Design, synthesis, and evaluation of orally available clioquinol-moracin M hybrids as multitarget-directed ligands for cognitive improvement in a rat model of neurodegeneration in Alzheimer's disease

Article abstract of DOI:10.1021/acs.jmedchem.5b01222

A novel series of clioquinol-moracin hybrids were designed and synthesized by fusing the pharmacophores of clioquinol and moracin M, and their activities as multitarget-directed ligands against Alzheimer's disease were evaluated. Biological activity results demonstrated that these hybrids possessed significant inhibitory activities against phosphodiesterase 4D (PDE4D) and Aβ aggregation as well as remarkable antioxidant effects and excellent blood-brain barrier permeability. The optimal compound, 18d (WBQ5187), exhibited excellent PDE4D inhibitory potency (IC₅₀ = 0.32 μM), significant antioxidant effects, appropriate biometal chelating functions, and interesting properties that modulated self- and metal-induced Aβ aggregation. Two-dimensional NMR studies revealed that 18d had significant interactions with Aβ_1-42 at the R5, H6, H14, Q15, and F20 residues. Furthermore, this typical hybrid possessed preeminent neuroprotective effects against inflammation in microglial cells. Most importantly, oral administration of 18d·HCl demonstrated marked improvements in cognitive and spatial memory in a rat model of Alzheimer's disease and protected hippocampal neurons from necrosis.

Full text of DOI:10.1021/acs.jmedchem.5b01222

Subscriber access provided by CMU Libraries - http://library.cmich.edu
Article
Design, Synthesis and Evaluation of Orally Available Clioquinol-
Moracin M Hybrids as Multi-Target-Directed Ligands for Cognitive
Improvement in a Rat Model of Neurodegeneration in Alzheimer’s Disease
Zhiren Wang, Yali Wang, Bo Wang, Wenrui Li, Ling Huang, and Xingshu Li
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.5b01222 • Publication Date (Web): 16 Oct 2015
Downloaded from http://pubs.acs.org on October 21, 2015
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Design, Synthesis and Evaluation of Orally Available
Clioquinol-Moracin M Hybrids as Multi-Target-Directed Ligands for
Cognitive Improvement in a Rat Model of Neurodegeneration in
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
Alzheimer’s Disease
Zhiren Wang, Yali Wang, Bo Wang, Wenrui Li, Ling Huang,* and Xingshu Li*
School of Pharmaceutical Sciences, Sun Yatꢀsen University, Guangzhou, 510006, China
ABSTRACT
A novel series of clioquinolꢀmoracin hybrids were designed and synthesized by fusing
the pharmacophores of clioquinol and moracin M, and their activities as
multiꢀtargetꢀdirected ligands against Alzheimer's disease were evaluated. Biological
activity results demonstrated that these hybrids possessed significant inhibitory
activities against phosphodiesterase 4D (PDE4D) and Aβ aggregation as well as
remarkable antioxidant effects and excellent bloodꢀbrain barrier permeability. The
optimal compound, 18d, exhibited excellent PDE4D inhibitory potency (IC50 = 0.32
ꢁM), significant antioxidant effects, appropriate biometal chelating functions, and
interesting properties that modulated selfꢀ and metalꢀinduced Aβ aggregation.
Twoꢀdimensional NMR studies revealed that 18d had significant interactions with
Aβ1–42 at the R5, H6, H14, Q15 and F20 residues. Furthermore, this typical hybrid
possessed preeminent neuroprotective effects against inflammation in microglial cells.
Most importantly, oral administration of 18d•HCl demonstrated marked
improvements in cognitive and spatial memory in a rat model of Alzheimer’s disease
and protected hippocampal neurons from necrosis.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 90
1
2
3
4
Introduction
5
6
7
8
9
Alzheimer’s disease (AD) is one of the most dreadful and prevalent
neurodegenerative diseases and affects approximately 44 million people worldwide.1
AD has been demonstrated to possess an extremely complex interconnected network
of multiple factors and etiological hallmarks, including abnormal βꢀamyloid (Aβ)
deposition and accumulation, low levels of acetylcholine, oxidative stress,
neuroinflammation of the central nervous system (CNS) and dyshomeostasis of
biometals.2ꢀ4 Thus, there has been a major growth in efforts to develop
multiꢀtargetꢀdirected ligands (MTDLs) for AD networks.5ꢀ14
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
The aberrant accumulation of misfolded Aβ peptides is a major pathological
hallmark of AD.15 Aβ peptides can form oligomers, protofibrils, fibrils, and ultimately,
insoluble plaques through selfꢀassembly and hydrophobic interactions.16 In particular,
the soluble low molecular weight aggregated Aβ species can cause damage to
neuronal and mitochondrial functions and lead to oxidative stress and
neuroinflammation.17 In addition, high levels and miscompartmentalization of metal
ions such as Cu2+, Zn2+, Fe2+ and Fe3+ can readily bind to Aβ via three histidine
residues, H6, H13 and H14, and facilitate Aβ aggregation.4, 18ꢀ21 Meanwhile, under
physiological conditions, the interaction between redoxꢀactive metal ions and Aβ was
demonstrated to generate reactive oxygen species (ROS) via Fentonꢀlike reactions,
which led to oxidative stress and, eventually, neuronal death in AD patients.4, 16, 21
Therefore, the modulation of metal–Aβ interactions as well as the metal distribution
in the brain has become a promising approach to inhibit Aβ aggregation.16, 18, 21 The
ACS Paragon Plus Environment

Page 3 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
metal chelating agent, clioquinol (CQ, Fig. 1), and its second generation agent
8ꢀhydroxyquinoline ionophore, PBT2 (Fig. 1), demonstrated a significant
improvement in cognition and memory by redistributing metal ions and decreasing
plasma Aβ levels in Phase II clinical trials.22ꢀ27 Recently, the multiꢀtargetꢀdirected
metalꢀchelating agents M30, HLA20, tacrineꢀ8ꢀhydroxyquinoline hybrids, L2ꢀb and
J2326 were developed and studies demonstrated significant inhibition of Aβ
aggregation in vitro or improvements in cognition in vivo in mouse models (Fig.
1).28ꢀ35
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. Several reported multiꢀtargetꢀdirected metalꢀchelating agents, PDE4D inhibitors and Moracin M.
Phosphodiesterase 4D (PDE4D), one of the primary enzymes that hydrolyzes
intracellular cyclic adenosine monophosphate (cAMP) to control neuronal
stimulationꢀinduced signal transduction, is widely expressed in the hippocampus and
cortex.36ꢀ38 Gene deletion studies have demonstrated that a lack of PDE4D in mice
significantly enhanced longꢀterm hippocampalꢀdependent memory.39, 40 More recent
studies clearly indicate that PDE4D plays a leading role among the PDE subtypes that
mediate memory formation.41 Rolipram, a PDE4D inhibitor, has beneficial effects on
hippocampalꢀdependent memory tasks, but PDE4Dꢀdeficient mice did not show
altered memory with the administration of rolipram, which further demonstrated the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 90
1
2
3
4
5
6
7
8
9
crucial role of PDE4D in memory consolidation.40 The newly developed PDE4D
inhibitor GEBRꢀ7b enhanced spatial and object memory performance in mice and rats
(Fig. 1).42ꢀ44 In addition, MKꢀ0952 (Fig. 1), another PDE4 inhibitor that enhanced
cognition in preclinical studies, prompted clinical trials in patients with
mildꢀtoꢀmoderate AD from phase II clinical studies.45, 46 Moracin M is a natural
product that is isolated from the root bark of Morus alba L, and has a broad scope of
biological activity such as antioxidant and antiꢀinflammatory activity, and in
particular, the inhibition of PDE4D, which is directly related to AD pathology.47, 48
Considering that both inflammation and biometal dyshomeostasis are important
etiological hallmarks of AD, we fused the pharmacophores of CQ and moracin M into
one molecule without major structural modifications (Fig. 2). We then evaluated their
biological activities with regards to the modulation of Aβ aggregation and bloodꢀbrain
barrier (PAMPAꢀBBB) permeability, the inhibition of human PDE4D and the
prevention of neuroinflammation. Furthermore, an optimal compound, 18d•HCl, was
further assessed in a rat model of AD by conducting behavioral performance and
histopathological studies of hippocampal neurons.
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 2. Multiꢀtargetꢀdirected design strategy by fusing the pharmacophore moieties of CQ and Moracin M.
Results and Discussion
ACS Paragon Plus Environment

Page 5 of 90
Journal of Medicinal Chemistry
1
2
3
4
Chemistry
5
6
7
8
9
Syntheses of the Wittig reagents (4a-4c) are described in Scheme 1. Compound 2,
which was prepared as previously described,49 was reacted with triethyl phosphite to
give the key intermediate 3a. Compounds 3b and 3c were obtained by treating 3a
with NaClO and I2, respectively. Then, the reaction of compounds 3a-3c with methyl
iodide produced the Wittig reagents 4a-4c in good yields. The syntheses of the target
compounds 16a-16g, 17a-17d, 17f and 18c-18e are described in Scheme 2. First, the
hydroxyl groups of the corresponding salicylic aldehydes were protected by reacting
them with chloro(methoxy)methane to give compounds 6a-6g. The subsequent
reactions of compounds 6a-6g with Wittig reagents (4a-4c) provided compounds
7a-7g, 8a-8d, 8f and 9c-9e, respectively, which were treated with HCl (37%) in
CH3OH to provide compounds 10a-10g, 11a-11d, 11f and 12c-12e. Finally, the target
compounds were produced by cyclizing the phenol intermediates with I2, followed by
deprotection of the methyl groups with BBr3. The hydrochloride of 18d•HCl was
obtained by treating target compound 18d with HCl gas in ethyl acetate. The
syntheses of the target compounds 16h and 16i are presented in Scheme 3. The
starting material 8ꢀhydroxyquinaldine (3d) was first protected using methyl iodide to
generate 4d, followed by condensation with corresponding salicylic aldehydes to
produce 7h and 7i. The interesterification of 7h and 7i in the presence of CH3OH and
K2CO3 gave the intermediates 10h and 10i, which were treated with I2 to yield 13h
and 13i. Finally, the target compounds 16h and 16i were obtained by deprotecting the
methyl groups with BBr3.
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 90
1
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
Scheme 1. Synthesis of 2, 3a-3c, 4a-4c. Reagents and conditions: a. HCHO, HCl(g), rt, 72 h; b.P(OEt)3, 110 °C,
12 h; c. DMF, KOH, CH3I, rt, 12 h; d. H2O, KOH, NaClO, rt, 2 h; e. H2O, KOH, KI, I2, rt, 6 h.
Scheme 2. Synthesis of 6a-6g, 7a-7g, 8a-8d, 8f, 9c-9e, 10a-10g, 11a-11d, 11f, 12c-12e, 13a-13g, 14a-14d, 14f,
15c-15e and 16a-16g, 17a-17d, 17f, 18c-18e, 18d•HCl. Reagents and conditions: a. CH2Cl2, DIPEA, ClMOM, rt,
5 h; b. DMF, NaH, rt, 4 h; c. CH3OH, HCl (37%), rt, 2 h; d. THF, K2CO3, I2, rt, 16 h; e. anhydrous CH2Cl2, BBr3,
ꢀ78 °C, 12 h. f. EtOAc, HCl (g), rt, 12 h.
ACS Paragon Plus Environment

Page 7 of 90
Journal of Medicinal Chemistry
1
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
Scheme 3. Synthesis of 4d, 7h-7i, 10h-10i, 13h-13i, and 16h-16i. Reagents and conditions: a. DMF, K2CO3, CH3I,
rt, 12 h; b. acetic anhydride, 150 °C, 12 h; c. CH3OH, K2CO3, rt, 1 h; d. THF, K2CO3, I2, rt, 16 h; e. anhydrous
CH2Cl2, BBr3, ꢀ78 °C, 12 h.
Inhibition of PDE4D
The inhibition of PDE4D2 by the target compounds was determined using a
previously described method, with rolipram and moracin M as the reference
compounds.50, 51 The results in Table 1 indicated that most of the target compounds
exhibited more potent inhibitory activities against PDE4D2 compared to moracin M.
Among the three series of compounds 16 (X = H), 17 (X = Cl) and 18 (X = I), 16e
(IC50 = 1.21 ꢁM) and 18e (IC50 = 0.23 ꢁM), which bear a phenolic dihydroxyl group
on the benzofuran moiety, demonstrated better activities than the other analogs in the
series. The removal of the phenolic hydroxyl group on the benzofuran moiety in
compounds 16a (IC50 > 10 ꢁM), 16c (> 10 ꢁM) and 18c (0.89 ꢁM) reduced the
potency of these compounds, indicating that the phenolic hydroxyl group in the
benzofuran moiety was crucial. Compound 16h (8.86 ꢁM) and 16i (2.01 ꢁM), with
the benzofuran moiety on the pyridine ring, also exhibited moderate activities. The
structureꢀactivity relationship study revealed that the phenolic hydroxyl group on the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 90
1
2
3
4
5
6
7
8
9
R3 position of the benzofuran moiety was more favorable for the activity than the
other positions. Compounds 16d (IC50 = 2.31 ꢁM), 17d (0.96 ꢁM) and 18d (0.32
ꢁM), which bear a phenolic hydroxyl group on the R3 position, exhibited more
potent activity than compounds 16b (IC50 > 10 ꢁM), 16c (> 10 ꢁM), 17b (2.08 ꢁM),
17c (1.07 ꢁM) and 18c (0.89 ꢁM), respectively. Compounds 16f (3.67 ꢁM) and 17f
(2.92 ꢁM), with a methyl group on the R5 position of the benzofuran moiety, showed
lower activity than the parent compounds 16d and 17d, respectively. The presence of
a halogen, particularly iodine, on the oxine moiety was demonstrated to be
indispensable for the inhibitory activity. Compounds 16d, 17d and 18d, with or
without the chlorine and iodine group in the oxine moiety, exhibited IC50 values of
2.31, 0.96 and 0.32 ꢁM, respectively. It is worth noting that compound 16g, which
contains a diiodine on the benzofuran moiety, displayed IC50 values of 0.75 ꢁM.
Table 1. Inhibition of PDE4D and Aβ Selfꢀaggregation, Oxygen Radical Absorbance Capacity (ORAC, Trolox
Equivalents), Permeability (Pe×10ꢀ6 cm sꢀ1) Determined by the PAMPAꢀBBB Assay for Target Compounds and
Predicted Penetration of the CNS.
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
Aβ1ꢀ42
PDE4D2
selfꢀinduced
aggregation
(%Inhib.)b
12.4 ± 1.5
Pe
Comd.
R
H
X
H
ORACc
Pred.
IC50 (ꢁM)a
(×10ꢀ6 cm sꢀ1)d
16a
>10
0.5 ± 0.2
4.3 ± 0.4
CNS±
ACS Paragon Plus Environment

Page 9 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
16b
16c
16d
16e
R1=OH
R2=OH
R3=OH
R2=R4=OH
R3=OH,
R5=Me
R1=R3=I
R=H
H
H
H
H
>10
35.1 ± 2.1
46.3 ± 5.3
57.7 ± 7.8
61.2 ± 6.1
3.3 ± 0.1
3.5 ± 0.1
3.9 ± 0.2
3.1 ± 0.3
14.5 ± 0.5
15.8 ± 0.7
15.5 ± 0.1
2.6 ± 0.1
CNS+
CNS+
CNS+
CNS±
>10
2.31 ± 0.32
1.21 ± 0.07
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
16f
H
3.67 ± 0.43
53.1 ± 3.4
4.4 ± 0.1
16.8 ± 0.7
CNS+
16g
16h
16i
H
ꢀ
0.75 ± 0.05
8.86 ± 0.39
2.01 ± 0.23
13.2 ± 3.2
33.8 ± 0.9
48.7 ± 9.8
19.2 ± 4.4
48.1 ± 2.6
52.0 ± 4.3
59.1 ± 5.1
0.2 ± 0.1
4.3 ± 0.3
3.1 ± 0.3
0.3 ± 0.1
2.9 ± 0.3
2.4 ± 0.1
2.9 ± 0.3
3.6 ± 0.2
15.8 ± 0.7
13.3 ± 0.2
2.6 ± 0.1
CNS±
CNS+
CNS+
CNS±
CNS+
CNS+
CNS+
R=OH
ꢀ
17a
17b
17c
17d
H
Cl 3.05 ± 0.36
Cl 2.08 ± 0.12
Cl 1.07 ± 0.08
Cl 0.96 ± 0.11
R1=OH
R2=OH
R3=OH
R3=OH,
R5=Me
R2=OH
R3=OH
R2=R4=OH
16.4 ± 0.1
17.9 ± 1.5
21.4 ± 2.6
17f
Cl 2.92 ± 0.35
51.6 ± 3.9
3.6 ± 0.2
3.5 ± 0.4
CNS±
18c
I
I
I
ꢀ
ꢀ
ꢀ
ꢀ
0.89 ± 0.09
0.32 ± 0.02
0.23 ± 0.01
0.61 ± 0.03
2.91 ± 0.16
>10
63.5 ± 8.7
67.5 ± 1.9
69.2 ± 8.3
ꢀ e
3.5 ± 0.1
3.6 ± 0.2
3.1 ± 0.2
ꢀ e
16.6 ± 0.9
CNS+
CNS+
CNS–
ꢀ e
18d
19.1 ± 0.8
18e
0.8 ± 0.1
ꢀ e
ꢀ e
ꢀ e
ꢀ e
rolipram
moracin M
clioquinol
curcumin
ꢀ e
43.1 ± 5.6
< 5
2.5 ± 0.2
0.6 ± 0.1
ꢀ e
ꢀ e
ꢀ e
35.2 ± 4.8
ꢀ e
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 90
1
2
3
4
5
6
7
8
9
resveratrol
ꢀ
ꢀ
ꢀ
ꢀ e
ꢀ e
ꢀ e
39.1 ± 6.1
ꢀe
ꢀe
ꢀe
ꢀ e
ꢀ e
chlorpromazine
hydrocortisone
ꢀ e
ꢀ e
5.8 ± 0.6
1.0 ± 0.2
CNS+
CNS–
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
a The concentration (ꢁM) required for 50% inhibition was determined from doseꢀresponse curves, independent
experiments were performed in triplicate. Compounds 16a, 16b and 16c exhibited 32.3 ± 2.8, 35.4 ± 2.8 and 36.1
± 2.1% of inhibition at 10 ꢁM concentration, respectively (in duplicate). b The A
β1ꢀ42/inhibitor ratio was equal to
5/1, with inhibitor at 5 ꢁM and Aβ1ꢀ42 at 25 ꢁM. Values are the mean of three independent experiments in triplicate.
c The means ± SD of three independent experiments. Data are expressed as ꢁmol of trolox equivalent/ꢁmol of
tested compound. d Compounds were dissolved in DMSO at 5 mg mLꢀ1 and diluted with PBS / EtOH (70:30).
Values are expressed as the means ± SD at least three independent experiments, compounds with permeabilities Pe
4.7 × 10ꢀ6 cm sꢀ1 could cross the bloodꢀbrain barrier by passive diffusion. e n.t. = not tested.
>
Inhibition of Aβ Self-Aggregation
The inhibitory activities of the target compounds on Aβ1–42 selfꢀinduced
aggregation were first determined by a thioflavin T (ThT) fluorescence assay using
curcumin and resveratrol as reference compounds (Table 1). None of the tested
compounds exhibited significant fluorescence signals under the experimental
conditions. Most of the target compounds significantly inhibited Aβ1–42 selfꢀinduced
aggregation at the tested concentrations. A structureꢀactivity relationship analysis
indicated that a relationship between the position of phenolic hydroxyl groups on the
benzofuran moieties and PDE4D inhibitory activity was also observed for Aβ1−42
inhibitory activity. Compounds 16d, 17d and 18d, which have a phenolic hydroxyl
group at the R3 position, exhibited much higher inhibitory activities (57.7%, 59.1%
ACS Paragon Plus Environment

Page 11 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
and 67.5%, respectively) than moracin M (43.1%). Comparatively, compounds 16a
and 17a, which do not have phenolic hydroxyl groups on the benzofuran moieties,
exhibited only 12.4% and 19.2% inhibition, respectively. The types of halogens in the
oxine moieties were also important for proper biological activity; compounds 18cꢀ18e,
which have iodine on the oxine moieties, exhibited 63.5%, 67.5%, and 69.2%
inhibition, respectively, and were more potent than their chloroꢀ or nonꢀsubstituted
counterparts (16cꢀ16e: 46.3%, 57.7% and 61.2%; 17cꢀ17d: 52.0% and 59.1%).
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
2-D NMR Study of the Direct Interaction between Compound 18d and A
Solution
β in
Twoꢀdimensional (2ꢀD) Translational Relaxation Optimized Spectroscopy
1
(TROSY) Hꢀ15N Heteronuclear Single Quantum Correlation (HSQC) NMR was
performed to investigate the direct atomicꢀlevel interaction between compound 18d
15
15
and Nꢀlabeled Aβ1ꢀ42 (Fig. 3A). When compound 18d was titrated into Nꢀlabeled
Aβ1ꢀ42, significant chemical shift perturbations (CSP) and broadening of some of the
peaks for the Aβ1–42 residues were observed, particularly for Aβ1–42 residues R5, H6,
H14, Q15 and F20 (Fig. 3A and 3B). Because these residues are part of and in close
contact with the putative metal coordination site of Aβ (H6, H13, and H14), this
finding demonstrated that compound 18d was in close proximity to the metal binding
residues. Moreover, compound 18d not only possessed a strong targeting ability for
Aβ but also may chelated the metal ions surrounded by Aβ (Fig. 3C).19, 20, 28, 31 These
results were similar to those of previously reported metaꢀchelating compounds with
bifunctionality in metal chelation and Aβ interaction, which demonstrates the
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 90
1
2
3
4
significantly ability of this compound to modulate the formation of metaꢀinduced Aβ
5
6
7
29, 31, 52, 53
aggregates and promote disaggregation.28,
Substantial chemical shift
8
9
perturbations were observed when 5.0 equivalents of compound 18d was added into
15Nꢀlabeled Aβ1ꢀ42, which were attributed to a possible change in the conformation of
Aβ1–42 (Fig. 3A). Small chemical shifts were detected for the methionine 35 (M35)
residue, demonstrating that compound 18d did not induce Aβ1–42 oxidation.54
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
1
Figure 3. 2ꢀD NMR study the interactions of 18d with Aβ1–42 in solution. (A) 2D Hꢀ15N TROSYꢀHSQC NMR
spectra of 18dꢀtitrated Aβ1–42 (freshly dissolved Aβ1–42 (50 ꢁM) in 5 mM phosphate buffer, 0.5 mM Na2EDTA, and
ACS Paragon Plus Environment

Page 13 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
0.05 mM NaN3, 7% D2O (v/v), pH 7.5, 950 MHz). (B) Chemical shift perturbations for Aβ1–42 residues in the
presence of 18d (Aβ1–42 : 18d = 1 : 2). * Denotes absent or overlapping signals. (C) Residues with the largest
7
8
9
changes in chemical shifts mapped onto the structure of Aβ1–42 (PDB 1Z0Q).
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
Free Oxygen Radical Scavenging Ability
To evaluate the antioxidant activities of the target compounds, the free oxygen
radical Absorbance Capacity (ORAC) assay was performed and trolox and moracin M
were used as the reference compounds.55ꢀ57 The results presented in table 1 showed
that most of the target compounds demonstrated a higher antioxidant ability compared
to moracin M and clioquinol. As expected, the pharmacophore for ORAC of the target
compounds was primarily rooted in the benzofuran moiety. The structureꢀactivity
relationship between the position of phenolic hydroxyl group on benzofuran moiety
and the PDE4D inhibitory activity was also observed in the antioxidant activity assay.
Compounds possessing a phenolic hydroxyl group on the benzofuran moiety (16b-16f,
16h-16i, 17b-17f, 18c-18e) exhibited significantly higher ORAC values (4.4ꢀ2.4) than
those lacking the phenolic hydroxyl group (16a, 0.5; 17a, 0.3) or with a replacement
of the phenolic hydroxyl with halogen (16g, 0.2). Notably, compounds 16f and 17f,
which contain a methyl group on the R5 position of the benzofuran moiety, provided
the best ORAC values (4.4 and 3.6, respectively). The presence of a halogen,
particularly chlorine, on the oxine moiety was slightly unfavorable for antioxidant
activity. Compounds 16d, 17d and 18d, without or with a chlorine or iodine in the
oxine moiety, showed ORAC values of 3.9, 2.9 and 3.6, respectively.
Blood-Brain Barrier Permeability Assay
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 90
1
2
3
4
5
6
7
8
9
The parallel artificial membrane permeability (PAMPA) assay, which can
successfully predict passive BBB permeation, was used to determine the BBB
penetration of the target compounds.55ꢀ57 From the results presented in Table 1, it
could be seen that most of the target compounds exhibited significant BBB
permeability. Among them, compounds 16b-16d, 17b-17d and 18c-18d, which bear a
single phenolic hydroxyl group on the benzofuran moiety, exhibited much better BBB
permeability than compounds without or with diꢀphenolic hydroxyl groups, such as
16a, 16e, 16g, 17a and 18e, which exhibited questionable BBB permeability. As brain
penetration is the first requirement for developing antiꢀAD drugs, compound 18d was
chosen for further evaluation.
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
Neuroprotective Effects of the Compounds against LPS-Stimulated
Inflammation in BV2 Microglial Cells
Neuroinflammation plays a crucial role in causing neuronal death and damage,
which in turn leads to neurodegenerative diseases such as AD, PD and multiple
sclerosis.58 The activation of brain microglial cells in the CNS and the subsequent
excess production of inflammatory mediators, such as nitric oxide (NO), may result in
uncontrolled neuroinflammation in neurodegenerative diseases.59 Therefore, by
inhibiting the production of inflammatory mediators NO in microglial cells,
antineuroinflammatory therapy could delay or halt the disease progression prior to
irreversible damage and the occurrence of clinical symptoms. The
antineuroinflammatory properties of target compound 18d were determined by the
Griess assay, which detects the suppression of NO production following LPSꢀinduced
ACS Paragon Plus Environment

Page 15 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
inflammatory events in BV2 microglia cells using quercetin as a positive control. At
first, all of the compounds promoted more than 95% cellular viability at the tested
7
8
9
concentrations,
as
determined
by
the
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
3ꢀ(4,5ꢀdimethylꢀ2ꢀthiazolyl)ꢀ2,5ꢀdiphenylꢀ2Hꢀtetrazolium bromide (MTT) assay. As
the results in Fig. 4 and Table 2 indicated, compound 18d exhibited significantly
higher inhibition (EC50 = 1.5 ꢁM) of NO production in the LPSꢀstimulated BV2 cells
than moracin M (15.1 ꢁM) and clioquinol (> 30 ꢁM). Therefore, this result further
demonstrated that the unique structure of compound 18d resulted in improved
antineuroinflammatory activities compared to the leading compounds moracin M and
clioquinol.
Figure 4. Nitric oxide production inhibiting effect of control (A) and 18d (B).
Table 2. The EC50 Values for Inhibition Effects of NO Production in BV2 Cell.
Compound
quercetin
EC50 (ꢁM) a
11.0 ± 1.0
15.1 ± 0.8
Compound
clioquinol
18d
EC50 (ꢁM) a
> 30
moracin M
1.50 ± 0.2
a Data are expressed as the mean ± SD at least three independent experiments.
Metal-Chelating Properties of 18d
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 90
1
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
Figure 5. (A) UV absosobance spectrum of CQ (50 ꢁM) alone or in the presence of CuSO4 (50 ꢁM), ZnCl2 (50
ꢁM) FeSO4 (50 ꢁM) or FeCl3 (50 ꢁM) in buffer (20 mM HEPES, 150 mM NaCl, pH 7.4). (B). Determination of
the stoichiometry of complex Cu2+ꢀCQ by Job’s method. (C) UV absosobance spectrum of compound 18d (50 ꢁM)
alone or in the presence of CuSO4 (50 ꢁM), ZnCl2 (50 ꢁM) FeSO4 (50 ꢁM) or FeCl3 (50 ꢁM) in buffer (20 mM
HEPES, 150 mM NaCl, pH 7.4). (D) UVꢀvis titration of compound 18d (50 ꢁM) with Cu2+ in buffer (20 mM
HEPES, 150 mM NaCl, pH 7.4) at room temperature. The concentration of Cu2+ varied from 0 to 50 ꢁM. A
breakpoint was observed at 0.5:1 ratio. (E) highꢀresolution mass spectroscopy of the Cu2+ꢀ18d complex in
solution.
ACS Paragon Plus Environment

Page 17 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
A UVꢀvisible spectroscopy assay was performed to evaluate the ability of the
target compounds to chelate biometals such as Cu2+, Zn2+, Fe2+ and Fe3+.28, 55, 57, 60, 61
The results presented in Fig. 5A and 5C show that similar to parent compound CQ
(Fig. 5A), new optical bands were detected at 456 and 426 nm after the addition of
CuSO4 or ZnCl2 to the solution of compound 18d (Fig. 5C), which demonstrated the
production of the corresponding complex via metal chelation. Although there was no
obvious optical shift with the addition of FeSO4 or FeCl3, the dramatic increase in
absorption indicated a possible interaction between these biometals and the ligand.
The results of moracin M indicated that it chelated with Cu2+, but no optical bands
shifted in the presence of Zn2+, Fe2+and Fe3+(Fig. S1). Following the absorption at 456
nm, a series of UVꢀvis spectra were collected of compound 18d titrated with Cu2+,
and the isosbestic point demonstrated a 1:2 Cu2+/ligand molar ratio for the unique
Cu2+ꢀ18d complex shown in Fig. 5D. This result was in accordance with the
stoichiometry of the Cu2+ꢀCQ complex (break at a mole fraction of 0.659, indicating
1:2 Cu2+/ligand), which was determined by Job’s method (Fig. 5B). To further
confirm the components of the complex, highꢀresolution mass spectroscopy was
performed in solution, and the results in Fig. 5E show the presence of [(18d)2Cu – H]–
(calculated: 856.8452, detected: 856.8441) and [(18d)2Cu + Cl]– (calculated:
903.8239, detected: 903.8224), which again indicated a 1:2 Cu2+/ligand molar ratio in
the Cu2+ꢀ18d complex.
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
Modulation of Self- and Metal-Induced A
aggregates by Compound 18d
β Aggregation and Disaggregation of
A
β
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 90
1
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
Figure 6. Experiments of inhibition selfꢀinduced Aβ1–42 aggregation and disaggregation of selfꢀinduced Aβ1–42
aggregates in the presence of 18d. (a and b) Top: Scheme for the inhibition or disaggregation experiments. Bottom:
The fluorescence intensity of the ThT binding assay, data are expressed as the mean ± SD at least three
independent experiments. Statistical significance was analyzed by ANOVA: ∗
p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001.
(c and d) TEM image analysis of the inhibition of Aβ1–42 aggregation and disaggregation of selfꢀinduced Aβ1–42
aggregates. Serial number: (1) fresh Aβ1–42, (2) Aβ1–42 alone, (3) Aβ1–42 + moracin M, (4) Aβ1–42 + CQ, (5) Aβ1–42
+ 18d. Experimental conditions: Aβ1–42 (25 ꢁM); compound/Aβ1ꢀ42 = 1/1; PBS (50 mM); pH 7.4; 37 °C.
The regulation of selfꢀinduced Aβ aggregation and the disassembly of preformed
Aβ1ꢀ42 aggregates by compound 18d were studied using the ThT assay, and
transmission electron microscopy (TEM) was then performed to determine the
morphological changes in the Aβ species (Fig. 6). The ThT assay showed that
compound 18d inhibited 86.3 ± 4.1% of the selfꢀinduced Aβ1ꢀ42 aggregation and
disassembled 83.9 ± 3.6% of the preformed Aβ1ꢀ42 aggregates (Fig. 6a and 6b), which
ACS Paragon Plus Environment

Page 19 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
were significantly higher than CQ (20.2 ± 7.2% and 13.5 ± 3.2%, respectively, p <
0.001) and moracin M (62.3 ± 5.7% and 60.4 ± 5.3%, respectively, p < 0.05 and 0.01).
In accord with the ThT assay, fewer and shorter Aβ1ꢀ42 aggregates (Fig. 6c and 6d,
sequence 5) were observed after compound 18d was added to the samples compared
to moracin M (Fig. 6c and 6d, sequence 3) and CQ (Fig. 6c and 6d, sequence 4).
These results strongly indicated 18d could effectively control selfꢀinduced Aβ
aggregation and disassemble the preformed selfꢀinduced Aβ aggregates, which should
mainly be derived from its overall framework rather than the individual moieties.
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 7. Experiments of inhibition Cu2+ꢀinduced Aβ1–42 aggregation and disaggregation of Cu2+ꢀinduced Aβ1–42
aggregates in the presence of 18d. (a and b) Top: Scheme for the inhibition or disaggregation experiments. Bottom:
The fluorescence intensity of the ThT binding assay, data are expressed as the mean ± SD at least three
independent experiments. Statistical significance was analyzed by ANOVA: ∗
(c and d) TEM images analysis of the inhibition of Cu2+ꢀinduced Aβ1–42 aggregation and disaggregation of
Cu2+ꢀinduced Aβ1–42 aggregates. Serial number: (1) fresh A 1–42, (2) Aβ1–42 alone, (3) Aβ1–42 + Cu2+, (4) Aβ1–42
p < 0.01, ∗∗ p < 0.01, ∗∗∗ p < 0.001.
β
+
Cu2+ + CQ, (5) Aβ1–42 + Cu2+ + 18d, (6) Aβ1–42 + Cu2+ + moracin M. Experimental conditions: Aβ1–42 (25 ꢁM);
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 90
1
2
3
4
chelator/A
β
1ꢀ42/Cu2+ = 2/1/1; HEPES (20 ꢁM) and NaCl (150 ꢁM); pH 6.6; 37 °C.
5
6
7
8
9
In addition, compound 18d also showed remarkable effects on the modulation of
Cu2+ꢀinduced Aβ aggregation (Fig. 7). Excellent Cu2+ꢀinduced Aβ aggregation
inhibition (80.4 ± 5.5%) was obtained compared to moracin M and CQ (50.8 ± 4.8%
and 64.4 ± 5.0%, respectively, p < 0.01,). Similarly, 81.0 ± 5.6% of the of the
preformed Cu(II)ꢀinduced aggregates were disaggregated, which was much better
than moracin M and CQ (37.3 ± 8.7% and 37.6 ± 5.5%, respectively, p < 0.01).
The TEM assay showed that the Cu2+ꢀtreated sample of fresh Aβ produced more
fibrils than did the nonꢀtreated sample (Fig. 7c, sequence 2 and 3). When compound
18d and Cu2+ were incubated with Aβ, fewer Aβ fibrils were detected (Fig. 7c,
sequence 5). Fig. 7d shows the TEM assay results for disaggregation in the presence
of compound 18d, moracin M and CQ. As indicated, compound 18d also exhibited
stronger disaggregation activity on the preformed Cu(II)ꢀinduced Aβ fibrils. 18d and
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
1
CQ possess similar metal binding properties, but a 2D Hꢀ15N TROSYꢀHSQC NMR
demonstrated that the CQ scaffold had poor interactions with Aβ residues.28
Consequently, the improved effects of 18d on the modulation of Cu(II)ꢀinduced
Aβ1−42 aggregation could be attributed to the introduction of a benzofuran moiety to
CQ, which may have increased the interactions between 18d and the metal
coordination residues (especially H6 and H14, Fig. 3) of Aβ1−42. Overall, because of
its unique structure, compound 18d could significantly modulate Cu(II)ꢀinduced
Aβ1−42 aggregation and disassembled the preformed Cu(II)ꢀinduced aggregates into
smaller, amorphous conformations.
ACS Paragon Plus Environment

Page 21 of 90
Journal of Medicinal Chemistry
1
2
3
4
Metabolic Stability
5
6
7
8
9
The metabolic stability of compound 18d•HCl in rat liver microsomes was
examined and compared with donepezil and CQ using testosterone as a positive
control. As shown in Table 3, donepezil, CQ and testosterone had a T1/2 of 73.3, 34.8
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
and 2.1 min, respectively, in liver microsomes; these values were consistent with
63
previous reports and internal validation data.62,
Comparatively, 18d•HCl
demonstrated significant stability after incubation with rat liver microsomes (T1/2
=
293.8 min) under the same conditions. These data suggest that compound 18d•HCl
possesses considerably greater metabolic stability in vitro compared with the
reference compounds.
Table 3. Metabolic Stability of 18d•HCl in Liver Microsomes of SD Rat.
compound
testosterone b
donepezil c
CQ
k (minꢀ1)
T1/2 (min) a
2.1 ± 0.3
0.31513 ± 0.04378
0.00896 ± 0.00033
0.01988 ± 0.00178
0.00236 ± 0.00008
73.3 ± 2.6
34.8 ± 3.1
293.8 ± 7.2
18d•HCl
a Results are expressed as the mean ± SD of at least three independent experiments performed in triplicate. b, c The
positive control (testosterone) and the compound donepezil exhibited metabolic stability that was consistent with
the literature and internal validation data.62, 63
Compound 18d•HCl Demonstrated no Acute Toxicity
Twenty KM mice (KM mice, which are common closed colony mice and most
widely used in biomedical research in China) were randomly allocated into 4 groups,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 90
1
2
3
4
5
6
7
8
9
and the test compound 18d•HCl was given in single oral doses of 0, 677, 1333, or
2000 mg/kg. After administration of the compound, mice were monitored
continuously for the first 4 h for any abnormal behavior and mortality changes,
intermittently for the next 24 h, and occasionally thereafter for 14 days to monitor the
onset of any delayed effects. During the experimental period, no acute toxicity such as
mortality, significant abnormal changes in water or food consumption or body weight
were observed. Furthermore, all mice were sacrificed on the 14th day after drug
administration, and no damage to the heart, liver or kidneys was macroscopically
detected. Overall, compound 18d•HCl was nonꢀtoxic and well tolerated at doses up to
2000 mg/kg.
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
Cognitive and Memory Improvement in a Rat Model of AD
Figure 8. (A) The mean daily body weight profile of each group rats during drug administration period (n = 11).
(B) The escape latency time of each group was counted every day during the period of training trial (mean ± SD, n
= 11). (C) The representative tracks of the rats in the Morris water maze during the training trial period. The
location of the platform and effective region (twofold diameter of the platform) were represented as a blue and
ACS Paragon Plus Environment

Page 23 of 90
Journal of Medicinal Chemistry
1
2
3
4
bright green circle, respectively.
5
6
7
8
9
The learning and memory assessments were performed using the Morris water
maze by administering compound 18d•HCl to the Aβꢀinduced AD rat, which suffers
from learning and memory dysfunction, with clioquinol and donepezil as the positive
control treatments. Fiftyꢀfive male Wistar rats (270ꢀ300 g) were randomly allocated
into 5 groups (n = 11 for each group): sham, model, donepezil (2 mg/kg/day), CQ (30
mg/kg/day) and compound 18d•HCl (30 mg/kg/day).22, 23 The rats for model, CQ,
donepezil and compound 18d•HCl groups received intrahippocampal injections of an
aggregated Aβ solution, while the sham group of rats were injected with saline. After
the surgery, the rats in each group were administered the corresponding drug dosage
in a 0.5% CMCꢀNa solution (4 mL/kg) by intragastric infusion throughout the entire
assay (a blank 0.5% CMCꢀNa solution was used for the sham and model groups).
Twentyꢀsix days later, behavioral performance was evaluated using the Morris water
maze task, which demanded incremental learning of the location of a fixed, hidden
platform throughout the training period (5 consecutive days). The results presented in
Fig. 8 showed that the administration of compound 18d•HCl (Fig. 8A) during the
drug treatment period (32 consecutive days) did not influence the mean daily body
weight profile of the rats, and, in particular, it did not cause any adverse or abnormal
events (such as emesisꢀlike or diarrhea behavior) or affect the survival. This
demonstrated that longꢀterm administration of compound 18d•HCl is safe at doses of
30 mg/kg/day.
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
During the training trials (Fig. 8B), the mean escape latency and searching
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 90
1
2
3
4
5
6
7
8
9
distance for the rats in each group declined progressively; however, the model rats
normally spent more time and required more distance to find the hidden platform,
which was located in the center of the first quadrant (Fig. 8B and 8C).
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 9. (A) Effect of memory retention on the spatial probe trial (with the platform removed from the pool) in
the water maze test. The results were expressed as the means ± SD (n = 11). Statistical significance was analyzed
by twoꢀway ANOVA: ns
p > 0.05, ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001 versus the model and compound 18d•HCl
groups. (1) Number of virtual platform (the original platform location) crossings. (2) The average swimming speed
for the rats. (3) The swimming path length in the virtual platform test. (4) The time spent in the virtual platform
quadrant. (5) The swimming path length in the effective region (twoꢀfold diameter of the platform). (6) The time
spent in the effective region. (B) The representative tracks of the rats in Morris water maze during the spatial probe
trial period. The location of the platform and the effective region (twofold diameter of the platform) were
represented as a blue and bright green circle, respectively.
Twentyꢀfour hours after the last training trial, memory retention was assessed by
ACS Paragon Plus Environment

Page 25 of 90
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
performing a spatial probe trial, with the platform removed from the pool and the
twoꢀfold diameter of the platform was set as the effective region. (The statistical
significance of each was analyzed by ANOVA versus the model and compound
18d•HCl groups.) The results in Fig. 9 indicated that the sham group had significantly
higher numbers of virtual platform (the original platform location) crossings (5.0 ±
0.8, p < 0.001) compared to the model group rats (2.1 ± 0.9), which strongly
suggested that intrahippocampal injection of Aβ led to a spatial memory deficiency in
the rats (Fig. 9A1). The numbers of virtual platform crossings for the rats that were
administered donepezil (3.9 ± 1.1, p < 0.01), CQ (4.0 ± 0.9, p < 0.01) and compound
18d•HCl (6.0 ± 1.7, p < 0.001) were remarkably improved compared to the model
group. Interestingly, the rats treated with compound 18d•HCl (6.0 ± 1.7)
demonstrated a more favorable amelioration of the cognitive and memory functions
compared to the donepezil (3.9 ± 1.1, p < 0.01) and CQ (4.0 ± 0.9, p < 0.01) groups
(Fig. 9A1). Meanwhile, the average swimming speed for each group of rats was
virtually equivalent, which further demonstrated that the longꢀterm uptake of
compound 18d•HCl did not impair the animals’ motility and exploratory activities
(Fig. 9A2, p > 0.05, sham = 20.3 ± 2.9, model = 21.0 ± 3.1, donepezil = 21.9 ± 4.1,
CQ = 22.9 ± 2.0 and compound 18d•HCl = 22.2 ± 4.0 cm/s). On one hand, the
swimming path length in the virtual platform location of the compound 18d•HCl
(33.4 ± 18.1, p < 0.001, p < 0.05 and p < 0.01) and sham (25.7 ± 8.0, p < 0.01) groups
were longer than the model (10.8 ± 5.4), donepezil (25.0 ± 11.5) and CQ (19.9 ± 2.1)
groups (Fig. 9A3). On the other hand (Fig. 9A4), the rats treated with compound
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
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 26 of 90
1
2
3
4
5
6
7
8
9
18d•HCl (1.9 ± 1.0, p < 0.05) spent more time in the virtual platform location than
the model rats (0.9 ± 0.6). These results further revealed that administration of
18d•HCl led to a substantial improvement of the conventional reference spatial
memory and cognitive abilities compared to the CQꢀtreated, donepezilꢀtreated or
untreated littermates. Similarly, the sham group (102.6 ± 6.7, p < 0.05), as well as the
compound 18d•HCl group (121.3 ± 45.4, p < 0.001 and p < 0.05), demonstrated an
obviously longer swimming path length in the effective region than the model (72.1 ±
6.0) and donepezil (92.6 ± 17.2) groups (Fig. 9A5). The time spent in the effective
region for the sham (5.7 ± 1.2, p < 0.01) and compound 18d•HCl (7.1 ± 2.7, p <
0.001, p < 0.01 and p < 0.05) groups were significantly difference from the model (3.3
± 1.1), donepezil (5.0 ± 0.9) and CQ (5.1 ± 0.9) groups (Fig. 9A6), which further
suggested that the compound 18d•HCl-treated rats, unlike the rats in the model,
donepezil or CQ groups, formed a more clear and definite spatial preference for the
correct quadrant of the platform (Fig. 9B). Taken together, these overall behavioral
performance observations and results demonstrated that compound 18d•HCl can
markedly improve spatial memory and cognition compared to CQ and donepezil at
the tested dosage.
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
Histopathological Studies in the Hippocampus
ACS Paragon Plus Environment

Page 27 of 90
Journal of Medicinal Chemistry
1
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
Figure 10. Histopathological changes in rat hippocampal neurons (HE staining × 200, n = 11). CA1, CA2, CA3
and DG indicate the CA1, CA2, CA3 and Dentate Gyrus (DG) regions of the hippocampus. Representative 200
×magnification images of HEꢀstained hippocampus for each group are shown. The rats in the sham and compound
18d•HCl groups did not demonstrate histopathological abnormalities or neuronal loss; the hippocampal neurons
were clear and intact, and the cells were properly aligned. In the model group, the neurons were loosely arranged,
with a condensed cytoplasm, an irregular shape and a pyknotic nucleus. In the CQ and donepezil groups, the
damage was less than the model group, but greater than the sham and compound 18d•HCl groups.
The histopathological changes of the hippocampal neurons were evaluated by
hematoxylin and eosin (HE) staining, which demonstrated significant neuronal
abnormalities in the hippocampus of the rats in the model group (Fig. 10). The
pyramidal neurons in the CA1, CA2, CA3 and dentate gyrus (DG) regions of the
hippocampus were disintegrated, with remarkable nuclear pyknosis, neuronal
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 90
1
2
3
4
5
6
7
8
9
shrinkage, an irregular shape and more darkly stained cells, which indicated a necrotic
morphology. In contrast, the hippocampal neurons from the sham group were clear
and intact, and the cells were properly aligned. In the compound 18d•HCl-treated
group, no significant neuronal abnormalities were detected in the hippocampus; the
pyramidal cells in each region were arranged neatly and tightly, the nucleolus was
visible, the boundary between the cytoplasm and nucleus was clear, and no cell loss
was found, which strongly indicated that compound 18d•HCl significantly protected
the hippocampal neurons from damage and death. Additionally, the cells in the
donepezil and CQ groups possessed a better morphology than the model group, but
not as good as the sham and compound 18d•HCl groups. The denatured cell index
(DCI = the number of denatured cells/the number of total cells) in each region of the
hippocampus demonstrated that the oral intake of compound 18d•HCl significantly
reduced the ratio of dead neurons in the CA1 (11.2% ± 5.7, p < 0.001), CA2 (10.3% ±
5.6, p < 0.001), CA3 (6.9% ± 2.1, p < 0.001) and DG (12.2% ± 6.5, p < 0.001) regions
compared to the model (87.5% ± 9.5, 70.3% ± 13.6, 86.8% ± 8.1 and 80.1% ± 10.8,
respectively), donepezil (46.8% ± 8.5, 51.3% ± 10.0, 45.6% ± 5.3 and 55.6% ± 10.9,
respectively) and CQ (47.8% ± 5.3, 36.3% ± 5.8, 33.6% ± 8.5 and 62.0% ± 18.5,
respectively) groups (Fig. 11). In general, the compound 18d•HCl treatment
significantly reduced the number of apoptotic neurons in the hippocampus compared
to the donepezil and CQ treatments at the dosage used here.
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
ACS Paragon Plus Environment

Page 29 of 90
Journal of Medicinal Chemistry
1
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
Figure 11. The denatured cell index (DCI = the number of denatured cells/the number of total cells) of the rat
hippocampal neurons (HE staining, n = 11). CA1, CA2, CA3 and DG indicate the CA1, CA2, CA3 and Dentate
Gyrus (DG) regions of the hippocampus. The results were expressed as the means ± SD (n = 11). Statistical
significance was analyzed by twoꢀway ANOVA: ns
model and compound 18d•HCl groups.
CONCLUSION
p > 0.05, ∗ p < 0.05, ∗∗ p < 0.01, ∗∗∗ p < 0.001 versus the
In summary, this study involved the design, synthesis and biological evaluation of
a novel series of multiꢀtargetꢀdirected ligands against AD by fusing the
pharmacophores of clioquinol and moracin M. Most of the target compounds
displayed excellent inhibition of Aβ aggregation and PDE4D activity as well as
remarkable BBB penetration and antioxidant effects. The optimal candidate
compound, 18d, directly interacted with crucial residues of Aβ in solution, as shown
by 2D NMR, and modulated metalꢀinduced Aβ aggregation, and distinctively
disaggregated selfꢀ or metalꢀinduced Aβ aggregates. This unique hybrid demonstrated
significant effects in protecting neuronal cells against LPSꢀstimulated inflammation.
Most importantly, an oral intake of 30 mg/kg/day of compound 18d•HCl by AD
ACS Paragon Plus Environment