol solution of compound 9d (B) with
ascending amounts of CuCl . The final concentration of tested compound was 37.5 μM, and the final
2
2
+
concentration of Cu ranged from 3.75 to 150 μM.
2
+
2
.2.6 Inhibition of self- and Cu -induced Aβ aggregation of DHTA derivatives
1-42
2
+
The inhibitory activities of the target compounds on self- and Cu -induced Aβ
aggregation
-42
1
were determined by a thioflavin T (ThT) fluorescence assay, with curcumin as a reference
3
7-40
compound.
The results summarized in Table 2 indicated that all of the DHTA derivatives showed
2
+
excellent inhibitory effects on self- and Cu -induced Aβ
aggregation (71.6~99.9% and
-42
1
5
7.6~89.9% at 25 μM respectively), compared with curcumin (40.2 ± 0.9% and 66.0 ± 1.3%
respectively). In view of the excellent self-induced Aβ1-42 aggregation inhibition activity of DHTA
derivatives, we further determined their IC50 values, and the results were all at micromolar level
(2.31-5.47 μM). However, compounds 3 and 4 almost had no inhibitory activity on Aβ1-42
aggregation inhibition at 25μM, indicating that the introduction of benzyl-alkyl side chain could
indeed enhance their Aβ1-42 aggregation inhibitory activities. By comparison with lead compound
25
Memoquin (IC50 = 5.93 ± 0.33 μM), all of the derivatives exhibited better inhibitory activities.
Different from the conclusion of AChE inhibition assay, the different alkyl side chain length and
species of terminal benzylamines had no significant influence on the aggregation inhibition activity,
and no obvious regularity of structure-activity relationship had been found.
Table 2. In vitro inhibition of Aβ1-42 aggregation and disaggregation of Aβ1-42 aggregation fibrils by
DHTA derivatives and reference compound.
a
%
Inhibition of Aβ1-42 aggregation
IC (μM)
50
Compd.
n
d
b,d
2+
c,d self-induced aggregation
Self-induced
6.5 ± 0.2
4.7 ± 0.1
85.0 ± 1.8
81.0 ± 2.0
93.5 ± 2.6
98.2 ± 3.0
90.0 ± 1.1
Cu -induced
e
3
4
—
—
3
30.7 ± 1.2
8.1 ± 0.2
n.t.
e
n.t.
8
8
a
81.2 ± 3.3
57.6 ± 2.0
70.4 ± 2.8
88.8 ± 4.2
77.6 ± 2.6
3.98 ± 0.10
4.20 ± 0.16
3.63 ± 0.08
2.41 ± 0.06
3.55 ± 0.09
b
3
8c
3
8
d
a
3
9
4
13

9
b
4
4
88.0 ± 2.9
98.6 ± 2.2
96.2 ± 1.9
99.7 ± 4.0
99.9 ± 3.3
71.6 ± 1.7
99.8 ± 2.9
91.4 ± 2.0
99.7 ± 4.1
99.6 ± 2.6
40.2 ± 0.9
84.1 ± 1.7
65.4 ± 2.9
71.7 ± 3.0
69.4 ± 3.3
82.5 ± 4.0
74.0 ± 1.0
89.1 ± 2.6
74.3 ± 1.8
85.3 ± 2.1
89.9 ± 2.4
66.0 ± 1.3
3.86 ± 0.13
3.45 ± 0.06
3.05 ± 0.12
3.71 ± 0.11
3.16 ± 0.07
5.47 ± 0.10
2.37 ± 0.04
2.80 ± 0.03
2.31 ± 0.05
3.31 ± 0.09
9
c
9d
4
9
e
4
9
f
4
9
g
4
1
0a
0b
6
1
6
1
0c
0d
Curcumin
6
1
6
e
—
n.t.
a
b
c
d
e
For inhibition of Aβ aggregation assay, the thioflavin-T fluorescence method was used.
Inhibition of self-induced Aβ1 aggregation (25 μM) by tested inhibitors at 25 μM.
-42
2+
2+
Inhibition of Cu -induced Aβ1 aggregation. The concentration of tested compounds and Cu were 25 μM.
-42
Data are presented as the mean ± SD of three independent experiments.
n.t. = not tested.
2
.2.7 Effect on the disaggregation of Aβ1-42 aggregation fibrils of DHTA derivatives
Subsequently, we still used ThT method to determine the disaggregation ability to the formed
Aβ fibers with DHTA derivatives. The results were summarized in Table 3. The results showed that
DHTA derivatives also manifested good disaggregation ability to self- induced Aβ1-42 aggregation
fibrils. Similarly, the length of alkyl side chain and the type of terminal benzylamine moieties had
little effect on the activity. It showed that the designed DHTA derivatives could not only inhibit
Aβ1-42 aggregation, but also depolymerize the polymeric protein fiber, with the potency to suppress
and eliminate Aβ deposit plaque in AD brain, or to prevent and treat AD.
Table 3. In vitro disaggregation of Aβ1-42 aggregation fibrils by DHTA derivatives and reference
compound.
a
%
Disaggregation of Aβ1-42 aggregation
Compd.
n
b,d
2+
c,d
Self-induced
Cu -induced
8
8
a
3
3
3
3
68.5 ± 1.2
62.7 ± 1.6
70.1 ± 2.2
77.3 ± 2.6
71.3 ± 1.7
70.6 ± 1.9
73.5 ± 2.5
73.1 ± 3.0
b
8
c
8d
14

9
9
a
4
4
67.7 ± 2.3
81.1 ± 3.0
80.3 ± 3.3
75.2 ± 3.1
64.4 ± 1.1
60.4 ± 2.0
67.4 ± 1.5
80.9 ± 2.7
83.5 ± 2.3
76.3 ± 1.4
83.8 ± 1.6
77.9 ± 3.1
80.0 ± 2.9
76.3 ± 2.6
77.2 ± 1.6
83.3 ± 2.1
84.7 ± 1.9
85.3 ± 2.3
89.3 ± 2.4
82.3 ± 3.1
89.8 ± 4.0
85.3 ± 3.2
54.7 ± 1.0
b
9
c
4
9d
4
9
e
4
9
f
4
9
g
4
1
0a
0b
6
1
6
1
0c
0d
Curcumin
6
1
6
e
—
n.t.
a
For disaggregation of Aβ aggregation fibrils, the thioflavin-T fluorescence method was used.
b
For disaggregation of self-induced Aβ1-42 aggregation fibrils, the concentration of tested compounds and Aβ1-42
were 25 μM.
c
2+
2+
For disaggregation of Cu -induced Aβ1 aggregation fibrils, the concentration of tested compounds and Cu
-42
were 25 μM.
d
Data are presented as the mean ± SD of three independent experiments.
e
n.t. = not tested.
2
.2.8 Evaluation of in vitro blood-brain barrier (BBB) permeability
Passing the blood-brain barrier (BBB) and reaching the therapeutic targets is critical for drugs to
act on neurodegenerative processes. The parallel artificial membrane permeation assay of the
blood-brain barrier (PAMPA-BBB) was performed in order to determine the BBB permeability of the
4
1
target compound. First, the experimental permeabilities of 11 commercial drugs with reported
values were compared to verify the assay (Table S1, Supplementary Material). The results of
experimental data consistent with bibliographic values gave a strong linear correlation: P (exp.) =
e
2
0
.9163 × P (bibl.) − 0.2247 (R = 0.9558) (Figure S1, Supplementary Material). From this formula
e
4
2
as well as the limit established by Di et al. for BBB permeation, we determined that compounds
-6
with P values above 3.44 × 10 cm/s could cross the BBB (Table S2, Supplementary Material). As
e
it shown in Table 4, compound 8d, 9d and 10d had moderate BBB permeabilities and would be able
to reach the therapeutic targets in central nervous system (CNS).
-6
Table 4. Permeability results P (× 10 cm/s) from the PAMPA-BBB assay for selected DHTA
e
derivatives with their predicted penetration into the CNS.
15

a
-6
b
Compd.
P (× 10 cm/s)
Prediction
CNS +
e
8
9
d
d
4.23 ± 0.09
4.45 ± 0.11
5.54 ± 0.16
CNS +
10d
CNS +
a
Compounds 8d, 9d and 10d were dissolved in DMSO at 5 mg/mL and diluted with PBS/EtOH (70:30). The final
concentration of each compound was 100 μg/mL.
b
Data are the mean ± SD of three independent experiments.
2
.2.9 Evaluation of in vitro anti-neuroinflammatory activities of DHTA derivatives
Neuroinflammation in AD is mainly characterized by the activation of microglia in the brain
and the release of multiple proinflammatory cytokines. To evaluate the anti-neuroinflammatory
activity of the target compounds, representative compounds 8d, 9c, 9d, 10d and parent compound 4
were tested with the inflammatory microglial BV-2 cells stimulated by lipopolysaccharide (LPS).
Flurbiprofen was used as reference compound. The in vitro neurocytotoxicity, inhibition of LPS
4
3-46
induced NO and TNF-α production were investigated.
2
.2.9.1 Effects of compounds on the survival rate of BV-2 microglial
First, we used MTT method to determine the cytotoxicity of representative compounds without
or with LPS on the BV-2 microglia cells. As the results shown in Figure 5, whether added LPS (1.0
μg/mL) or not, cell viability did almost not change after treatment with various concentrations (0.5,
2
.5 and 10.0 μM) of the compounds. It indicated compound 4, 8d, 9c, 9d, 10d and flurbiprofen have
no cytotoxicity toward the BV-2 cell at concentrations below 10 μM, and LPS (1.0 μg/mL) could not
alter cell viability also.
16

Figure 5. Effects of compound 4, 8d, 9c, 9d, 10d and flurbiprofen without LPS (A) or with LPS (B)
on the cell viability of microglia BV-2 cells. BV-2 cells were pre-incubated with indicated
concentrations of compounds for 30 min followed by 1.0 μg/mL LPS treatment for 24 h. Cell
viability was determined by MTT assay. The data are expressed as the mean ± SD from three
independent experiments.
2
.2.9.2 Effects of compounds on the release of NO in LPS-stimulated BV-2 cells
Inhibition of LPS-induced NO production was measured by using Griess reaction method. First,
we examined the effect of compound itself on NO release from BV-2 cells without LPS. As shown in
Figure 6, NO release volume didn’t change a lot compared with the control group, so compounds 8d,
9
c, 9d and 10d had no effect on it. Then in LPS-treated cells, we observed a massive induction of
NO production, which was significantly reduced by treatment with compounds in a dose-dependent
manner. Compared with reference compound flurbiprofen (11.8% and 23.0% when the cells were
treated with 2.5 and 10.0 μM), representative compounds showed good inhibitory activity on
LPS-induced NO production. 8d, 9c, 9d and 10d displayed better NO inhibitory activity with the
respective percentages of 35.0%, 36.7%, 30.0% and 33.3% when the cells were treated with 2.5 μM.
They also showed excellent inhibitory activity with the respective percentages of 50.8%, 57.5%,
5
8.3% and 63.3% when the cells were treated with 10.0 μM. Moreover, the entire representative
compounds showed far better activities in comparison with the parent compound 4 (8.3% and 12.5%
when the cells were treated with 2.5 and 10.0 μM). It indicated that the introduction of
17

alkylbenzylamine group by the amidation of carboxyl groups was beneficial to the reduction of NO
release in LPS-stimulated BV-2 cells.
Figure 6. Effects of compounds 4, 8d, 9c, 9d, 10d and flurbiprofen on NO release in BV-2 cells and
LPS-stimulated BV-2 cells. The data are expressed as the mean ± SD from three independent
experiments.
2
.2.9.3 Effects of compounds on TNF-α in LPS-stimulated BV-2 cells
Inhibition of LPS-induced TNF-α production was measured by the enzyme-linked
immunosorbent assay (ELISA). The result was shown in Figure 7. First we could also find that
compound 8d, 9c, 9d and 10d had no effect on TNF-α production without added LPS up to the
concentration of 10 μM. Then in the LPS-treated cells, the amount of TNF-α released in the BV-2
cells increased significantly. It was also reduced by treatment with compounds at the concentration of
0
.5-10 μM in a dose-dependent manner. Compared with the reference compound flurbiprofen (9.6%
and 30.3% when the cells were treated with 2.5 and 10.0 μM), all the representative compounds
showed much better inhibitory activities on LPS-induced TNF-a production. The inhibition activity
of the compounds was relatively weak at 0.5 μM, and they showed moderate activities at 2.5 μM
with the respective percentages of 28.2%, 32.9%, 37.1% and 20.6%. In addition, the cells were
treated with 10.0 μM, 8d, 9c, 9d and 10d displayed good inhibitory activity with the respective
percentages of 41.1%, 59.4%, 55.3% and 50.6%. Moreover, all the representative compounds
showed better activities in comparison with the parent compound 4 (5.9% and 18.2% when the cells
18

were treated with 2.5 and 10.0 μM).
Figure 7. Effects of compounds 4, 8d, 9c, 9d, 10d and flurbiprofen on TNF-α release in
LPS-stimulated BV-2 cells. The data are expressed as the mean ± SD from three independent
experiments.
Collectively, representative compounds 8d, 9c, 9d and 10d had no obvious toxicity to BV-2
cells at their effective concentration, and could not promote inflammatory effects to the activate
microglial cells. It may indicate certain medicinal safety of these compounds. And they also had
significant inhibitory activities on NO and TNF-α production in LPS-stimulated BV-2 cells, which
was beneficial to relieve the neuroinflammation in the brain of neurodegenerative disease.
3
. Conclusion
In summary, a series of DHTA derivatives were designed, synthesized and evaluated as
multifunctional agents for the treatment of AD. In vitro assays demonstrated that most of the
synthesized compounds were potent in inhibiting AChE activity and displayed high selectivity for
2
+
AChE over BuChE. They showed significant inhibitory activities of self- and Cu -induced Aβ
1–42
aggregation, moderate antioxidant activities and good anti-inflammatory properties. Among them,
compound 9d exhibited a useful inhibitory activity toward AChE with IC value of 0.56 μM. The
5
0
kinetic analysis suggested that 9d showed mixed-type inhibition, and could bind to both CAS and
PAS of AChE, which was consistent with the molecular modeling study. In addition, 9d displayed
19

2
+
good self- and Cu -induced Aβ
aggregation inhibitory potency (96.2% and 71.7% at 25 μM,
-42
1
respectively), and the IC value of self-induced Aβ aggregation was 3.05 μM. Moreover, 9d also
1-42
5
0
2
+
showed good disaggregation ability to self- and Cu -induced Aβ
aggregation fibrils (75.2% and
-42
1
7
7.2% at 25 μM, respectively), moderate antioxidant activity and good biometal chelating ability.
Furthermore, 9d possessed the prospective properties of acting as anti-neuroinflammatory agent.
More importantly, 9d could cross the BBB and penetrate into brain. The IC values of its various
5
0
activities are basically at the micromole level, and the activities are relatively balanced. These
multifunctional properties highlight compound 9d as a promising candidate for further studies
directed to the development of novel drugs against AD.
4
. Experimental section
4
.1 Chemistry
All materials, unless otherwise noted, were of analytical reagent grade and purchased from
commercial suppliers. Thin-layer chromatography (TLC) was performed on silica gel GF254 plates
from Qingdao Haiyang Chemical Co. Ltd. (China). The spots were detected by UV light (254 nm).
Column chromatography was performed using silica gel (230-400 mesh) purchased from Qingdao
Haiyang Chemical Co. Ltd. (China). Melting points were measured with YRT-3 melting-point
apparatus (China) and uncorrected. HPLC analysis was carried out on a Shimadzu LC-10Avp plus
1
13
system using a Kromasil C column (4.6 mm × 250 mm, 5 μm). The H NMR and C NMR spectra
18
were recorded in CDCl or DMSO-d on a Varian INOVA spectrometer at 25 °C with TMS as the
3
6
internal standard. Coupling constants are given in Hz. Mass spectra were recorded on Agilent-6210
TOF LC-MS Spectrometer.
4
.1.1. Diethyl 2,5-dioxocyclohexane-1,4-dicarboxylate (2)
Sodium crumb (5.82 g, 253 mmol) was added in ethanol (60 mL) at room temperature, and then
the reaction mixture was refluxed until the solid was dissolved. Diethyl succinate (1, 20 mL, 119
mmol) was added in the freshly prepared sodium ethanol/ethanol reagent. The mixture was refluxed
for another 5 h. Ethanol was removed under reduced pressure. The residue was acidified with 2N
sulfuric acid until the pH was about 2 with the precipitation of yellow solid. The reaction mixture
was stirred at room temperature for 5 h. The precipitate was filtered, washed with water and dried at
20

room temperature to afford crude product 2 which was recrystallized with ethyl acetate to obtain
2
6
off-white crystalline solid (9.25 g, 60.4%). mp 128-129 °C (Lit. 128 °C)
4
.1.2. Diethyl 2,5-dihydroxyterephthalate (3)
The suspension of compound 2 (9.25 g, 36.1 mmol) in glacial acetic acid (60 mL) was heated at
6
0 °C until the solid was dissolved. N-chlorosuccinimide (4.90 g, 36.7 mmol) was added in
portions .After that the reaction was heated at 80 °C for 5 h, and then cooled to room temperature
with a large amount of crystals precipitating. After filtration, the filter cake was washed with cold
methanol and dried at room temperature to give compound 3 (8.16 g 89.0%) as yellow columnar
2
7
1
crystals. mp 134-135 °C (Lit. 133-134 °C). H NMR (400 MHz, CDCl ) δ 10.15 (s, 2H), 7.48 (s,
3
2
H), 4.42 (q, J = 7.2 Hz, 4H), 1.43 (t, J = 7.2 Hz, 6H).
.1.3. 2,5-Dihydroxyterephthalic acid (4)
4
The mixture of compound 3 (2.00 g, 7.87 mmol) and sodium hydroxide (1.61 g, 40.3 mmol) in
water (40 mL) was heated at 80 °C under argon protection for 5 h. The reaction solution was cooled
to room temperature, and the pH was adjusted to approximately 1.0 with concentrated hydrochloric
acid. A large amount of bright yellow solids were precipitated, and the mixture was stirred at room
temperature for 1 h. The mixture was filtered, and the filter cake was washed with water until neutral
and dried at room temperature to obtain crude product 4 (1.50 g, 96.2%) as a bright yellow solid
1
which was used without further purification. H NMR (400 MHz, DMSO-d ) δ 10.59 (brs, 2H), 7.29
6
(s, 2H), 3.96 (brs, 2H).
4
.1.4. General procedure for the synthesis of 8-10
To a stirred solution of compound 4 (50 mg, 0.252 mmol), EDCI (169 mg, 0.882 mmol), HOBt
(119 mg, 0.881 mmol) and triethylamine (0.176 mL, 1.26 mmol) in tetrahydrofuran (3 mL),
corresponding primary amines 5–7 (0.882 mmol) was added. The reaction mixture was stirred at
room temperature overnight. After the reaction was completed, the mixture was concentrated. Then
dichloromethane (15 mL) was added to the residue and the mixture was washed with saturated
aqueous sodium bicarbonate solution (10 mL) and brine (10 mL), dried over sodium sulfate, filtered
and concentrated under reduced pressure to give the crude compound. The obtained residue was
purified by silica gel column chromatography to afford 8–10.
1
4
4
.1.4.1. N ,N -bis(3-(benzyl(methyl)amino)propyl)-2,5-dihydroxyterephthalamide (8a)
Compound 8a was synthesized from intermediate 4 and 5a according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:3) as eluent, the
1
pure product 8a was obtained as yellow oil, yield 62.6%. H NMR (400 MHz, CDCl ) δ 11.80 (brs,
3
21

2
4
1
4
H), 8.76 (brs, 2H), 7.33-7.26 (m, 10H), 6.87 (s, 2H), 3.64 (s, 4H), 3.55-3.45 (m, 4H), 2.65-2.60 (m,
1
3
H), 2.39 (s, 6H), 1.86-1.79 (m, 4H); C NMR (100 MHz, CDCl ) δ 168.7, 152.7, 136.9, 129.4,
3
+
28.6, 127.6, 119.2, 114.3, 62.9, 56.1, 41.8, 40.0, 24.6. ESI-MS m/z: 519.4 [M+H] .
1
4
.1.4.2. N ,N -bis(3-(benzyl(ethyl)amino)propyl)-2,5-dihydroxyterephthalamide (8b)
Compound 8b was synthesized from intermediate 4 and 5b according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:3) as eluent, the
1
pure product 8b was obtained as yellow oil, yield 44.5%. H NMR (400 MHz, CDCl ) δ 11.73 (brs,
3
2
8
1
4
H), 8.55 (brs, 2H), 7.37-7.26 (m, 10H), 6.91 (s, 2H), 3.70 (s, 4H), 3.50-3.45 (m, 4H), 2.75-2.65 (m,
13
H), 1.85-1.76 (m, 4H), 1.19 (t, J = 7.2Hz, 6H); C NMR (100 MHz, CDCl ) δ 168.8, 152.6, 137.1,
3
+
29.4, 128.5, 127.6, 119.1, 114.5, 58.0, 51.9, 47.1, 39.7, 24.5, 10.9. ESI-MS m/z: 547.3 [M+H] .
1
4
.1.4.3. 2,5-Dihydroxy-N ,N -bis(3-((2-methoxybenzyl)(methyl)amino)propyl)terephthalamide (8c)
Compound 8c was synthesized from intermediate 4 and 5c according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:5) as eluent, the
1
pure product 8c was obtained as yellow oil, yield 48.4%. H NMR (400 MHz, CDCl ) δ 11.50 (brs,
3
2
3
H), 9.23 (brs, 2H), 7.31-7.27 (m, 4H), 6.94-6.91 (m, 4H), 6.84 (d, J = 8.0 Hz, 2H), 3.76 (s, 4H),
1
3
.68 (s, 6H), 3.51-3.47 (m, 4H), 2.85-2.75 (m, 4H), 2.43 (s, 6H), 1.95-1.84 (m, 4H); C NMR (100
MHz, CDCl ) δ 168.7, 157.9, 152.6, 131.8, 129.4, 124.4, 120.2, 119.1, 114.4, 110.7, 57.1, 56.6, 55.0,
3
+
4
1.7, 40.2, 24.5. ESI-MS m/z: 579.3 [M+H] .
1
4
4
.1.4.4. N ,N -bis(3-(ethyl(2-methoxybenzyl)amino)propyl)-2,5-dihydroxyterephthalamide (8d)
Compound 8d was synthesized from intermediate 4 and 5d according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:5) as eluent, the
1
pure product 8d was obtained as yellow oil, yield 47.3%. H NMR (400 MHz, CDCl ) δ 11.80 (brs,
3
2
3
H), 9.07 (brs, 2H), 7.30-7.26 (m, 4H), 6.94-6.90 (m, 4H), 6.84 (d, J = 8.4 Hz, 2H), 3.75 (s, 4H),
13
.70 (s, 6H), 3.47-3.44 (m, 4H), 2.80-2.70 (m, 8H), 1.90-1.78 (m, 4H), 1.16 (t, J = 6.4 Hz, 6H); C
NMR (100 MHz, CDCl ) δ 168.7, 157.9, 152.4, 131.7, 129.3, 124.6, 120.3, 119.0, 114.6, 110.7, 55.0,
3
+
5
2.5, 52.4, 47.4, 40.1, 24.3, 10.2. ESI-MS m/z: 607.4 [M+H] .
1
4
4
.1.4.5. N ,N -bis(4-(benzyl(methyl)amino)butyl)-2,5-dihydroxyterephthalamide (9a)
Compound 9a was synthesized from intermediate 4 and 6a according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:4) as eluent, the
1
pure product 9a was obtained as yellow oil, yield 45.2%. H NMR (400 MHz, CDCl ) δ 11.01 (brs,
3
22

2
4
4
H), 8.00 (brs, 2H), 7.32-7.26 (m, 10H), 7.01 (s, 2H), 3.61 (s, 4H), 3.43-3.37 (m, 4H), 2.48-2.42 (m,
+
H), 2.27 (s, 6H), 1.70-1.65 (m, 8H). ESI-MS m/z: 547.3 [M+H] .
1
4
.1.4.6. N ,N -bis(4-(benzyl(ethyl)amino)butyl)-2,5-dihydroxyterephthalamide (9b)
Compound 9b was synthesized from intermediate 4 and 6b according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/methanol (8:1) as eluent,
1
the pure product 9b was obtained as yellow oil, yield 65.7%. H NMR (400 MHz, CDCl ) δ 8.67 (brs,
3
2
H), 7.76 (brs, 2H), 7.39-7.25 (m, 10H), 7.18 (s, 2H), 3.78 (s, 4H), 3.40 (t, J = 6.0 Hz, 4H), 2.71 (q, J
1
3
=
7.2 Hz, 4H), 2.65 (t, J = 6.8 Hz, 4H), 1.71-1.63 (m, 8H), 1.14 (t, J = 7.2 Hz, 6H); C NMR (100
MHz, CDCl ) δ 168.6, 152.1, 135.8, 129.6, 128.5, 127.8, 119.5, 115.0, 57.3, 52.1, 47.0, 39.2, 26.8,
3
+
2
3.6, 10.5. ESI-MS m/z: 575.4 [M+H] .
1
4
4
.1.4.7. 2,5-Dihydroxy-N ,N -bis(4-((2-methoxybenzyl)(methyl)amino)butyl)terephthalamide (9c)
Compound 9c was synthesized from intermediate 4 and 6c according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:5) as eluent, the
1
pure product 9c was obtained as yellow oil, yield 39.1%. H NMR (400 MHz, CDCl ) δ 8.59 (brs,
3
2
3
H), 7.27-7.24 (m, 4H), 7.03 (s, 2H), 6.92 (t, J = 7.2 Hz, 2H), 6.85 (d, J = 8.4 Hz, 2H), 3.76 (s, 6H),
1
3
.67 (s, 4H), 3.38-3.34 (m, 4H), 2.52-2.48 (m, 4H), 2.30 (s, 6H), 1.73-1.67 (m, 8H); C NMR (100
MHz, CDCl ) δ 168.8, 157.9, 152.3, 131.7, 129.1, 124.5, 120.2, 119.4, 114.6, 110.4, 56.6, 55.7, 55.1,
3
+
4
2.7, 39.3, 26.7, 24.4. ESI-MS m/z: 607.3 [M+H] .
1
4
4
.1.4.8. N ,N -bis(4-(ethyl(2-methoxybenzyl)amino)butyl)-2,5-dihydroxyterephthalamide (9d)
Compound 9d was synthesized from intermediate 4 and 6d according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:4) as eluent, the
1
pure product 9d was obtained as yellow oil, yield 47.5%. H NMR (400 MHz, CDCl ) δ 7.99 (brs,
3
2
8
6
1
2
4
H), 7.37 (d, J = 7.2 Hz, 2H), 7.26-7.23 (m, 2H), 7.06 (s, 2H), 6.92 (t, J = 7.2 Hz, 2H), 6.84 (d, J =
.4 Hz, 2H), 3.79 (s, 6H), 3.76 (s, 4H), 3.37 (t, J = 6.0 Hz, 4H), 2.71 (q, J = 6.8 Hz, 4H), 2.62 (t, J =
1
3
.0 Hz, 4H), 1.71-1.65 (m, 8H), 1.12 (t, J = 6.8 Hz, 6H, 2 × CH ); C NMR (100 MHz, CDCl ) δ
3
3
68.5, 157.8, 152.1, 131.3, 129.0, 124.3, 120.4, 119.6, 115.1, 110.4, 55.0, 52.3, 51.3, 47.6, 39.2, 26.9,
+
3.9, 10.4. ESI-MS m/z: 635.4 [M+H] .
1
4
.1.4.9.
9e)
N ,N -bis(4-((2-(dimethylamino)benzyl)(ethyl)amino)butyl)-2,5-dihydroxyterephthalamide
(
Compound 9e was synthesized from intermediate 4 and 6e according to the general procedure.
23

After purification by chromatography on silica gel using dichloromethane/acetone (1:6) as eluent, the
1
pure product 9e was obtained as yellow oil, yield 31.9%. H NMR (400 MHz, CDCl ) δ 8.56 (brs,
3
2
H), 8.00 (brs, 2H), 7.58 (d, J = 7.2 Hz, 2H), 7.29 (s, 2H), 7.24 (t, J = 8.0 Hz, 2H), 7.16 (d, J = 8.0
Hz, 2H), 7.05 (t, J = 7.2 Hz, 2H), 3.88 (s, 4H), 3.40-3.37 (m, 4H), 2.73-2.64 (m, 8H), 2.64 (s, 12H),
1
3
1
1
.67-1.62 (m, 8H), 1.11 (t, J = 6.8 Hz, 6H); C NMR (100 MHz, CDCl ) δ 168.4, 153.1, 151.9,
3
30.7, 130.3, 128.5, 123.7, 119.9, 119.5, 115.5, 52.5, 52.3, 47.4, 45.2, 39.1, 26.7, 23.5, 10.4. ESI-MS
+
m/z 661.4 [M+H] .
1
4
4
.1.4.10. N ,N -bis(4-((4-(dimethylamino)benzyl)(ethyl)amino)butyl)-2,5-dihydroxyterephthalamide
(9f)
Compound 9f was synthesized from intermediate 4 and 6f according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:6) as eluent, the
1
pure product 9f was obtained as yellow oil, yield 37.7%. H NMR (400 MHz, CDCl ) δ 8.61 (brs,
3
4
4
H), 7.30 (s, 2H), 7.22 (d, J = 8.4 Hz, 4H), 6.63 (d, J = 8.4 Hz, 4H), 3.80 (s, 4H), 3.40 (t, J = 5.6 Hz,
H), 2.90 (s, 12H), 2.77 (q, J = 6.8 Hz, 4H), 2.70 (t, J = 6.8 Hz, 4H), 1.82-1.72 (m, 4H), 1.68-1.58
13
(
m, 4H), 1.19 (t, J = 6.8 Hz, 6H); C NMR (100 MHz, CDCl ) δ 168.6, 152.0, 150.4, 131.1, 120.0,
3
+
1
19.7, 115.6, 112.2, 56.4, 51.5, 46.6, 40.3, 38.9, 26.6, 22.9, 9.9. ESI-MS m/z: 661.4 [M+H] .
4
.1.4.11.
1
4
N ,N -bis(4-(2,3-dihydrobenzo[f][1,4]oxazepin-4(5H)-yl)butyl)-2,5-dihydroxyterephthalamide (9g)
Compound 9g was synthesized from intermediate 4 and 6g according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:3) as eluent, the
1
pure product 9g was obtained as yellow oil, yield 55.3%. H NMR (400 MHz, CDCl ) δ 11.20 (brs,
3
2
4
H), 7.63 (brs, 2H), 7.19 (t, J = 8.0 Hz, 2H), 7.13 (d, J = 7.2 Hz, 2H), 7.01 (s, 2H), 7.01-6.98 (m, 4H),
.09 (ts, J = 4.0 Hz, 4H), 3.92 (s, 4H), 3.47-3.41 (m, 4H), 3.17 (ts, J = 4.0 Hz, 4H), 2.55 (t, J = 6.4
1
3
Hz, 4H), 1.73-1.64 (m, 8H); C NMR (100 MHz, DMSO-d ) δ 165.9, 157.7, 149.6, 129.7, 128.9,
6
1
26.6, 121.3, 118.5, 117.9, 113.7, 67.7, 55.9, 55.7, 51.1, 37.2, 24.7, 22.4. ESI-MS m/z: 603.3
+
[
M+H] .
1
4
4
.1.4.12. N ,N -bis(6-(benzyl(methyl)amino)hexyl)-2,5-dihydroxyterephthalamide (10a)
Compound 10a was synthesized from intermediate 4 and 7a according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:3) as eluent, the
1
pure product 10a was obtained as yellow oil, yield 47.2%. H NMR (400 MHz, CDCl ) δ 7.33-7.26
3
24

(m, 10H), 7.09 (s, 2H), 6.76 (brs, 2H), 3.57 (s, 4H), 3.42 (q, J = 6.8 Hz, 4H), 2.43 (t, J = 7.6 Hz, 4H),
+
2
.25 (s, 6H), 1.61-1.57 (m, 8H), 1.38-1.32 (m, 8H). ESI-MS m/z: 603.4 [M+H] .
1
4
4
.1.4.13. N ,N -bis(6-(benzyl(ethyl)amino)hexyl)-2,5-dihydroxyterephthalamide (10b)
Compound 10b was synthesized from intermediate 4 and 7b according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:1) as eluent, the
1
pure product 10b was obtained as yellow oil, yield 44.0%. H NMR (400 MHz, CDCl ) δ 8.50 (brs,
3
2
2
H), 7.61 (brs, 2H), 7.43-7.27 (m, 12H), 3.80 (s, 4H), 3.42-3.36 (m, 4H), 2.72 (q, J = 6.8 Hz, 4H),
1
3
.61 (t, J = 7.6 Hz, 4H), 1.64-1.54 (m, 8H), 1.35-1.25 (m, 8H), 1.17 (t, J = 6.8 Hz, 6H); C NMR
(
100 MHz, CDCl ) δ 168.2, 151.8, 134.8, 129.7, 128.5, 128.0, 119.9, 115.6, 57.0, 52.2, 46.8, 39.5,
3
+
2
8.9, 26.5, 26.4, 25.1, 10.2. ESI-MS m/z: 631.4 [M+H] .
1
4
4
.1.4.14. 2,5-Dihydroxy-N ,N -bis(6-((2-methoxybenzyl)(methyl)amino)hexyl)terephthalamide (10c)
Compound 10c was synthesized from intermediate 4 and 7c according to the general procedure.
After purification by chromatography on silica gel using dichloromethane/acetone (1:4) as eluent, the
1
pure product 10c was obtained as yellow oil, yield 49.2%. H NMR (400 MHz, CDCl ) δ 7.58 (brs,
3
2
3
8
1
H), 7.39 (d, J = 7.2 Hz, 2H), 7.32-7.28 (m, 4H), 6.93 (t, J = 7.2 Hz, 2H), 6.87 (d, J = 8.4 Hz, 2H),
.82 (s, 6H), 3.79 (s, 4H), 3.42-3.37 (m, 4H), 2.61 (t, J = 8.0 Hz, 4H), 2.39 (s, 6H), 1.68-1.54 (m,
1
3
H), 1.36-1.30 (m, 8H); C NMR (100 MHz, CDCl ) δ 168.0, 157.9, 151.7, 131.8, 129.7, 122.2,
3
20.5, 120.3, 116.1, 110.5, 56.6, 55.3, 54.5, 41.0, 39.4, 28.9, 26.5, 26.4, 25.6. ESI-MS m/z: 663.4
+
[
M+H]
1
4
4
.1.4.15. N ,N -bis(6-(ethyl(2-methoxybenzyl)amino)hexyl)-2,5-dihydroxyterephthalamide (10d)
Compound 10d was synthesized from intermediate 4 and 7d according to the general procedure.
After purification by chromatography on silica gel using petroleum ether/acetone (1:2) as eluent, the
1
pure product 10d was obtained as yellow oil, yield 46.1%. H NMR (400 MHz, CDCl ) δ 7.44 (d, J =
3
7
.2 Hz, 2H), 7.26-7.23 (m, 2H), 7.16 (s, 2H), 7.07 (brs, 2H), 6.94 (t, J = 7.2 Hz, 2H), 6.86 (d, J = 8.4
Hz, 2H), 3.82 (s, 6H), 3.74 (s, 4H), 3.42-3.39 (m, 4H), 2.66 (q, J = 6.8 Hz, 4H), 2.57 (t, J = 7.6 Hz,
1
3
4
1
2
H), 1.62-1.54 (m, 8H), 1.38-1.30 (m, 8H), 1.12 (t, J = 6.8 Hz, 6H); C NMR (100 MHz, CDCl ) δ
3
68.1, 157.8, 151.9, 131.2, 129.1, 123.9, 120.5, 120.2, 115.7, 110.4, 55.3, 52.7, 50.9, 47.3, 39.5, 29.0,
+
6.7, 26.5, 25.5, 10.5. ESI-MS m/z: 691.5 [M+H] .
4
.2. Biological evaluation
25

4
.2.1. Evaluation of in vitro AChE and BuChE inhibitiory activity
In order to evaluate the in vitro cholinesterase (ChE) activities of the compounds, the Ellman’s
2
4,30
spectrophotometrical method was performed.
We used AChE from 5% rat cortex homogenate,
purified AChE from Electrophorus electricus (Sigma-Aldrich Co.) and BuChE from rat serum.
Before using the brain homogenate, it was preincubated with tetraisopropyl pyrophosphoramide
(iso-OMPA, selective inhibitor of BuChE, 4.0 mmol/L, Sigma-Aldrich Co.) for 5 min. For the
inhibition assays of rat AChE or BuChE, the reaction was initiated by adding 30 µL acetylthiocholine
iodide (1 mmol/L) (J&K Scientific) or butyrylthiocholine iodide (1 mmol/L) (TCI Shanghai
Development), 10 µL 5% homogenate or 25% serum, 20 µL different concentrations of test
compounds and 40 µL phosphate-buffered solution (0.1 mmol/L, pH = 7.40). The mixture was then
incubated at 37 °C for 15 min, and ended with the addition of 30 µL 5,5’-dithiobis-2-nitrobenzoic
acid (DTNB, 0.2%) (J&K Scientific) to produce the yellow anion of 5-thio-2-nitrobenzoic acid. The
absorbance was measured at 405 nm in a Varioskan Flash Multimode Reader (Thermo Scientific).
For EeAChE inhibition assay, EeAChE was used at 0.05 U/mL (final concentration), and the assay
was performed in a phosphate buffer (0.01 mmol/L, pH= 8.00). The absorbance was measured at 412
nm. The other procedure was the same as above. IC values were calculated as the concentration of
5
0
compound that produces 50% inhibition of AChE or BuChE activity. Donepezil was applied as the
positive drug. All samples were assayed in triplicate.
4
.2.2. Evaluation of AChE inhibitor kinetics
Kinetic characterization of AChE inhibition was carried out based on a reported method using
3
1
EeAChE. The assay solution consisted of 30 μL of 0.2% DTNB, 10 μL of 0.5 U/mL EeAChE, 20
μL three different concentrations of inhibitors and 20 μL phosphate buffer (0.1 M, pH 8.00). The
mixture was pre-incubated with the EeAChE for 15 min at 37°C, followed by the addition of 20 μL
TACh in different concentrations. The absorbance was measured at 412 nm. The parallel control
experiments were performed without inhibitor. Slopes of reciprocal plots were then plotted against
the concentration of 9d in a weighted analysis, and K was determined as the intercept on the
i
negative x-axis.
4
.2.3. Molecular modeling study
The crystal structures of Torpedo AChE complexed with donepezil (PDB code: 1EVE) and
mouse AChE (mAChE, PDB code: 4B85) were obtained from the Protein Data Bank after removing
26

3
2
the original inhibitors and water molecules. The 3D structure of 9d was prepared as similarly as
2
8
previously described. Docking studies were performed using the AUTODOCK 4.2 program. The
enzyme structure was used as an input for the AutoGrid program, after the addition of polar hydrogen
atoms to amino acid residues and the distribution of Gasteiger charges to all atoms of the enzyme by
using Autodock Tools (version 1.5.6). Precalculated grid maps were obtained using AutoGrid, one for
each atom type present in the ligand. The grid point spacing for all maps is 0.375Å. The center of the
grid box was located in the center of donepezil with coordinates x = 2.023, y = 63.295, z = 67.062.
The size of box for the active site were set at 60 × 60 × 60 Å. Flexible ligand docking was performed
for the compounds. Each docked system was executed through 100 runs of AutoDock searches
conducted by the Lamarckian genetic algorithm (LGA). In addition to the above parameters, other
parameters were accepted as default. The docking results were clustered using a root mean square
(RMS) tolerance of 1.0, and the lowest energy conformation of the highest filled cluster was selected
for analysis. Autodock Tools or Discovery Studio 2.5 software was used to complete graphic
manipulations and visualizations.
4
.2.4. Evaluation of antioxidant activity
The antioxidant activity was measured by the oxygen radical absorbance capacity-fluorescein
3
4
(
ORAC-FL) assay. Fluorescein (FL) and 6-hydroxy-2,5,7,8-tetramethylchromane-2- carboxylic
acid (Trolox) were purchased from TCI (Shanghai) Development. 2,2’-Azo-bis(amidinopropane)
dihydrochloride (AAPH) was purchased from Accela ChemBio Co., Ltd. All the assays were
performed with 75 mM phosphate buffer (pH = 7.40). 120 μL FL (150 nM final concentration) and
2
0 μL antioxidant were added in the wells of a black 96-well plate using Trolox as a standard (1-8
μM, final concentration). The plate was incubated for 15 min at 37 °C and then placed in a Varioskan
Flash Multimode Reader (Thermo Scientific). 60 μL AAPH solution (12 mM final concentration)
was quickly added using an autosampler and the fluorescence was recorded for 90 minutes per
minute with an excitation wavelength of 485 nm and an emission wavelength of 535 nm. The plate
was automatically shaken before each reading. A buffer blank (FL + AAPH) instead of antioxidant or
Trolox calibration was carried out in each assay. The samples were evaluated at different
concentrations (1-10 μM). All the reaction solution was prepared in duplicate, and each sample was
tested at least three independent assays. The area under the fluorescence decay curve (AUC) was
automatically calculated by the instrument. The blank AUC was subtracted to obtain the net AUC of
27

the sample. ORAC-FL value of each sample was expressed as Trolox equivalents by using the
standard curve where the ORAC-FL value of Trolox was taken as.
4
.2.5. Evaluation of metal-chelating activity
The chelating studies were performed using a Varioskan Flash Multimode Reader (Thermo
Scientific). After the methanol solution of the tested compound alone or in the presence of CuCl2,
ZnCl , AlCl , and FeSO incubated for 30 min at room temperature, the UV absorption spectra was
2
3
4
recorded under wavelength ranging from 200 to 600 nm. The final concentrations of tested
compound and metals were 37.5 μM. The difference UV-vis spectra due to complex formation were
obtained by numerical subtraction of the spectra of the metal alone and the compound alone from the
spectra of the mixture.
Titrating the methanol solution of tested compound with ascending of CuCl to get the
2
2
+
stoichiometry of the compound-Cu complex. The final concentration of tested compound was also
3
7.5 μM, and the final concentration of CuCl ranged from 3.75 to 150 μM. The UV spectra were
2
recorded and treated by numerical subtraction of CuCl and tested compound at corresponding
2
concentrations, plotted versus the mole fraction of tested compound.
2
+
4
.2.6. Evaluation of self- and Cu -induced Aβ aggregation inhibitiory activity
1-42
2
+
The inhibitory activities of compounds on self- and Cu -induced Aβ
aggregation was
-42
1
3
7-40
performed using a Thioflavin T-based fluorometric assay.
Thioflavin T (ThT) was purchased
from TCI (Shanghai) Development. 1,1,1,3,3,3-hexafluoro-2-propanol (HFIP) was purchased from
Energy Chemical. β-Amyloid1-42 (Aβ1-42), supplied as trifluoroacetate salt, was purchased from GL
Biochem Ltd. Aβ1-42 was dissolved in HFIP (1 mg/mL) and incubated for 24 h at room temperature,
then the solvent was evaporated at room temperature. The residue was dissolved in dry DMSO to the
concentration of 200 μM and kept frozen at –80 °C. The DMSO solution of tested compounds in 2.5
mM were stored and diluted with phosphate buffer solution (pH 7.4) before use.
For the self-induced aggregation assay, 20 μL Aβ1-42 (25 μM, final concentration) and 20 μL test
compounds (25 μM, final concentration) diluted with 50 mM phosphate buffer solution (pH 7.40)
were added and incubated at 37 °C for 24 h. A transparent heat-resistant plastic film was used to seal
the plate to minimize evaporation effect. After the incubation, 160 μL of 5 μM ThT in 50 mM
glycine-NaOH buffer (pH 8.5) was added. Each assay was carried out in triplicate. Fluorescence was
measured on a Varioskan Flash Multimode Reader (Thermo Scientific) with excitation and emission
28

wavelengths of 446 nm and 490 nm, respectively. Calculate the percent inhibition by the following
expression: (1-IF /IF ) × 100. And IF and IF are the fluorescence intensities obtained for Aβ after
i
c
i
c
1-42
subtracting the background in the presence and absence of inhibitors, respectively.
2
+
2+
For the Cu -induced Aβ
aggregation assay, standard solutions of Cu was diluted with the
-42
1
HEPES buffer (20 mM, pH 6.60, 150 mM NaCl) to the concentration of 75 μM. The Aβ1-42 stock
solution was diluted with HEPES buffer. The mixture of 20 μL Aβ1-42 (25 μM, final concentration)
2
+
and 20 μL Cu (25 μM, final concentration), with or without 20 μL tested compound (25 μM, final
concentration) was incubated at 37 °C for 24h. Then, 190 μL of 5 μM ThT in 50 mM glycine-NaOH
buffer (pH 8.50) was added. Each assay was run in triplicate. The detection method was the same as
above.
4
.2.7. Evaluation of effect on the disaggregation of Aβ1-42 aggregation
For the disaggregation of self-induced Aβ fibrils experiment, the Aβ1-42 stock solution was
diluted with HEPES buffer (20 mM, pH 6.6, 150 mM NaCl). 20 μL Aβ1-42 (50 mM) was incubated at
7 °C for 24 h. Then 20 μL tested compound (25 mM, final concentration) was added and the
3
mixture was incubated at 37 °C for another 24 h. To reduce the evaporation effect, the plates were
sealed by a transparent heat-resistant plastic film. After incubation, 160 mL of 5 mM ThT in 50 mM
glycine-NaOH buffer (pH 8.5) was added. Each assay was run in triplicate. The detection method
was the same as above.
For the disaggregation of copper-induced Aβ fibrils experiment, the Aβ1-42 stock solution was
diluted with HEPES buffer. The mixture of 20 μL Aβ1-42 (25 mM, final concentration) and 20 μL
2
+
Cu (25 mM, final concentration) was incubated 37 °C for 24 h. 20 μL tested compound (25 mM,
final concentration) was then added and the mixture was incubated at 37 °C for another 24 h. The
plates were sealed by a transparent heat-resistant plastic film to reduce the evaporation effect. After
incubation, 190 mL of 5 mM ThT in 50 mM glycine-NaOH buffer (pH 8.5) was added. Each assay
was run in triplicate. The detection method was the same as above.
4
.2.8. In vitro blood-brain barrier permeation assay
The in vitro blood-brain barrier permeation assay was conducted using the artificial membrane
4
2
permeation assay of the blood-brain barrier (PAMPA-BBB) described by Di et al. The acceptor
plate (PVDF membrane, pore size 0.45 μm, MAIPN4550) and the donor plate (MATRNPS50) were
purchased from Millipore. And porcine brain lipid (PBL) was purchased from Avanti Polar Lipids.
29