1,2,3,4-Tetrahydrobenzo[h][1,6]naphthyridines as a new family of potent peripheral-to-midgorge-site inhibitors of acetylcholinesterase: Synthesis, pharmacological evaluation and mechanistic studies

Article abstract of DOI:10.1016/j.ejmech.2013.12.008

A series of 1,2,3,4-tetrahydrobenzo[h][1,6]naphthyridines differently substituted at positions 1, 5, and 9 have been designed from the pyrano[3,2-c]quinoline derivative 1, a weak inhibitor of acetylcholinesterase (AChE) with predicted ability to bind to the AChE peripheral anionic site (PAS), at the entrance of the catalytic gorge. Fourteen novel benzonaphthyridines have been synthesized through synthetic sequences involving as the key step a multicomponent Povarov reaction between an aldehyde, an aniline and an enamine or an enamide as the activated alkene. The novel compounds have been tested against Electrophorus electricus AChE (EeAChE), human recombinant AChE (hAChE), and human serum butyrylcholinesterase (hBChE), and their brain penetration has been assessed using the PAMPA-BBB assay. Also, the mechanism of AChE inhibition of the most potent compounds has been thoroughly studied by kinetic studies, a propidium displacement assay, and molecular modelling. We have found that a seemingly small structural change such as a double O → NH bioisosteric replacement from the hit 1 to 16a results in a dramatic increase of EeAChE and hAChE inhibitory activities (>217- and >154-fold, respectively), and in a notable increase in hBChE inhibitory activity (>11-fold), as well. An optimized binding at the PAS besides additional interactions with AChE midgorge residues seem to account for the high hAChE inhibitory potency of 16a (IC ₅₀ = 65 nM), which emerges as an interesting anti-Alzheimer lead compound with potent dual AChE and BChE inhibitory activities.

Full text of DOI:10.1016/j.ejmech.2013.12.008

European Journal of Medicinal Chemistry 73 (2014) 141e152
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
1,2,3,4-Tetrahydrobenzo[h][1,6]naphthyridines as a new family of
potent peripheral-to-midgorge-site inhibitors of acetylcholinesterase:
Synthesis, pharmacological evaluation and mechanistic studies
Ornella Di Pietro a, Elisabet Viayna a, Esther Vicente-García b, Manuela Bartolini c,
Rosario Ramón b, Jordi Juárez-Jiménez d, M. Victòria Clos e, Belén Pérez e,
,
g
,
a,
**
Vincenza Andrisano f, F. Javier Luque d, Rodolfo Lavilla b , Diego Muñoz-Torrero
*
a Laboratori de Química Farmacèutica (Unitat Associada al CSIC), Facultat de Farmàcia, and Institut de Biomedicina (IBUB), Universitat de Barcelona, Av.
Joan XXIII 27-31, E-08028 Barcelona, Spain
b Barcelona Science Park, Baldiri Reixac 10-12, E-08028 Barcelona, Spain
c Department of Pharmacy and Biotechnology, Alma Mater Studiorum, University of Bologna, Via Belmeloro 6, I-40126 Bologna, Italy
d Departament de Fisicoquímica, Facultat de Farmàcia, and IBUB, Universitat de Barcelona, Prat de la Riba 171, E-08921 Santa Coloma de Gramenet, Spain
e Departament de Farmacologia, de Terapèutica i de Toxicologia, Institut de Neurociències, Universitat Autònoma de Barcelona, E-08193 Bellaterra,
Barcelona, Spain
f Department for Life Quality Studies, University of Bologna, Corso d’Augusto 237, I-47921 Rimini, Italy
g Laboratori de Química Orgànica, Facultat de Farmàcia, Universitat de Barcelona, Av. Joan XXIII 27-31, E-08028 Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
A series of 1,2,3,4-tetrahydrobenzo[h][1,6]naphthyridines differently substituted at positions 1, 5, and 9
have been designed from the pyrano[3,2-c]quinoline derivative 1, a weak inhibitor of acetylcholines-
terase (AChE) with predicted ability to bind to the AChE peripheral anionic site (PAS), at the entrance of
the catalytic gorge. Fourteen novel benzonaphthyridines have been synthesized through synthetic se-
quences involving as the key step a multicomponent Povarov reaction between an aldehyde, an aniline
and an enamine or an enamide as the activated alkene. The novel compounds have been tested against
Electrophorus electricus AChE (EeAChE), human recombinant AChE (hAChE), and human serum butyr-
ylcholinesterase (hBChE), and their brain penetration has been assessed using the PAMPA-BBB assay.
Also, the mechanism of AChE inhibition of the most potent compounds has been thoroughly studied by
kinetic studies, a propidium displacement assay, and molecular modelling. We have found that a
seemingly small structural change such as a double O / NH bioisosteric replacement from the hit 1 to
16a results in a dramatic increase of EeAChE and hAChE inhibitory activities (>217- and >154-fold,
respectively), and in a notable increase in hBChE inhibitory activity (>11-fold), as well. An optimized
binding at the PAS besides additional interactions with AChE midgorge residues seem to account for the
high hAChE inhibitory potency of 16a (IC50 ¼ 65 nM), which emerges as an interesting anti-Alzheimer
lead compound with potent dual AChE and BChE inhibitory activities.
Received 14 October 2013
Received in revised form
5 December 2013
Accepted 6 December 2013
Available online 18 December 2013
Keywords:
Benzo[h][1,6]naphthyridines
Povarov reaction
Dual inhibitors
AChE PAS binding
AChE midgorge binding
Ó 2013 Elsevier Masson SAS. All rights reserved.
1. Introduction
health system worldwide. The number of people with AD increases
rapidly, and in line with this, both prevalence and costs are also
Alzheimer’s disease (AD) is a progressive and ultimately fatal
neurodegenerative disorder that is currently threatening every
increasing [1]. Currently, it is estimated that dementia, of which AD
is the most common type, is affecting 36 million people, with a total
cost amounting to as much as 1% of global gross domestic product
[1]. AD is among the top ten causes of death, but, worryingly, the
only one that cannot be prevented, cured or slowed down [2],
thereby making it imperative the development of efficacious drugs.
Current therapeutic options, i.e. the acetylcholinesterase (AChE)
inhibitors donepezil, galantamine and rivastigmine and the gluta-
mate NMDA receptor antagonist memantine, are regarded as
* Corresponding author. Barcelona Science Park, Baldiri Reixac 10-12, E-08028
Barcelona, Spain. Tel.: þ34 934037106; fax: þ34 934024539.
** Corresponding author. Laboratori de Química Farmacèutica (Unitat Associada al
CSIC), Facultat de Farmàcia, and IBUB, Universitat de Barcelona, Av. Joan XXIII, 27-
31, E-08028 Barcelona, Spain. Tel.: þ34 934024533; fax: þ34 934035941.
E-mail addresses: rlavilla@pcb.ub.es (R. Lavilla), dmunoztorrero@ub.edu
(D. Muñoz-Torrero).
merely symptomatic, and very promising b-amyloid (Ab)-directed
0223-5234/$ e see front matter Ó 2013 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2013.12.008

142
O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
drug candidates designed to confront the underlying mechanisms
of AD are inexorably failing in late stage clinical trials due to lack of
quinoline nitrogen atom, which would become protonatable at
physiological pH. This would enable the resulting benzo[h][1,6]
efficacy or safety. The increasingly accepted notion that A
cause but one of the causes of AD [3] is spurring the development of
multi-target drugs that simultaneously hit A formation and ag-
gregation as well as other important targets such as tau hyper-
phosphorylation and aggregation, oxidative stress and
cholinesterases, among others, as a more realistic option to effec-
tively treat AD [4].
In every case, any single-target or multi-target drug candidate
purported to modify AD progression would need to be adminis-
tered in the early presymptomatic or preclinical stage of AD, before
neurodegeneration is too widespread. Indeed, preclinical AD has
been proposed as the initial stage of AD in the new criteria and
guidelines for diagnosing AD [2]. Accurate and reliable biomarkers,
indicative of the earliest signs of the disease, are necessary both to
identify individuals in the presymptomatic stage of AD, amenable
to early interventions with disease-modifying drugs and to monitor
their effects. Many research endeavours are being made to select
the best diagnostic biomarkers or combinations thereof [5] but
much more work is still needed before preclinical AD can be
diagnosed [2]. Meanwhile, diagnosis of AD will remain based on
symptoms, i.e. on the occurrence of cognitive decline, for whose
alleviation AChE inhibitors (AChEIs) are the best therapeutic option
[6], thereby warranting the search for novel AChEIs.
b
is not the
naphthyridine system to establish catione
p interactions addition-
ally to the stacking, thereby potentially increasing its affinity
pep
b
for the PAS of AChE and the AChE inhibitory activity. These in-
teractions would be similar to those established by the phenan-
thridinium system of propidium, but unlike propidium, the non-
permanent character of the positive charge at the quinoline nitro-
gen atom of the benzo[h][1,6]naphthyridine system would not
preclude their penetration into the central nervous system (CNS).
Herein, we describe the synthesis, cholinesterase inhibitory
activity evaluation and a comprehensive assessment of the binding
mode to AChE by kinetic, propidium displacement and molecular
modelling studies of a series of 1,2,3,4-tetrahydrobenzo[h][1,6]
naphthyridines differently substituted at positions 1, 5, and 9.
Moreover, the brain penetration of these compounds has been
assessed using the parallel artificial membrane permeation assay
(PAMPA-BBB).
2. Results and discussion
2.1. Synthesis of the target compounds
To assess the effect on cholinesterase inhibitory activity of sub-
stitution at position 1, we initially planned the synthesis of 1,2,3,4-
benzo[h][1,6]naphthyridines bearing an ethyl ester group at posi-
tion 9 and a 4-chlorophenyl substituent at position 5, like in the
pyrano[3,2-c]quinoline analogue 1, and either a benzyl group (10a),
hydrogen atom (12a) or a 4-methoxybenzyl group (13a) on the ni-
trogen atom at position 1 (Scheme 1). Also, to ascertain the effect of
the substituent at position 5, we planned the synthesis of the 1-
benzylated ethyl ester analogues bearing a 3-pyridyl (10b) or 4-
methoxycarbonylphenyl (10c) substituent at position 5. Finally, to
shed light on the role of the substituent at position 9, we studied the
substitution of the ethyl carboxylic ester group of 10a by an N-ethyl
carboxamide (14a) and an ethylaminomethyl (17a) group. Following
the evaluation of the AChE inhibitory activity of this first generation
of 1,2,3,4-benzo[h][1,6]naphthyridine derivatives and the estab-
lishment of structureeactivity relationships, we additionally envi-
sioned the synthesis of compounds 12b, 16a, 18a, and 18b (Scheme
1) as second generation optimized analogues (see below).
Most known AChEIs have been designed to interact with the
ꢀ
catalytic site of the enzyme, which is placed at the bottom of a 20 A
deep narrow gorge. The entrance of the gorge contains the so-
called peripheral anionic site (PAS) [7], which can be also tar-
geted either separately or simultaneously by potential inhibitors
[8]. Recently, we have developed a new family of AChEIs that
consisted of a pyrano[3,2-c]quinoline moiety connected through
linkers of different lengths to a unit of the potent active site AChEI
6-chlorotacrine [9]. Among those hybrids, the most potent human
AChE (hAChE) inhibitors bore a 5-(4-chlorophenyl)pyrano[3,2-c]
quinoline moiety, which is present in their synthetic ester precur-
sor 1 (Fig. 1) and is reminiscent to the phenyl-substituted tricyclic
system of the AChE PAS inhibitor propidium (2, Fig. 1). Not unex-
pectedly, molecular dynamics simulations suggested that the 5-(4-
chlorophenyl)pyrano[3,2-c]quinoline moiety of the hybrids in-
teracts at the PAS of AChE, namely by establishing
pep stacking
interactions with residues Trp286 and Tyr72 (hAChE numbering),
whereas the 6-chlorotacrine unit interacts with the active site
residues Trp86 and Tyr337. Strikingly, despite the predicted ability
of the 5-(4-chlorophenyl)pyrano[3,2-c]quinoline moiety to interact
with the AChE PAS, compound 1 was found to be essentially inac-
The synthesis of compounds 10aec was envisaged through a
three-step sequence involving an initial Povarov multicomponent
reaction [10] between the known cyclic enamide 3 [11], as the
activated alkene, ethyl 4-aminobenzoate, 5, and the aromatic alde-
hydes 4aec, under Sc(OTf)3 catalysis in acetonitrile (Scheme 1).
These reactions afforded in moderate to good yields and 1:1 to 1.7:1
diastereomeric ratio the diastereomeric mixtures of cis-fused octa-
hydronaphthyridines 6aec, which were subjected to DDQ oxidation
[12] to yield the lactams 9aec in 73%, 5%, and 36% yield, respectively,
after silica gel column chromatography purification. In an attempt to
improve the yield of 9b, the oxidation of the mixture 6b with MnO2
[13] instead of DDQ only afforded unreacted material and open-ring
byproducts. Chemoselective reduction of lactams 9aec with
(EtO)3SiH under Zn(OAc)2 catalysis [14] provided the desired ben-
zonaphthyridines 10aec in low to moderate (15e52%) yields.
Analogously, the Povarov reaction of aniline 5, aldehydes 4a,b
and the commercially available N-Boc-protected cyclic enamine 7,
followed by DDQ or MnO2 oxidation of the resulting diastereomeric
mixtures 8a,b afforded the N-Boc-protected derivatives 11a and
11b in 57% and 24% overall yields (Scheme 1). Acidic deprotection of
11a quantitatively yielded the target benzonaphthyridine 12a,
which was also used as starting material for the synthesis of the 1-
(4-methoxybenzyl)-substituted benzonaphthyridine 13a (40%
yield) by deprotonation with NaH and alkylation with 1-
tive as inhibitor of hAChE (IC50 > 10 mM).
In the light of these results, we inferred that substitution of the
oxygen atom at position 1 of the pyrano[3,2-c]quinoline system of 1
by a nitrogen atom would result in increased basicity of the
CO2Me
9
NH2
1
O
Me
N
N
N
2 I
Et
H2N
5
Et
Propidium iodide, 2
1
Cl
Fig. 1. Structure of the pyrano[3,2-c]quinoline derivative 1 and the peripheral site
AChE inhibitor propidium iodide.

O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
143
CO2Et
CO2Et
CO2Et
Ph
Ph
Boc
N
Boc
N
H
H
H
H
O
N
O
N
+
+ 4 + 5
NH
X
NH
X
NH2
Sc(OTf)3
CH3CN
Sc(OTf)3
CH3CN
3
5
7
CHO
6
8
X
DDQ
CHCl3
4
R2
R2
(for 11)
DDQ, CHCl3
R2
H
or MnO2, pyridine, toluene
O
N
CO2Et
CO2Et
Ph
R1
N
R1
N
1) a) KOH, MeOH; b) HCl, Et2O
2) a) ClCO2Et, Et3N, CH2Cl2; b) EtNH2·HCl
O
N
(for 10)
N
X
N
X
N
X
Zn(OAc)2
(EtO)3SiH
THF
14, R1 = CH2Ph
9
15, R1 = Boc
16, R1 = H
R2
R2
R2
10, R1 = CH2Ph
11, R1 = Boc
12, R1 = H
LiAlH4
THF
HCl
NaH, DMSO
ClCH2(p-MeO-Ph)
dioxane
H
N
13, R1 = CH2(p-MeO-Ph)
R1
N
N
X
4, 6, 8-18a, R2 = Cl; X = CH
4, 6, 8-18b, R2 = H; X = N
17, R1 = CH2Ph
18, R1 = H
4, 6, 8-18c, R2 = CO2Me; X = CH
R2
Scheme 1. Synthesis of the target 1,2,3,4-tetrahydrobenzo[h][1,6]naphthyridines.
chloromethyl-4-methoxybenzene. In turn, acidic deprotection of
11b afforded the target benzonaphthyridine 12b in 76% yield.
Derivatization at position 9 was carried out following standard
procedures. Thus, ethyl carboxylic esters 10a, 11a and 12b were
converted into the corresponding N-ethylcarboxamides 14a, 15a
and 16b, respectively, in moderate overall yields, by alkaline hy-
drolysis followed by reaction of the resulting carboxylic acids, iso-
lated as naphthyridine hydrochlorides, with ethyl chloroformate in
the presence of Et3N, and reaction of the mixed anhydrides with
ethylamine (Scheme 1). Finally, LiAlH4 reduction of the amides 14a
and 16b afforded the amines 17a and 18b in 44% and 51% yields,
respectively, whereas reaction of the N-Boc-protected amide 15a
with LiAlH4 proceeded with both N-Boc-deprotection and reduc-
tion of the amide, directly affording the target N-Boc-deprotected
benzonaphthyridine 18a in 35% yield, together with a small amount
of the N-Boc-deprotected amide 16a (14% yield).
Ellman et al. [15]. Another cholinesterase that seems to play an
important role in the cognitive decline associated to AD is butyr-
ylcholinesterase (BChE). BChE exerts a compensatory effect in
response to the decrease of AChE in CNS as AD progresses, thereby
making dual inhibition of AChE and BChE a desirable property for
anti-Alzheimer drugs [16]. Thus, the inhibitory activity of these
compounds against human serum BChE (hBChE) was determined
as well [15].
In general, the novel first-generation benzonaphthyridines were
found to be moderately potent inhibitors of EeAChE, with IC50
values ranging from the submicromolar to the low micromolar
range (Table 1). The best substitution pattern at position 1 clearly
involves the presence of an unsubstituted secondary amino group,
compound 12a being 23- and >107-fold more potent EeAChE in-
hibitor than the N-benzylated and N-(4-methoxy)benzylated
counterparts 10a and 13a, respectively. With the sole exception of
12a, the rest of first generation benzonaphthyridines are N-ben-
zylated derivatives, among which two additional structureeactivity
relationship (SAR) trends leading to a higher EeAChE inhibitory
activity could be derived, namely the presence of a 3-pyridyl and an
ethylaminomethyl substituent at positions 5 and 9, respectively.
Thus, the 5-(3-pyridyl)-substituted ester 10b is 3-fold more potent
2.2. Biological activity assays
The inhibitory activity of the novel 1,2,3,4-tetrahydrobenzo[h]
[1,6]naphthyridines against Electrophorus electricus AChE (EeAChE)
and human recombinant (hAChE) was evaluated by the method of

144
O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
Table 1
two similar structural classes of inhibitors featuring small changes
in their aromatic rings have been recently reported [17].
Overall, amide 16a turned out to be the most potent benzo-
naphthyridine of the whole series as EeAChE inhibitor, exhibiting a
nanomolar IC50 value (46 nM).
Inhibitory activity of 1,2,3,4-tetrahydrobenzo[h][1,6]naphthyridines against AChE
and BChE.a
Compound
EeAChE
IC50 M)
hAChE
IC50 M)
hBChE IC50 (mM) or
% inhibition at 30 mM
(
m
(
m
First generation
9a
9b
9c
10a
10b
10c
12a
When tested on hAChE, significant inter-species differences
relative to EeAChE were found in some cases, even though similar
general SAR trends were observed. Thus, benzonaphthyridines
bearing an amide or an amine functionality in the side chain at
position 9 were found to be potent inhibitors, with IC50 values in
the submicromolar range in all cases except for the 5-(3-pyridyl)-
substituted derivative 18b, whereas most ester derivatives were
weakly active (Table 1). Among the most potent derivatives, the
higher hAChE inhibitory activity was associated to the presence of
an unsubstituted secondary amino group at position 1 (16a and 18a
being 12- and 2-fold more potent than 14a and 17a), an amide at
position 9 (14a and 16a being 1.2- and 9-fold more potent than the
amines 17a and 18a) and a 4-chlorophenyl substituent at position 5
(18a being 28-fold more potent than 18b). Again, benzonaphthyr-
idine 16a turned out to be the most interesting compound of the
series, emerging as a very potent inhibitor of hAChE (IC50 65 nM).
Noteworthy, 16a is 500-fold more potent than the specific PAS in-
hibitor propidium and 6-fold more potent than the active site in-
hibitor tacrine, the second most potent hAChEI among the
approved anti-Alzheimer drugs.
5.21 ꢀ 0.33
>30b
4.15 ꢀ 0.16
13.0 ꢀ 0.8
>25c
8.41%
4.42%
5.54%
8.68%
nde
6.61%
20.3%
12.4%
25.3%
nde
13.6 ꢀ 1.8
6.33 ꢀ 0.96
1.97 ꢀ 0.17
6.62 ꢀ 0.62
0.281 ꢀ 0.031
>30h
>25d
1.06 ꢀ 0.09
>25f
>25g
13a
14a
17a
>25i
5.48 ꢀ 0.51
0.801 ꢀ 0.069
0.147 ꢀ 0.014
0.942 ꢀ 0.038
Second generation
12b
16a
18a
18b
Tacrine
Propidium
0.148 ꢀ 0.017
0.046 ꢀ 0.006
0.532 ꢀ 0.030
2.15 ꢀ 0.20
nde
22.8 ꢀ 1.6
27.5%
0.065 ꢀ 0.003
0.556 ꢀ 0.024
15.8 ꢀ 0.9
0.92 ꢀ 0.03
1.37 ꢀ 0.07
2.59 ꢀ 0.14
0.046 ꢀ 0.003j
13.2 ꢀ 0.4j
0.424 ꢀ 0.021j
nde
32.3 ꢀ 2.2j
a
IC50 inhibitory concentration (
m
M) of Electrophorus electricus or human recom-
M) or % inhibition at 30 M of
binant AChE and IC50 inhibitory concentration (
m
m
human serum BChE. IC50 values are expressed as mean ꢀ standard error of the mean
(SEM) of at least four experiments (n ¼ 4), each performed in duplicate.
b
43.7% Inhibition of EeAChE activity at 30
mM.
c
d
e
f
13.8% Inhibition of hAChE at 25
10.0% Inhibition of hAChE at 25
Not determined.
15.7% Inhibition of hAChE at 25
17.8% Inhibition of hAChE at 25
33.7% Inhibition of EeAChE activity at 30
10.2% Inhibition of hAChE at 25 M.
Data taken from Ref. [18], involving the same experimental conditions.
m
m
M.
M.
Regarding the inhibition of hBChE, most compounds displayed
very weak inhibitory activity (4e28% inhibition at 30 mM). Inter-
estingly, the N-debenzylated amide or amine derivatives 16a, 18a,
and 18b were found to be more potent inhibitors of hBChE, with
m
m
M.
M.
g
h
i
mM.
IC50 values around 1e3
propidium but less potent than tacrine.
Inhibition of A aggregation is another valuable property for
mM (Table 1), they being more potent than
m
j
b
anti-Alzheimer compounds, which is additionally investigated in
many cholinesterase inhibitors [19]. Unfortunately, these benzo-
that its 5-(4-chlorophenyl)- and 5-(4-methoxycarbonylphenyl)-
substituted analogues 10a and 10c, whereas the benzonaphthyr-
idine 17a, bearing an amine functionality in the side chain at po-
sition 9, is about 40-fold more potent than the ester and amide
derivatives 10a and 14a (Table 1). Not unexpectedly, the benzo-
naphthyridines 10b,c are more potent EeAChE inhibitors than their
less basic lactam precursors 9b,c (>15- and 2-fold, respectively),
with the exception of compounds 10a and 9a, which were roughly
equipotent.
naphthyridines turned out to be rather weak inhibitors of A
self-aggregation, displaying percentages of inhibition up to 16% at
10 M (data not shown).
Overall, benzonaphthyridine 16a emerges as a promising anti-
b42
m
Alzheimer agent, by virtue of its dual potent hAChE and hBChE
inhibitory activities. Because interactions of ligands at the PAS of
AChE are not as tight as those that can be established at the active
site, peripheral site AChEIs do not usually display high affinities and
potencies [20]. Thus, the potent hAChE inhibitory activity of some
of the benzonaphthyridines, particularly 16a, which were designed
as peripheral site AChEIs, was somehow astonishing. To shed light
on the binding mode of these compounds within AChE, a set of
mechanistic studies was performed, encompassing kinetic experi-
ments (LineweavereBurk and CornisheBowden plots), propidium
displacement assay, and molecular modelling studies (docking and
molecular dynamics simulations, and Solvated Interaction Energy
calculations).
In the light of the SAR derived from the first-generation ben-
zonaphthyridines, starting from 10a we designed a second gener-
ation of analogues bearing simultaneously several or all of the
structural features that were found to lead to higher EeAChE
inhibitory activity, all of them N-debenzylated at position 1 and
additionally bearing either a 3-pyridyl group at position 5 (12b) or
an ethylaminoethyl chain at position 9 (18a) or both groups (18b).
To further explore the role of N-debenzylation at position 1, com-
pound 16a, the debenzylated analogue of the amide 14a, was also
included among the second generation benzonaphthyridines.
With the exception of 18a, which is 4-fold less potent than its N-
benzylated analogue 17a, the second-generation N-debenzylated
benzonaphthyridines 12b and 16a were clearly more potent than
their N-benzylated counterparts 10b and 14a (13- and 119-fold,
respectively). However, the two SAR trends seen for the first-
generation N-benzylated benzonaphthyridines were not apparent
in the second-generation N-debenzylated analogues, in which the
presence of a 4-chlorophenyl group at position 5 and an amide
functionality at the side chain in position 9 were the structural
features leading to an optimal EeAChE inhibitory activity. These
results seem to suggest a different orientation of the N-benzylated
and N-debenzylated compounds within EeAChE. Worthy of note,
divergent SARs and binding modes of the PAS-binding moiety of
2.3. Kinetic studies
To investigate the mode of inhibition of the most active AChEI
benzonaphthyridine, 16a, and its N-benzylated analogue 14a,
LineweavereBurk double reciprocal plots were generated. The
interception of the lines in the LineweavereBurk plot above the x-
axis (Fig. 2) demonstrated that both compounds serve as mixed-
type inhibitors of AChE. Mixed-type of inhibition was further
confirmed by CornisheBowden plots (S/v versus [I]) [21].
The inhibition constant (Ki) and the K0i (dissociation constant for
the enzymeesubstrateeinhibitor complex) estimated for 14a were
0.785
0.073
m
m
M and 2.34
mM, respectively, and for 16a were 0.065 mM and
M, respectively. These findings show that introduction of a

O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
145
benzyl substituent at position 1 decreases the affinity not only for
the enzyme active site (12-fold higher dissociation constant of the
EI complex) but likely also for the PAS (32-fold higher K0i value).
Moreover, the similar values of Ki and K0i found for 16a suggest that
the high inhibitory potency of this compound might arise from the
ability to tightly bind both sites, and not only at the PAS.
To get further insights into the mechanism of inhibition and
confirm the ability to interact with the AChE PAS, the affinity of the
four most interesting derivatives (14a, 16a, 17a, and 18a) for the PAS
of AChE was investigated by displacement studies with propidium
iodide, using the method of Taylor et al. [22]. Propidium selectively
associates with the PAS of AChE exhibiting an eight-fold enhance-
ment of fluorescence [22a,23]. Back-titration experiments with
increasing concentration of all selected compounds but 14a showed
a concentration-dependent decrease in the fluorescence intensity
associated with the propidiumeAChE complex, suggesting that
they can effectively displace propidium from the AChE’s PAS
(Fig. 3A). Solubility of 14a in the assay conditions was insufficient to
allow a full back-titration experiment. At 1/1 ratio with propidium
iodide 14a was able to reduce fluorescence intensity associated
with the AChEepropidium complex by only 5%, confirming a lower
affinity for the PAS than the N-debenzylated analogue 16a, as also
suggested by the K0i values. In general, the affinity trend was
16a > 18a > 17a with concentrations required for decreasing initial
fluorescence
intensity
of
AChEepropidium
complex
([propidium] ¼ 8
m
M) equal to 13, 23, and 33 m
M, respectively.
Back-titration experiments for the most active derivative in the
series, 16a, predicted a dissociation constant of 1.76
This value is consistent with a quite tight binding to the PAS, only
2.5-fold weaker than that of propidium (KD on EeAChE ¼ 0.7
[23]). Values for the other tested analogues were slightly higher,
being 2.18 and 3.20 M for 18a and 17a, respectively. The slightly
lower value obtained for the N-benzylated derivative, 17a, further
confirms an unfavourable effect of the benzyl substituent at posi-
tion 1.
mM (Fig. 3B).
Fig. 3. (A) Back-titration of the propidiumeAChE complex by compounds 16a, 17a, and
18a (2.0 mM EeAChE, 8.0 mM propidium, Tris HCl 1.0 mM, pH 8.0); (B) determination of
mM
KD value for most active derivative 16a. KD value is calculated from the antilog of the Y-
intercept value [22b]. P stands for propidium iodide and I stands for tested inhibitor; Fe
is the initial fluorescence intensity when enzyme sites are saturated with P, FP is the
fluorescence intensity when propidium is completely displaced from the enzyme, and
F denotes the fluorescence intensity after adding a determined amount of displacing
agent during the titration experiment.
m
of a number of AChE inhibitors to the enzyme [9]. Docking was
performed using three models of hAChE that differ in the orienta-
tion of Trp286, which was arranged to reflect the three major
conformations adopted by this residue upon inspection of the
available X-ray crystallographic structures (see Experimental part)
[25].
2.4. Molecular modelling studies
The binding of compounds 16a and 18a to hAChE (Fig. 4) was
firstly explored by docking calculations carried out with rDock [24].
It is worth noting that previous studies strongly support the per-
formance of this docking program for predicting the binding mode
Fig. 2. LineweavereBurk plots illustrating mixed-type inhibition of AChE-mediated acetylthiocholine hydrolysis by compound (A) 14a and (B) 16a. ATCh ¼ acetylthiocholine;
v ¼ initial velocity rate.

146
O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
The docking results revealed a preferential binding to the AChE
To further validate the binding mode of 16a and 18a, the binding
affinities were determined using the Solvated Interaction Energy
(SIE) calculations. The SIE method relies on MM/PBSA calculations
of the ligandereceptor complex, but the free energy components
are weighted by a scaling factor parameterized to reproduce the
experimental binding affinities for a diverse set of proteineligand
complexes [26]. The predicted binding affinities for 16a and 18a
are ꢁ8.5 ꢀ 0.4 and ꢁ8.8 ꢀ 0.6 kcal/mol (see Table S1 in
Supplementary material), which compare with the experimental
value determined from the inhibition constant for 16a (ꢁ9.7 kcal/
mol). Thus, within the uncertainty of the SIE method, the predicted
binding affinities reflect the similar inhibitory potency of 16a and
18a. Since they exhibit a similar binding mode, it can be concluded
that the large desolvation penalty of the protonated amine present
in the side chain at position 9 counterbalances the enhanced
coulombic stabilization found for 18a, the net effect leading to an
inhibitory potency close to the potency of the amide derivative 16a.
model where Trp286 retains the orientation found in the AChEe
propidium complex (PDB ID 1N5R [20b]). This finding is not un-
expected keeping in mind the size of the heteropolycyclic ring
system present in compounds 16a and 18a and in propidium. Thus,
a distinctive binding mode was clearly identified on the basis of the
most populated cluster of docked poses and the docking score,
where the 5-(4-chlorobenzyl) substituent of 16a and 18a stacks
against the indole ring of Trp286 and the CONHEt (16a) and
CH2NH2Et (18a) substituents penetrate along the gorge towards the
catalytic site.
A 100 ns MD simulation was run to refine the binding mode of
compounds 16a and 18a. Simulations yielded structurally and
energetically stable trajectories, which showed an initial rear-
rangement of the ligand without significant alterations in the res-
idues that shape the binding site (see Fig. S1 in Supplementary
material). The results obtained for compound 16a point out that
the central pyridine ring of the benzonaphthyridine system stacks
ꢀ
against Trp286 (average distance of 3.74 A; Fig. 4A), thus enabling
2.5. Bloodebrain barrier permeation assay
the cationep interaction between the protonated pyridine nitrogen
atom of 16a and the indole system of Trp286. Furthermore, binding
is assisted by the formation of hydrogen bonds between the pyri-
dine nitrogen atom and the hydroxyl group of Tyr72 (average dis-
Brain penetration is an essential property for every anti-
Alzheimer drug candidate. The ability of the synthesized benzo-
naphthyridines to cross the bloodebrain barrier (BBB) and there-
fore to reach the CNS was assessed using the known parallel
artificial membrane permeation assay (PAMPA-BBB) as an in vitro
model of passive transcellular permeation [27]. The in vitro
permeability (Pe) of the novel 1,2,3,4-benzo[h][1,6]naphthyridines
through a lipid extract of porcine brain was determined using a
mixture of phosphate-buffered saline (PBS)/EtOH (70:30). Assay
validation was carried out by comparison of the experimental and
reported permeability values of 14 commercial drugs (see Table S2
in Supplementary material), which provided a good linear corre-
lation: Pe (exp) ¼ 1.4974 Pe (lit) ꢁ 0.8434 (R2 ¼ 0.9428). Using this
equation and the limits established by Di et al. for BBB permeation
[27], the following ranges of permeability were established: Pe
(10ꢁ6 cm sꢁ1) > 5.1 for compounds with high BBB permeation
(CNSþ); Pe (10ꢁ6 cm sꢁ1) < 2.15 for compounds with low BBB
permeation (CNSꢁ); and 5.1 > Pe (10ꢁ6 cm sꢁ1) > 2.15 for com-
pounds with uncertain BBB permeation (CNSþ/ꢁ). All the tested
benzonaphthyridines were predicted to be able to cross the BBB,
with the exception of amine 18b, for which an uncertain brain
penetration was predicted. Indeed, amines 18b and 18a, whose
permeability value was near the minimum threshold for high BBB
permeation, seemed to be the most polar benzonaphthyridines of
the series, which, as mentioned above might account for the
apparently high desolvation penalty detrimental for their AChE
inhibitory potencies (Table 2).
ꢀ
tance of 3.20 A) and between the amide NH group and the hydroxyl
ꢀ
group of Tyr124 (average distance of 3.17 A). Compared to 16a,
binding of 18a involves the formation of a water-mediated bridge
between the protonated amine of the side chain at position 9 and
ꢀ
Asp72 (average distance of 5.9 A) and a catione
p
interaction with
ꢀ
the benzene ring of Tyr341 (average distance of 3.54 A), besides the
catione interaction with Trp286. Finally, the NH group at position
p
1 forms water-mediated hydrogen bonds with the carbonyl groups
of Ser293 and Phe338. Clearly, the similar binding mode of com-
pounds 16a and 18a (Fig. 4B) must be drastically perturbed by the
attachment of the N-benzyl group at position 1 due to the steric
clash with the neighbouring residues, which likely explains the
weaker potency measured for compounds 14a and 17a. Overall, MD
simulations show that compounds 16a and 18a are capable of
forming a network of diverse interactions with residues at the PAS
and midgorge sites.
3. Conclusion
We have carried out the optimization of the AChE PAS-binding
affinity of the initial hit 1, ethyl 5-(4-chlorophenyl)-3,4-dihydro-
2H-pyrano[3,2-c]quinoline-9-carboxylate, neutral at physiological
pH, by replacement of the oxygen atom at position 1 by a nitrogen
atom [of a NeH, N-benzyl or N-(4-methoxybenzyl) group]. The
main aim of this structural change was to increase the basicity, and
therefore the protonation ability, of the resulting 1,2,3,4-
tetrahydrobenzo[h][1,6]naphthyridine derivatives, thereby mak-
ing it possible the establishment of catione
p interactions besides
pep
stacking with the PAS residue Trp286. Moreover, the effect on
Fig. 4. (A) Representation of the binding mode of compound 16a (in orange) obtained
at the end of the 100 ns MD trajectory. The side chains or backbone units of the res-
idues involved in interactions are shown as green-coloured sticks. Water molecules
that mediate interactions of the ligand are shown as red spheres. Propidium is shown
as grey sticks. (B) Superposition of the compounds 16a (orange) and 18a (cyan) as
found at the end of the MD trajectories. (For interpretation of the references to colour
in this figure legend, the reader is referred to the web version of this article.)
the AChE inhibitory activity of replacements of the 4-chlorophenyl
and ethyl carboxylate groups at positions 5 and 9, present in 1, by 4-
(methoxycarbonyl)phenyl or 3-pyridyl groups at position 5 and by
N-ethylcarboxamido or ethylaminomethyl groups at position 9
were investigated to explore potential additional interactions

O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
147
nearby the PAS. Despite some significant inter-species differences,
the substitution pattern leading to a higher inhibitory activity both
in EeAChE and hAChE involves the presence of a debenzylated ni-
trogen atom at position 1, and 4-chlorophenyl and N-ethyl-
carboxamido groups at positions 5 and 9, respectively. Overall, the
hit-to-lead optimization process from 1 to 16a simply involves a
double bioisosteric O / NH replacement at position 1 and in the
side chain at position 9, but results in a dramatic increase in EeAChE
(>217-fold) and hAChE (>154-fold) inhibitory activities. Interest-
ingly, such a change also leads to a noticeable increase in hBChE
inhibitory activity (>11-fold). Because most AChE PAS inhibitors
exhibit potencies in the micromolar range, the very potent hAChE
inhibitory activity of the lead 16a (IC50 ¼ 65 nM) might arise from
additional interactions other than those established with PAS res-
idues, as supported by the results derived from a comprehensive
mechanistic study. On the one hand, kinetic studies and propidium
displacement studies have confirmed the ability of 16a to tightly
bind the AChE PAS. On the other hand, molecular modelling studies
have suggested the ability of the heteroaromatic system of 16a to
Thin-layer chromatography was performed with aluminium-
backed sheets with silica gel 60 F254 (Merck, ref 1.05554), and
spots were visualized with UV light and 1% aqueous solution of
KMnO4. NMR spectra of all of the new compounds were per-
formed at the Centres Científics i Tecnològics of the University of
Barcelona (CCiTUB), while elemental analyses and high resolution
mass spectra were carried out at the Mycroanalysis Service of the
IIQAB (CSIC, Barcelona, Spain) with a Carlo Erba model 1106
analyzer, and at the CCiTUB with a LC/MSD TOF Agilent Technol-
ogies spectrometer, respectively. The HPLC measurements were
performed using a HPLC Waters Alliance HT apparatus comprising
a pump (Edwards RV12) with degasser, an autosampler, a diode
array detector and a column as specified below. The reverse phase
HPLC determinations were carried out on a YMC-Pack ODS-AQ
column (50 ꢂ 4.6 mm, D S. 3
mm, 12 nm). Solvent A: water with
0.1% formic acid; Solvent B: acetonitrile with 0.1% formic acid.
Gradient: 5% of B to 100% of B within 3.5 min. Flux: 1.6 mL/min at
50 ꢃC. The analytical samples of all of the compounds that were
subjected to pharmacological evaluation were dried at 65 ꢃC/
2 Torr (standard conditions) and possess a purity ꢄ95% as evi-
denced by their elemental analyses and/or HPLC measurements.
The synthetic procedures for the preparation of the intermediate
and target compounds are exemplified through the synthesis of
the most potent compound of the series, 16a. The synthesis of the
rest of compounds is included in the Supplementary material.
establish cationep and pep interactions with the PAS residue
Trp286 but also the ability of the amide functionality at position 9
to penetrate along the gorge towards the catalytic site and to
establish additional hydrogen bond interactions with AChE midg-
orge residues. The tight binding of 16a to the AChE PAS and the
additional midgorge interactions seem to account for its very
potent hAChE inhibitory activity. Overall, the potent dual hAChE
and hBChE inhibitory activities of 16a make it a very interesting
anti-Alzheimer lead compound.
4.1.1. Ethyl 1-(tert-butoxycarbonyl)-5-(4-chlorophenyl)-1,2,3,4,4a
,5,6,10b-octahydrobenzo[h][1,6]naphthyridine-9-carboxylate,
diastereomeric mixture 8a
To a stirred solution of p-chlorobenzaldehyde, 4a (1.36 g,
9.67 mmol) and ethyl 4-aminobenzoate, (1.60 g, 9.69 mmol) in
anhydrous CH3CN (25 mL), 4 A molecular sieves and Sc(OTf)3
4. Experimental part
ꢀ
4.1. Chemistry. General methods
(0.95 g, 1.93 mmol) were added. The mixture was stirred at room
temperature under argon atmosphere for 5 min and then treated
with a solution of enamine 7 (1.80 mL, 1.78 g, 9.70 mmol) in
anhydrous CH3CN (12 mL). The resulting suspension was stirred
at room temperature under argon atmosphere for 3 days. Then,
the resulting mixture was diluted with sat. aq. NaHCO3 (150 mL)
and extracted with EtOAc (3 ꢂ 200 mL). The combined organic
extracts were dried over anhydrous Na2SO4, filtered and evapo-
rated under reduced pressure to give a solid residue (4.75 g),
Melting points were determined in open capillary tubes with a
MFB 595010M Gallenkamp melting point apparatus. 400 MHz 1H/
100.6 MHz 13C NMR spectra were recorded on a Varian Mercury
400 spectrometer. The chemical shifts are reported in ppm (
d
scale) relative to internal tetramethylsilane, and coupling con-
stants are reported in Hertz (Hz). Assignments given for the NMR
spectra of the new compounds have been carried out by com-
parison with the NMR data of 9c, 10c, 11a, 17a, and 18b, which in
turn, were assigned on the basis of DEPT, COSY 1H/1H (standard
procedures), and COSY 1H/13C (gHSQC or gHMBC sequences) ex-
periments. IR spectra were run on a PerkineElmer Spectrum RX I
or on a Thermo Nicolet Nexus spectrophotometer. Absorption
values are expressed as wave-numbers (cmꢁ1); only significant
absorption bands are given. Column chromatography was per-
formed on silica gel 60 AC.C (35e70 mesh, SDS, ref 2000027).
which was purified by column chromatography (35e70 mm silica
gel, hexane/EtOAc mixtures, gradient elution). On elution with
hexane/EtOAc 80:20 to 70:30, the diastereomeric mixture 8a
(2.86 g, 63% yield, 3:1 diastereomeric ratio (1H NMR)) was iso-
lated as a white solid.
4.1.2. Ethyl 1-(tert-butoxycarbonyl)-5-(4-chlorophenyl)-1,2,3,4-
tetrahydrobenzo[h][1,6]naphthyridine-9-carboxylate 11a
To a solution of diastereomeric mixture 8a (1.41 g, 2.99 mmol) in
anhydrous CHCl3 (37 mL), DDQ (1.36 g, 5.99 mmol) was added. The
reaction mixture was stirred at room temperature under argon
atmosphere overnight, diluted with CH2Cl2 (150 mL) and washed
with sat. aq. NaHCO3 (3 ꢂ 200 mL). The combined organic extracts
were dried over anhydrous Na2SO4, filtered and evaporated under
reduced pressure to give an orange solid residue (1.46 g), which was
Table 2
Permeability values and predicted brain penetration of the novel 1,2,3,4-benzo[h]
[1,6]naphthyridines from the PAMPA-BBB assay.
a
Compound
Pe (10ꢁ6 cm sꢁ1
)
Prediction
9a
9b
9c
13.3 ꢀ 3.75
12.2 ꢀ 1.54
16.8 ꢀ 1.29
8.10 ꢀ 1.13
9.50 ꢀ 1.05
9.70 ꢀ 1.03
14.6 ꢀ 1.15
26.0 ꢀ 3.87
7.70 ꢀ 0.94
22.9 ꢀ 0.78
5.60 ꢀ 0.58
2.40 ꢀ 0.73
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ
CNSþ/ꢁ
purified through column chromatography (35e70
mm silica gel,
10a
10c
12a
12b
13a
14a
16a
18a
18b
hexane/EtOAc mixtures, gradient elution). On elution with hexane/
EtOAc 80:20, compound 11a (1.25 g, 90% yield) was isolated as a
white solid; Rf 0.61 (hexane/EtOAc 1:1).
A solution of 11a (100 mg, 0.21 mmol) in CH2Cl2 (8 mL) was
filtered through a 0.2 mm PTFE filter and evaporated at reduced
pressure. The solid was washed with pentane (3 ꢂ 4 mL) to give,
after drying under standard conditions, the analytical sample of 11a
a
Values are expressed as the mean ꢀ SD of three independent experiments.
(97 mg): mp 154e155 ꢃC (CH2Cl2); IR (KBr)
n 1713, 1697 (C]O st),

148
O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
1618, 1592, 1577, 1566 (AreCeC and AreCeN st) cmꢁ1
;
1H NMR
1.39 [s, 9H, C(CH3)3], 1.43 (t, J ¼ 7.2 Hz, 3H,
4.1.4. N-{{5-(4-Chlorophenyl)-1,2,3,4-tetrahydrobenzo[h][1,6]
naphthyridin-9-yl}methyl}ethanamine 18a and 5-(4-chlorophenyl)-
N-ethyl-1,2,3,4-tetrahydrobenzo[h][1,6]naphthyridine-9-
(400 MHz, CDCl3)
d
CO2CH2CH3), 2.00 (br signal, 2H, 3-H2), 2.80 (m, 2H, 4-H2), 3.20e
3.60 (br signal, 2H, 2-H2), 4.45 (q, J ¼ 7.2 Hz, 2H, CO2CH2CH3), 7.48
[ddd, J ¼ 8.4 Hz, J0 z J00 z 2.0 Hz, 2H, 5-AreC3(5)eH], 7.55 [ddd,
J z 8.4 Hz, J0 z J00 z 2.0 Hz, 2H, 5-AreC2(6)eH], 8.08 (d, J z 8.8 Hz,
1H, 7-H), 8.24 (dd, J ¼ 8.8 Hz, J0 ¼ 1.6 Hz 1H, 8-H), 8.59 (d, J ¼ 1.6 Hz,
carboxamide 16a
A solution of amide 15a (0.65 g, 1.39 mmol) in anhydrous THF
(32 mL) was cooled to 0 ꢃC with an ice bath, and treated portion-
wise with LiAlH4 (0.17 g, 4.48 mmol). The resulting suspension was
stirred under reflux overnight, cooled to 0 ꢃC with an ice bath and
treated dropwise with 1 N NaOH (20 mL), then diluted with H2O
(25 mL), and extracted with EtOAc (3 ꢂ 30 mL). The combined
organic extracts were dried over anhydrous Na2SO4, filtered and
evaporated under reduced pressure to give a solid residue (0.51 g),
1H, 10-H); 13C NMR (100.6 MHz, CDCl3)
d 14.3 (CH3, CO2CH2CH3),
24.1 (CH2, C3), 25.4 (CH2, C4), 27.9 [3CH3, C(CH3)3], 44.7 (CH2, C2),
61.3 (CH2, CO2CH2CH3), 82.1 [C, C(CH3)3], 122.9 (C, C10a), 123.8 (C,
C4a), 127.0 (C, C9),127.6 (CH, C10), 128.2 (CH, C8),128.6 [2CH, 5-Are
C3(5)], 129.7 (CH, C7), 130.3 [2CH, 5-AreC2(6)], 134.8 (C, 5-AreC4),
138.3 (C, 5-AreC1), 145.7 (C, C10b), 148.8 (C, C6a), 153.9 (C, NCOO),
160.5 (C, C5), 166.3 (C, CO2CH2CH3); HRMS (ESI), calcd for
½C26H2735ClN2O4 þ Hþꢅ 467.1732, found 467.1723.
which was purified through column chromatography (35e70 mm
silica gel, CH2Cl2/MeOH/50% aq. NH4OH mixtures, gradient elution).
On elution with CH2Cl2/MeOH/50% aq. NH4OH 99:1:0.2, N-Boc-
deprotected amide 16a (72 mg, 14% yield) was isolated as a
yellowish solid. On elution with CH2Cl2/MeOH/50% aq. NH4OH
97:3:0.2 to 95:5:0.2, N-Boc-deprotected amine 18a (170 mg, 35%
yield) was isolated as a yellowish solid; Rf(16a) 0.15 (CH2Cl2/MeOH/
NH4OH 9:1:0.05); Rf(18a) 0.49 (CH2Cl2/MeOH/NH4OH 9:1:0.05).
A solution of 16a (64 mg, 0.17 mmol) in CH2Cl2 (5 mL) was
4.1.3. 1-(tert-Butoxycarbonyl)-5-(4-chlorophenyl)-N-ethyl-1,2,3,4-
tetrahydrobenzo[h][1,6]naphthyridine-9-carboxamide 15a
A suspension of ester 11a (2.54 g, 5.44 mmol) and KOH (85%
purity, 1.08 g, 16.3 mmol) in MeOH (140 mL) was stirred under
reflux for 24 h. The resulting solution was cooled down at room
temperature and concentrated under reduced pressure. The solid
residue (3.35 g) was treated with a solution of HCl in Et2O (0.8 N,
138 mL, 110 mmol) and the resulting suspension was concentrated
under reduced pressure to give the corresponding aminoquinoline
carboxylic acid, in the form of hydrochloride, as a white solid
(3.77 g). This crude product was used in the next step without
further purification.
A solution of this crude product (3.60 g) in anhydrous CH2Cl2
(45 mL) was cooled to 0 ꢃC with an ice bath and treated dropwise
with freshly distilled Et3N (2.89 mL, 2.10 g, 20.7 mmol) and ClCO2Et
(0.49 mL, 556 mg, 5.12 mmol). The resulting suspension was thor-
oughly stirred at 0 ꢃC for 30 min and treated with EtNH2$HCl
(0.42 g, 5.15 mmol). The reaction mixture was stirred at room
temperature for 3 days, diluted with 10% aq. Na2CO3 (200 mL), and
extracted with CH2Cl2 (3 ꢂ 300 mL). The combined organic extracts
were washed with H2O (3 ꢂ 200 mL), dried over anhydrous Na2SO4,
filtered and concentrated under reduced pressure to give a solid
residue (2.36 g), which was purified through column chromatog-
filtered through a 0.2 mm PTFE filter and treated with a methanolic
solution of HCl (0.75 N, 2.2 mL, 1.65 mmol). The resulting solution
was evaporated at reduced pressure and the solid was washed with
pentane (3 ꢂ 4 mL) to give, after drying under standard conditions,
16a$HCl (57 mg) as a brown solid: mp 320e321 ꢃC (CH2Cl2/MeOH
69:31); IR (ATR)
2929, 2865, 2810, 2640, þNH, NH, OH and CH st), 1655, 1647, 1629,
1586, 1545 (C]O, AreCeC and AreCeN st) cmꢁ1 1H NMR
(400 MHz, CD3OD)
n 3500e2500 (max at 3395, 3231, 3090, 3028,
;
d
1.29 (t, J ¼ 7.2 Hz, 3H, CONHCH2CH3), 1.99 (tt,
J z J0 z 6.0 Hz, 2H, 3-H2), 2.75 (t, J z 6.0 Hz, 2H, 4-H2), 3.49 (q,
J ¼ 7.2 Hz, 2H, CONHCH2CH3), 3.71 (t, J ¼ 5.6 Hz, 2H, 2-H2), 4.84 (s,
NH and þNH), 7.66 [complex signal, 4H, 5-AreC2(6)eH and 5-Are
C3(5)eH], 7.85 (d, J z 9.2 Hz, 1H, 7-H), 8.24 (dd, J ¼ 9.2 Hz,
J0 ¼ 1.6 Hz, 1H, 8-H), 8.86 (d, J ¼ 1.6 Hz, 1H, 10-H); 13C NMR
(100.6 MHz, CD3OD)
d 14.8 (CH3, CONHCH2CH3), 20.0 (CH2, C3),
25.0 (CH2, C4), 36.2 (CH2, CONHCH2CH3), 43.0 (CH2, C2), 109.8 (C,
C4a), 116.2 (C, C10a), 120.9 (CH, C7), 123.6 (CH, C10), 130.4 (2CH),
131.8 (2CH) [5-AreC2(6) and 5-AreC3(5)], 132.3 (C, 5-AreC1), 132.5
(CH, C8), 133.8 (C, C9), 138.3 (C, 5-AreC4), 140.2 (C, C6a), 151.5 (C,
C5), 155.8 (C, C10b), 168.1 (C, CONHCH2CH3); HRMS (ESI), calcd for
raphy (35e70
mm silica gel, hexane/EtOAc mixtures, gradient
elution). On elution with hexane/EtOAc 60:40, amide 15a (1.22 g,
50% overall yield) was obtained as a white solid; Rf 0.23 (hexane/
EtOAc 1:1).
A solution of 15a (50 mg, 0.11 mmol) in CH2Cl2 (4 mL) was
filtered through a 0.2 mm PTFE filter and evaporated at reduced
h
i
C
21H2035ClN3O þ Hþ 366.1368, found 366.1364; Elemental
analysis, calcd for C21H20ClN3O$HCl$H2O C 60.01%, H 5.52%, N
10.00%, Cl 16.87%, found C 60.35%, H 5.81%, N 8.93%, Cl 16.05%. HPLC
purity: 94%.
pressure. The solid was washed with pentane (3 ꢂ 4 mL) to give,
A solution of 18a (106 mg, 0.30 mmol) in CH2Cl2 (8 mL) was
filtered through a 0.2 mm PTFE filter and treated with a methanolic
solution of HCl (0.75 N, 3.6 mL, 2.70 mmol). The resulting solution
was evaporated at reduced pressure and the solid was washed with
pentane (3 ꢂ 4 mL) to give, after drying under standard conditions,
18a$2HCl (96 mg) as a yellowish solid: mp 323e324 ꢃC (CH2Cl2/
after drying under standard conditions, the analytical sample of 15a
(45 mg) as a white solid: mp 203e204 ꢃC (CH2Cl2); IR (ATR)
n
3391,
3316 (NH st), 1711, 1687, 1654, 1639, 1617, 1597, 1583, 1568, 1532
(C]O, AreCeC and AreCeN st) cmꢁ1 1H NMR (400 MHz, CDCl3)
;
d
1.27 (t, J ¼ 7.2 Hz, 3H, CONHCH2CH3), 1.40 [s, 9H, C(CH3)3], 1.99 (br
signal, 2H, 3-H2), 2.79 (t, J ¼ 6.4 Hz, 2H, 4-H2), 3.10e3.50 (br signal,
2H, 2-H2), 3.54 (tt, J ¼ 7.2 Hz, J0 z 5.2 Hz, 2H, CONHCH2CH3), 6.27 (t,
J ¼ 5.2 Hz, 1H, CONHCH2CH3), 7.47 [ddd, J ¼ 8.4 Hz, J0 z J00 z 2.0 Hz,
2H, 5-AreC3(5)eH], 7.53 [ddd, J z 8.4 Hz, J0 z J00 z 2.0 Hz, 2H, 5-
AreC2(6)eH], 7.96 (dd, J ¼ 8.4 Hz, J0 ¼ 2.0 Hz, 1H, 8-H), 8.08 (d,
J z 8.4 Hz, 1H, 7-H), 8.25 (br s, 1H, 10-H); 13C NMR (100.6 MHz,
MeOH 69:31); IR (KBr)
3028, 2926, 2863, 2767, 2667, 2552, 2422, þNH, NH and CH st),
1639, 1587, 1504 (AreCeC and AreCeN st) cmꢁ1 1H NMR
(400 MHz, CD3OD)
n 3500e2400 (max at 3379, 3198, 3095,
;
d
1.41 (t, J ¼ 7.6 Hz, 3H, 9-CH2NHCH2CH3), 1.99
(tt, J z J0 z 6.0 Hz, 2H, 3-H2), 2.76 (t, J ¼ 6.0 Hz, 2H, 4-H2), 3.23 (q,
J ¼ 7.6 Hz, 2H, 9-CH2NHCH2CH3), 3.73 (t, J ¼ 5.6 Hz, 2H, 2-H2), 4.44
(s, 2H, 9-CH2NHCH2CH3), 4.84 (s, NH and þNH), 7.67 [complex
signal, 4H, 5-AreC2(6)-H and 5-AreC3(5)eH], 7.92 (d, J ¼ 8.8 Hz,
1H, 7-H), 8.06 (dd, J ¼ 8.8 Hz, J0 ¼ 1.6 Hz, 1H, 8-H), 8.58 (d, J ¼ 1.6 Hz,
CDCl3)
d 14.9 (CH3, CONHCH2CH3), 24.0 (CH2, C3), 25.4 (CH2, C4),
28.0 [3CH3, C(CH3)3], 35.1 (CH2, CONHCH2CH3), 44.8 (CH2, C2), 82.1
[C, C(CH3)3], 123.0 (C, C10a), 124.0 (C, C4a), 124.2 (CH), 126.3 (CH)
(C8 and C10), 128.6 [2CH, 5-AreC3(5)], 129.9 (CH, C7), 130.3 [2CH,
5-AreC2(6)], 131.4 (C, C9), 134.7 (C, 5-AreC4), 138.3 (C, 5-AreC1),
145.4 (C, C10b), 148.0 (C, C6a), 154.0 (C, NCOO), 159.9 (C, C5), 167.0
(C, CONHCH2CH3); HRMS (ESI), calcd for ½C26H2835ClN3O3 þ Hþꢅ
466.1892, found 466.1887.
1H, 10-H); 13C NMR (100.6 MHz, CD3OD)
d 11.6 (CH3, 9-
CH2NHCH2CH3), 20.0 (CH2, C3), 25.1 (CH2, C4), 43.0 (CH2, C2),
44.2 (CH2, 9-CH2NHCH2CH3), 51.5 (CH2, 9-CH2NHCH2CH3), 109.8 (C,
C4a), 116.7 (C, C10a), 121.7 (CH, C7), 126.0 (CH, C10), 130.4 (2CH),
131.8 (2CH) [5-AreC2(6) and 5-AreC3(5)], 131.0 (C, C9), 132.3 (C, 5-

O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
149
AreC1), 135.5 (CH, C8), 138.2 (C, 5-AreC4), 139.1 (C, C6a), 151.4 (C,
C5), 155.2 (C, C10b); HRMS (ESI) calcd for ½C21H2235ClN3 þ Hþꢅ
352.1575, found 352.1574; Elemental analysis, calcd for
reciprocal plots were constructed at relatively low concentration of
substrate (0.11e0.55 mM) and using the same experimental con-
ditions reported for the hAChE assay at Section 4.2.1. The plots were
assessed by a weighted least square analysis that assumed the
C
21H22ClN3$1.5HCl$2.75H2O C 55.30%, H 6.41%, N 9.21%, Cl 19.43%,
found C 55.23%, H 6.16%, N 8.93%, Cl 19.10%. HPLC purity >99%.
variance of n to be a constant percentage of n for the entire data set.
To confirm the mode of inhibition, CornisheBowden plots were
obtained by plotting S/v (substrate/velocity ratio) versus inhibitor
concentration [21]. Data analysis was performed with GraphPad
Prism 4.03 software (GraphPad Software Inc.).
Calculation of the inhibitor constant (Ki) value was carried out
by re-plotting slopes of lines from the LineweavereBurk plot versus
the inhibitor concentration and Ki was determined as the intersect
on the negative x-axis. K0i (dissociation constant for the enzymee
substrateeinhibitor complex) value was determined by plotting the
apparent 1/vmax, app versus inhibitor concentration [29].
4.2. Biological assays
4.2.1. Determination of inhibitory effect on AChE and BChE activity
The inhibitory activity against E. electricus (Ee) AChE (Sigmae
Aldrich) and human serum BChE (SigmaeAldrich) was evaluated
spectrophotometrically by the method of Ellman et al. [15]. The
reactions took place in a final volume of 300
buffered solution pH 8.0, containing EeAChE (0.03 u/mL) or hBChE
(0.02 u/mL) and 333
M 5,50-dithiobis(2-nitrobenzoic) acid (DTNB;
mL of 0.1 M phosphate-
m
SigmaeAldrich) solution used to produce the yellow anion of 5-
thio-2-nitrobenzoic acid. Inhibition curves were performed in du-
plicates using at least 10 increasing concentrations of inhibitors and
pre-incubated for 20 min at 37 ꢃC before adding the substrate [28].
One duplicate sample without inhibitor was always present to yield
100% of AChE or BChE activities. Then substrates, acetylthiocholine
4.2.3. Propidium displacement studies
The affinity of selected inhibitors for the peripheral binding site
of EeAChE (type VI-S, SigmaeAldrich) was tested using propidium
iodide (P) (SigmaeAldrich), a known PAS-specific ligand as previ-
ously described by Taylor et al. [22]. A shift in the excitation
wavelength follows the complexation of propidium iodide and
AChE [22a]. Fluorescence intensity was monitored by a Jasco 6200
spectrofluorometer (Jasco Europe, Italy) using a 0.5 mL quartz
cuvette at room temperature. EeAChE (2
molecular mass subunits) was first incubated with 8
iodide in 1 mM TriseHCl, pH 8.0 at room temperature. Stock so-
lutions (4 mM) of each inhibitor were prepared in methanol. In the
back titration experiments of the propidiumeAChE complex by the
iodide (450
mM; SigmaeAldrich) or butyrylthiocholine iodide
(300 M; SigmaeAldrich), were added and the reaction was
m
developed for 5 min at 37 ꢃC. The colour production was measured
at 414 nm using a labsystems Multiskan spectrophotometer.
Data from concentrationeinhibition experiments of the in-
hibitors were calculated by non-linear regression analysis, using
the GraphPad Prism program package (GraphPad Software; San
Diego, USA), which gave estimates of the IC50 (concentration of
drug producing 50% of enzyme activity inhibition). Results are
expressed as mean ꢀ S.E.M. of at least 4 experiments performed in
duplicate.
m
M, assuming 82,000
mM propidium
tested inhibitor, aliquots of inhibitor (2e40
mM, 4e68 mM and 2e
44 M for 16a, 18a and 14a, respectively) were added successively,
m
and fluorescence emission was monitored at 602 nm upon excita-
tion at 535 nm. Blanks containing propidium alone, inhibitor plus
propidium and EeAChE alone were prepared and fluorescence
emission determined. Raw data were processed following the
method of Taylor and Lappi [22b] to estimate KD values assuming a
dissociation constant value for propidium for acetylcholinesterase
The inhibitory activity against human recombinant AChE was
also assessed using the method of Ellman et al. [15]. Initial rate
assays were performed at 37 ꢃC with a Jasco V-530 double beam
Spectrophotometer. The rate of increase in the absorbance at
412 nm was followed for 240 s. AChE stock solution was prepared
by dissolving human recombinant AChE (E.C.3.1.1.7) lyophilized
powder (SigmaeAldrich) in 0.1 M phosphate buffer (pH ¼ 8.0)
containing Triton X-100 0.1%. Stock solutions of inhibitors (1e
3 mM) were prepared in methanol and diluted in methanol. In-
from E. electricus equals to 0.7 mM [23].
4.2.4. PAMPA-BBB assay
To evaluate the brain penetration of the synthesized com-
pounds, a parallel artificial membrane permeation assay for bloode
brain barrier was used, following the method described by Di et al.
[27]. The in vitro permeability (Pe) of fourteen commercial drugs
through lipid extract of porcine brain membrane together with the
test compounds was determined. Commercial drugs and the syn-
thesized compounds were tested using a mixture of PBS:EtOH
(70:30). Assay validation was made by comparing the experimental
permeability of the different compounds with the reported bibli-
ography values of the commercial drugs, which showed a good
correlation: Pe (exp) ¼ 1.4974 Pe (lit) ꢁ 0.8434 (R2 ¼ 0.9428). From
this equation and taking into account the limits established by Di
et al. for BBB permeation, we established the ranges of permeability
as compounds of high BBB permeation (CNSþ): Pe
(10ꢁ6 cm sꢁ1) > 5.1; compounds of low BBB permeation (CNSꢁ): Pe
(10ꢁ6 cm sꢁ1) < 2.15, and compounds of uncertain BBB permeation
(CNSþ/ꢁ): 5.1 > Pe (10ꢁ6 cm sꢁ1) > 2.15.
hibitors were first screened at a single concentration (25
for compounds showing a percentage of inhibition higher than 50%
at the screening concentration (25 M), the IC50 values were
mM). Then,
m
determined. In this case, five/six increasing concentrations of the
inhibitor were used, able to give an inhibition of the enzymatic
activity in the range of 20e80%. The assay solution consisted of a
0.1 M phosphate buffer pH 8.0, with the addition of 340
0.02 unit/mL of human recombinant AChE and 550 M of substrate
(acetylthiocholine iodide, ATCh). 50 L aliquots of increasing con-
mM DTNB,
m
m
centration of the tested inhibitor (or methanol) were added to the
assay solution and pre-incubated with the enzyme for 20 min at
37 ꢃC before the addition of the substrate. Assays were carried out
with a blank containing all components except AChE in order to
account for the non-enzymatic hydrolysis of the substrate. The
reaction rates were compared and the percent inhibition due to the
presence of inhibitor was calculated. Each concentration was
analyzed in duplicate, and IC50 values were determined graphically
from log concentrationeinhibition curves (GraphPad Prism 4.03
software, GraphPad Software Inc.). Two/three independent exper-
iments were performed for the determination of each IC50 value.
4.3. Molecular modelling
4.3.1. Setup of the system
Molecular modelling was performed using the X-ray crystallo-
graphic structure of the recombinant human AChE (PDB ID: 3LII)
4.2.2. Kinetic inhibition studies
To estimate the mode of inhibition of compound 16a and the
corresponded benzylated analogue 14a, LineweavereBurk double
[30]. The structure was refined by removal of N-acetyl-
amine and sulphate anions and addition of missing hydrogen
D-glucos-

150
O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
atoms. Furthermore, the missing loop that comprises residues
259e263 (PGGTG) was modelled from the X-ray structure of hAChE
complexed with huprine W (PDB ID: 4BDT) [31].
Then, the position of hydrogen atoms was optimized using 4500
steps of conjugate gradient and 500 steps of steepest descent al-
gorithm. At the third stage, hydrogen atoms, water molecules and
counterions were further optimized using 11,500 steps of conjugate
gradient and 3500 steps of steepest descent algorithm. Finally, the
whole system was optimized using 8500 steps of conjugate
gradient and 2500 steps of steepest descent algorithm. Thermali-
zation of the system was performed in five steps of 25 ps, increasing
the temperature from 50 to 298 K. Concomitantly, the residues
that define the binding site were restrained during thermalization
The enzyme was modelled in its physiological active form with
neutral His447 and deprotonated Glu334, which together with
Ser203 form the catalytic triad. The ionization state for the rest of
ionizable residues was assessed from PROPKA3 calculations [32].
Accordingly, the standard ionization state at neutral pH was
considered but for residues Glu285, Glu450 and Glu452, which
were protonated. Finally, three disulfide bridges were defined be-
tween Cys residues 257e272, 529e409, and 69e96, respectively.
Structural waters were retrieved from those found in the AChEe
donepezil complex 1EVE [33].
using a variable restraining force. Thus, a force constant of
ꢀꢁ2
25 kcal molꢁ1
A
was used in the first stage of the thermalization
and was subsequently decreased by increments of 5 kcal molꢁ1
A
ꢀꢁ2
Since Trp286 can adopt three main conformations in the pe-
ripheral binding site [9], three models were built up by re-orienting
the side chain of Trp286 as found in the X-ray structures of the
AChE complexes with propidium, bis(7)-tacrine and syn-TZ2PA6
(PDB ID: 1N5R, 2CKM and 1Q83, respectively). These models were
energy minimized using the AMBER force field (see below).
in the next stages. Then, an additional step of 250 ps was performed
in order to equilibrate the system density at constant pressure
(1 bar) and temperature (298 K). Finally, a 100 ns trajectory was run
using a time step of 2 fs. SHAKE was used for those bonds con-
taining hydrogen atoms in conjunction with periodic boundary
conditions at constant volume and temperature, particle mesh
Ewald for the treatment of long range electrostatic interactions, and
ꢀ
4.3.2. Docking
a cutoff of 10 A for nonbonded interactions.
Docking of AChE inhibitors was performed using the rDock
program, which is an extension of the program RiboDock and uti-
lizes an empirical scoring function calibrated on the basis of pro-
teineligand complexes [24]. It is worth noting that previous studies
strongly support the excellent performance of rDock for predicting
the binding mode of a variety of AChE inhibitors to the enzyme
The structural analysis was performed using in-house software
and standard codes of AMBER12. The solvent interaction energies
(SIE) technique developed by Purisima and co-workers was used to
estimate the interaction free energies for the AChE inhibitors [26].
Calculations were performed for a set of 150 snapshots taken along
the last 30 ns of the MD trajectory.
ꢀ
gorge [9]. A cavity of radius 17 A, centred on the structure of a
superligand containing huprine X, donepezil and propidium (as
found in the X-ray structures 1E66 [34], 1EVE and 1N5R) was used
to define the docking volume. Since huprine X and propidium are
bound to the catalytic and peripheral binding sites, and donepezil is
aligned along the gorge, this definition guarantees the exploration
of the binding mode along the whole volume accessible for binding.
Calculations were performed with no structural waters. Confor-
mational flexibility around rotatable bonds of the ligand was
allowed. Docking calculations were performed separately for the
three models of the human enzyme, which differ in the relative
orientation of the side chain of Trp286 (see above). Conformational
adjustments of other residues in the binding site were accounted
for indirectly by rescaling (by a factor of 0.9) the van der Waals
volume of atoms. Each compound was subjected to 100 docking
runs and the poses were sorted according to its docking score. The
top 50 best scored poses were clustered and further analyzed by
visual inspection.
Acknowledgements
This work was supported by Ministerio de Ciencia e Innovación
(MICINN) (CTQ2011-22433, CTQ2012-30930, SAF2011-27642,
SAF2009-10553) and Generalitat de Catalunya (GC) (2009SGR1396,
2009SGR1024, 2009SGR249). This work was supported by MIUR
(PRIN 2009Z8YTYC-003). Fellowship from GC to E.V. and from
Leonardo da Vinci Project Unipharma-Graduates 6 to O.D.P. are
gratefully acknowledged. The Center for Scientific and Academic
Services of Catalonia (CESCA) is acknowledged for providing
access to computational facilities. We thank Dr. Marc Revés
(Universitat de Barcelona) for his helpful assistance in the HPLC
purity measurements.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2013.12.008.
4.3.3. Molecular dynamics simulations
Molecular dynamics (MD) simulations were run to further check
the stability of the proposed binding mode of AChE inhibitors.
Starting from the initial poses obtained from docking calculations, a
100 ns MD simulation was performed using the PMEMD module of
AMBER12 [35] software package and the parm99SB [36] force field
for the protein and GAFF [37]-derived parameters for the ligand.
The geometry of the ligand was optimized at the B3LYP/6-31G(d)
level [38]. The charge distribution of the inhibitors was defined
from the electrostatic charges determined by fitting the B3LYP/6-
31G(d) electrostatic potential using the RESP procedure [39]. Naþ
cations were added to neutralize the negative charge of the system
with the XLEAP module of AMBER12. The system was immersed in
an octahedral box of TIP3P water molecules [40], preserving the
crystallographic waters inside the binding cavity. The final system
contained around 53,000 atoms.
References
[1] N.L. Batsch, M.S. Mittelman, World Alzheimer Report 2012. Overcoming the
Stigma of Dementia, Alzheimer’s Disease International, London, 2012. http://
www.alz.co.uk.
[2] Alzheimer’s Association, Alzheimer’s disease facts and figures, Alzheimers
Dementia 9 (2013) 208e245.
[3] S.W. Pimplikar, Reassessing the amyloid cascade hypothesis of Alzheimer’s
disease, Int. J. Biochem. Cell Biol. 41 (2009) 1261e1268.
[4] (a) W.J. Geldenhuys, C.J. Van der Schyf, Rationally designed multi-targeted
agents against neurodegenerative diseases, Curr. Med. Chem. 20 (2013)
1662e1672;
(b) X. Chen, M. Decker, Multi-target compounds acting in the central nervous
system designed from natural products, Curr. Med. Chem. 20 (2013) 1673e
1685;
(c) P. Russo, A. Frustaci, A. Del Bufalo, M. Fini, A. Cesario, Multitarget drugs of
plants origin acting on Alzheimer’s disease, Curr. Med. Chem. 20 (2013)
1686e1693;
(d) A. Rampa, F. Belluti, S. Gobbi, A. Bisi, Hybrid-based multi-target ligands for
the treatment of Alzheimer’s disease, Curr. Top. Med. Chem. 11 (2011) 2716e
2730;
The geometry of the system was minimized in four steps. First,
water molecules and counterions were refined through 7000 steps
of conjugate gradient and 3000 steps of steepest descent algorithm.
(e) A. Cavalli, M.L. Bolognesi, A. Minarini, M. Rosini, V. Tumiatti, M. Recanatini,

O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
151
ꢁ
C. Melchiorre, Multi-target-directed ligands to combat neurodegenerative
diseases, J. Med. Chem. 51 (2008) 347e372.
[5] (a) R.A. Sperling, J. Karlawish, K.A. Johnson, Preclinical Alzheimer disease e the
challenges ahead, Nat. Rev. Neurol. 9 (2013) 54e58;
[20] (a) Z. Radic, P. Taylor, Interaction kinetics of reversible inhibitors and sub-
strates with acetylcholinesterase and its fasciculin 2 complex, J. Biol. Chem.
276 (2001) 4622e4633;
ꢁ
(b) Y. Bourne, P. Taylor, Z. Radic, P. Marchot, Structural insights into ligand
(b) D.C. Anderson, Alzheimer’s disease biomarkers: more than molecular di-
agnostics, Drug Dev. Res. 74 (2013) 92e111;
interactions at the acetylcholinesterase peripheral anionic site, EMBO J. 22
(2003) 1e12;
(c) S.J. Teipel, O. Sabri, M. Grothe, H. Barthel, D. Prvulovic, K. Buerger,
A.L. Bokde, M. Ewers, W. Hoffmann, H. Hampel, Perspectives for multimodal
neurochemical and imaging biomarkers in Alzheimer’s disease, J. Alzheimers
Dis. 33 (Suppl. 1) (2013) S329eS347;
(c) B. Kaboudin, S. Emadi, M.R. Faghihi, M. Fallahi, V. Sheik-Hasani, Synthesis
of a-oxycarbanilinophosphonates and their anticholinesterase activities: the
most potent derivative is bound to the peripheral site of acetylcholinesterase,
J. Enzyme Inhib. Med. Chem. 28 (2013) 576e582;
(d) J.F. Quinn, Biomarkers for Alzheimer’s disease; showing the way or leading
us astray? J. Alzheimers Dis. 33 (Suppl. 1) (2013) S371eS376;
(e) R.J. Caselli, E.M. Reiman, Characterizing the preclinical stages of Alz-
heimer’s disease and the prospect of presymptomatic intervention,
J. Alzheimers Dis. 33 (Suppl. 1) (2013) S405eS416;
(f) V.T. Papaliagkas, The role of cerebrospinal fluid biomarkers for Alzheimer’s
disease diagnosis. Where are we now? Recent Patents CNS Drug Discov. 8
(2013) 70e78.
(d) S. Young, K. Fabio, C. Guillon, P. Mohanta, T.A. Halton, D.E. Heck,
R.A. Flowers II, J.D. Laskin, N.D. Heindel, Peripheral site acetylcholinesterase
inhibitors targeting both inflammation and cholinergic dysfunction, Bioorg.
Med. Chem. Lett. 20 (2010) 2987e2990;
(e) P. de la Torre, L. Astudillo Saavedra, J. Caballero, J. Quiroga, J.H. Alzate-
Morales, M. Gutiérrez Cabrera, J. Trilleras, A novel class of selective acetyl-
cholinesterase inhibitors: synthesis and evaluation of (E)-2-(benzo[d]thiazol-
2-yl)-3-heteroarylacrylonitriles, Molecules 17 (2012) 12072e12085.
[21] A. Cornish-Bowden, A simple graphical method for determining the inhibition
constants of mixed, uncompetitive and non-competitive inhibitors, Biochem.
J. 137 (1974) 143e144.
[22] (a) P. Taylor, J. Lwebuga-Mukasa, S. Lappi, J. Rademacher, Propidium e a
fluorescent probe for a peripheral anionic site on acetylcholinesterase, Mol.
Pharmacol. 10 (1974) 703e708;
[6] (a) L. Cheewakriengkrai, S. Gauthier, A 10-year perspective on donepezil,
Expert Opin. Pharmacother. 14 (2013) 331e338;
(b) J.L. Cummings, S.J. Banks, R.K. Gary, J.W. Kinney, J.M. Lombardo, R.R. Walsh,
K. Zhong, Alzheimer’s disease drug development: translational neuroscience
strategies, CNS Spectr. 18 (2013) 128e138;
(c) M.B. Colovic, D.Z. Krstic, T.D. Lazarevic-Pasti, A.M. Bondzic, V.M. Vasic,
Acetylcholinesterase inhibitors: pharmacology and toxicology, Curr. Neuro-
pharmacol. 11 (2013) 315e335;
(b) P. Taylor, S. Lappi, Interaction of fluorescence probes with acetylcholin-
esterase, the site and specificity of propidium binding, Biochemistry 14 (1975)
1989e1997.
(d) P. Anand, B. Singh, A review on cholinesterase inhibitors for Alzheimer’s
disease, Arch. Pharm. Res. 36 (2013) 375e399.
[23] N. Nunes-Tavares, A. Nery da Matta, C.M. Batista
e Silva, G.M. Araújo,
[7] J.L. Sussman, M. Harel, F. Frolow, C. Oefner, A. Goldman, L. Toker, I. Silman,
Atomic structure of acetylcholinesterase from Torpedo californica e a proto-
typic acetylcholine-binding protein, Science 253 (1991) 872e879.
[8] (a) M.L. Bolognesi, A. Minarini, M. Rosini, V. Tumiatti, C. Melchiorre, From dual
binding site acetylcholinesterase inhibitors to multi-target-directed ligands
(MTDLs): a step forward in the treatment of Alzheimer’s disease, Mini-Rev.
Med. Chem. 8 (2008) 960e967;
(b) A. Castro, P. Muñoz, A. Martínez, Dual binding site acetylcholinesterase
inhibitors: design, SAR, and potential role as new disease-modifying agents
for Alzheimer’s disease, in: A. Martínez Gil (Ed.), Medicinal Chemistry of
Alzheimer’s Disease, Transworld Research Network, Kerala, 2008, pp. 21e44;
(c) D. Muñoz-Torrero, Acetylcholinesterase inhibitors as disease-modifying
therapies for Alzheimer’s disease, Curr. Med. Chem. 15 (2008) 2433e2455.
[9] P. Camps, X. Formosa, C. Galdeano, D. Muñoz-Torrero, L. Ramírez, E. Gómez,
N. Isambert, R. Lavilla, A. Badia, M.V. Clos, M. Bartolini, F. Mancini,
V. Andrisano, M.P. Arce, M.I. Rodríguez-Franco, O. Huertas, T. Dafni, F.J. Luque,
Pyrano[3,2-c]quinoline-6-chlorotacrine hybrids as a novel family of acetyl-
S.R. Louro, A. Hassón-Voloch, Inhibition of acetylcholinesterase from Elec-
trophorus electricus (L.) by tricyclic antidepressants, Int. J. Biochem. Cell Biol.
34 (2002) 1071e1079.
[24] (a) S.D. Morley, M. Afshar, Validation of an empirical RNA-ligand scoring
function for fast flexible docking using RiboDock, J. Comput. Aided Mol. Des.
18 (2004) 189e208;
(b) X. Barril, R.E. Hubbard, S.D. Morley, Virtual screening in structure-based
drug discovery, Mini-Rev. Med. Chem. 4 (2004) 779e791.
[25] C. Galdeano, E. Viayna, P. Arroyo, A. Bidon-Chanal, J.R. Blas, D. Muñoz-Torrero,
F.J. Luque, Structural determinants of the multifunctional profile of dual
binding site acetylcholinesterase inhibitors as anti-Alzheimer agents, Curr.
Pharm. Des. 16 (2010) 2818e2836.
[26] (a) M. Naïm, S. Bhat, K.N. Rankin, S. Dennis, S.F. Chowdhury, I. Siddiqi,
P. Drabik, T. Sulea, C.I. Bayly, A. Jakalian, E.O. Purisima, Solvated interaction
energy (SIE) for scoring proteineligand binding affinities. 1. Exploring the
parameter space, J. Chem. Inf. Model. 47 (2007) 122e133;
(b) Q. Cui, T. Sulea, J.D. Schrag, C. Munger, M.-N. Hung, M. Naïm, M. Cygler,
E.O. Purisima, Molecular dynamics-solvated interaction energy studies of
proteineprotein interactions: the MP1-p14 scaffolding complex, J. Mol. Biol.
379 (2008) 787e802.
cholinesterase- and
Chem. 52 (2009) 5365e5379.
[10] L.S. Povarov, -Unsaturated ethers and their analogues in reactions of diene
b-amyloid-directed anti-Alzheimer compounds, J. Med.
a,b
synthesis, Russ. Chem. Rev. 36 (1967) 656e670.
[27] L. Di, E.H. Kerns, K. Fan, O.J. McConnell, G.T. Carter, High throughput artificial
membrane permeability assay for blood-brain barrier, Eur. J. Med. Chem. 38
(2003) 223e232.
[28] E. Viayna, T. Gómez, C. Galdeano, L. Ramírez, M. Ratia, A. Badia, M.V. Clos,
E. Verdaguer, F. Junyent, A. Camins, M. Pallàs, M. Bartolini, F. Mancini,
V. Andrisano, M.P. Arce, M.I. Rodríguez-Franco, A. Bidon-Chanal, F.J. Luque,
P. Camps, D. Muñoz-Torrero, Novel huprine derivatives with inhibitory ac-
[11] (a) E. Vicente-García, F. Catti, R. Ramón, R. Lavilla, Unsaturated lactams:
new inputs for Povarov-type multicomponent reactions, Org. Lett. 12 (2010)
860e863;
(b) D.A. Powell, R.A. Batey, Total synthesis of the alkaloids martinelline and
martinellic acid via a hetero DielseAlder multicomponent coupling reaction,
Org. Lett. 4 (2002) 2913e2916.
[12] O. Jiménez, G. de la Rosa, R. Lavilla, Straightforward access to a structurally
diverse set of oxacyclic scaffolds through a four-component reaction, Angew.
Chem., Int. Ed. 44 (2005) 6521e6525.
[13] E. Vicente-García, R. Ramón, S. Preciado, R. Lavilla, Multicomponent reaction
access to complex quinolines via oxidation of the Povarov adducts, Beilstein J.
Org. Chem. 7 (2011) 980e987.
[14] S. Das, D. Addis, S. Zhou, K. Junge, M. Beller, Zinc-catalyzed reduction of
amides: unprecedented selectivity and functional group tolerance, J. Am.
Chem. Soc. 132 (2010) 1770e1771.
[15] G.L. Ellman, K.D. Courtney, V. Andres Jr., R.M. Featherstone, A new and rapid
colorimetric determination of acetylcholinesterase activity, Biochem. Phar-
macol. 7 (1961) 88e95.
[16] R.M. Lane, S.G. Potkin, A. Enz, Targeting acetylcholinesterase and
butyrylcholinesterase in dementia, Int. J. Neuropsychopharmacol. 9 (2006)
101e124.
[17] C.D. Anderson, N. Forsgren, C. Akfur, A. Allgardsson, L. Berg, C. Engdahl,
W. Qian, F.J. Ekström, A. Linusson, Divergent structureeactivity relationships
of structurally similar acetylcholinesterase inhibitors, J. Med. Chem. 56 (2013)
7615e7624.
tivity toward b-amyloid aggregation and formation as disease-modifying anti-
Alzheimer drug candidates, ChemMedChem 5 (2010) 1855e1870.
[29] R.B. Silverman, The Organic Chemistry of Enzyme-catalyzed Reactions, Aca-
demic Press, San Diego, 2000.
[30] H. Dvir, I. Silman, M. Harel, T.L. Rosenberry, J.L. Sussman, Acetylcholinesterase:
from 3D structure to function, Chem. Biol. Interact. 187 (2010) 10e22.
[31] F. Nachon, E. Carletti, C. Ronco, M. Trovaslet, Y. Nicolet, L. Jean, P.-Y. Renard,
Crystal structures of human cholinesterases in complex with huprine W and
tacrine: elements of specificity for anti-Alzheimer’s drugs targeting acetyl-
and butyrylcholinesterase, Biochem. J. 453 (2013) 393e399.
[32] M.H.M. Olsson, C.R. Sondergard, M. Rostkowski, J.H. Jensen, PROPKA3:
consistent treatment of internal and surface residues in empirical pKa pre-
dictions, J. Chem. Theor. Comput. 7 (2011) 525e537.
[33] G. Kryger, I. Silman, J.L. Sussman, Structure of acetylcholinesterase complexed
with E2020 (Aricept): implications for the design of new anti-Alzheimer
drugs, Structure 7 (1999) 297e307.
[34] H. Dvir, D.M. Wong, M. Harel, X. Barril, M. Orozco, F.J. Luque, D. Muñoz-Tor-
rero, P. Camps, T.L. Rosenberry, I. Silman, J.L. Sussman, 3D structure of Torpedo
ꢀ
californica acetylcholinesterase complexed with huprine X at 2.1 A resolution:
[18] M.L. Bolognesi, V. Andrisano, M. Bartolini, R. Banzi, C. Melchiorre, Propidium-
based polyamine ligands as potent inhibitors of acetylcholinesterase and
kinetic and molecular dynamics correlates, Biochemistry 41 (2002) 2970e
2981.
acetylcholinesterase-induced amyloid-
24e27.
[19] E. Viayna, R. Sabate, D. Muñoz-Torrero, Dual inhibitors of
b
aggregation, J. Med. Chem. 48 (2005)
[35] D.A. Case, T.A. Darden, T.E. Cheatham III, C.L. Simmerling, J. Wang, R.E. Duke,
R. Luo, R.C. Walker, W. Zhang, K.M. Merz, B. Roberts, S. Hayik, A. Roitberg,
G. Seabra, J. Swails, A.W. Goetz, I. Kolossváry, K.F. Wong, F. Paesani, J. Vanicek,
R.M. Wolf, J. Liu, X. Wu, S.R. Brozell, T. Steinbrecher, H. Gohlke, Q. Cai, X. Ye,
J. Wang, M.-J. Hsieh, G. Cui, D.R. Roe, D.H. Mathews, M.G. Seetin, R. Salomon-
b-amyloid aggre-
gation and acetylcholinesterase as multi-target anti-Alzheimer drug candi-
dates, Curr. Top. Med. Chem. 13 (2013) 1820e1842.

152
O. Di Pietro et al. / European Journal of Medicinal Chemistry 73 (2014) 141e152
Ferrer, C. Sagui, V. Babin, T. Luchko, S. Gusarov, A. Kovalenko, P.A. Kollman,
AMBER 12, University of California, San Francisco, 2012.
[36] V. Hornak, R. Abel, A. Okur, B. Strockbine, A. Roitberg, C. Simmerling, Com-
parison of multiple Amber force fields and development of improved protein
backbone parameters, Proteins 65 (2006) 712e725.
[37] J. Wang, R.M. Wolf, J.W. Caldwell, P.A. Kollman, D.A. Case, Development and
testing of a general AMBER force field, J. Comput. Chem. 25 (2004) 1157e
1174.
[38] (a) A.D. Becke, Density-functional thermochemistry. III. The role of exact ex-
change, J. Chem. Phys. 98 (1993) 5648e5652;
(b) C. Lee, W. Yang, R.G. Parr, Development of the ColleeSalvetti correlation-
energy formula into a functional of the electron density, Phys. Rev. B 37
(1988) 785e789.
[39] C.I. Bayly, P. Cieplak, W.D. Cornell, P.A. Kollman, A well-behaved electrostatic
potential based method using charge restraints for deriving atomic charges,
J. Phys. Chem. 97 (1993) 10269e10280.
[40] W.L. Jorgensen, J. Chandrasekhar, J.D. Madura, R.W. Impey, M.L. Klein, Com-
parison of simple potential functions for simulating liquid water, J. Chem.
Phys. 79 (1983) 926e935.