Dual GSK-3β/AChE Inhibitors as a New Strategy for Multitargeting Anti-Alzheimer's Disease Drug Discovery

Article abstract of DOI:10.1021/acsmedchemlett.7b00463

Designing multitarget-directed ligands (MTDLs) is considered to be a promising approach to address complex and multifactorial maladies such as Alzheimer's disease (AD). The concurrent inhibition of the two crucial AD targets, glycogen synthase kinase-3β (GSK-3β) and human acetylcholinesterase (hAChE), might represent a breakthrough in the quest for clinical efficacy. Thus, a novel family of GSK-3β/AChE dual-target inhibitors was designed and synthesized. Among these hybrids, 2f showed the most promising profile as a nanomolar inhibitor on both hAChE (IC₅₀ = 6.5 nM) and hGSK-3β kinase activity (IC₅₀ = 66 nM). It also showed good inhibitory effect on β-amyloid self-aggregation (inhibitory rate = 46%) at 20 μM. Western blot analysis revealed that compound 2f inhibited hyperphosphorylation of tau protein in mouse neuroblastoma N2a-Tau cells. In vivo studies confirmed that 2f significantly ameliorated the cognitive disorders in scopolamine-treated ICR mice and less hepatotoxicity than tacrine. This study provides new leads for assessment of GSK-3β and AChE pathway dual inhibition as a promising strategy for AD treatment.

Full text of DOI:10.1021/acsmedchemlett.7b00463

Subscriber access provided by University of Florida | Smathers Libraries
Letter
Dual GSK-3#/AChE Inhibitors as a New Strategy for
Multitargeting Anti-Alzheimer’s Disease Drug Discovery
Xueyang Jiang, Tingkai Chen, Junting Zhou, Siyu He, Hongyu
Yang, Wei Qu, Yao Chen, Feng Feng, and Haopeng Sun
ACS Med. Chem. Lett., Just Accepted Manuscript • DOI: 10.1021/acsmedchemlett.7b00463 • Publication Date (Web): 09 Feb 2018
Downloaded from http://pubs.acs.org on February 11, 2018
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Medicinal Chemistry Letters is published by the American Chemical Society. 1155
Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 7
ACS Medicinal Chemistry Letters
1
2
3
4
5
6
7
8
9
Dual GSKꢀ3β/AChE Inhibitors as a New Strategy for Multitargeting
AntiꢀAlzheimer’s Disease Drug Discovery
XueꢀYang Jianga, TingꢀKai Chena, JunꢀTing Zhoua, SiꢀYu Heb, HongꢀYu Yangb, Yao Chenc, Wei Qua,
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Feng Fenga and HaoꢀPeng Sunb
a Department of Natural Medicinal Chemistry, China Pharmaceutical University, Nanjing 210009, China;
b Department of Medicinal Chemistry, China Pharmaceutical University, Nanjing, 210009, China;
c School of Pharmacy, Nanjing University of Chinese Medicine, Nanjing, 210023, China;
Supporting Information
KEYWORDS: Dual GSKꢀ3β/AChE inhibitor, hybrid, multitarget direct ligands (MTDLs)
ABSTRACT: Designing multiꢀtargetꢀdirected
ligands (MTDLs) is considered to be a promisꢀ
ing approach to address complex and multifactoꢀ
rial maladies such as Alzheimer’s disease (AD).
The concurrent inhibition of the two crucial AD
targets, glycogen synthase kinaseꢀ3β (GSKꢀ3β)
and human acetylcholinesterase (hAChE), might
represent a breakthrough in the quest for clinical
efficacy. Thus, a novel family of GSKꢀ3β/AChE
dualꢀtarget inhibitors was designed and syntheꢀ
sized. Among these hybrids, 2f showed the most
promising profile as a nanomolar inhibitor on both hAChE (IC50 = 6.5 nM) and hGSKꢀ3β kinase activity (IC50 = 66 nM). It also
showed good inhibitory effect on βꢀamyloid selfꢀaggregation (Inhibitory rate = 46%) at 20 ꢁM. Western blot analysis revealed that
compound 2f inhibited hyperphosphorylation of tau protein in mouse neuroblastoma N2aꢀTau cells. In vivo studies confirmed that
2f significantly ameliorated the cognitive disorders in scopolamineꢀtreated ICR mice and less hepatotoxicity than tacrine. This
study provides new leads for assessment of GSKꢀ3β and AChE pathway dual inhibition as a promising strategy for AD treatment.
Among all the neurodegenerative diseases, Alzheimer's disꢀ
ease (AD), an irreversible progressive neurodegenerative
disorder of the central nervous system (CNS), is described
by a progressive loss of cognitive abilities in the old age peoꢀ
ple [1]. Due to the very limited number of available drugs and
their low efficacy, AD remains incurable. Aberrant protein
processing is a distinctive feature of AD, in which are βꢀ
amyloid (Aβ) and abnormally hyperꢀphosphorylated tau proꢀ
tein, upon misfolding and selfꢀaccumulate in the brains of
affected individuals as aggregates, namely amyloid senile
plaques and neurofibrillary tangles (NFTs), respectively [2,3].
These assemblies represent the most relevant histopathological
hallmarks of AD and have been considered to play crucial
roles in its pathogenesis. A series of hypotheses have been
proposed, among which tau hypothesis, amyloid cascade hyꢀ
pothesis and cholinergic hypothesis are of the central imꢀ
portance [4].
[6]. Aberrant hyperꢀphosphorylation of the tau protein loses its
ability to stabilize microtubules, which leads to accumulation
and formation of NFTs [7]. As for the tau hypothesis, glycoꢀ
gen synthase kinaseꢀ3β (GSKꢀ3β), a multiꢀtasking serꢀ
ine/threonine kinase largely expressed in CNS, has been
proved to play a significant role in regulating tau phosphorylaꢀ
tion mainly at Ser396, Ser199, and Ser413, which causes tau
to be detached from the microtubules and precipitates as NFTs
under both physiological and pathological conditions [8]. Furꢀ
thermore, increased GSKꢀ3β activity also induces Aβ forꢀ
mation in a unique fashion by regulating γꢀsecretase, and reꢀ
sults in toxicity to the cultured neurons [9]. A conditional
GSKꢀ3β overexpressing transgenic mouse exhibited microtuꢀ
bule destabilization, NFTs formation and cognitive deficits
[10]. Both preclinical and early phase clinical studies have
validated GSKꢀ3β as a therapeutic target for AD. Recent findꢀ
ings have shown that GSKꢀ3β inhibitors (GSKꢀ3βIs) can be a
route to shift the equilibrium from neurodegeneration to neuꢀ
rogenesis both in vitro and in vivo [11]. Clearly, hindering
generation of Aβ and hyperꢀphosphorylated tau protein by
inhibiting GSKꢀ3β activity has been envisaged as a suitable
strategy for AD treatment, and thus the development of selecꢀ
Tau is a microtubuleꢀassociated protein mostly found in neuꢀ
ronal axons in CNS [5]. Under normal physiological condiꢀ
tions, tau is associated with microtubules by preventing their
dynamic instability, contributing to morphogenesis of neurons
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 2 of 7
tive GSKꢀ3βIs has been popularly pursued (Figure S1)
provides a sound basis for the design of MTDLs [23, 24]. In
addition, the more potent AChEI 6ꢀClꢀtacrine and the less
hepatotoxic 7ꢀMeOꢀtacrine have provided useful scaffolds to
generate new hybrids [25].
1
2
3
[12,13]. Up till now, a series of new promising high affinity
ATPꢀcompetitive GSKꢀ3βI with basis of pyridothiazole core
have been synthesized [14].
4
5
6
7
8
9
On the other hand, the decline of acetylcholine (ACh) levels
leads to cognitive and memory deficits, thus recovering choꢀ
linergic function is considered to be clinically beneficial [15].
Several studies have also indicated that AChE seems to be
implicated in AD pathogenesis by promoting the formation of
both Aβ fibrils [16]. Fortunately, acetylcholinesterase inhibiꢀ
tors (AChEIs) possibly affect metabolic processing of the
APP, and thus may influence Aβ generation [17]. It is worth
emphasizing that although many hypotheses and strategies are
currently being proposed for the treatment of AD, AChEIs still
remain to be a supreme clinical success in AD treatment,
which fully demonstrates the value of this target [18]. Howevꢀ
er, the cholinergic deficit is just one of the hallmarks of the
AD pathology. One of the reasons that the singleꢀtarget diꢀ
rected drugs have failed to reach clinical trials is the pathologꢀ
ical complexity found in AD. An openꢀlabel study found that
the combination of cholinesterase inhibitors and memantine
was well tolerated in dementia therapy [19]. As a new paraꢀ
digm in drug discovery for AD, compared to combination
therapies, multifunctional molecules as multitarget direct ligꢀ
ands (MTDL) avoid drugꢀdrug interactions, offꢀtarget adverse
effects, poor patient compliance, and high development costs
[20]. Given the fact that AD is a systemic disorder of the cenꢀ
tral nervous system, multiꢀtarget strategies will be more promꢀ
ising. As an AChEI with definite efficacy and binding mode,
tacrine is a good scaffold for the design of MTDLs due to its
simple structure but high ligand efficiency [21, 22]. Moreover,
tacrine has a good endurance against substantial structural
modification while retaining the targetꢀbased activity, further
Given the aforeꢀmentioned evidence, GSKꢀ3β and AChE,
the two main pathways of AD, are ideal candidates for such
multitarget approach. Their activities are deeply involved in
AD pathogenesis and progression: the cholinergic deficit and
NFTs, the two main AD pathological hallmarks. Therefore,
the simultaneous modulation of both GSKꢀ3β and AChE, by
blocking hyperꢀphosphorylated tau protein and aggregation of
Aβ plaques, as well as improving cognition, might serve as a
promising strategy for AD treatment. Herein, we designed the
first GSKꢀ3β/AChE bifunctional inhibitors by hybridizing the
pharmacophores of GSKꢀ3βI with AChEI. The rationale for
the MTDL design originated from the coꢀcrystal structure of
the selective GSKꢀ3βI 1 with the ATPꢀbinding site of GSKꢀ3β
kinase domain. The pyridine carboxamide of 1 acted as the
hinge binding head forming hydrogen bonds with the V135
backbone amide, thiazole ring as hinge group, and the carbonꢀ
yl oxygen of the thiazolyl primary amide formed a critical
hydrogen bond with the Lys85 (Figure 1B) [14]. It should be
noted that the thiazolyl methoxy moiety and the primary amꢀ
ide are located in the solvent exposed sites, to be the best posiꢀ
tion to attach tacrine to compound 1 by alkylenediamine tethꢀ
ers so that critical binding interactions were not disrupted. It is
important that the designed hybrids have no significant effect
on the binding of tacrine to the catalytic active site (CAS) of
AChE and the pyridoxathiazole fragment acts as a binder for
the peripheral anionic site (PAS) and thus may enhance AChE
inhibition. Therefore, we incorporated tacrine at the thiazolyl
ring of 1 with a proper linker to design a new MTDL (Figure
1A).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Design strategy of dual GSKꢀ3β/AChE inhibitors. (A) Design of a merged GSKꢀ3β/AChE pharmacophore taking adꢀ
vantage of the solvent exposed sites in GSKꢀ3β and the probable interactions to be gained by a merged tacrine at the entrance to
AChE substrate binding pocket. (B) Xꢀray coꢀcrystal structure of 1 in the kinase domain of GSKꢀ3β. Intermolecular interactions are
shown as dot lines with different colors according to the type of the interaction: green, conventional hydrophobic contact; dark
pink, πꢀπ stacked; light pink, πꢀalkyl contact. The thiazolyl methoxy and the primary amide moiety extends into the solventꢀ
exposed region of the protein (red circle).
The route employed to synthesize target compounds 2a−2k
is outlined in Scheme 1. Pyridothiazole ester 3 was obtained
from 2ꢀaminoꢀ4ꢀcyanopyridine by acylation, addition and cyꢀ
clization [26], which was then used in Mitsunobu reaction
with Nꢀethanol tacrine [27], giving compound 2a. Hydrolysis
of the ethyl ester in 2a to afford 2b, followed by coupling with
ACS Paragon Plus Environment

Page 3 of 7
ACS Medicinal Chemistry Letters
NH4Cl afforded compound 2c or methylation with MeI gave
zone of 1. The inhibitory potency was significantly reduced
when the amino group (1) was replaced by ethoxy group (2a),
carboxyl group (2b) and methoxy group (2d). This indicated
that the amino group was crucial for the enzymatic inhibition
of 1, probably because the NH of the amide bond forms an
intramolecular hydrogen bond with the oxygen atom of the
thiazole ring to stabilize the orientation of the amide bond.
However, 2c did not show potency against hAChE activity.
Given the importance of the amino group on GSKꢀ3β inhibitoꢀ
ry activity, tacrine was introduced in the position of the amide
group by alkyl chain, resulting in compounds 2eꢀ2k. As shown
in Table 1, compound 2e showed potent inhibitory activity
against GSKꢀ3β, with IC50 value of 68 nM, while 6ꢀ
chlorotacrine and 7ꢀmethoxytacrine hybrids (2hꢀ2k) showed
similar GSKꢀ3β inhibitory effects to 2e, with activities ranging
from 63 to 98 nM. These results suggested that GSKꢀ3β inhiꢀ
bition could rarely be affected by changing the length of the
side chain, or by addition of the tacrine moiety. However, 2hꢀ
2k exhibited potent antiproliferative activities against human
neuroblastoma SHꢀSY5Y cell lines (IC50 values ranging from
3.1 to 8.9 ꢁM), the improved cytotoxicity might limit the furꢀ
ther cellꢀbased evaluation. It is noteworthy that 2f and 2g, with
an nꢀpropane or an nꢀhexane linker, displayed potent GSKꢀ3β
inhibition with IC50 of 66 and 18 nM, respectively. In addition,
2g exhibited an IC50 of 30 ꢁM against SHꢀSY5Y cells while 2f
exhibited a slightly lower IC50 value of 18 ꢁM. According to
their singleꢀ or doubleꢀdigit nanomolar activities on ChEs and
GSKꢀ3β, these two compounds were thought to own the best
safety among all the target molecules to exert their biological
effects as dual GSKꢀ3β/AChE inhibitors. Here we chose 2f as
1
2
3
4
5
6
7
8
out compound 2d. Amination of 4 with the corresponding
diamines afforded intermediates 5a–5g. Mitsunobu reaction of
3 with MeOH, followed by hydrolysis to yield carboxyl 4 [14].
Finally, the carboxylic acid 4 was coupled with tacrine interꢀ
mediates 5a–5g to produce the target compounds 2e–2k, reꢀ
spectively.
Scheme 1. Preparation of Compounds 2a−2ka
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
aReagents and conditions: (a) Nꢀethanol tacrine, PPh3, DIAD,
THF, rt, overnight, 52%; (b) 1.5N LiOH, MeOH/THF/H2O, rt,
5h, 87%; (c) NH4Cl, HATU, DIPEA, DMF, rt, overnight,
38%; (d) CH3I, K2CO3, DMF, rt, 5h, 43%; (e) i: anhydrous
MeOH, PPh3, DIAD, THF, rt, 18h; ii: 1N LiOH, MeOH/THF,
rt, 3h, 80%; (f) HATU, DIPEA, DMF, rt, 10h, 31ꢀ48%.
The GSKꢀ3β enzymatic inhibition of all the target comꢀ
pounds was analyzed using human recombinant GSKꢀ3βusing
a luminescence method [28]. Compounds 2aꢀ2k displayed
potent GSKꢀ3β kinase inhibitory activities. The IC50 values
ranged from 18 to 270 nM, similar to compound 1 (Table 1).
These results suggested that the GSKꢀ3β inhibitory effect was
durable by introducing tacrine moiety at the solvent exposure
the
representative
for
further
investigations.
Table 1. Inhibition of hGSKꢀ3β, hAChE and hBuChE, Aβ1ꢀ42 SelfꢀAggregation, and AntiꢀProliferative Activitiesa
hGSKꢀ3β
hAChE
hBuChE
Aβ1ꢀ42
SHꢀSY5Y
Compd
R
R’
n
SIc
(IC50 nM)a
(IC50 nM)a
(IC50 nM)a
IR (%)d
(IC50 ꢁM)e
OEt
OH
NH2
OMe
180 ± 20
270 ± 30
34 ± 3.1
99 ± 10
68 ± 5.3
66 ± 6.2
18 ± 1.4
63 ± 4.9
98 ± 9.3
83 ± 6.5
65 ± 4.7
310 ± 20
580 ± 50
300 ± 10
130 ± 7.1
6.3 ± 0.2
6.4 ± 0.3
22 ± 2.1
2.1 ± 0.9
3.6 ± 0.3
23 ± 1.7
38 ± 4.2
230 ± 31
610 ± 50
450 ± 50
150 ± 10
28 ± 2.3
2
30 ± 2.9
25 ± 2.1
15 ± 1.8
47 ± 5.1
40 ± 4.3
46 ± 4.1
39 ± 4.1
40 ± 3.7
42 ± 4.1
47 ± 4.5
45 ± 4.2
< 5
7.1 ± 0.6
21 ± 1.8
19 ± 2.5
13 ± 0.9
6.3 ± 0.5
18 ± 1.2
30 ± 2.8
2.3 ± 0.3
3.1 ± 0.2
8.9 ± 0.9
2.9 ± 0.2
120 ± 15
2a
2b
2c
2d
2e
0.8
0.5
0.2
9
43
5
190
80
91
60
0.17
H
H
H
6ꢀCl
6ꢀCl
1
2
5
1
2
1
2
51 ± 6.1
260 ± 32
100 ± 7.9
410 ± 39b
290 ± 23b
2100 ± 180b
2300 ± 170b
40 ± 3.7
2f
2g
2h
2i
2j
2k
Tac.
7ꢀOMe
7ꢀOMe
a
Recombinant human AChE, BuChE and GSKꢀ3β were used. Data are the mean of at least three independent determinaꢀ
b
c
d
tions. BuChE (EC 3.1.1.8) from horse serum. Selectivity index (SI) = IC50 hBuChE/IC50 hAChE. Inhibition of Aβ1ꢀ42 selfꢀ
aggregation investigated by the thioflavinꢀT fluorescence assay. Assays were carried out in the presence of 20 ꢁM inhibitor
and 50 ꢁM Aβ1ꢀ42. SHꢀSY5Y cells exposing for 48 h to compound.
e
The inhibitory potency of compounds 2aꢀ2k on hAChE and
BuChE from human serum (hBuChE) was tested by Ellman’s
method (Table 1) [29]. All tested derivatives showed an IC50
against hAChE in nanomolar range, which were consistent
with the data on the inhibition of cholinesterase from the elecꢀ
tric eel (see Table S1). We retained the amino group of the
GSKꢀ3βI 1 or introduced other groups, including carboxylic
acid, methyl ester and ethyl ester, at the formyl side chain.
Hybrids exhibited slight hAChE inhibitory activities from 130
to 580 nM. Concerning the effect of the spacer chain length on
the antiꢀAChE activity, we introduced various lengths of alkyl
chains as linker to the side chain based on the GSKꢀ3βI 1. As
shown in Table 1, the compounds with various linkers exhibitꢀ
ed potent the antiꢀAChE activities (from 2.1 to 280 nM),
which verified the flexibility of the side chain in AChE inhibiꢀ
tory activity. It should be noted that 2hꢀ2k, bearing 6ꢀClꢀ
tacrine or 7ꢀMeOꢀtacrine fragments, were superior AChEIs to
tacrine. Besides, these four compounds exhibited much higher
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 4 of 7
inhibitory potency on AChE than on BuChE, indicating that
cation; purple, πꢀalkyl contact. (B) LineweaverꢀBurk plots
resulting from subvelocity curves of AChE activity with difꢀ
ferent substrate concentrations (25ꢀ450 µM) in the absence
and presence of 1, 5, 10, 20 nM of 2f.
1
2
3
4
5
6
7
8
such modifications could help to design highly selective
hAChE inhibitors. The most potent hAChE inhibitor was 2h
(IC50 = 2.1 nM), which also had the best selectivity towards
AChE with 190ꢀfold better than BuChE. Unfortunately, these
hybrids possessed a strong antiproliferative activity against
SHꢀSY5Y cell lines, giving IC50 values spanning from 2.3 to
8.9 ꢁM. The optimal tethers were twoꢀthree carbon chain (2e,
2f) for the thiazolylpyridine hybrids bearing tacrine unit,
showing IC50 values of 6.4 and 6.5 nM, respectively.
manner by occupying both CAS and PAS. The location and
orientation of 2f in AChE activeꢀsite gorge was consistent
with the critical role of tacrine fragments, which binds at the
anionic site by stacking against the aromatic rings of Trp86
and Tyr124. The endocyclic nitrogen of 2f was hydrogen
bonded to the mainꢀchain carbonyl oxygen of the catalytic
residue His447, indicating that the catalytic triad was disruptꢀ
ed. It was noticeable that pyridothiazole moiety was located at
the PAS of hAChE binding groove, which formed πꢀπ stacking
contacts with the sidechain of Trp286. The LineweavereBurk
plots for AchE showed the point of intersection was in the
third quadrant, but very close to the X axis. It is a mixed but
noncompetitive combination, suggesting that 2f should comꢀ
bine the PAS site as a nonꢀcompetitive inhibitor, similarly to
donepezil.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Considering that roles of AChE and BuChE in the modulaꢀ
tion of the central cholinergic tone vary with the progression
of the disease, it is conceivable that selective AChEIs are more
effective in the early stages of AD. The best selectivity for
hAChE was demonstrated by 2h, which was 190ꢀfold more
active toward hAChE versus BuChE. Targeting BuChE has
become a viable alternative for the treatment of patients with
mildꢀtoꢀlate stages of AD. BuChE is primarily expressed and
secreted by glial cells and it is thought to play a compensatory
role in response to the decrease of AChE activity as AD proꢀ
gresses. Selective inhibition of BuChE thus represents a useful
strategy to ameliorate the cholinergic deficit and improve the
cognitive performance of patients in the late stages of AD.
Hybrid 2d was the most potent and selective hBuChEI (IC50 =
28 nM) of this series, with a 4.6ꢀfold better inhibition. The
accumulation of oligomeric aggregates of Aβ is considered to
play an important role in the pathogenesis of AD. Thus, the
inhibition of Aβ oligomerization has been investigated as an
attractive therapeutic strategy for AD treatment. All comꢀ
pounds were evaluated for their inhibitory capacity on selfꢀ
induced Aβ1ꢀ42 aggregation based on a thioflavin Tꢀbased fluoꢀ
rometric assay. Most of the hybrids showed moderate inhibiꢀ
tion on Aβ1ꢀ42 selfꢀaggregation (ranging from 15ꢀ47%, Table
1), while tacrine was not able to inhibit Aβ1ꢀ42 selfꢀ
aggregation. Considering its better inhibition activity of
AChE, Aβ1ꢀ42 selfꢀaggregation, and smaller molecular weight
compared to 2g, we assumed that the most promising
achievement of the present study is 2f, which showed a dual
lowꢀnanomolar inhibition profile in vitro GSKꢀ3β/AChE (IC50
values of 66 nM and 6.5 nM, respectively).
It is widely believed that elevated levels of hyperphosꢀ
phorylated tau (pꢀtau) protein are causal in the process of
forming the neurofibrillatory tangles, which are one of the
pathologic features in the brains of AD patients [30]. Encourꢀ
aged by the potent inhibitory activity of 2f against GSKꢀ3β,
Western blot assays were conducted to further confirm the
effect of the dualꢀaction targeting of GSKꢀ3β and AChE on pꢀ
tau levels. We treated mouse neuroblastoma N2aꢀTau cells
with increasing concentrations of 2f for 6 hours and the lysates
were subjected to westernꢀblot analyses. The results in Figure
3B showed that the phosphorylation of tau protein at Ser396
site decreased after treatment with 10 µM of 2f. Meanwhile,
the total tau protein level was not changed resulting in an atꢀ
tenuation of pꢀtau/total tau. These data suggested that 2f could
significantly inhibit tau protein hyperphosphorylation, thus
offering promises for reducing NFTs in the development of
thiazolylpyridineꢀbased multitarget drug candidates for AD
treatment.
To investigate the binding pattern of 2f with hAChE, moꢀ
lecular docking studies were performed using Discovery Stuꢀ
dio (DS). As shown in Figure 2, 2f bound to AChE in a dualꢀ
s
i
t
e
Figure 3. (A) Binding mode prediction of 2f with hGSKꢀ3β
(PDB id: 4PTC). 2f was shown in stick mode colored in green.
(B) Compound 2f effectively reduce tau protein hyperphosꢀ
phorylation in mouse neuroblastoma N2aꢀTau cells. Protein
expression was detected by immunoblot analysis with a specifꢀ
ic antibody. The effect of 2f on phosphorylation status of tau.
Cognitionꢀimproving potency is the most important profile
of antiꢀAD agents. Given the interesting multipotent activity
profile of 2d and 2f, their ability to ameliorate scopolamineꢀ
induced cognition impairment in ICR mice was investigated in
a behavioral study by a Morris water maze test. Scopolamineꢀ
induced cognitionꢀimpaired adult mice were used as an animal
model to measure the cognitive improvement effects of 2d and
2f, compared with tacrine (20 ꢁmol/kg body weight). 5 days of
memory training and learning and a probe trial on the sixth
Figure 2. Binding mode prediction of 2f with hAChE (PDB
id: 4EY7). (A) 2f was shown in stick mode colored in green.
Key residues were labeled as thin stick mode colored in orꢀ
ange. Intermolecular interactions were shown with different
colors dot lines: light green, hydrophobic contact; yellow, πꢀ
ACS Paragon Plus Environment

Page 5 of 7
ACS Medicinal Chemistry Letters
day were performed in the test. The mean escape latency valꢀ
amineꢀinduced cognitive impairment in ICR mice determined
by the Morris water maze test. (A) The latency to target. (B)
The distance to target. Data are presented as the mean ± SEM
(n = 6; ***p < 0.001, ****p < 0.0001 vs scopolamine group).
1
2
3
4
5
6
7
8
ues and distance to target of all the groups on the sixth day
were shown in Figure 4 (the mean escape latency values for
each training day are provided on Table S2 and the trajectories
of the mice in each group were also shown in Figure S5).
Compared to the control group, scopolamineꢀinduced mice
exhibited a prominent delay of mean escape latency compared
to the control group (41 seconds vs. 8.6 seconds). Meanwhile,
the 2fꢀtreated group had better performance than the tacrineꢀ
treated group, displaying the superior latency to target (17
seconds) to the tacrine group (26 seconds). Furthermore, 2d
significantly reduced the latency to target (11 seconds), closed
to the control group. The trend of distance to target was basiꢀ
cally consistent with the latency to target (Figure 5B). The
results suggested 2d and 2f considerably ameliorated the cogꢀ
nitive impairment of the treated mice and was much better
than tacrine, and also suggested that 2d and 2f may penetrate
the bloodꢀbrain barrier and target the central nervous system.
Given that the serious hepatotoxicity of tacrine for its longꢀ
term use, the possible drugꢀinduced hepatotoxicity associated
with the presence of the tetrahydroacridine nucleus in 2d and
2f were investigated by comparing to tacrine. Heparinized
serum was collected 8, 22, and 36 h after the administration of
tacrine, 2d and 2f, and the levels of alanine aminotransferase
(ALT) and aspartate aminotranferase (AST) were evaluated
(Table 2). ALT is primarily distributed in liver, and damaged
hepatocytes release ALT and AST into the extracellular space.
Thus, the serum levels of AST and ALT are directly related to
the extent of the liver damage. 8 h after tacrine administration,
the levels of ALT and AST were slightly induced. On the othꢀ
er hand, 2d and 2f did not showed any hepatotoxicity at all
timeꢀpoints, the level of ALT and AST even slightly reduced
at 8 h, compared to those of tacrine group. To confirm the
hepatotoxicity of 2d and 2f ulteriorly, morphologic studies by
immunohistochemical staining were performed. Treatment of
tacrine (Figure 5B), 2d (Figure 5C) or 2d (Figure 5D) did not
result in remarkable morphologic changes, such as hepatic cell
gap and number in liver compared to the control group (Figure
5A). Cytotoxicity profile was also demonstrated in human
hepatocellular HepG2 cells, where 2f showed low impairment
of the cell viability (IC50 = 35 ꢁM) (see Figure S4). In general,
2f showed hepatic safety profile, ensuring its further developꢀ
ment.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 4. Effects of oral administration of 2d (15 mg/kg), 2g
(15 mg/kg), 2f (15 mg/kg) and tacrine (15 mg/kg) on scopolꢀ
Table 2. ALT and ASAT Activity after the Administration of 2d, 2f and Reference Compound Tacrine (30 mg/kg)a
Group
ALT (U/L)
22 h
AST (U/L)
22 h
120 ± 20
120 ± 30
120 ± 19
120 ± 24
8 h
36 h
8 h
36 h
Control
Tacrine
2d
35 ± 8.4
32 ± 4.9
25 ± 7.9
28 ± 7.6
39 ± 9.4
25 ± 7
30 ± 8
25 ± 7
30 ± 4
29 ± 7
120 ± 20
130 ± 22
110 ± 24
110 ± 24
120 ± 13
110 ± 28
130 ± 32
120 ± 21
31 ± 12
29 ± 7.9
2f
aValues were expressed as the Mean ± SD (n=6).
In conclusion, dual GSKꢀ3β/AChE inhibitors were successꢀ
fully designed, which showed nanomolar potency on both
targets. To the best of our knowledge, it is the first attempt to
design GSKꢀ3β/AChE dual inhibitor as antiꢀAD agents. The
representative compound 2f significantly inhibited tau protein
hyperphosphorylation, while it also reduced the selfꢀ
aggregation of Aβ1ꢀ42. Furthermore, ICR male mice treated
with scopolamine exhibited significantly ameliorated memory
performance in a Morris water maze test when 2f was adminꢀ
istrated. Additionally, 2f exhibited preliminary safety in hepaꢀ
totoxicity studies, without improving the level of ALT and
AST. Such biological properties highlight 2f as a very interestꢀ
ing prototype in the search for new diseaseꢀmodifying drugs in
the treatment of AD.
ASSOCIATED CONTENT
Supporting Information
Figure 5. Histomorphological appearance of livers of male
mice after treatment with the solvent only (control, A), or 22 h
after administration of tacrine (B), 2d (C), or 2f (D). HE stainꢀ
ing, original magnification×200.
Synthetic procedures, characterization data, binding mode predicꢀ
tion of 2f with eeAChE and hGSKꢀ3β, and biological assay proꢀ
cedures (PDF)
Corresponding Author
*Eꢀmail: fengfeng@cpu.edu.cn
ACS Paragon Plus Environment

ACS Medicinal Chemistry Letters
Page 6 of 7
*Eꢀmail: sunhaopeng@163.com
(13) Giacobini, E.; Gold, G. Alzheimer disease therapy—
1
2
3
4
5
6
7
8
moving from amyloidꢀβ to tau. Nat. Rev. Neurol. 2013, 9,
677−686.
Author Contributions
The manuscript was written through contributions of all authors.
(14) Sivaprakasam, P.; Han, X.; Civiello, R. L.; JacutinꢀPorte,
S.; Kish, K.; Pokross, M.; Lewis, H. A.; Ahmed, N.;
Newitt, J. A.; Baldwin, E. T. Discovery of new acylaꢀ
minopyridines as GSKꢀ3 inhibitors by a structure guided
inꢀdepth exploration of chemical space around a pyrꢀ
rolopyridinone core. Bioorg. Med. Chem. Lett. 2015, 25,
1856−1863.
(15) EsquiviasꢀPérez, M.; Maalej, E.; Romero, A.; Chabꢀ
choub, F.; Samadi, A.; MarcoꢀContelles, J.; OsetꢀGasque,
M. J. Nontoxic and neuroprotective βꢀnaphthotacrines for
Alzheimer’s disease. Chem. Res. Toxicol. 2013, 26,
986−992.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENT
We gratefully thank the support from the grants (Nos.
81402851, 81373956 and 81573281) of National Natural Science
Foundation of China, BK20140957 of Natural Science Foundaꢀ
tion of Jiangsu Province. We also thank the support from Fundaꢀ
mental Research Funds for the Central Universities (2015ZD009),
Jiangsu Qing Lan Project.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(16) Inestrosa, N. C.; Dinamarca, M. C.; Alvarez, A. Amyꢀ
loidcholinesterase interactions. Implications for Alzꢀ
heimer’s disease. FEBS J. 2008, 275, 625−632.
(17) Nitsch, R. M.; Slack, B. E.; Wurtman, R. J.; Growdon, J.
H. Release of Alzheimer amyloid precursor derivatives
stimulated by activation of muscarinic acetylcholine reꢀ
ceptors. Science 1992, 258, 304−307.
(18) Zemek, F.; Drtinova, L.; Nepovimova, E.; Sepsova, V.;
Korabecny, J.; Klimes, J.; Kuca, K. Outcomes of Alzheiꢀ
mer's disease therapy with acetylcholinesterase inhibitors
and memantine. Expert Opin. Drug Saf. 2014, 13, 759ꢀ74.
(19) Hartmann, S.; Möbius, H. J. Tolerability of memantine in
combination with cholinesterase inhibitors in dementia
therapy. Int. Clin. Psychopharm. 2003, 18, 81−85.
(20) Leon, R.; Garcia, A. G.; MarcoꢀContelles, J. Recent adꢀ
vances in the multitargetꢀdirected ligands approach for the
treatment of Alzheimer’s disease. Med. Res. Rev. 2013,
33, 139−189.
(21) Spilovska, K.; Korabecny, J.; Nepovimova, E.; Dolezal,
R.; Mezeiova, E.; Soukup, O.; Kuca, K. Multitarget taꢀ
crine hybrids with neuroprotective properties to confront
Alzheimer's Disease. Curr. Top. Med. Chem. 2017, 17,
1006−1026.
(22) Romero, A.; Cacabelos, R.; OsetꢀGasque, M. J.; Samadi,
A.; MarcoꢀContelles, J. Novel tacrineꢀrelated drugs as poꢀ
tential candidates for the treatment of Alzheimer’s disꢀ
ease. Bio. Med. Chem. Lett. 2013, 23, 1916−1922.
(23) Nepovimova, E.; Korabecny, J.; Dolezal, R.; Babkova,
K.; Ondrejicek, A.; Jun, D.; Kuca, K. Tacrineꢀtrolox hyꢀ
brids: a novel class of centrally active, nonhepatotoxic
multiꢀtargetꢀdirected ligands exerting anticholinesterase
and antioxidant activities with low in vivo toxicity. J.
Med. Chem. 2015, 58, 8985−9003.
(24) Nepovimova, E.; Uliassi, E.; Korabecny, J.; Peñaꢀ
Altamira, L. E.; Samez, S.; Pesaresi, A.; Bolognesi, M. L.
Multitarget drug design strategy: quinoneꢀtacrine hybrids
designed to block amyloidꢀβ aggregation and to exert anꢀ
ticholinesterase and antioxidant effects. J. Med. Chem.
2014, 57, 8576−89.
(25) Singh, M.; Kaur, M.; Chadha, N.; Silakari, O. Hybrids: a
new paradigm to treat Alzheimer’s disease. Mol. Divers.
2016, 20, 271−297.
ABBREVIATIONS
MTDLs, multiꢀtargetꢀdirected ligands; AD, Alzheimer’s disease;
GSKꢀ3β, glycogen synthase kinaseꢀ3β; hꢀAChE, human acetylꢀ
cholinesterase; CAS, catalytic active site; PAS, the peripheral
anionic site; Aβ, βꢀamyloid; NFTs, neurofibrillary tangles; CNS,
central nervous system; BuChE, butyrylcholinesterase; PAS, peꢀ
ripheral anionic site; CPCCI, cyclopropanecarbonyl chloride;
TFA, trifluoroacetic acid; DIAD, diisopropyl azodiformate;
DIPEA, N,NꢀDiisopropylethylamine; ALT, alanine aminotransꢀ
ferase; AST, aspartate aminotranferase.
REFERENCES
(1) Lazarov, O.; Marr, R. A. Neurogenesis and Alzheimer’s
disease: at the crossroads. Exp. Neurol. 2010, 223,
267−281.
(2) Querfurth, H. W.; LaFerla, F. M. Alzheimer’s disease. N.
Engl. J. Med. 2010, 362, 329−344.
(3) Walsh, D. M.; Selkoe, D. J. Deciphering the molecular
basis of memory failure in Alzheimer’s disease. Neuron
2004, 44, 181−193.
(4) Kepp, K. P. Bioinorganic chemistry of Alzheimer’s disꢀ
ease. Chem. Rev. 2012, 112, 5193−5239.
(5) Dolan, P. J.; Johnson, G. V. The role of tau kinases in Alzꢀ
heimer’s disease. Curr. Opin. Drug Disc. 2010, 13,
595−603.
(6) Weingarten, M. D.; Lockwood, A. H.; Hwo, S. Y.;
Kirschner, M. W. A protein factor essential for microtuꢀ
bule assembly. P. Natl. Acad. Sci. 1975, 72, 1858−1862.
(7) Hardy, J. The relationship between amyloid and tau. J.
Mol. Neurosci. 2003, 20, 203−206.
(8) Avila, J.; Wandosell, F.; Hernández, F. Role of glycogen
synthase kinaseꢀ3 in Alzheimer’s disease pathogenesis
and glycogen synthase kinaseꢀ3 inhibitors. Expert. Rev.
Neurother. 2010, 10, 703−710.
(9) Phiel, C. J.; Wilson, C. A.; Lee, V. M.; Klein, P. S. GSKꢀ
3alpha regulates production of Alzheimer's disease amyꢀ
loidꢀbeta peptides. Nature 2003, 423, 435−439.
(10) Hernández, F.; Borrell, J.; Guaza, C.; Avila, J.; Lucas, J.
J. Spatial learning deficit in transgenic mice that condiꢀ
tionally overꢀexpress GSKꢀ3β in the brain but do not form
tau filaments. J. Neurochem. 2002, 83, 1529−1533.
(11) MoralesꢀGarcia, J. A.; LunaꢀMedina, R.; AlonsoꢀGil, S.;
SanzꢀSanCristobal, M.; Palomo, V.; Gil, C.; Santos, A.;
Martinez, A.; PerezꢀCastillo, A. Glycogen synthase kinase
3 inhibition promotes adult hippocampal neurogenesis in
vitro and in vivo. ACS Chem. Neurosci. 2012, 3, 963−971.
(12) Maqbool, M.; Mobashir, M.; Hoda, N. Pivotal role of
glycogen synthase kinaseꢀ3: A therapeutic target for Alzꢀ
heimer's disease. Eur. J. Med. Chem. 2016, 107, 63−81.
(26) Reichelt, A.; Bailis, J. M.; Bartberger, M. D.; Yao, G.;
Shu, H.; Kaller, M. R.; Allen, J. G.; Weidner, M. F.; Keeꢀ
gan, K. S.; Dao, J. H. Synthesis and structure–activity reꢀ
lationship of trisubstituted thiazoles as Cdc7 kinase inhibꢀ
itors. Eur. J. Med. Chem. 2014, 80, 364−382.
(27) But, T. Y. S.; Toy, P. H. Organocatalytic mitsunobu reacꢀ
tions. J. Am. Chem. Soc. 2006, 128, 9636−9637.
(28) Baki, A.; Bielik, A.; Molnar, L.; Szendrei, G.; Keseru ́ ̈,
G. M. A high throughput luminescent assay for glycogen
6
ACS Paragon Plus Environment

Page 7 of 7
ACS Medicinal Chemistry Letters
synthase kinaseꢀ3beta inhibitors. Assay Drug Dev. Techꢀ
nol. 2007, 5, 75−83.
1
(29) Ellman, G. L.; Courtney, K. D.; Andres, V., Jr.; Featherꢀ
Stone, R. M. A new and rapid colorimetric determination
of acetylcholinesterase activity. Biochem. Pharmacol.
1961, 7, 88−95.
(30) Cai, Z. Y.; Yu Z.; Bin Z. Roles of glycogen synthase kiꢀ
nase 3 in Alzheimer's disease. Curr. Alzheimer Res. 2012,
9, 864−879.
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
7
ACS Paragon Plus Environment