New multi-target-directed small molecules against Alzheimer's disease: A combination of resveratrol and clioquinol

Article abstract of DOI:10.1039/c4ob00998c

Alzheimer's disease (AD) is currently one of the most difficult and challenging diseases to treat. Based on the 'multi-target-directed ligands' (MTDLs) strategy, we designed and synthesised a series of new compounds against AD by combining the pharmacophores of resveratrol and clioquinol. The results of biological activity tests showed that the hybrids exhibited excellent MTDL properties: a significant ability to inhibit self-induced β-amyloid (Aβ) aggregation and copper(ii)-induced Aβ aggregation, potential antioxidant behaviour (ORAC-FL value of 0.9-3.2 Trolox equivalents) and biometal chelation. Among these compounds, (E)-5-(4-hydroxystyryl)quinoline-8-ol (10c) showed the most potent ability to inhibit self-induced Aβ aggregation (IC₅₀ = 8.50 μM) and copper(ii)-induced Aβ aggregation and to disassemble the well-structured Aβ fibrils generated by self- and copper(ii)-induced Aβ aggregation. Note that 10c could also control Cu(i/ii)-triggered hydroxyl radical (OH) production by halting copper redox cycling via metal complexation, as confirmed by a Cu-ascorbate redox system assay. Importantly, 10c did not show acute toxicity in mice at doses of up to 2000 mg kg^-1 and was able to cross the blood-brain barrier (BBB), according to a parallel artificial membrane permeation assay. These results indicate that compound 10c is a promising multifunctional compound for the development of novel drugs for AD. This journal is the Partner Organisations 2014.

Full text of DOI:10.1039/c4ob00998c

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
PAPER
New multi-target-directed small molecules against
Alzheimer’s disease: a combination of resveratrol
and clioquinol†
Cite this: Org. Biomol. Chem., 2014,
12, 5936
Fei Mao,a Jun Yan,a Jianheng Li,a Xian Jia,b Hui Miao,a Yang Sun,a Ling Huang*a and
Xingshu Li*a
Alzheimer’s disease (AD) is currently one of the most difficult and challenging diseases to treat. Based on
the ‘multi-target-directed ligands’ (MTDLs) strategy, we designed and synthesised a series of new com-
pounds against AD by combining the pharmacophores of resveratrol and clioquinol. The results of bio-
logical activity tests showed that the hybrids exhibited excellent MTDL properties: a significant ability to
inhibit self-induced β-amyloid (Aβ) aggregation and copper(II)-induced Aβ aggregation, potential anti-
oxidant behaviour (ORAC-FL value of 0.9–3.2 Trolox equivalents) and biometal chelation. Among these
compounds, (E)-5-(4-hydroxystyryl)quinoline-8-ol (10c) showed the most potent ability to inhibit self-
induced Aβ aggregation (IC50 = 8.50 μM) and copper(II)-induced Aβ aggregation and to disassemble the
well-structured Aβ fibrils generated by self- and copper(II)-induced Aβ aggregation. Note that 10c could
also control Cu(I/II)-triggered hydroxyl radical (OH•) production by halting copper redox cycling via metal
complexation, as confirmed by a Cu–ascorbate redox system assay. Importantly, 10c did not show acute
toxicity in mice at doses of up to 2000 mg kg−1 and was able to cross the blood–brain barrier (BBB),
according to a parallel artificial membrane permeation assay. These results indicate that compound 10c is
a promising multifunctional compound for the development of novel drugs for AD.
Received 14th May 2014,
Accepted 13th June 2014
DOI: 10.1039/c4ob00998c
www.rsc.org/obc
been developed as the main therapeutic options for the treat-
ment of AD), the following are also important in the aetiology
Introduction
Alzheimer’s disease (AD), which was first described by Alois of AD: β-amyloid (Aβ) deposits, τ-protein aggregation, oxidative
Alzheimer in 1906,1–3 is currently one of the most difficult and stress and the dyshomeostasis of biometals.5–7 Therefore, a
baffling diseases to treat. The characteristic symptom of AD is more appropriate approach to addressing the multifaceted
the progressive loss of memory and other cognitive functions nature of AD may be the development of multi-target-directed
that invariably cause a decline in an individual’s ability to ligands (MTDLs).8,9
function in society. AD occurs more frequently in elderly
The extracellular senile plaques and intracellular neuro-
people, and it is believed that 35.6 million people suffered fibrillary tangles formed by hyperphosphorylated τ-protein are
from AD in 2010. Note that the number of patients is predicted the key pathological markers of AD. Because β-amyloid peptide
to nearly double every 20 years to 65.7 million in 2030 and (Aβ) is the main component of senile plaques, the “amyloid
115.4 million in 2050.4
hypothesis” proposes that the production and accumulation of
In recent decades, much evidence has suggested that AD is the oligomeric aggregates of Aβ in the brain are the central
a multifaceted illness. In addition to the low levels of acetyl- event in the pathogenesis of AD and these aggregates initiate
choline (on the basis of which acetylcholinesterase inhibitors, the pathogenic cascade that ultimately leads to neuronal loss
such as tacrine, donepezil, rivastigmine and galantamine, have and dementia.10 Therefore, some anti-Aβ aggregation com-
pounds have been identified for the treatment of AD.11–14
Recent studies have indicated that the dysregulation of
aSchool of Pharmaceutical Sciences, Sun Yat-sen University, Guangzhou 510006,
China. E-mail: huangl72@mail.sysu.edu.cn, lixsh@mail.sysu.edu.cn;
brain metal ion homeostasis, especially that of copper and
zinc, may play a very important role in the pathogenesis of AD.
Zinc(II) and copper(II) were found to be markedly enriched in
plaque amyloid, and were confirmed to coordinate to histidine
residues in senile plaque (SP) cores by Raman microscopy.15
While copper and zinc accumulate in the extracellular plaque,
Fax: +86-20-3994-3051+86-20-3994-3050; Tel: +86-20-3994-3051+86-20-3994-3050
bSchool of Pharmaceutical Engineering, Shenyang Pharmaceutical University,
Shenyang 110016, China
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c4ob00998c
5936 | Org. Biomol. Chem., 2014, 12, 5936–5944
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 1 Chemical structure of metal-chelating agents.
the intracellular copper stores are deficient in AD patients and
can promote the generation of Aβ, leading to a vicious cycle of
plaque formation and metal ion accumulation.16 Therefore,
the modulation of these biometals in the brain has been pro-
posed to be a potential therapeutic strategy for the treatment
of AD.17 Considering this proposal, biometal chelators such as
desferrioxamine (DFO), EDTA, clioquinol, and 5-[N-methyl-N-
propargylaminomethyl]-8-hydroxyquinoline (M30) (Fig. 1) have
been studied.18
Fig. 2 Design strategy for multifunctional chelators.
(CQ) to obtain a series of 8-hydroxyquinoline-resveratrol deriva-
tives that are expected to acts as biometal chelators, anti-
oxidants, and inhibitors of Aβ aggregation (Fig. 2).
Intracellular oxidative stress is one of the earliest events in
the pathogenesis of AD.19 Neuronal tissue is very sensitive to
oxidative stress, and an imbalance in pro-oxidant vs. anti-
oxidant homeostasis in the CNS results in the production of
several potentially toxic reactive oxygen species (ROS), which
play a prominent role in several neurodegenerative diseases,
similar to microorganisms in infectious diseases.20 Anti-
oxidant protection is necessary during aging, especially in AD
patients, because endogenous antioxidant protection rapidly
Results and discussion
Chemistry
The target compounds 10a–10e, 10g, 13a–e and 13h–13i were
synthesised according to the synthetic approaches depicted in
Schemes 1 and 2. First, aldehyde 2 was protected with a
methoxymethyl (MOM) group through the reaction of 4-hydroxy-
benzaldehyde with chloro(methoxy)methane in the presence of
diisopropylethylamine. The reaction of methyl 3,5-dihydroxy-
benzoate (3) and chloro(methoxy)methane provided an inter-
mediate 4, which was reduced with LiAlH4 and then oxidised
using PCC to give the MOM-protected aldehyde 5. Alter-
natively, 8-hydroxyquinoline or its analogue 2-methyl-8-hydro-
xyquinoline reacted with formaldehyde and concentrated
declines.
A recent statistical study involving developed
countries indicated that a higher consumption of dietary anti-
oxidants, such as flavonoids, is associated with the lower rates
of dementia.21 Thus, drugs that specifically scavenge oxygen
radicals could be useful for either the prevention or the treat-
ment of AD.22,23
The antioxidant properties of resveratrol have been
thoroughly demonstrated and are associated with a wide range
of biological effects. Fortunately, this compound has also been
reported to cause no adverse effects.24 In addition, recent
papers emphasise the Aβ anti-aggregative and cytoprotective
properties of resveratrol in human neuroblastoma cells.25,26
Despite these positive effects, the poor bioavailability of
dietary supplements of resveratrol severely limits its use.27
Evidence suggests that the most successful treatment strat-
egy will likely incorporate a sequential and multifactorial
approach.9 Recently, multifunctional compounds targeting
metal-induced Aβ aggregation have been extensively studied
for AD treatments28–30 such as a hybrid of 6-chlorotacrine and
a metal-Aβ modulator reported by Lim and coworkers that
inhibits acetylcholinesterase and metal-free/metal-induced Aβ
aggregation.31 In addition, Green et al. reported a pyclen deri-
vative with enhanced antioxidant properties.32 Our approach to
the identification of novel multi-target-directed drug candidates
for the treatment of AD is to develop multifunctional com-
pounds that can simultaneously inhibit Aβ aggregation, modu-
late the dyshomeostasis of biometals, and manage oxidative
stress. In this paper, we combined resveratrol with the pharma-
cophore moiety of the well-known metal chelator clioquinol
Scheme 1 Reagents and conditions: (a) MOMCl, (i-Pr)2NEt, CH2Cl2, 0 °C
to r.t.; (b) LiAlH4, THF, 0 °C to r.t.; (c) PCC, CH2Cl2; (d) HCl (37% in water),
HCHO (37% in water), 0 °C to r.t.; (e) P(OEt)3, reflux; (f) CH3ONa, DMF,
0 °C to 80 °C; (g) 6 M HCl, CH3OH, reflux; (h) BBr3, CH2Cl2, −78 °C to r.t.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5936–5944 | 5937

View Article Online
Paper
Organic & Biomolecular Chemistry
Inhibition of self-mediated Aβ1–42 aggregation
To evaluate the inhibitory activities of 8-hydroxyquinoline-
resveratrol derivatives against self-mediated Aβ1–42 aggregation,
a thioflavin-T (ThT) fluorescence binding assay33 was per-
formed, and resveratrol, a known amyloid aggregation inhibi-
tor, was used as a reference compound. The results in Table 1
show that most of the synthesised compounds exhibited mod-
erate to good potencies compared to resveratrol. Among them,
compounds 10a–10d and 10g, which feature amino groups or
hydroxyl groups on the B-ring and do not have a substituent
(R2 = H) on the A-ring, exhibited similar or better inhibition of
Aβ aggregation (52.85%–71.69%, 20 µM) than resveratrol
(64.57%, 20 µM). Alkyl groups on the A-ring appear to be
unfavourable for inhibition, as illustrated by compound 10e,
which has a methyl group at the R1-position of the A-ring and
a hydroxyl at the R2-position, and showed only 49.38% inhi-
bition of Aβ aggregation at 20 µM. Compound series 13,
derived from 2-methyl-8-hydroxyquinoline and aromatic alde-
hydes, exhibited a similar inhibition compared with com-
pound series 10. For instance, compounds 13e and 13h–13i,
Scheme 2 Reagents and conditions: (a) aldehydes, Ac2O, Ar, 130 °C;
(b) HCl (37% water), DMF, 100 °C; (c) BBr3, CH2Cl2, −78 °C to r.t.
hydrochloric acid and provided 7. Compound 7 was reacted
with triethyl phosphite and then chloro(methoxy)methane to
give the Wittig reagent 8, which was then reacted with 2, 5,
amino-substituted aldehydes or 3,4,5-trimethoxybenzaldehyde
to provide 9. Finally, compounds 10a–10f were obtained by
deprotecting 9 in the presence of hydrochloric acid, and com-
pound 10g was obtained by the demethylation of compound
10f in the presence of boron tribromide at −78 °C (Scheme 1).
2-Methyl-8-hydroxyquinoline was treated with nitro-,
bromine-, hydroxyl- or methoxy-substituted benzaldehydes in
the presence of acetic anhydride, followed by deacetylation
with hydrochloric acid to provide 13a–13g. Compounds 13h
and 13i were obtained by the demethylation of the corres-
ponding dimethoxy compounds 13f and 13g in the presence of
boron tribromide at −78 °C (Scheme 2).
which have hydroxyl groups on the C-ring, showed
a
67.46–75.31% inhibition at 20 µM. The results also showed
that the inhibitory activity of para-hydroxyl substitutions at the
C-ring was better than hydroxyl substitutions at the ortho- or
meta-positions (the ability of compounds 13a, 13b and 13e to
inhibit Aβ aggregation was 49.21%, 45.87% and 67.46%,
Table 1 Inhibition of Aβ1–42 aggregation and oxygen radical absorbance capacity (ORAC, Trolox Equivalents) by resveratrol, CQ and 8-hydroxy-
quinoline-resveratrol derivatives 10a–10e, 10g, 13a–e and 13h–13i
Inhibition of Aβ1–42
Compounds
Substituents
aggregationa (%)
IC50 b (μM)
ORACd
10a
10b
10c
10d
10e
10g
13a
13b
13c
13d
13e
13h
13i
Resveratrol
CQ
R2 = NMe2, R1 = R3 = R4 = H
R2 = NEt2, R1 = R3 = R4 = H
R2 = OH, R1 = R3 = R4 = H
R1 = R2 = H, R3 = R4 = OH
R1 = Me, R2 = OH, R3 = R4 = H
R1 = R2 = R3 = OH
R5 = 2-OH
52.85 4.21
57.76 3.43
71.69 2.42
69.52 3.27
49.38 2.72
69.53 3.70
49.21 3.36
45.87 4.22
18.18 1.35
22.33 2.14
67.46 5.37
75.31 4.82
70.09 2.72
64.57 3.52
30.22 2.21
n.t.c
18.14 1.51
12.19 1.02
8.50 0.35
9.26 0.56
n.t.c
1.92 0.16
2.81 0.22
2.18 0.12
2.52 0.10
1.87 0.13
n.t.c
8.98 0.82
n.t.c
3.16 0.28
3.22 0.22
0.93 0.05
0.91 0.07
n.t.c
R5 = 3-OH
n.t.c
R5 = 4-Br
n.t.c
R5 = 4-NO2
n.t.c
R5 = 4-OH
9.17 0.52
8.72 1.02
9.32 1.22
15.10 1.4
n.t.c
R5 = 3,5-OH
n.t.c
R5 = 3,4,5-OH
—
n.t.c
5.73 0.27
0.61 0.05
2.42 0.21
—
—
8-Hydroxyquinoline
n.t.c
a The thioflavin-T fluorescence method was used. Values are expressed as means SD from at least two independent measurements. All the
values were obtained with 20 μM of the tested compounds. b The thioflavin-T fluorescence method was used. Values are expressed as mean SD
from at least two independent measurements. c n.t. = not tested. d The mean SD of three independent experiments. Data are expressed in μmol
of Trolox equivalent/μmol of tested compound.
5938 | Org. Biomol. Chem., 2014, 12, 5936–5944
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
respectively). Electron-withdrawing groups were unfavourable
for inhibition, as indicated by compounds 13c and 13d, which
feature a 4-nitro or 4-bromo substitution on the C-ring,
showing only a slight Aβ aggregation inhibition at 20 µM.
The complete dose–response curves of compounds with
greater than 50% inhibition were also evaluated, and the
results are shown in Table 1. The curves show that compounds
with para-hydroxyl, 3,5-dihydroxyl or 3,4,5-trihydroxyl substi-
tution at the B-/C-ring exhibited similar Aβ aggregation inhibi-
tory activity (IC50 value: 8.50 μM–9.32 μM). Among these
compounds, compound 10c most potently inhibited self-
mediated Aβ1–42 aggregation (10c: IC50 = 8.50 μM; resveratrol:
IC50 = 15.10 μM).
Antioxidant activity in vitro
The antioxidant activities of the synthesised compounds were
determined by measuring the oxygen radical absorbance
capacity of fluorescein (ORAC-FL)34,35 using a vitamin E ana-
logue, Trolox (6-hydroxy-2,5,7,8-tetramethyl chroman-2-
carboxylic acid), as a standard. The results in Table 1 indicate
that most compounds had potent oxygen radical absorbance
capacities, ranging from 1.8- to 3.2-fold of the value of Trolox,
except compounds 13c and 13d, which feature 4-nitro or
4-bromo substitution on the B-ring, and showed activity of
0.9-fold that of Trolox.
Inhibition of Aβ1–42 fibril formation monitored by
transmission electron microscopy (TEM)
Fig. 4 Top: Scheme of the disaggregation experiment. Bottom: Dis-
aggregation of self-induced Aβ1–42 aggregation fibrils by 10c, resveratrol
and CQ ([Aβ1–42] = [10c] = [resveratrol] = [CQ] = 25 μM, 37 °C, 24 h). (A)
Results of the ThT binding assay. Statistical comparisons with control
(Aβ1–42 alone) were performed by a one-way ANOVA followed by the
Dunnett’s multiple comparison test using GraphPad Prism 5.0 Software
(levels of significance **p < 0.01, ***p < 0.001); (B) TEM images of
To further examine the inhibitory ability of the 8-hydroxy-
quinoline-resveratrol derivatives on Aβ1–42 aggregation, we
selected 10c, which showed the most optimal Aβ1–42 aggrega-
tion inhibition potency (IC50 = 8.50 μM) in the TEM assay. Fol-
lowing incubation at 37 °C for 24 h, the sample of Aβ1–42 alone
aggregated into well-defined Aβ fibrils (Fig. 3b). In contrast,
only small bulk aggregates were visible and no characteristic
samples. (a) Aβ1–42, 0 h; (b) Aβ1–42 alone; (c) Aβ1–42 + 10c; (d) Aβ1–42
resveratrol; (e) Aβ1–42 + CQ.
+
fibrils were observed in the presence of compound 10c
(Fig. 3c) under the same experimental conditions, which
proved that 10c can inhibit Aβ1–42 fibrils formation.
Disaggregation of self-mediated Aβ1–42 aggregation fibrils by 10c
The disaggregation ability of 10c toward the self-mediated Aβ
aggregation of fibrils was assayed as follows. First, Aβ fibrils
were generated by incubating fresh Aβ (25 μM) for 24 h at
37 °C (Fig. 4b). Compound 10c (25 μM) was then added to the
sample before incubating for another 24 h at the same temp-
erature. The ThT binding assay and examination by TEM
(Fig. 4c) showed that 10c effectively disaggregates Aβ fibrils
(88.73% disaggregation at 25 µM as determined by the ThT
binding assay).
Fig. 3 Top: Scheme of the inhibition experiment. Bottom: TEM images
of the samples of Aβ1–42 (25 μM) in the presence or absence of com-
Metal-chelating properties of compound 10c
The ability of compound 10c to chelate biometals, such as
copper(II), iron(II) and zinc(II), was investigated by UV-vis
pounds (25 μM) (37 °C, 24 h). (a) Aβ1–42, 0 h; (b) Aβ1–42 alone; (c) Aβ1–42
10c; (d) Aβ1–42 + resveratrol; (e) Aβ1–42 + CQ.
+
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5936–5944 | 5939

View Article Online
Paper
Organic & Biomolecular Chemistry
Fig. 6 Determination of the stoichiometry of complex 10c–copper(II)
by the Job’s method.
straight lines, intersecting at a mole fraction of 0.32, indicating a
2 : 1 stoichiometry for complex 10c–copper(II).
The effects of compound 10c on copper(II)-induced Aβ1–42
aggregation
To investigate the effects of 10c on metal-induced Aβ aggrega-
tion, we examined the inhibitory activity of compound 10c on
copper(II)-induced Aβ1–42 aggregation and the disaggregation
effects of 10c on copper(II)-induced Aβ1–42 aggregation fibrils.
The ThT binding assay and transmission electron microscopy
(TEM) were used to identify the degree of Aβ aggregation.
The results in Fig. 7 indicate that copper(II) accelerates the
aggregation of Aβ. The fluorescence of Aβ treated with
copper(II) is 1.63 times that of Aβ alone, according to the ThT
binding assay. In contrast, the fluorescence of Aβ treated with
copper(II) and the test compounds decreased significantly
(fluorescence after treatment with 10c, resveratrol and CQ was
21.93%, 42.82% and 54.13%, respectively), which indicated
that these compounds could reduce the rate of Aβ aggregation
treated with copper(II). TEM images demonstrated that the Aβ
aggregates treated with copper(II) were significantly more
intense than the Aβ aggregates alone but the Aβ aggregates
obviously decreased when 10c, resveratrol and CQ were added
to the samples, which was consistent with the ThT binding
assay results.
Fig. 5 UV spectra of compound 10c (A) and resveratrol (B) (40 μM)
alone and in the presence of 20 μM CuSO4, FeSO4 or ZnCl2. All the solu-
tions were prepared using a HEPES buffer solution (20 mM HEPES,
150 mM NaCl, pH 7.4).
spectrometry,36 and the results are shown in Fig. 5A. The specific
absorbance peaks of 10c can be seen at 254 and 290 nm.
After copper(II) was added to the solution of 10c, specific
absorbance at 254 nm shifted to 260 nm, the peak at 290 nm
shifted to 294 nm, and a new peak was observed at 445 nm.
These changes in absorbance indicate the formation of a 10c–
copper(II) complex. Similarly, the specific absorption of the
10c–iron(II) complex was shifted to a maxima at 261 nm and
309 nm, and one new absorption peak at 537 nm was
observed. When the same experiment was performed using
10c and zinc(II) to form 10c–zinc(II), two specific absorptions
overlaid at 285 nm without the generation of new peaks. In
addition, we evaluated the chelating ability of resveratrol to
biometals, such as copper(II), iron(II) and zinc(II), by UV-vis
spectrometry. As shown in Fig. 5B, resveratrol interacts
Disaggregation of copper(II)-induced Aβ1–42 aggregation
fibrils by 10c
The disaggregation studies of copper(II)-induced Aβ1–42 aggre-
strongly with copper(II) but only weakly with iron(II) and shows gation fibrils by 10c are shown in Fig. 8. Compound 10c,
almost no interaction with zinc(II), which is in agreement with
the reported literature.26
resveratrol and CQ (50 µM) were added individually to Aβ
fibrils that were generated by reacting Aβ with 1 equiv. of
To determine the stoichiometry of the complex 10c–copper(II), copper(II) for 24 h at 37 °C under constant agitation. The ThT
a series of solutions that maintains the total concentrations of
compound 10c and CuSO4 constant but vary their proportions
were prepared according to the Job’s method.37,38 UV spec-
troscopy was used to determine the absorbance of the CuSO4
complex and 10c at different concentrations. As indicated in
Fig. 6, plotting the absorbance changes at 445 nm yielded two
binding assay indicated that all the compounds could disag-
gregate Aβ fibrils at 50 μM (percentage of disaggregation was
as follows: 10c, 84.46%; resveratrol, 82.83%; CQ, 43.13%).
Fig. 8c shows the visualisation of the Aβ species by TEM, and
clearly indicates that 10c was capable of disassembling the Aβ
fibrils resulting from copper(II)-induced aggregation.
5940 | Org. Biomol. Chem., 2014, 12, 5936–5944
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Fig. 7 Top: Scheme of the inhibition experiment. Bottom: Inhibition of
copper(II)-induced Aβ1–42 aggregation by 10c, resveratrol and CQ
([Aβ1–42] = copper(II) = 25 μM, [10c] = [resveratrol] = [CQ] = 50 μM,
37 °C, 24 h). (A) Results of the ThT binding assay. Statistical comparisons
with control (Aβ1–42 aggregation in the presence of copper(II)) were per-
formed by a one-way ANOVA followed by the Dunnett’s multiple com-
parison test using GraphPad Prism 5.0 Software (level of significance
Fig. 8 Top: Scheme of the disaggregation experiment. Bottom: Disag-
gregation of copper(II)-induced Aβ1–42 fibrils by 10c, resveratrol and CQ
([Aβ1–42] = copper(II) = 25 μM, [10c] = [resveratrol] = [CQ] = 50 μM,
37 °C, 24 h). (A) Results of the ThT binding assay, Statistical comparisons
with control (untreated with compound) were performed by a one-way
ANOVA followed by the Dunnett’s multiple comparison test using
GraphPad Prism 5.0 Software (level of significance, ***p < 0.001); (B)
TEM images of samples. (a) Aβ1–42, 0 h; (b) Aβ1–42 + copper(II); (c) Aβ1–42
***p < 0.001). (B) TEM images of samples. (a) Aβ1–42, 0 h; (b) Aβ1–42
+
copper(II); (c) Aβ1–42 + copper(II) + 10c; (d) Aβ1–42 + copper(II) + resvera-
trol; (e) Aβ1–42 + copper(II) + CQ; (f) Aβ1–42 alone.
+ copper(II) + 10c; (d) Aβ1–42 + copper(II) + resveratrol; (e) Aβ1–42
+
copper(II) + CQ.
Ability to halt copper redox cycling via metal complexation
As reported, redox-active Cu(II) is implicated in the generation
of ROS, leading to an increase in oxidative stress, which is a
proposed neuropathology of AD. Therefore, we utilised the Cu–
ascorbate redox system (Scheme 3), described by Faller and co-
workers, as a model to determine whether the ligand 10c
could halt copper based redox activity under aerobic con-
ditions.39 The production of hydroxyl radicals (OH•) by copper
redox cycling in the presence of ascorbate was measured by co-
incubation with coumarin-3-carboxylic acid (CCA), which gen-
Scheme 3 Production of hydroxyl radical (OH•) by copper redox
cycling in the presence of ascorbate.
Prediction of the BBB penetration and absorption of
compound 10c and 10d
erates fluorescent 7-hydroxy-CCA in the presence of OH•. As Brain penetration is an essential element for successful anti-
shown in Fig. 9, the OH• in the copper–ascorbate system AD drugs.40 To evaluate the ‘drug-like-quality’ and brain pene-
increases steadily with time and reaches a plateau at approxi- tration ability of the synthetic compounds, we used Lipinski’s
mately 12 min. When 10c was coincubated with the Cu–ascor- rules41 and calculated log BB using the equation log BB =
bate system, this process was prevented almost entirely, −0.0148 × PSA + 0.152 × c log P + 0.130 as a criteria. The results
indicating that the ligand is capable of halting copper redox in Table 2 indicate that compound 10c fulfils the necessary
cycling involved in oxidative stress via metal complexation.
drug-like and prediction of brain penetration criteria.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5936–5944 | 5941

View Article Online
Paper
Organic & Biomolecular Chemistry
Table 2 Physical properties and prediction of BBB penetration and
absorption of 10c and 10d
Calculationa
10c
10d
Lipinski’s rules
MW
C log P
HBA
263.29
3.52
3
279.29
3.28
4
≤450
≤5.0
≤10
HBD
2
3
≤5
PSA
52.89
−0.12
0
73.71
−0.46
0
≤90
≥−1.0
Log BBb
Absorption levelc
a MW: molecular weight; c log P: calculated logarithm of the octanol–
water partition coefficient; HBA: hydrogen-bond acceptor atoms; HBD:
hydrogen-bond donor atoms; PSA: polar surface area. b Compounds
with log BB > 0.3 are able to readily cross the BBB, compounds with
log BB < −1.0 are only poorly distributed to the brain. c Data are
predicted by discovery studio 2.1 soft (Accelrys). 0: good absorption, 1:
moderate absorption, 2: low absorption, 3: very low absorption.42
Fig. 9 Fluorescence intensity of the copper–ascorbate and copper–
ascorbate–10c system. CCA [50 μM] and ascorbate [150 μM] were incu-
bated in each system. [Cu(II)] = [5 μM], [10c] = 15 μM. PBS buffer,
pH = 7.4.
In vitro blood–brain barrier permeation assay
The parallel artificial membrane permeation assay (PAM-
PA-BBB) for predicting passive BBB permeation was also per-
formed for our compounds. This method developed by Di
et al.43 was capable of identifying compounds as either BBB
permeable or non-permeable with high success, high through-
put, and reproducibility. First, the in vitro permeability (Pe) of 13
quality control standards (with known BBB permeability) were
determined via a lipid extract of porcine brain using phosphate
buffered saline (PBS)–EtOH (70 : 30). The experimental data
versus bibliographic values showed a good linear correlation,
Pe(exp.) = 1.4574Pe(bibl.)–1.0773 (R2 = 0.9427) (see ESI Fig. S1†).
From this equation and by considering the limit established by
Di et al. for blood–brain barrier permeation, permeability values
of over 4.7 × 10−6 cm s−1 (PBS–EtOH, 70 : 30) were considered
sufficient for compounds to cross the BBB (see ESI Table S1†).
The selected compounds were then tested in the PAMPA-BBB
assay. The results (Table 3) showed that compound 10c could
cross BBB but resveratrol could not. Comparing the per-
meability values of 10c and 10d, it appears that increasing
number of hydroxyl groups are against crossing the BBB.
Table 3 Permeability results (Pe 10−6 cm s−1) from the PAMPA-BBB
assay for selected 8-hydroxyquinoline-resveratrol derivatives with their
predicted penetration into the CNS
Compounds
Permeability (Pe 10−6 cm s−1 a
)
Prediction
10c
10d
Resveratrol
14.7 1.4
4.8 0.4
1.05 0.05
CNS+
CNS+
CNS−
a Values are expressed as mean
experiments.
SD of three independent
Conclusion
Based on the knowledge that resveratrol possesses a variety of
bioactivities, we used a fusing strategy to develop a novel series
of compounds containing the pharmacophores of resveratrol
and clioquinol as multi-target agents for the treatment of AD.
Among the synthesised compounds, 10c exhibited good inhi-
bition of self-induced Aβ aggregation and the ability to dis-
assemble the well-structured Aβ fibrils generated by self-induced
Aβ aggregation, both of which can be attributed to the features
of resveratrol. In addition, as a result of the 8-hydroxyquinoline
moiety, 10c showed good metal-chelating ability and the ability
to inhibit copper(II)-induced Aβ aggregation and disassemble Aβ
fibrils generated by copper(II)-induced Aβ aggregation. Further-
more, 10c exhibited very potent antioxidant activity compared to
Trolox. When tested for toxicity in mice, 10c showed no acute
toxicity at doses of up to 2000 mg kg−1. Moreover, according to a
parallel artificial membrane permeation assay, 10c can cross the
blood–brain barrier (BBB). In summary, these results show that
compound 10c is worthy of further studies directed toward the
development of novel drugs for the treatment of AD.
Acute toxicity of compound 10c
To investigate the safety profile of 8-hydroxyquinoline-resvera-
trol derivatives, the acute toxicity of 10c was determined in KM
mice at doses of 0, 677, 1333, and 2000 mg kg−1 (each group,
n = 5) by oral administration. No signs of acute toxicity were
observed at these doses within the first 4 h after adminis-
tration, and no animal died in the following 14 days. All the
mice were sacrificed on the 14th day of drug administration
and were examined macroscopically for any possible damage
to the heart, liver, and kidneys. The results showed that the
animals treated with compound 10c did not show any acute
toxicity and mortality, either immediately or during the post-
treatment period. Furthermore, no significant abnormal
changes were observed during the experimental period in
terms of water or food consumption or body weight. Therefore,
Acknowledgements
compound 10c was proven to be non-toxic and well-tolerated We thank the National Natural Science Foundation of China
at doses of up to 2000 mg kg−1
.
(no. 21302235, 20972198), Ph.D. Programs Foundation of
5942 | Org. Biomol. Chem., 2014, 12, 5936–5944
This journal is © The Royal Society of Chemistry 2014

View Article Online
Organic & Biomolecular Chemistry
Paper
Ministry of Education of China (20120171120045), and Distin- 21 K. Beking and A. Vieira, Public Health Nutr., 2010, 13, 1403–
guished Young Talents in Higher Education of Guangdong
(2012LYM_003) for financially supporting this study.
1409.
22 H. P. Lee, G. Casadesus, X. Zhu, H. G. Lee, G. Perry,
M. A. Smith, K. Gustaw-Rothenberg and A. Lerner, Expert
Rev. Neurother., 2009, 9, 1615–1621.
23 H. Y. Zhang, D. P. Yang and G. Y. Tang, Drug Discovery
Today, 2006, 11, 749–754.
Notes and references
24 D. J. Boocock, G. E. Faust, K. R. Patel, A. M. Schinas,
V. A. Brown, M. P. Ducharme, T. D. Booth, J. A. Crowell,
M. Perloff, A. J. Gescher, W. P. Steward and
D. E. Brenner, Cancer Epidemiol. Biomarkers Prev., 2007,
16, 1246–1252.
25 A. R. Ladiwala, J. C. Lin, S. S. Bale, A. M. Marcelino-Cruz,
M. Bhattacharya, J. S. Dordick and P. M. Tessier, J. Biol.
Chem., 2010, 285, 24228–24237.
1 N. C. Berchtold and C. W. Cotman, Neurobiol. Aging, 1998,
19, 173–189.
2 J. Hardy, Neuron, 2006, 52, 3–13.
3 J. Lageé and M. Manuelínez, J. Alzheimer’s Dis., 2006, 9,
15–26.
4 http://www.alz.co.uk/research/files/WorldAlzheimerReport.
World Alzheimer Report. 2009, Alzheimer’s disease
international.
26 A. Granzotto and P. Zatta, PLoS One, 2011, 6, e21565.
5 G. Pepeu and M. G. Giovannini, Curr. Alzheimer Res., 2009, 27 I. M. Kapetanovic, M. Muzzio, Z. Huang, T. N. Thompson
6, 86–96.
and D. L. McCormick, Cancer Chemother. Pharmacol., 2011,
68, 593–601.
6 M. Villarroya, A. G. Garcia, J. Marco-Contelles and
M. G. Lopez, Expert Opin. Invest Drugs, 2007, 16, 1987–1998. 28 J. Geng, M. Li, L. Wu, J. Ren and X. Qu, J. Med. Chem.,
7 E. Scarpini, P. Scheltens and H. Feldman, Lancet Neurol.,
2003, 2, 539–547.
8 M. B. H. Youdim and J. J. Buccafusco, Trends Pharmacol.
Sci., 2005, 26, 27–35.
2012, 55, 9146–9155.
29 M. G. Savelieff, Y. Liu, R. R. Senthamarai, K. J. Korshavn,
H. J. Lee, A. Ramamoorthy and M. H. Lim, Chem.
Commun., 2014, 50, 5301–5303.
9 A. Cavalli, M. L. Bolognesi, A. Minarini, M. Rosini, 30 A. Kochi, T. J. Eckroat, K. D. Green, A. S. Mayhoub,
V. Tumiatti, M. Recanatini and C. Melchiorre, J. Med.
Chem., 2008, 51, 347–372.
M. H. Lim and S. Garneau-Tsodikova, Chem. Sci., 2013, 4,
4137–4145.
10 J. Hardy and D. J. Selkoe, Science, 2002, 297, 353–356.
11 P. A. Novick, D. H. Lopes, K. M. Branson, A. Esteras-Chopo,
I. A. Graef, G. Bitan and V. S. Pande, J. Med. Chem., 2012,
55, 3002–3010.
31 S. Lee, X. Zheng, J. Krishnamoorthy, M. G. Savelieff,
H. M. Park, J. R. Brender, J. H. Kim, J. S. Derrick,
A. Kochi, H. J. Lee, C. Kim, A. Ramamoorthy,
M. T. Bowers and M. H. Lim, J. Am. Chem. Soc., 2014,
136, 299–310.
12 R. Di Santo, R. Costi, G. Cuzzucoli Crucitti, L. Pescatori,
F. Rosi, L. Scipione, D. Celona, M. Vertechy, O. Ghirardi, 32 K. M. Lincoln, P. Gonzalez, T. E. Richardson, D. A. Julovich,
P. Piovesan, M. Marzi, S. Caccia, G. Guiso, F. Giorgi and
P. Minetti, J. Med. Chem., 2012, 55, 8538–8548.
R. Saunders, J. W. Simpkins and K. N. Green, Chem.
Commun., 2013, 49, 2712–2714.
13 M. Catto, R. Aliano, A. Carotti, S. Cellamare, F. Palluotto, 33 M. Rosini, E. Simoni, M. Bartolini, A. Cavalli, L. Ceccarini,
R. Purgatorio, A. De Stradis and F. Campagna, Eur. J. Med.
Chem., 2010, 45, 1359–1366.
14 A. Kumar, L. Moody, J. F. Olaivar, N. A. Lewis, R. L. Khade,
N. Pascu, D. W. McClymont, A. Tarozzi, M. L. Bolognesi,
A. Minarini, V. Tumiatti, V. Andrisano, I. R. Mellor and
C. Melchiorre, J. Med. Chem., 2008, 51, 4381–4384.
A. A. Holder, Y. Zhang and V. Rangachari, ACS Chem. Neuro- 34 B. Ou, M. Hampsch-Woodill and R. L. Prior, J. Agric. Food
sci., 2010, 1, 691–701. Chem., 2001, 49, 4619–4626.
15 J. Dong, C. S. Atwood, V. E. Anderson, S. L. Siedlak, 35 A. Dávalos, C. Gómez-Cordovés and B. Bartolomé, J. Agric.
M. A. Smith, G. Perry and P. R. Carey, Biochemistry, 2003,
42, 2768–2773.
16 T. A. Bayer, S. Schafer, A. Simons, A. Kemmling, T. Kamer,
R. Tepest, A. Eckert, K. Schussel, O. Eikenberg,
Food Chem., 2003, 52, 48–54.
36 J.-S. Choi, J. J. Braymer, R. P. R. Nanga, A. Ramamoorthy
and M. H. Lim, Proc. Natl. Acad. Sci. U. S. A., 2010, 107,
21990–21995.
C. Sturchler-Pierrat, D. Abramowski, M. Staufenbiel and 37 S. S. Hindo, A. M. Mancino, J. J. Braymer, Y. Liu,
G. Multhaup, Proc. Natl. Acad. Sci. U. S. A., 2003, 100,
14187–14192.
S. Vivekanandan, A. Ramamoorthy and M. H. Lim, J. Am.
Chem. Soc., 2009, 131, 16663–16665.
17 A. I. Bush, J. Alzheimer’s Dis., 2008, 15, 223–240.
18 T. Amit, Y. Avramovich-Tirosh, M. B. H. Youdim and
S. Mandel, FASEB J., 2007, 22, 1296–1305.
38 M. a. I. Fernández-Bachiller, C. n. Pérez, G. C. González-
Muñoz, S. Conde, M. G. López, M. Villarroya, A. G. García
and M. a. I. Rodríguez-Franco, J. Med. Chem., 2010, 53,
4927–4937.
19 D. J. Bonda, X. Wang, G. Perry, A. Nunomura, M. Tabaton,
X. Zhu and M. A. Smith, Neuropharmacology, 2010, 59, 290– 39 L. Guilloreau, S. Combalbert, A. Sournia-Saquet,
294.
H. Mazarguil and P. Faller, ChemBioChem, 2007, 8, 1317–
20 T. M. Bray, Proc. Soc. Exp. Biol. Med., 1999, 222, 195.
1325.
This journal is © The Royal Society of Chemistry 2014
Org. Biomol. Chem., 2014, 12, 5936–5944 | 5943

View Article Online
Paper
Organic & Biomolecular Chemistry
40 H. van de Waterbeemd and E. Gifford, Nat. Rev. Drug Dis- 42 W. J. Egan, K. M. Merz and J. J. Baldwin, J. Med. Chem.,
covery, 2003, 2, 192–204. 2000, 43, 3867–3877.
41 D. E. Clark and S. D. Pickett, Drug Discovery Today, 2000, 5, 43 L. Di, E. H. Kerns, K. Fan, O. J. McConnell and G. T. Carter,
49–58.
Eur. J. Med. Chem., 2003, 38, 223–232.
5944 | Org. Biomol. Chem., 2014, 12, 5936–5944
This journal is © The Royal Society of Chemistry 2014