Toxicological and pharmacological evaluation, antioxidant, ADMET and molecular modeling of selected racemic chromenotacrines {11-amino-12-aryl-8,9, 10,12-tetrahydro-7H-chromeno[2,3-b]quinolin-3-ols} for the potential prevention and treatment of Alzheimer's disease

Article abstract of DOI:10.1016/j.ejmech.2013.12.021

The pharmacological analysis of racemic chromenotacrines (CT) 1-7, bearing the 11-amino-12-aryl-8,9,10,12-tetrahydro-7H-chromeno[2,3-b]quinolin-3-ol ring skeleton, in a series of experiments targeted to explore their potential use for the treatment of Alzheimer's disease (AD), is reported. The toxicological evaluation showed that among all these chromenotacrines, CT6 is much less hepatotoxic than tacrine in a range of concentrations from 1 to 300 μM, measured as cell viability in HepG2 cells. Moreover, CT6 did not significantly increase lactate dehydrogenase, aspartate transaminase, and alanine transaminase release in HepG2 cells. Besides, CT6 treatment exerts a high protective effect against the lipid peroxidation induced after H₂O₂-treated SH-SY5Y cells, in a concentration-dependent manner. CT6 showed an excellent antioxidant profile in the AAPH test, and protects against the decrease in cell viability induced by respiratory chain inhibitors (Oligomicyn A/Rotenone) and NO donors in neuronal cultures. This effect could be due to a mixed antiapoptotic and antinecrotic neuroprotective effect at low and intermediate CT6 concentrations, respectively. CT1-7 are potent and selective inhibitors of EeAChE in the submicromolar range. CT3 [IC₅₀ (EeAChE) = 0.007 ± 0.003 μM], and CT6 [IC₅₀ (EeAChE) = 0.041 ± 0.001 μM] are the most potent AChE inhibitors. Kinetic studies on the non-toxic chromenotacrine CT6 showed that this compound behaves as a non-competitive inhibitor (Ki = 0.047 ± 0.003 μM), indicating that CT6 binds at the peripheral anionic site, a fact confirmed by molecular modeling analysis. In silico ADMET analysis showed also that CT6 should have a moderate BBB permeability. Consequently, non-toxic chromenotacrine CT6 can be considered as an attractive multipotent molecule for the potential treatment of AD.

Full text of DOI:10.1016/j.ejmech.2013.12.021

European Journal of Medicinal Chemistry 74 (2014) 491e501
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Short communication
Toxicological and pharmacological evaluation, antioxidant, ADMET
and molecular modeling of selected racemic chromenotacrines {11-
amino-12-aryl-8,9,10,12-tetrahydro-7H-chromeno[2,3-b]quinolin-3-
ols} for the potential prevention and treatment of Alzheimer’s disease
,
b
,
a
a
María Jesús Oset-Gasque a , María Pilar González , Javier Pérez-Peña ,
Nuria García-Font a, Alejandro Romero c, Javier del Pino c, Eva Ramos c,
Dimitra Hadjipavlou-Litina d, Elena Soriano e, Mourad Chioua f, Abdelouahid Samadi f,
*
Dushyant S. Raghuvanshi g, Krishna N. Singh g , José Marco-Contelles
,
f,b,
**
***
a Departamento de Bioquímica y Biología Molecular II, Facultad de Farmacia, Universidad Complutense de Madrid (UCM), 28040 Madrid, Spain
b Instituto Universitario de Investigación en Neuroquímica (IUIN), Universidad Complutense de Madrid (UCM), 28040 Madrid, Spain
c Departamento de Toxicología y Farmacología, Facultad de Veterinaria, Universidad Complutense de Madrid, 28040 Madrid, Spain
d Department of Pharmaceutical Chemistry, School of Pharmacy, Aristotle University of Thessaloniki, Thessaloniki 54124, Greece
e SEPCO (IQOG, CSIC), Juan de la Cierva, 3, 28006 Madrid, Spain
f Laboratorio de Química Médica (IQOG, CSIC), C/ Juan de la Cierva 3, 28006 Madrid, Spain
g Department of Chemistry, Centre of Advanced Study, Faculty of Science, Banaras Hindu University, Varanasi 221 005, India
a r t i c l e i n f o
a b s t r a c t
Article history:
The pharmacological analysis of racemic chromenotacrines (CT) 1e7, bearing the 11-amino-12-aryl-8,9,10,12-
tetrahydro-7H-chromeno[2,3-b]quinolin-3-ol ring skeleton, in a series of experiments targeted to explore
their potential use for the treatment of Alzheimer’s disease (AD), is reported. The toxicological evaluation
showed that among all these chromenotacrines, CT6 is much less hepatotoxic than tacrine in a range of con-
Received 3 September 2013
Received in revised form
9 December 2013
Accepted 19 December 2013
Available online 8 January 2014
centrations from 1 to 300 mM, measured as cell viability in HepG2 cells. Moreover, CT6 did not significantly
increase lactate dehydrogenase, aspartate transaminase, and alanine transaminase release in HepG2 cells. Be-
sides, CT6 treatmentexerts a high protective effect against the lipid peroxidation induced after H2O2-treated SH-
SY5Y cells, in a concentration-dependent manner. CT6 showed an excellent antioxidant profile in the AAPH test,
and protects against the decrease in cell viability induced by respiratory chain inhibitors (Oligomicyn A/Rote-
none) and NO donors in neuronal cultures. This effect could be due to a mixed antiapoptotic and antinecrotic
neuroprotective effect at low and intermediate CT6 concentrations, respectively. CT1-7 are potent and selective
Keywords:
11-Amino-12-aryl-8,9,10,12-tetrahydro-7H-
chromeno[2,3-b]quinolin-3-ols
Tacrine analogs
Toxicity
EeAChE
hBuChE
inhibitors of EeAChE in the submicromolar range. CT3 [IC50 (EeAChE) ¼ 0.007 ꢀ 0.003
(EeAChE) ¼ 0.041 ꢀ 0.001 M] are the most potent AChE inhibitors. Kinetic studies on the non-toxic chrome-
M),
indicating that CT6 binds at the peripheral anionic site, a fact confirmed by molecular modeling analysis. In silico
ADMET analysis showed also that CT6 should have a moderate BBB permeability. Consequently, non-toxic
chromenotacrine CT6 can be considered as an attractive multipotent molecule for the potential treatment of AD.
Ó 2014 Elsevier Masson SAS. All rights reserved.
mM], and CT6 [IC50
m
Kinetic analysis
Inhibition mechanism
Antioxidant
Neuroprotection
Molecular modeling
ADMET
notacrine CT6 showed that this compound behaves as a non-competitive inhibitor (Ki ¼ 0.047 ꢀ 0.003
m
Alzheimer’s disease
Abbreviations: ACh, acetylcholinesterase; AChEI, acetylcholinesterase inhibitor; AD, Alzheimer’s disease; ADMET, absorption, distribution, metabolism, excretion, and
toxicity; BBB, bloodebrain barrier; CSF, cerebrospinal fluid; CNS, central nervous system; CT, chromenotacrine; HIA, human intestinal absorption; log P, calculated logarithm
of the octanol/water partition coefficient; MTDL, multi-target-directed-ligand approach; MW, molecular weight; RB, rotatable bond; ROS, radical oxygenated species; TPSA,
topological polar surface area.
* Corresponding author. Departamento de Bioquímica y Biología Molecular II, Facultad de Farmacia, Universidad Complutense de Madrid (UCM), 28040 Madrid, Spain.
Fax: þ34 91 3941779.
** Corresponding author. Laboratorio de Química Médica (IQOG, CSIC), C/Juan de la Cierva 3, 28006 Madrid, Spain. Fax: þ34 91 5644853.
*** Corresponding author. Department of Chemistry, Centre of Advanced Study, Faculty of Science, Banaras Hindu University, Varanasi 221 005, India. Fax: þ91 542 2368127.
E-mail addresses: mjoset@ucm.es (M.J. Oset-Gasque), knsinghbhu@yahoo.co.in (K.N. Singh), iqoc21@iqog.csic.es (J. Marco-Contelles).
0223-5234/$ e see front matter Ó 2014 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.12.021

492
M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
1. Introduction
than tacrine, but they blocked voltage-dependent Ca2þ channels
[15].
With these precedents, and in order to validate the proposed
binding mode, modify the structure of the previous active com-
pounds, and looking for more equipotent AChE vs BuChE inhibitory
activities, and increased antioxidant capacities, we reported the
synthesis, the biological evaluation, and molecular modeling of a
Alzheimer’s disease (AD) is an age-related neurodegenerative
process characterized by diverse cognitive impairments [1].
Although the etiology of AD is not yet understood, several factors
such as amyloid-
b (Ab) deposits [2], s-protein aggregation, and
deficits of acetylcholine (ACh) play significant roles [3]. The
cholinergic theory suggests that the selective loss of cholinergic
neurons in AD results in low levels of ACh in specific regions of the
brain that mediate learning and memory functions [4]. Acetylcho-
linesterase inhibitors (AChEI), such as tacrine [5], donepezil, riva-
stigmine, and galanthamine, are well known to improve AD
symptoms by inhibiting AChE rising the levels of ACh in the syn-
aptic cleft. Accordingly, these molecules have been approved for
commercial use, but with limited success [6], mainly due to the
multifactorial nature of AD [7,8].
New therapeutic strategies, such as the Multi-Target-Directed-
Ligand approach (MTDL) [9e13], are sought. In this context,
some years ago we embarked in a project targeted at the synthesis
of a series of multipotent compounds designed to target AChE
inhibition and neuronal Ca2þ modulation [14]. Thus, we have
synthesized and evaluated a number of pyridotacrine (I) and
pyranotacrine (II) derivatives, which combine the tetrahy-
droaminoquinoline moiety present in tacrine with a pyridine or a
4H-pyran ring system (Fig. 1), respectively, a substitution pattern
similar to that found in the isosteric 1,4-dihydropyridines [15]. In
general, compounds of type I and II were less potent as AChEIs
number of racemic
10,11,12,14-tetrahydro-9H-benzo
b
-naphthotacrines (III) bearing the 14-aryl-
[5,6]chromeno[2,3-b]quinolin-
13-amine ring skeleton (Fig. 1) [16]. From this work, compound 4-
(13-amino-10,11,12,14-tetrahydro-9H-benzo [5,6]chromeno[2,3-b]
quinolin-14-yl)phenol emerged as a selective, potent and mixed
type EeAChE (Electrophorus electricus acetylcholinesterase) inhibi-
tor (IC50 ¼ 7 ꢀ 2 nM), 4-fold more active than tacrine, but unable to
displace propidium iodide, suggesting that the inhibitor does not
strongly bind to the PAS of AChE, an observation confirmed later by
docking, molecular dynamic simulations, and MM-GBSA calcula-
tions [16]. More recently, and based on these results, we have
described the synthesis and pharmacological evaluation of a
number of racemic, isomeric a-naphthotacrines (IV) bearing the 7-
aryl-9,10,11,12-tetrahydro-7H-benzo [7,8]chromeno[2,3-b]quino-
lin-8-amine structure (Fig. 1) [17]. We hypothesized that changing
the location of the fused benzene ring A, on going from compounds
of type III to IV, should have consequences in the binding and,
consequently, in the AChE inhibition power. From this research we
identified 4-(8-amino-9,10,11,12-tetrahydro-7H-benzo [7,8]chro-
meno[2,3-b]quinolin-7-yl)-2-methoxyphenol as a very potent and
selective
hAChE
(human
acetylcholinesterase)
inhibitor
(IC50 ¼ 0.33 ꢀ 0.04
mM) with antioxidant capacity (1.5 ꢀ 0.1 Trolox
equivalents) [17].
Now, in this context and in the present work, we have focussed
our interest in the related 12-aryl-8,9,10,12-tetrahydro-7H-chro-
meno[2,3-b]quinolin-11-amines (V) [18] (Fig. 1), and report here
the toxicological analysis on hepatoma cells, and the inhibition of
EeAChE/hBuChE by a series of selected compounds from this family
such as the chromenotacrines CT1-7 (Fig. 2). From this study we
have
identified
chromenotacrine
11-amino-12-(3,4,5-
trimethoxyphenyl)-8,9,10,12-tetrahydro-7H-chromeno[2,3-b]qui-
nolin-3-ol (CT6) (Fig. 2), as non-toxic in human liver cells, potent
and selective EeAChE inhibitor showing antioxidant and neuro-
protective properties.
2. Results and discussion
2.1. Chemistry
Racemic CT1-7 (Fig. 2) has been previously synthesized by one
of us by Friedländer-type reaction of 2-amino-3-cyano-4H-chro-
menes and cyclohexanone in the presence of aluminum chloride,
under controlled microwave irradiation [18]. CT1-5 and 7 bear
different type of substituents at C40 in the aromatic ring at C12,
while CT6 is a polysubstituted derivative with the three methoxy
groups located at C30, C40 and C50.
The pyranopyridine ring motif embedded in CT1-7 is present in
a number of heterocyclic compounds showing anti-allergic, anti-
inflammatory, and estrogenic activities [19]. Among the different
substitution patterns known, benzopyrano[2,3-b]pyridines are
known to exhibit anti-proliferative [20], cancer chemopreventive
[21], anti-bacterial [22], antimyopic [23], anti-hystaminic [24], and
hypotensive [25] activities, but to the best of our knowledge their
potential as tacrine analogs for AD has never been addressed.
Consequently, and based on the easy availability of molecules of
this type, we selected chromenotacrines CT1-7 (Fig. 2) to investi-
gate their pharmacological profile starting with the toxicological
and cholinergic properties.
Fig. 1. Structures of tacrine, pyridotacrines (I) and pyranotacrines (II), developed in
Macro-Contelles’ laboratory [14,15]; tacrine analogs bearing the 14-aryl-10,11,12,14-
tetrahydro-9H-benzo [5,6]chromeno[2,3-b] quinolon-13-amine (III) [16], and 7-aryl-
9,10,11,12-tetrahydro-7H-benzo [7,8]chromeno[2,3-b]quinolin-8-amine (IV) [17] ring
skeletons, and the 12-aryl-8,9,10,12-tetrahydro-7H-chromeno[2,3-b]quinolin-11-
amines (V) described in this work.

M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
493
Fig. 2. Structure of the chromenotacrines CT1-7 analyzed in this work, and 7-METO (8).
2.2. Pharmacology and Biochemistry
whereas tacrine treatment significantly increased cell death in a
concentration-dependent manner. This biochemical test in HepG2
proves that CT6 is less hepatotoxic than tacrine, as showed in the
cell viability study.
2.2.1. Toxicological studies of CT1-7
The human hepatoma cell line HepG2 is the best-characterized
and most frequently used cell line respect to hepatotoxic endpoints
and has been used to test metabolism and liver toxicity of several
drugs [26]. The evaluation results, using tacrine [27] as a reference
compound, are summarized in Table 1. After a 24 h incubation
period, at increasing concentrations of each compound tested (1e
2.2.2. AChE/BuChE inhibitory activity
CT1-7 were evaluated as inhibitors of EeAChE and horse serum
BuChE according to Ellman’s protocol [31]. The observed IC50 values
for EeAChE inhibition are shown in Table 3.
As shown, CT1-7 are potent EeAChE inhibitors (Table 3), in the
low micromolar range, and very selective, as no inhibition was
observed for eqBuChE. The most potent chromenotacrine corre-
300
mM), a decreased concentration dependent pattern was
observed in the cell viability. Very interestingly, CT6 was not toxic
at any concentration assayed. The presence of three methoxy
groups in the aromatic ring at C12, makes CT6 a very potent
electron-donor polysubstituted, non-toxic chromenotacrine, a fact
that clearly confirms that the non-toxic effects of tacrine analogs
are associated to the substitution with a methoxy electron-donor
substituent, as it is the case of non-toxic tacrine analog 7-MEOTA
(8) [28] (Fig. 2). Not surprisingly, CT7 and CT4, in this order,
bearing the good N-dimethylamino and methoxy electron-donor
groups, respectively at C40 position, were the less toxic chrome-
notacrines investigated here. Finally, note that electron-
withdrawing substituted chromenotacrines at C40, such as CT3,
bearing a bromine atom at this position, show less potent cell-
viability activity power.
On the other hand, it is known that tacrine has a limited ther-
apeutic window due to the hepatotoxicity associated with the
release of aminotransferases [29] and lactate dehydrogenase (LDH)
from liver [30]. In order to determine aspartate aminotransferase
(AST), alanine aminotransferase (ALT) and LDH levels, we incubated
the less toxic chromenotacrine CT6 and tacrine in HepG2 cells for
24 h. The effects of CT6 at the three concentrations on these en-
sponded to toxic CT3 (IC50 ¼ 0.007 ꢀ 0.003
m
M), but the non-toxic
inhibitory profile
CT6 also showed good AChE
a
(IC50 ¼ 0.041 ꢀ 0.001). However, from the structureeactivity rela-
tionship perspective, no clear trends could be withdrawn from
these data. Particularly interesting is the observed effect of a fluo-
rine atom at C40 compared with bromine at the same position, as
CT2 is 14.7-fold less active than CT3. Similarly, CT5, bearing a
modest methyl electron-donor group, is 8.1-fold more active than
CT4, bearing only one methoxy group, and in the same range that
Table 2
Aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels in
HepG2 cells treated with CT6 and tacrine.
Compounds (
m
M)
AST (IU/L)
ALT (IU/L)
Control
50.3 ꢀ 2.90
60.3 ꢀ 4.10ns
83.3 ꢀ 3.18**
118.7 ꢀ 8.41***
53.0 ꢀ 2.64ns
66.0 ꢀ 2.52ns
78.3 ꢀ 2.40**
11.7 ꢀ 0.88
Tacrine (10)
Tacrine (30)
Tacrine (100)
CT6 (10)
CT6 (30)
CT6 (100)
12.7 ꢀ 0.89ns
42.7 ꢀ 3.93***
68.3 ꢀ 1.76***
13.0 ꢀ 2.00ns
16.7 ꢀ 1.86ns
20.3 ꢀ 2.33ns
zymes were not significant, except at the dose of 100
However, tacrine induced a statistically significant increase of ALT
and AST at 30 and 100 M respect to control group.
mM (Table 2).
m
Data are expressed as the means ꢀ SEM of four independent experiments of
Furthermore, we carried out the LDH test (Fig. 3), and com-
pound CT6 did not show LDH release at any concentration assayed,
different cultures. All compounds were assayed at increasing concentrations (10, 30
and 300
mM). ***P < 0.01, **P < 0.01, ns non significant, with respect to control.
Statistical analysis: one-way ANOVA followed by the NewmaneKeuls post-hoc test.
Table 1
In vitro toxicity of compounds C1e7 and tacrine, in HepG2 cells.
CT
Viability (%) HepG2 cells
1
m
M
3
m
M
10
m
M
30
mM
100
m
M
300 mM
1
2
3
4
5
6
7
99.5 ꢀ 1.49ns
91 ꢀ 4.00ns
99 ꢀ 3.51ns
97.3 ꢀ 5.03ns
96.9 ꢀ 2.55ns
67.4 ꢀ 3.59***
75.1 ꢀ 4.49**
87.4 ꢀ 2.05ns
96.8 ꢀ 3.98ns
66.2 ꢀ 2.04***
62.8 ꢀ 3.45***
84.8 ꢀ 2.60*
94.1 ꢀ 3.37ns
85.6 ꢀ 1.83ns
100.7 0.66ns
86.1 ꢀ 2.37*
88.7 ꢀ 3.42ns
62.4 ꢀ 1.41***
55.7 ꢀ 1.81***
76.9 ꢀ 2.06***
95 ꢀ 1.63ns
55.8 ꢀ 2.29***
55.5 ꢀ 2.02***
74.4 ꢀ 2.86***
85.9 ꢀ 2.44ns
71.7 ꢀ 3.35**
98.3 0.68ns
85.5 ꢀ 0.82**
64.3 ꢀ 4.54***
49.6 ꢀ 3.58***
50.1 ꢀ 3.22***
61.5 ꢀ 1.99***
75.5 ꢀ 2.97**
47.9 ꢀ 3.88***
102.6 0.68ns
80.3 ꢀ 2.44***
40 ꢀ 2.20***
89.48 ꢀ 4.55ns
98.7 1.09ns
88.5 ꢀ 3.22*
90 ꢀ 2.95ns
74.3 ꢀ 2.26**
100.4 0.99ns
85.5 ꢀ 1.60**
81.6 ꢀ 4.88*
101.9 1.42ns
99 ꢀ 0.48ns
Tacrine
93.4 ꢀ 4.69ns
Data are expressed as the means ꢀ s.e.m. of quadruplicate cell viability measurements in at least four different cultures. All compounds were assayed at increasing con-
centrations (1e300 M). ***P < 0.01, **P < 0.01, *P < 0.05, ns non significant, with respect to control.
m

494
M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
Fig. 3. LDH release in HepG2 cells exposed to compound CT6 and tacrine. Bars show necrotic cell death, measured as % of LDH release, induced by indicated concentrations of
tacrine (A) or CT6 (B). Data represent the mean ꢀ SEM of four independent experiments in triplicate. n.s (non significant).*p < 0.05; **p < 0.01 and ***p < 0.001, compared to
control.
CT6, bearing three methoxy groups. Finally, note that the parent
CT1 and the methoxy substituted CT4 showed a very similar inhi-
bition power.
2.2.4. Antioxidant activity of CT6
2.2.4.1. Lipid peroxidation
Malonyldialdehyde (MDA) is one of the most important in-
termediates produced during lipid peroxidation. Incubation with
inhibitory
effect
of
CT6.
H2O2 (300 mM) (Fig. 5) in SH-SY5Y cells for 8 h induced a significant
2.2.3. Kinetics of EeAChE inhibition by CT6
increase in MDA levels (8-fold above basal) compared to control
group. As shown, CT6 provided a significant decrease of MDA levels
in a concentration-dependent manner.
Lipid peroxidation induced by the water soluble azo compound
2,20-azobis-2-methyl-propanimidamide dihydrochloride (AAPH)
has been used as a clean and controllable source of alkylperoxyl
free radicals. In our study, AAPH was used as a free radical initiator
to follow oxidative changes of linoleic acid to conjugated diene
hydroperoxides. CT6 showed significant inhibition on lipid perox-
In order to get a deeper insight in the mechanism of the inhi-
bition of these chromenotacrines, the most non-toxic and potent
EeAChEI CT6 was selected for the analysis of the kinetic of the
inhibition.
Graphical analysis of the reciprocal Lineweaver Burk plots
(Fig. 4A) showed increased slopes (decreased Vmax) but not in-
tercepts (similar Km) at increasing concentration of the inhibitor.
This pattern indicates a non-competitive-type inhibition. This was
confirmed by statistical analysis of the Km and Vmax modifications at
different concentrations of CT6, which indicate that Km’s does not
significantly change while Vmax decreased in a statistically signifi-
cant way in a dose-dependent manner (Table 4). Replots of the 1/V
versus concentration of CT6 (Dixon analysis) gave an estimate of
idation (LPO) at 100 mM (Table 5) compared to Trolox, used as a
standard (63%). Recent findings [33] show that the activation of
brain lipoxygenases (LOX) is an early event in the pathogenesis of
AD. LOXs play a role in membrane lipid peroxidation by forming
hydroperoxides in the lipid bilayer. Inhibition of soybean LOX was
performed by the UV absorbance based enzyme assay [34]. Un-
fortunately, CT6 showed very low inhibition in comparison to the
reference compound nordihydroguaiaretic acid (NDGA).
the inhibition constant, Ki ¼ 0.047 ꢀ 0.003
mM (Fig. 4B), a con-
centration very close to IC50 value. These data indicate that CT6 is
bound to the enzyme in a different site than the substrate, such as
the peripheral anionic site (PAS) [32]. These results are in good
agreement with data deduced from the computational docking
analysis (see below Section 2.2.6).
Next, and in order to know whether CT6 was able to revert the
ROS formation in cortical neurons in culture, two toxic stimuli, so-
dium nitroprussiate (SNP) and oligomycin-A 10
30 M (OR), were evaluated.
m
M þ rotenone
The general structure of CT6 bearing a phenol group at C3,
clearly suggests the possibility that this compound may act as a
good radical scavenger, and consequently, able to trap radical
oxygenated species (ROS), that are known to play an important role
in the progress and development of AD. Accordingly, next, a series
of experiments have been designed to analyze this capacity on this
chromenotacrine.
m
2.2.4.2. Effect of CT6 on ROS levels in cortical neurons. As shown in
Table 6, SNP (0.5 mM), administered to cortical neurons during
24 h, increased the ROS formation by about 2.5 times respect
to control. CT6 at 10 and 25 mM was able to revert this effect in a
25-40%. These data agree with the weak effect of CT6 on lipid
peroxidation neuroprotection, although in the case of cortical
neurons, contrary to the in vitro assays, the effects were obtained at
Table 3
IC50
(mM) values for the inhibition of Electrophorus elec-
tricus AChE (EeAChE) by racemic CT1-7.a
lower doses (10e25 mM). At higher CT6 doses, we observed an in-
crease in the ROS formation, indicating a toxic effect of this com-
pound at these high concentrations.
CT
IC50 EeAChE (mM)
1
2
3
4
5
6
7
0.55 ꢀ 0.11
0.103 ꢀ 0.03
0.007 0.003
0.53 ꢀ 0.006
0.065 ꢀ 0.001
0.041 0.001
0.29 ꢀ 0.06
2.2.5. Neuroprotective effect of CT6
Once the ability of CT6 to scavenge ROS was evaluated, we
turned our attention to investigate the capacity of this chromeno-
tacrine to act as a neuroprotectant agent. Thus, we investigated the
neuroprotection profile of CT6 on primary cortical neurons treated
Tacrine
0.04 ꢀ 0.002
a
with Oligomycin-A (10
mM) and rotenone (30 mM) (OR), two
Values are expressed as mean ꢀ standard error of the
mean of at least three different experiments in triplicated.
mitochondrial respiratory chain inhibitors, which block complex V

M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
495
Fig. 4. Kinetics of inhibition of EeAChE hydrolysis of acetylthiocholine (ATCh) by CT6. (A) LineweavereBurk reciprocal plots of initial velocity at substrate concentrations (0.05e
10 M) are presented. Lines were derived from a weighted least-squares analysis of data. (B) Ki calculation for CT6 by Dixon analysis (P < 0.001).
m
Table 4
Table 5
Kinetics of the EeAChE inhibition for CT6 (Km’s and Vmax’s).
% Inhibition of lipid peroxidation (AAPH %) and in vitro inhibition of soybean lip-
oxygenase (LOX %) by CT6, Trolox, and NGGA.
[CT6],
m
M
Vmax
(D
DO/min)
P<
Km (
m
M)
P<
Compound
(%) AAPH (100
m
M)
(%) LOX Inh. (100 mM)
C
0.18 ꢀ 0.056
0.18 ꢀ 0.033
0.17 ꢀ 0.043
0.20 ꢀ 0.09
0.11 ꢀ 0.008
0.10 ꢀ 0.017
0.05 ꢀ 0.014
0.03 ꢀ 0.005
e
0.46 ꢀ 0.023
0.48 ꢀ 0.021
0.54 ꢀ 0.105
0.55 ꢀ 0.068
0.54 ꢀ 0.009
0.57 ꢀ 0.019
0.49 ꢀ 0.026
0.53 ꢀ 0.017
e
0.01
0.05
0.1
0.5
1
ns
ns
ns
***
**
ns
ns
ns
ns
ns
ns
ns
CT6
Trolox
NDGA
85
63
-
18
e
84
Values are expressed as mean ꢀ standard error of the mean of at least three different
experiments.
5
10
***
***
Statistical comparisons were carried out against the control (one way ANOVA;
n ¼ 6). **P < 0.01; ***P < 0.001; ns ¼ non significant.
effect began at 10e25
mM CT6, while in the case of SNP the toxic
effect of the drug was only appreciated from 50
mM concentration.
The different neuroprotective profile of CT6 in both cases could be
due to the fact that different ROS and RNS could be formed in each
case, and to the different ability of CT6 to scavenge these species in
the corresponding oxidative metabolism.
Alternatively, the different neuroprotective effect of CT6 in the
decrease in cell viability induced by OR and SNP could be explained
by the cell death type. Thus, while SNP induces both necrosis and
apoptosis, OR induces mainly apoptosis at the used doses. In our
hands CT6 was able to completely abolish the apoptosis induced by
and I respectively, as previously described [35] by using the XTT
(2,3-bis(2-methoxi-4-nitro-5-sulfophenyl)-2H-tetrazolyl-5-
carboxyanilide) assay to determine neuronal viability.
The CT6 neuroprotective effect was assayed against the decrease
in cell viability induced by both OR and SNP, and was added 15 min
before these toxic agents, at concentrations between 0.1
mM and
100 M. Data in Fig. 6 indicate that CT6 has a neuroprotective effect
m
against cell death induced by OR (Fig. 6A) and by SNP (Fig. 6B), both
conditions producing a decrease in the cellular viability of about a
OR (10
m
M/30 microM) (1.83 ꢀ 0.08-fold the control values), while
40%. EC50 are very similar, 1.11 ꢀ 0.07
m
M and 3.47 ꢀ 0.81
mM for OR
its capacity to reduce 50% of the necrosis induced by 1 mM SNP
and SNP, respectively. However, the highest neuroprotective effects
were obtained against SNP-reduced cell viability, since maximal
neuroprotective activity was 148.12 ꢀ 17.63% against SNP but only
81.08 ꢀ 3.95% against OR. In the case of the OR treatment, a toxic
(3.43 ꢀ 0.41-fold the control values) requires higher concentrations
of at least 25e50 mM. In conclusion, the neuroprotective effect of
CT6 could be antiapoptotic at low concentration and antinecrotic at
higher concentrations. However, doses higher than 50 M seem to
m
be neurotoxic for neurons. These effects are in accordance with the
decrease in lipid peroxidation induced by CT6.
Table 6
Action of CT6 on SNP-induced ROS formation in cortical
neurons.a
CT6 [
m
M]
ROS (ratio over C)
Control
SNP (0.5 mM)
1
10
25
1 ꢀ 0.12
2.65 ꢀ 0.46***
2.31 ꢀ 0.34***ns
1.93 ꢀ 0.38***ꢁꢁ
1.81 ꢀ 0.25***ꢁꢁꢁ
a
Cortical neuron cultures were challenged with 0.5 mM SNP
during 24 h and ROS levels (H2O2) measured as indicated in
Material and Methods. Data are expressed as ratios over control
and are means ꢀ SEM of two experiments from different cell
cultures, each one performed in triplicate. Statistical compari-
sons were performed against control (***) or SNP (ꢁꢁꢁ) by one-
way ANOVA.*** ¼ p < 0.001, ns ¼ non significant.
Fig. 5. MDA production in SH-SY5Y cells induced by 300 mM H2O2 co-incubated with
or without CT6 (1e30
mM) for 8 h. Data represent the mean ꢀ s.e.m of four inde-
pendent experiments in triplicate. &&&p < 0.001 comparing control respect to H2O2.
*p < 0.05;**p < 0.01 and ***p < 0.001 comparing H2O2 respect to H2O2 plus CT6.

496
M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
Fig. 6. Neuroprotective effects of CT6 based on the decrease in cellular viability induced by mitochondrial chain blockers OligA10 mM/Rot 30 mM (A) or 0.5 mM SNP (B) in primary
cultures of cortical neurons. The values are the mean ꢀ SEM of three independent experiments, each one carried out in triplicate, in different cell cultures. The statistical analysis
compares the differences against O/R or SNP, in the absence of drugs, (***) or against controls (ooo) at * ¼ P < 0.05, ** ¼ P < 0.01, *** ¼ P < 0.001 (one way ANOVA).
2.2.6. Molecular modeling of CT6
To shed some light on the suitability of the designed compounds
shown in Fig. 2 to act as AChE inhibitors, we have performed
docking simulations of the (R)- and (S)-enantiomers for compounds
CT6 and CT3.
As protein target, the crystallographic structure of the AChE of
the electrophorus electricus (PDB: 1C2B) has been selected [36].
Attempts to dock the ligands in the catalytic site of EeAChE were
unsuccessful which pointed out that the catalytic site of AChE is not
flexible enough to easily accommodate these bulky compounds. In
contrast, the most favored binding mode accommodates the en-
antiomers (R) of both ligands in the opening of the PAS, located at
the rim of the gorge. Thus, the chromenotacrine moiety stacks
against Trp286, Phe297 and Tyr72 residues (for CT6, see Fig. 7; for
CT3, see Supplementary material). The tetracyclic skeleton is
engaged in hydrophobic interactions with the aromatic residue
Tyr124, and with Phe299, Val300 and Leu289. A close examination
reveals that the pyridinic N is doubly hydrogen bonded to the
Ser298 side chain and backbone, whereas the hydroxylic substit-
uent can establish a hydrogen bond with the backbone of Trp286.
Fig. 8. Putative binding mode of inhibitor (S)-CT6 at the EeAChE gorge predicted by the
The amino group can be probably engaged in a H-bond with
docking simulation. The ligand is rendered as thick with carbon atoms in pink. Selected
residues of the protein are displayed in green (the most relevant residues Trp286,
Tyr72, Tyr124, Glu285, Phe295, Arg296, Phe297 and Ser298 are shown as sticks
colored according to atoms, carbon atoms in green; other residues are shown as thin
sticks in green). Hydrogen-bond interactions are shown by dashed lines.
Fig. 7. Putative binding mode of inhibitor (R)-CT6 at the EeAChE gorge predicted by the
docking simulation. The ligand is rendered as thick with carbon atoms in cyan.
Selected residues of the protein are displayed in green (the most relevant residues
Trp286, Tyr72, Tyr124, Phe295, Arg296, Phe297 and Ser298 are shown as sticks colored
according to atoms, carbon atoms in green; other residues are shown as thin sticks in
green). Hydrogen-bond interactions are shown by dashed lines.
Fig. 9. Superposition of enantiomers of CT6 in the putative binding site [(R)- in cyan,
(S)- in pink)].

M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
497
Table 7
Calculated physicochemical properties for chromenetacrines CT1-7.
CT1
CT2
CT3
CT4
CT5
CT6
CT7
tacrine
Molecular weight
344.41
3
4
1
0
3.61
4.28
68.37
0.04
0.66
ꢂ0.01
ꢂ1.9
ꢂ0.01
ꢂ3.3
94.80
23.27
Hepatotoxicity
362.4
3
4
1
0
3.72
4.46
68.37
0.05
378.85
3
4
1
0
3.83
4.84
68.37
0.02
0.59
0.15
L1.7
0.15
L3.4
95.42
21.64
374.43
3
5
2
0
3.04
4.18
77.61
0.04
0.54
0.04
ꢂ2.0
0.04
ꢂ3.4
94.93
25.64
358.43
3
4
1
0
3.83
4.67
68.37
0.04
434.48
3
7
4
0
1.93
3.83
96.06
0.05
0.27
L0.03
L2.3
L0.03
L3.6
95.36
26.20
387.47
3
5
2
0
3.25
4.34
71.61
0.03
0.46
0.13
ꢂ2.0
0.09
ꢂ3.5
95.21
33.24
198.26
2
2
0
0
2.21
2.72
38.91
0.13
0.30
0.21
ꢂ2.00
0.21
ꢂ2.9
96.52
25.86
No. of H-bond donors
No. of H-bond acceptor
No. of Rotatable Bonds
No. violations Lipinski’s rule
Log P (Moriguchi)a
Log Pb
2
ꢀ
TPSA (A )
Fraction unbound in plasma
Log BBa,c
0.57
0.04
0.63
0.18
Log BBb,c
Log PSd
ꢂ1.9
ꢂ1.8
Log PBd
0.04
0.18
Log (PS*fu, brain)d
Human intestinal absorption (%)
In vitro Caco-2 perm (nm/sec)
Toxicitya,e
ꢂ3.3
94.81
22.82
Hepatotoxicity
ꢂ3.4
94.94
23.69
Hepatotoxicity
Acute toxicity in rats,
hepatotoxicity
a
AMET Predictor, v.6.5.
ACD/Percepta 14.0.
b
c
According to the classification made by Ma et al. [45]: High absorption to CNS: log BB more than 0.3; Middle absorption to CNS: log BB 0.3ꢂ1.0; Low absorption to CNS: log
BB less than ꢂ1.0.
d
Other estimated parameters related to brain penetration were used to classify the compounds as CNS permeable or non-permeable [49]: rate of brain penetration (Log PS)
is the rate of passive diffusion/permeability; brain/plasma equilibration rate (Log(PS*fu, brain)); fu, brain e fraction unbound in plasma.
e
Hepatotoxicity is indicated if alkaline phosphatase level is predicted as elevated.
Gly122. The phenyl group is extended to the central region of the
gorge forming T-shaped interactions with Phe297; in this
each pair of enantiomers and ligands is very similar, thus sup-
porting the values of inhibitory activity. This putative binding mode
was also proposed for tacrine-dihydropyridine hybrids, structures
closely related to these chromenotacrines, and previously reported
by some of us [37]. The rigid tetracyclic core and the presence of a
phenyl substituent on a sp3-hybridized carbon as well as the pre-
dicted interactions with the rich aromatic residues may obstruct a
deeper penetration of the ligand along the gorge to reach the cat-
alytic anionic site. Hence, the docking study strongly suggests a
binding mode in the PAS and a non-competitive inhibition mech-
anism for these chromenotacrines.
On the other hand, the lack of a tryptophan residue mimicking
Trp286 in the peripheral site of hBuChE (replaced by Ala277), as
well as the replacement of Tyr72 in AChE by Asn68 in hBuChE,
could also explain the selectivity on AChE over BuChE of these
compounds.
pep
orientation, two of the methoxy substituents of CT6 are involved in
H-bond interactions with the backbone of Phe295 and Arg296. In
the case of CT3, an H-bond is also detected between Br and Phe295
backbone. Additionally, hydrophobic interactions with the aro-
matic side chains of Phe295, Tyr337, Phe338 and Tyr341 are
observed for both (R)-ligands, along with Ser293 and Ile294. To
note that no interactions with the catalytic triad residues Ser203,
Glu334 and His447 have been found.
On other hand, most favorable docking pose for the (S)-enan-
tiomer of CT6 and CT3 was also found at the PAS (see Fig. 8, and
Supplementary material for CT3). Binding of (S)-enantioner ap-
pears to be as favorable as the (R)-enantiomer (binding energy
differs just by 0.2 and 0.1 kcal/mol for CT6 and CT3, respectively)
because a highly equivalent orientation and mode are observed.
Thus, the inhibitor is stacked against Trp286, Phe297 and Tyr72.
The ligandeenzyme interaction would be probably further
strengthened by a hydrogen bond between the amino group and
the hydroxy group of Tyr72. The tetracyclic core forms hydrophobic
interactions with the aromatic residue Tyr124, and with Phe299,
Val300 and Leu289. The hydroxylic group is doubly hydrogen
bonded to the Glu285 side chain and backbone of Ser298, whereas
the endocyclic O atom is also predicted to mediate a hydrogen-
bond interaction with the hydroxy group of Ser298. The phenyl
2.2.7. ADMET Analysis of the chromenotacrines
Various well known AChE inhibitors, such as ensaculine, done-
pezil, propidium, rivastigmine, and tacrine, have shown slight
improvement in cognitive and memory disorders. However, these
available nitrogen containing AChEI drugs have certain side effects
and lesser central nervous system (CNS) permeability. So, the
new drugs developed for the treatment of AD should present a
better CNS penetration profile and decreased toxic effects. To this
end, some relevant ADMET (Absorption, Distribution, Metabolism,
Excretion and Toxicity) properties were calculated, with special
emphasis on the requirements of CNS. The drugs used for neuro-
logical disorder treatment are generally CNS acting drugs, so
factors that are important to the success of CNS drugs were
analyzed. Computer predictions were performed with ADMET
Predictor 6.5 [38] and ACD/Percepta 14.0.0 [39] software packages
(see Table 7).
group interacts by means of
pep T-shaped interactions with
Phe297. The methoxy substituents of CT6 form H-bonds with the
backbone of Phe295 and Arg296, whereas the bromine of CT3 H-
bonds with the backbone of Phe295. Finally, hydrophobic in-
teractions with the aromatic side chains of Phe295, Tyr337, Phe338
and Tyr341, and also with Ser293 and Ile294 could stabilize the
complex. Analogously to the (R)-enantiomer, no interactions with
the catalytic triad residues have been detected.
A superposition of both enantiomers of CT6 on the binding site
in the PAS (Fig. 9; for CT3 see Supplementary material) shows the
highly equivalent binding mode with the EeAChE.
Drugs that penetrate CNS should have lower polar surface areas
than other kinds of molecules [40]. In the case of CNS, penetrating
drugs are estimated at 60 to 90 A [41,42]. In our study, all the
ꢀ
In summary, the most probable binding mode for both enan-
tiomers of both ligands is found at the PAS. The binding energy for
compounds present a proper value, except CT6 which shows a high
TPSA, slightly above the reference value (96.06 A ).
2
ꢀ

498
M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
All the compounds showed significant drug-like characteristics
4.2.2. Culture of HepG2 cells and treatment
according to the Lipinski’s rule of five [43]. A more rigid rule for CNS
drugs [42,44] (MW ꢃ 450, HB donor ꢃ3, HB acceptors ꢃ7, log P ꢃ5,
TPSA ꢃ90, and number of rotatable bonds ꢃ8) is also satisfied for
these CT, although CT6 is slightly outlier due to the high polar
surface area.
The human hepatoma cell line HepG2 was cultured in Eagle’s
minimum essential medium (EMEM) supplemented with 15
nonessential amino acids, 1 mM sodium pyruvate, 10% heat-
inactivated fetal bovine serum (FBS), 100 units/mL penicillin, and
100 mg/mL streptomycin (reagents from Invitrogen, Madrid, Spain).
The bloodebrain barrier (BBB) is a separation of circulating
blood and cerebrospinal fluid (CSF) in the central nervous system
(CNS). BloodeBrain Barrier (BBB) penetration is represented as
BB ¼ [Brain]/[Blood], where [Brain] and [Blood] are the steady-state
concentration of radiolabeled compounds in brain and peripheral
blood. Predicting BBB penetration means predicting whether
compounds pass across the bloodebrain barrier. This is crucial in
pharmaceutical sphere because CNS-active compounds must pass
across it and CNS-inactive compounds mustn’t pass across it in
order to avoid of CNS side effects. According to the computed
values, these chromenotacrines show a brain penetration profile
not high but sufficient for CNS activity, although CT6 presents the
poorer penetration [45,46].
Cultures were seeded into flasks containing supplemented medium
and maintained at 37 ꢁC in a humidified atmosphere of 5% CO2 and
95% air. For assays, HepG2 cells were subcultured in 96-well plates
at a seeding density of 1x105 cells per well. When the HepG2 cells
reached 80% confluence, the medium was replaced with fresh
medium containing 1e300
mM compounds or 0.1% DMSO as a
vehicle control.
4.2.3. MTT Assay and cell viability
Cell viability, virtually the mitochondrial activity of living cells,
was measured by quantitative colorimetric assay with 3-[4,5-
dimethylthiazol-2-yl]-2,5-diphenyl-tetrazolium bromide (MTT)
(Sigma Aldrich), as described previously [50]. Briefly, 20 mL of the
On other hand, a middle Caco-2 cell permeability is suggested
for all these chromenotacrines [47], as well as a high percent hu-
man intestinal absorption (%HIA) [48]. Finally, the prediction of
parameters related to toxicity concluded that CT1, CT2 and CT5
would probably show hepatotoxicity [49], a prediction that has
been confirmed by the experimental results (see Section 2.2.1.
Toxicological studies of CT1-7).
MTT labeling reagent, at a final concentration of 0.5 mg/mL, was
added to each well at the end of the incubation period and the plate
was placed in a humidified incubator at 37 ꢁC with 5% CO2 and 95%
air (v/v) for an additional 2 h period. Then, the insoluble formazan
was dissolved with dimethylsulfoxide; colorimetric determination
of MTT reduction was measured at 540 nm. Control cells treated
with 0.1% DMSO were taken as 100% viability.
4.2.4. Measurement of IC50 with EeAChE
3. Conclusions
IC50’s were determined using 0.036 U/mL of EeAChE. Enzymatic
activity was evaluated by the Ellman’s method [31]. The reaction
was performed on multiwell Petri dishes of 48 wells in a total
In this work we have investigated several pharmacological is-
sues of the well known chromenotacrines CT1-7, targeted for their
potential application for the treatment of AD, such as the toxicity on
HepG2 cells, ChE inhibition, the antioxidant properties and neu-
roprotection activities. From all the data obtained, we initially have
clearly identified racemic chromenotacrine 11-amino-12-(3,4,5-
trimethoxyphenyl)-8,9,10,12-tetrahydro-7H-chromeno[2,3-b]qui-
nolin-3-ol (CT6) as a non-toxic, compared with tacrine, showing
moderate BBB permeability, and a potent, non-competitive, in the
low micromolar range, EeAChEI, and totally selective since CT6 does
not inhibit hBuChE. Molecular modeling analysis confirmed that
chromenotacrine CT6 is a non-competitive EeAChE inhibitor, and
that both enantiomers are readily accommodated in the opening of
the PAS of the enzyme. Calculations also showed that the binding of
the (S)-enantioner appears to be as favorable as the (R)-enantiomer,
due to a highly equivalent orientation observed. In addition, CT6
behaves as an antioxidant and neuroprotective agent. To sum up,
volume of 500 mL. 0.036 U/mL of enzyme in phosphate buffer 0.1 M
pH 8 was incubated for 15 min with different drug concentrations
at 37 ꢁC. After that the enzymatic reaction was triggered by the
addition of 0.35 mM acetylthiocholine and 0.35 mM 5,50-DiThiobis-
(2-NitroBenzoic acid) (DTNB). Absorbance was measured for 5 min
in a spectrophotometer BioTek Power Wave XS at 410 nm.
4.2.5. Kinetic analysis of the AChE inhibition by CT6
To obtain estimates of the inhibition constant Ki, reciprocal plots
of 1/V versus 1/[S] were constructed at different concentrations of
the substrate acetylthiocholine (0.05e1 mM), at different concen-
trations of CT6 (range 0.01e10 mM), by using Ellman’s method [31].
1/Vmax of the reciprocal plots were then plotted against the con-
centrations of CT6, to evaluate Ki data, as described [35]. Data
analysis was performed with SigmaPlot 11.0 software (Systat Soft-
ware, a subsidiary of Cranes Software International Ltd.).
chromenotacrine
11-amino-12-(3,4,5-trimethoxyphenyl)-8,9,10,
12-tetrahydro-7H-chromeno[2,3-b]quinolin-3-ol (CT6) can be
considered as a very attractive multipotent drug for the potential
treatment of AD.
4.2.6. Statistical analysis
Data are shown as mean ꢀ SEM of results obtained from two or
three independent experiments from different cultures each one
performed in triplicate. Statistical analyses were performed by one
way ANOVA test, as indicated in each case. A p-value of 0.05 was
considered statistically significant. Fit curves for Ki determinations
were performed with SigmaPlot 11.0 software.
4. Experimental part
4.1. Chemistry
CT1-7 have been prepared as reported [18].
4.2.7. Determination of lipid peroxidation
MDA is a breakdown product of the oxidative degradation of cell
membrane lipids and it is generally considered an indicator of lipid
peroxidation. We evaluated lipid peroxidation induced by H2O2
4.2. Pharmacology
4.2.1. Materials
(300 mM) with or without CT6 (1e30 mM) for an 8 h incubation
DMEM/F-12 (Ham) Medium by Life Technologies, Fetal calf
serum (FCS) were purchased from Sera-Lab (Sussex, England).
SNAP. Other chemicals were reactive grade products from Merck
(Darmstadt, Germany).
period. Intracellular MDA production was quantified using a thio-
barbituric acid reactive substance (TBARS) assay kit (Cell Biolabs
Inc., San Diego, CA). Briefly, 1 ꢄ 106 cells per well were seeded in a
six-well plate, then collected in 200
mL of culture medium and

M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
499
sonicated for 3 ꢄ 5 s intervals at 40V over ice. SDS Lysis solution
(100 L) was added to the sample solution and the MDA standards
in a microcentrifuge tube and mix thoroughly. Then, 250 l of TBA
reagent were added to each sample and standard to be tested, and
incubate at 95 ꢁC for 45e60 min. Each sample and standard (200
l)
treated neurones were washed with PBS and incubated with the
XTT solution (final concentration 0.3 mg/mL) according to the Kit
specifications. After this incubation period, orange dye solution was
spectrophotometrically quantified. Results are expressed as per-
centages with respect to the control cells.
m
m
m
were loaded (in duplicate) into a clear 96-well plate and the
absorbance at 532 nm was recorded using a microplate reader
(Biochrom ASYS UVM 340, Cambridge, UK). The content of MDA
was calculated for each sample from a standard curve.
4.2.12. Measurement of the ROS formation
To assay the ROS formation, 2,7-dichlorodihydrofluorescein
diacetate (H2DCF-DA), a non-fluorescent lipophilic reagent, was
used. H2DCF-DA enters into the cells, where it is transformed into
2,7-dichlorodihydrofluorescein (H2DCF) by the action of intracel-
lular esterase. H2DCF is oxidized to fluorescent DCF by hydrogen
4.2.8. Inhibition of linoleic acid lipid peroxidation [33]
Production of conjugated diene hydroperoxide by oxidation of
linoleic acid in an aqueous dispersion is monitored at 234 nm.
AAPH is used as a free radical initiator. Ten microliters of the 16 mM
sodium linoleate dispersion was added to the UV cuvette contain-
ing 0.93 mL of 0.05 M phosphate buffer, pH 7.4 pre-thermostated at
37 ꢁC. The oxidation reaction was initiated at 37 ꢁC under air by the
addition of 50 mL of 40 mM AAPH solution. Oxidation was carried
out in the presence of aliquots (10 mL) in the assay without anti-
oxidant, lipid oxidation was measured in the presence of the same
level of DMSO. The rate of oxidation at 37 ꢁC was monitored by
recording the increase in absorption at 234 nm caused by conju-
gated diene hydroperoxides.
peroxide. H2DCF-DA (5 mM) was added to the cells, at the moment
in which they were subjected to the different treatments. After each
treatment the incubation medium was removed and the cells were
washed twice with PBS and the fluorescence was measured in an
FL600-BioTek spectrofluorometer with filters of 485 nm exc and
530 nm em. Results were expressed as arbitrary fluorescence units
(AFU) in percentage respect to control cells.
4.2.13. Assessment of cell viability after exposure of cell cultures to
oligomycin-rotenone treatments
To investigate the neuroprotective effect of tacrine and analo-
gous compounds, several concentrations of these compounds be-
4.2.9. Soybean lipoxygenase inhibition study in vitro
tween 0.01 and 100
measuring the increase in cell viability after 24 h treatment with a
mixture of 30 M rotenone and 10 M Oligomycin-A, which
mM were used. Neuroprotection was assayed by
[33]. Compound CT6 dissolved in DMSO were incubated at room
temperature with sodium linoleate (0.1 mL) and 0.2 mL of enzyme
solution (1/9 ꢄ 10ꢂ4 w/v in saline). The conversion of sodium
linoleate to 13-hydroperoxylinoleic acid at 234 nm was recorded
and compared with the appropriate standard inhibitor.
m
m
induced neuronal cell death. The mixture of rotenone plus
oligomycin-A blocks mitochondrial electron transport chain com-
plexes I and V, respectively, inducing cell death by oxidative stress.
Cell viability, specifically the mitochondrial activity of living
cells, was measured by quantitative colorimetric assay with the
mitochondrial probe XTT [2,3-bis(2-methoxy-4-nitro-5-sulfo
phenyl)-2H-tetrazolium-5-carboxanilide sodium salt] (Roche,
Madrid, Spain), based on the ability of living metabolically active
cells to reduce the yellow tetrazolium salt (XTT) to form an orange
formazan dye, whose quantity directly correlates to the number of
living cells. Measurements were carried out on neuronal cell cul-
tures seeded into 48-well culture plates as described [35]. Briefly,
control and treated neuronal cultures (2 ꢄ 105 cells/well) were
incubated with the XTT solution at 0.3 mg/ml final concentration
for 3 h in a humidified incubator at 37 ꢁC with 5% CO2 and 95% air
(v/v) and the soluble orange formazan dye formed was spectro-
photometrically quantified, using a Biotek PowerWave XS spec-
trophotometer microplate-reader at 450 nm (reference 650 nm).
All XTT assays were performed in triplicate in cells of different
batches. Control cells treated with EMEM alone were regarded as
100% viability. Controls containing different DMSO concentrations
(0.001e1% DMSO) were performed in all assays.
4.2.10. Cell isolation and culture
Primary neuronal cultures from rat cerebral cortex were pre-
pared as previously described [35] with minor modifications. All
procedures associated with animal experiments were approved by
the Ethics Committee of Universidad Complutense de Madrid
(UCM), Madrid (Spain). Cell suspensions from cerebral cortex were
prepared from 19-day-old Wistar rat embryos. Living cells in cell
suspension were counted by trypan blue exclusion method. Cells
were seeded on plastic 48 well multidishes, precoated with
0.05 mg/mL poly
D
-lysine at a density of 2 ꢄ 105 cells/well, and
were kept at 37 ꢁC in a 5% CO2 atmosphere in high glucose Dul-
becco’s medium supplemented with 15% heat-inactivated (56 ꢁC for
30 min) fetal calf serum. After 48 h, cultured cells were placed and
maintained in a serum-free medium (Dulbecco’s: Ham’s F12, 1:1
[vol/vol] containing 3.15 mg/mL glucose, 2.5 mM Glutamax, and
0.5 mM sodium pyruvate (DMEM/F-12, GlutaMAXÔ; GIBCO, Life
Technologies, Madrid (Spain), 1% AntibioticeAntimycotic (Gibco;
Life Technologies, Madrid, Spain) at 100 units of penicillin,100
mg of
streptomycin, and 0.25 g of amphotericin B final concentrations,
m
and supplemented with 1% B27 medium (Gibco; Life Technologies,
Madrid (Spain)). Six- to nine day cultures were used in the exper-
iments. Glial contamination was measured following the usual
protocol [51]. Under these conditions the glial cells represented
9% ꢀ 3% of the total cell population.
4.2.14. Determination of hepatic aspartate aminotransferase (AST)
and alanine aminotransferase (ALT) levels
The ALT and AST levels were tested as a marker of hepatotox-
icity. HepG2 cells were maintained in a 25 cm3 culture flasks and
incubated in a humidified incubator at 37 ꢁC with 5% CO2 and 95%
air (v/v). After 48 h, cells were treated with CT6 and tacrine at
4.2.11. Cell viability
different concentrations (10, 30, and 100 mM). Control cells received
It was determined by the XTT method. This assay is based on the
ability of living metabolically active cells to reduce the yellow
tetrazolium salt (XTT) to form an orange formazan dye. Therefore,
the conversion only occurs in living cells. The formazan dye formed
is directly quantified using a scanning multiwell spectrophotom-
eter at wavelength 492 nm (reference wavelength 690). The
amount of orange formazan formed, as monitored by the absor-
bance, directly correlates to the number of living cells. Control and
0.1% of DMSO instead of CT6 or tacrine. After incubation for 24 h,
cell supernatants were collected and centrifuged at 1000 g for
5 min at 4 ꢁC. ALA and AST levels were measured with an automatic
biochemical analyzer Cobas Integra 400 Plus (Roche, Madrid,
Spain). The results were expressed as unit per liter (U/L). One unit of
AST and ALT is defined as the amount of enzyme that generates
1.0
mmol of glutamate and pyruvate respectively per minute at
37 ꢁC.

500
M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
4.2.15. Lactate dehydrogenase (LDH) release
The receptor residues Trp286, Tyr124, Tyr337, Tyr72, Asp74, Thr75,
Trp86, and Tyr341 were selected to keep flexible during docking
simulation with the AutoTors module. The program searched until a
maximum of 100 conformations and the procedure was repeated
100 times (runs). After docking, the 100 solutions were clustered in
groups with RMSD less than 1.0 A. The clusters were ranked by the
lowest energy representative of each cluster. For all other param-
eters, the default values were used with AutoDock Tools.
LDH activity was measured as the rate of decrease in the
absorbance at 340 nm, resulting from the oxidation of NADH, ac-
cording to the following reaction
Piruvate þ NADHLDHLactate þ NADþ
ꢀ
!
For the determination of LDH, the culture medium from the
control and treated cells was collected. The neurones were ho-
mogenized with 0.1 M TriseHCl (pH 7.4), containing 0.1% Triton X-
100. Homogenates were centrifuged at 13.000 g for 10 min. The
amount of proteins in cells lysated was monitored by the Bradford
technique [51]. LDH activity was measured in the cell supernatants
and in the collected culture medium according to [52]. LDH release
is given as percentage of LDH release with respect to the total LDH
content (LDH in the supernatant þ LDH inside the cells).
Acknowledgments
JMC thanks MINECO (SAF2012-33304) for support. MJOG thanks
MICINN (SAF2010-20337) and UCM-Santander (GR35/10-B) for
financial support.
Appendix A. Supplementary data
4.2.16. Caspase-3 activity measurement
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2013.12.021.
Caspase-3 activity was measured as described [53]. Briefly,
control and treated cortical neurons were washed rapidly with PBS
and homogenized with cell lysis buffer (10 mM TriseHCl, 10 mM
NaH2PO4/Na2HPO4, pH 7.5, 130 mM NaCl, 0.5% Triton X-100, 10 mM
Na4P2O7 and 2 mM DTT). Homogenates were centrifuged at 13,000
g for 5 min. Caspase-3 activity was measured in the supernatants.
References
[1] M. Goedert, M.G.A. Spillantini, Science 314 (2006) 777e781.
[2] A. Castro, A. Martínez, Curr. Pharm. Des. 12 (2006) 4377e4387.
[3] J.L. Cummings, Rev. Neurol. Dis. 1 (2004) 60e69.
[4] V.N. Talesa, Mech. Ageing Dev. 122 (2001) 1961e1969.
[5] A. Romero, R. Cacabelos, M.J. Oset-Gasque, A. Samadi, J. Marco-Contelles,
Bioorg. Med. Chem. Lett. 23 (2013) 1916e1922.
[6] M. Racchi, M. Mazzucchelli, E. Porrello, C. Lanni, S. Govoni, Pharmacol. Res. 50
(2004) 441e451.
[7] N.C. Inestrosa, A. Álvarez, C.A. Pérez, R.D. Moreno, M. Vicente, C. Linker,
O.I. Casanueva, C. Soto, J. Garrido, Neuron 16 (1996) 881e891.
[8] M. Bartolini, C. Bertucci, V. Cavrini, V. Andrisano, Biochem. Pharmacol. 65
(2003) 407e416.
Supernatants with at least 20
for 2e4 h in caspase-3 assay buffer (20 mM HEPES, pH 7.5, with
2 mM DTT) and 20 M Ac. DEVD-AMC [N-acetyl-Asp-Glu-Val-Asp-
m
g of protein were incubated at 37 ꢁC
m
(7-amino-4-methylcoumarin)]. The fluorogenic AMC liberated
from Ac-DEVD-AMC was monitored using a spectrofluorometer
(Bio-Tek FL 600) at an excitation wavelength of 360 nm and an
emission wavelength of 460 nm. Under these conditions the
emission was linear between, at least, 0e6 h, depending on the
caspase-3 activity. Enzymatic activity was expressed as arbitrary
fluorescence unit after 1 h per mg protein (AFU/h/mg protein).
[9] A. Cavalli, M.L. Bolognesi, S. Capsoni, V. Andrisano, M. Bartolini, E. Margotti,
A. Cattaneo, M. Recanatini, C. Melchiorre, Angew. Chem. Int. Ed. 46 (2007)
3689e3692.
[10] D. Muñoz-Torrero, P. Camps, Curr. Med. Chem. 13 (2006) 399e422.
[11] L. Savini, A. Gaeta, C. Fattorusso, B. Catalanotti, G. Campiani, L. Chiasserini,
C. Pellerano, E. Novellino, D. McKissic, A. Saxena, J. Med. Chem. 46 (2003) 1e4.
[12] P.R. Carlier, E.S. Chow, Y. Han, J. Liu, J. El Yazal, Y.P. Pang, J. Med. Chem. 42
(1999) 4225e4231.
4.3. Docking studies
The geometry of the ligands were optimized using the ab initio
quantum chemistry program Gaussian 09 and the B3LYP/3-21G*
basis set. A set of atom-centered RHF 6-31G* charges was then
obtained by using the RESP methodology [54].
The crystallographic structure of the acetylcholinesterase with
sequence from Electrophorus electricus was retrieved from the
Protein Data Bank (PDB code: 1C2B) [36] as target protein. Missing
atoms were reconstructed with SwissPDB Viewer 4.1.0 [55].
The docking experiments were carried out using the Lamarckian
genetic algorithm implemented in the AutoDock 4.2 program [56].
Two consecutive docking runs were carried out. Initially, a box
encompassing the whole protein target was defined with a
[13] A. Cavalli, M.L. Bolognesi, A. Minarini, M. Rosini, V. Tumiatti, M. Recanatini,
C. Melchiorre, J. Med. Chem. 51 (2008) 347e372.
[14] R. León, A.G. García, J. Marco-Contelles, Med. Res. Rev. 33 (2013) 139e189.
[15] R. León, J. Marco-Contelles, Curr. Med. Chem. 11 (2011) 552e576.
[16] E. Maalej, F. Chabchoub, A. Samadi, C. de los Ríos, A. Perona, A. Morreale,
J. Marco- Contelles, Bioorg. Med. Chem. 21 (2011) 2384e2388.
[17] E. Maalej, F. Chabchoub, M.J. Oset-Gasque, M. Esquivias-Pérez, M.P. González,
L. Monjas, C. de los Ríos, M.I. Rodríguez-Franco, I. Iriepa, I. Moraleda,
M. Chioua, A. Romero, J. Marco-Contelles, A. Samadi, Eur. J. Med. Chem. 54
(2012) 750e763.
[18] D.S. Raghuvanshi, K.N. Singh, ARKIVOC (2010) 305e317.
[19] S. Ahmad, J. Nat. Prod. 47 (1984) 391e392.
[20] G. Kolokythas, N. Pouli, P. Marakos, H. Pratsinis, D. Kletsas, Eur. J. Med. Chem.
41 (2006) 71e79.
[21] M.A. Azuine, H. Tokuda, J. Takayasu, F. Enjyo, T. Mukainaka, T. Konoshima,
H. Nishino, G. Kapadia, J. Pharmacol. Res. 49 (2004) 161e169.
[22] S.K. Srivastava, R.P. Tripathi, R. Ramachandran, J. Biol. Chem. 280 (2005)
30273e30281.
[23] S. Toshiro, W. Noriko, Eur. Pat. Appl. EP 647445 A1 19950412, 1995.
[24] Y. Ito, H. Kato, S. Yasuda, N. Kato, N. Iwasaki, H. Nishino, M. Takeshita, Jpn.
Kokai Tokkyo Koho, JP 06107664 A2 19940419, 1994.
[25] K. Goto, O. Yaoka, T. Oe, PCT Int. Appl. WO 8401711 A1 19840510, 1984.
[26] F. Noor, J. Niklas, U. Müller-Viera, E. Heinzle, Toxicol. Appl. Pharmacol. 237
(2009) 221.
[27] S.I. Gracon, M.J. Knapp, W.G. Berghoff, M. Pierce, R. de Jong, S.J. Lobbestael,
J. Symons, S.L. Dombey, F.A. Luscombe, D. Kraemer, Alzheimer Dis. Assoc.
Disord. 12 (1998) 93e101.
ꢀ
distance-dependent dielectric constant and a grid spacing of 1 A for
the calculation of energy maps by means of AutoGrid4. The results
clearly showed the ligand poses around the PAS. Therefore, a sec-
ond run was performed on the region of interest at a higher reso-
lution to get more reliable results. A box encompassing both the
CAS and the PAS site was defined for the exploration of possible
binding modes. A volume for exploration was defined in the shape
of a three-dimensional cubic grid (60 ꢄ 72 ꢄ 60 Å3) at a resolution
ꢀ
of 0.3 A and centered on the gorge that enclosed the residues that
are known to make up the inhibitors binding pockets and modes.
At each grid point, the receptor’s atomic affinity potentials for
carbon, oxygen, nitrogen, and hydrogen atoms present in the ligand
were pre-calculated for rapid intra- and intermolecular energy
evaluation of the docking solution.
[28] J. Fusek, J. Patocka, J. Bajgar, M. Koupilova, V. Hrdina, Act. Nerv. Super. 28 (4)
(1986) 327e328.
[29] P.B. Watkins, H.J. Zimmerman, M.J. Knapp, S.I. Gracon, K.W. Lewis, JAMA 271
(1994) 992e998.
[30] M.A. Xiao-Chao, X. Jian, W. Hai-Xue, Z. Ting, T.U. Zeng-Hong, Acta Pharmacol.
Sin. 24 (2003) 247e250.
[31] G.L. Ellman, K.D. Courtney, B.J. Andres, R.M. Featherstone, Biochem. Pharma-
col. 7 (1961) 88e95.
Different conformers of the ligands were docked by randomly
changing the torsion angles and overall orientation of the molecule.

M.J. Oset-Gasque et al. / European Journal of Medicinal Chemistry 74 (2014) 491e501
501
[32] S.K. Fisher, S. Wonnacott, Acetylcholine, in: S.T. Brady, G.J. Siegel (Eds.), Basic
[44] H. Pajouhesh, G.R. Lenz, NeuroRx 2 (2005) 541e553.
Neurochemistry. Principles of Molecular, Cellular and Medical Neurobiology,
eighth ed., Elsevier, 2012, pp. 258e282.
[33] M. Chioua, D. Sucunza, E. Soriano, D. Hadjipavlou-Litina, A. Alcázar, I. Ayuso,
M.J. Oset-Gasque, M.P. González, L. Monjas, M.I. Rodríguez-Franco, J. Marco-
Contelles, A. Samadi, J. Med. Chem. 55 (2012) 153e168.
[34] Y. Yao, C.M. Clark, J.Q. Trojanowski, V.M.-Y. Lee, D. Praticò, Ann. Neurol. 58
(2005) 623e626.
[35] M. Esquivias-Pérez, E. Maalej, A. Romero, F. Chabchoub, A. Samadi, J. Marco-
Contelles, M.J. Oset-Gasque, Chem. Res. Toxicol. 26 (2013) 986e992.
[36] Y. Bourne, J. Grassi, P.E. Bougis, P. Marchot, J. Biol. Chem. 274 (1999) 30370e
30376.
[37] R. León, C. de los Ríos, J. Marco-Contelles, O. Huertas, X. Barril, F.J. Luque,
M.G. López, A.G. García, M. Villarroya, Bioorg Med. Chem. 16 (2008) 7759e7769.
[38] AMET Predictor, v.6.5, Simulations Plus, Inc., Lancaster, CA, 2013.
[39] ACD/Percepta 14.0.0, Advanced Chemistry Development, 2013.
[40] U. Kragh-Hansen, Pharmacol. Rev. 33 (1981) 17e22.
[45] X. Ma, C. Chen, J. Yang, Acta Pharmacol. Sin. 26 (2005) 500e512.
[46] The same trend is also suggested by other theoretical models: (a) F. Cheng,
W. Li, Y. Zhou, J. Shen, Z. Wu, G. Liu, P.W. Lee, Y. Tang, J. Chem. Inf. Model 52
(2012) 3099e3105;
(b) PreADMET 2.0, Bioinformatics
(BMDRC), Yonsei University (Seoul).
& Molecular Design Research Center
[47] S. Yamashita, T. Furubayashi, M. Kataoka, T. Sakane, H. Sezaki, H. Tokuda, Eur.
J. Pharm. Sci. 10 (2000) 195e204.
[48] S. Yee, Pharm. Res. 14 (1997) 763e766.
[49] The details can be found on the FDA website: http://www.fda.gov.
[50] F. Denizot, R. Lang, J. Immunol. Methods 89 (1986) 271e277.
[51] M.A. Bradford, Anal. Biochem. 72 (1976) 248e253.
[52] S. Figueroa, M.J. Oset-Gasque, C. Arce, C. Martínez-Honduvilla, M.P. González,
J. Neurosci. Res. 83 (2006) 441e449.
[53] E. López, S. Figueroa, M.J. Oset-Gasque, M.P. González, Br. J. Pharmacol. 138
(2003) 901e911.
[41] T.T. Wager, R.Y. Chandrasekaran, X. Hou, M.D. Troutman, P.R. Verhoest,
A. Villalobos, Y. Will, ACS Chem. Neurosci. 1 (2010) 420e434.
[42] D. Meanwell, Chem. Res. Toxicol. 24 (2011) 1420e1456.
[43] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Adv. Drug. Deliv. Rev. 46
(2001) 3e26.
[54] A.J. Knox, M.J. Meegan, D.G. Lloyd, Curr. Top. Med. Chem. 6 (2006) 217e
243.
[55] SwissPdb Viewer 4.1.0, GlaxoSmithKline R&D, Switzerland, 2012.
[56] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell,
A.J. Olson, J. Comp. Chem. 16 (2009) 2785e2791.