Donepezil + propargylamine + 8-hydroxyquinoline hybrids as new multifunctional metal-chelators, ChE and MAO inhibitors for the potential treatment of Alzheimer's disease

European Journal of Medicinal Chemistry 80 (2014) 543e561

Contents lists available at ScienceDirect

European Journal of Medicinal Chemistry

journal homepage: h ttp://www.elsevier.com/locate/ejmech

Original article

Donepezil þ propargylamine þ 8-hydroxyquinoline hybrids as new

multifunctional metal-chelators, ChE and MAO inhibitors for the

potential treatment of Alzheimer’s disease

,

Li Wang ^{a 1}, Gerard Esteban ^{b 1}, Masaki Ojima ^a, Oscar M. Bautista-Aguilera ^c,

Tsutomu Inokuchi ^a, Ignacio Moraleda , Isabel Iriepa , Abdelouahid Samadi ,

,

d

c

*

Moussa B.H. Youdim ^e, Alejandro Romero ^f, Elena Soriano ^g, Raquel Herrero ^h,

Ana Patricia Fernández Fernández ^h, Ricardo-Martínez-Murillo ^h, José Marco-Contelles ^c

,

*

,

Mercedes Unzeta ^b

,

*

^aDivision of Chemistry and Biotechnology, Graduate School of Natural Science and Technology, Okayama University, 3.1.1 Tsushima-Naka, Kita-ku,

Okayama 700-8530, Japan

^bDepartament de Bioquímica i Biologia Molecular, Facultat de Medicina, Institut de Neurociències, Universitat Autònoma de Barcelona, 08193 Bellaterra,

Barcelona, Spain

^cLaboratorio de Química Médica (IQOG, CSIC), C/Juan de la Cierva 3, 28006 Madrid, Spain

^dDepartamento de Química Orgánica y Química Inorgánica, Universidad de Alcalá, Ctra. Madrid-Barcelona, Km. 33,6, 28871 Alcalá de Henares,

Madrid, Spain

^eEve Topf Centers of Excellence for Neurodegenerative Diseases Research and Department of Pharmacology, Rappaport Family Research Institute,

Technion-Faculty of Medicine, Haifa 31096, Israel

^fDepartamento de Toxicología y Farmacología, Facultad de Veterinaria, Universidad Complutense de Madrid, 28040 Madrid, Spain

^gSEPCO, IQOG (CSIC), Juan de la Cierva, 3, 28006 Madrid, Spain

^hNeurovascular Research Group, Department of Molecular, Cellular and Developmental Neurobiology, Instituto Cajal (CSIC) Av. Doctor Arce 37,

28002 Madrid, Spain

a r t i c l e i n f o

a b s t r a c t

Article history:

The synthesis, biochemical evaluation, ADMET, toxicity and molecular modeling of novel multi-target-

directed Donepezil þ Propargylamine þ 8-Hydroxyquinoline (DPH) hybrids 1e7 for the potential pre-

vention and treatment of Alzheimer’s disease is described. The most interesting derivative was racemic

Received 10 February 2014

Received in revised form

24 April 2014

Accepted 26 April 2014

Available online 29 April 2014

a

-aminotrile4-(1-benzylpiperidin-4-yl)-2-(((8-hydroxyquinolin-5-yl)methyl)(prop-2-yn-1-yl)amino)

butanenitrile (DPH6) [MAO A (IC₅₀

¼

6.2

ꢀ

0.7

m

M; MAO

B

(IC₅₀

¼

10.2

ꢀ

0.9 mM); AChE

(IC₅₀¼ 1.8 ꢀ 0.1 M); BuChE (IC₅₀¼ 1.6 ꢀ 0.25

m

M)], an irreversible MAO A/B inhibitor and mixed-type

AChE inhibitor with metal-chelating properties. According to docking studies, both DPH6 enantiomers

interact simultaneously with the catalytic and peripheral site of EeAChE through a linker of appropriate

length, supporting the observed mixed-type AChE inhibition. Both enantiomers exhibited a relatively

similar position of both hydroxyquinoline and benzyl moieties with the rest of the molecule easily

accommodated in the relatively large cavity of MAO A. For MAO B, the quinoline system was hosted at the

cavity entrance whereas for MAO A this system occupied the substrate cavity. In this disposition the

quinoline moiety interacted directly with the FAD aromatic ring. Very similar binding afﬁnity values were

also observed for both enantiomers with ChE and MAO enzymes. DPH derivatives exhibited moderate to

good ADMET properties and brain penetration capacity for CNS activity. DPH6 was less toxic than

donepezil at high concentrations; while at low concentrations both displayed a similar cell viability

proﬁle. Finally, in a passive avoidance task, the antiamnesic effect of DPH6 was tested on mice with

experimentally induced amnesia. DPH6 was capable to signiﬁcantly decrease scopolamine-induced

learning deﬁcits in healthy adult mice.

Keywords:

Donepezil þ Propargylamine þ 8-

Hydroxyquinoline hybrids

Multifunctional drugs

Fe/Cu/Zn chelators

ChE and MAO inhibitors

In vivo scopolamine-induced long-term

memory deﬁcit

Alzheimer’s disease

* Corresponding authors.

E-mail addresses: Mercedes.Unzeta@uab.cat, mercedes.unzeta@uab.es (M. Unzeta).

G.E. and L.W. have equally contributed to this work.

1

http://dx.doi.org/10.1016/j.ejmech.2014.04.078

544

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

1. Introduction

Alzheimer’s disease (AD) is an age-related neurodegenerative

process characterized by progressive memory loss and other

cognitive impairments [1]. Although its etiology has not yet been

elucidated, b-amyloid (Ab) deposits [2], s-protein phosphorylation,

oxidative stress [3] and deﬁcits of acetylcholine (ACh) [4] are

considered to play signiﬁcant roles in the pathophysiology of AD

[5]. Consequently, AD patients have been treated with acetylcho-

linesterase inhibitors (AChEI) [6] with limited therapeutic success.

This might be due to the multifactorial nature of AD, a fact that has

prompted the hunt for new Multi-Target-Directed Ligands (MTDL),

based on the “one molecule, multiple target” paradigm [7]. Thus, in

this context, multifunctional molecules able to simultaneously bind

both cholinesterases and monoamineoxidases have been investi-

gated [8,9].

Monoamine oxidase (MAO; EC 1.4.3.4) is an important target to

be considered for the treatment of AD, as it catalyzes the oxidative

deamination of a variety of biogenic and xenobiotic amines with

the concomitant production of hydrogen peroxide, a key interme-

diate via Fenton reaction in the production of toxic radical

oxygenated species implicated in the progress of AD. MAO is a FAD

(Flavin-Adenine Dinucleotide)-containing enzyme bound to mito-

chondrial outer membranes of neuronal, glial and other cells [10].

MAO exists as two isoenzymes, MAO A/B, displaying different

substrate speciﬁcity, sensitivity to inhibitors, and amino acid se-

quences. While MAO A preferentially oxidizes neurotransmitters

norepinephrine and serotonin and it is selectively inhibited by

clorgyline, MAO B preferentially deaminates b-phenylethylamine

and it is irreversibly inhibited by l-deprenyl [11]. X-ray crystal

structures of rat MAO A [12] and human MAO B have been reported

[13].

Taken these facts into account, we previously reported the

synthesis and biological evaluation of new multifunctional MAO/

ChE inhibitor ASS234 [8,9] (Chart 1) by combining the N-benzyl-

piperidine and the N-propargylamine moieties present in done-

pezil, and PF9601N [14], ChE and MAO inhibitors, respectively.

In this work we report the design, synthesis, biochemical eval-

Chart 1. General structure of donepezil, M30, M30A, M30B, HLA20A, ASS234, and the

new multifunctional ChE/MAO DPH inhibitors described in this work.

have been similarly transformed into racemic

DPH2, DPH4 and DPH6. In order to conﬁrm the structure of DPH6,

an alternative, unequivocal synthesis of -aminonitrile 21 has been

a-aminonitriles

a

uation

Donepezil þ Propargylamine þ 8-Hydroxyquinoline (DPH) hybrids

1e7 and the identiﬁcation of the racemic -aminonitrile 4-(1-

and

molecular

modeling

of

achieved from aldehyde 19 via Strecker-type reaction (Scheme 5),

followed by standard N-alkylation as shown above.

All new compounds showed analytical and spectroscopic data,

in good agreement with their structure (Experimental Part).

a

benzylpiperidin-4-yl)-2-(((8-hydroxyquinolin-5-yl)methyl)(prop-

2-yn-1-yl)amino)butanenitrile (DPH6) (Table 1), as a multifunc-

tional lead molecule for the potential treatment of AD. These hy-

brids are the result of the juxtaposition of donepezil, a selective

AChE inhibitor currently used in the pharmacological treatment of

AD, and M30 (Chart 1), a potent brain selective MAO A/B inhibitor

and neuroprotective biometal-chelator [15]. HLA20A (Chart 1) is a

novel pro-chelator with improved cytotoxicity exhibiting poor af-

ﬁnity for metal ions while not being activated to an active chelator

by binding and inhibiting AChE, being able to modulate amyloid

2.2. Pharmacological evaluation

2.2.1. Cholinesterase inhibition

The in vitro activity of DPHs 1e7 derivatives as EeAChE and

eqBuChE inhibitors was assessed using the Ellman’s method [18]

(Table 1).

From these data some interesting structureeactivity relation-

ships (SAR) were obtained. DPH1, bearing no methyl in the linker

connecting the N-benzyl-piperidin-4⁰-yl residue to the N-propargyl

core, was inactive. In comparison, DPHs 2e7 derivatives were non-

selective moderate ChE inhibitors, DPH6 (n ¼ 3 in the linker)

precursor protein (APP) and to reduce A

2. Results and discussion

2.1. Chemistry

b aggregation [16].

exhibiting the most potent proﬁle [EeAChE (IC₅₀¼ 1.8 ꢀ 0.1

m

M);

eqBuChE (IC₅₀¼ 1.6 ꢀ 0.25

mM)]. Regarding tertiary amines DPHs 3,

DPHs 1e7 were prepared as shown in Schemes 1e5 by using

reductive amination of suitable and different carbaldehydes

bearing the N-benzylpiperidine nucleus and propargylamine, fol-

lowed by N-alkylation with easily available 5-(chloromethyl)qui-

nolin-8-ol (8) [17] (Experimental Part).

5 and 7, activity at inhibiting both cholinesterases gradually

increased from DPH3 (n ¼ 1) to DPH7 (n ¼ 3), which displayed the

most active proﬁle [EeAChE (IC₅₀¼ 2.7 ꢀ 0.4

mM); eqBuChE

mM)]. A similar trend was detected with the a-

(IC₅₀¼ 4.5 ꢀ 0.3

aminonitrile series (DPHs 2, 4 and 6), with DPH6 (n ¼ 3) as the most

In the reductive amination leading to the intermediates for the

preparation of DPH3, DPH5 and DPH7, we have also isolated and

potent anticholinesterasic compound. Generally, for equivalent

linker length,

a-aminonitriles derivatives proved to be more potent

characterized the unexpected

a-aminonitriles 11, 16 and 21 that

than amines at inhibiting both enzymes, with the exception of

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

545

Table 1

Dose-response values [IC₅₀

standard inhibitors.

(mM)] of membrane-bound MAO A and MAO B from rat liver, electric eel AChE (EeAChE) and equine BuChE (eqBuChE) with DPH derivatives and

DPH

EeAChE^a

eqBuChE^a

SI^c

ratMAO A^b

ratMAO B^b

SI^d

1

2

3

4

5

6

7

ꢁ100

e

ꢁ100

ꢁ1

14.5 ꢀ 3.7

17.0 ꢀ 4.7

5.5 ꢀ 0.5

5.0 ꢀ 0.8

1.8 ꢀ 0.1

2.7 ꢀ 0.4

0.0067 ꢀ 0.0004

ꢁ100

13.5 ꢀ 1.8

12.8 ꢀ 1.4

6.1 ꢀ 0.6

14.5 ꢀ 3.6

1.6 ꢀ 0.2

4.5 ꢀ 0.3

7.4 ꢀ 0.1

e

0.92

0.75

1.1

2.8

0.92

1.6

1100

0.037 ꢀ 0.02

22.1 ꢀ 0.7

85.4 ꢀ 3.7

9.7 ꢀ 1.5

ꢁ100

39.5 ꢀ 1.4

19.4 ꢀ 3.2

12.4 ꢀ 2.5

50.1 ꢀ 5.0

10.2 ꢀ 0.9

34.5 ꢀ 3.5

15 ꢀ 2.2

1.5

1.7

0.22

1.2

ꢁ0.5

6.2 ꢀ 0.7

ꢁ100

1.7

ꢁ0.34

0.02

Donepezil

M-30 [15]

113.9 ꢀ 14.05

l-Deprenyl [14]

ꢁ500

850 ꢀ 13

0.057 ꢀ 0.02

ꢁ500

e

3

0.02

0.03

0.0067

8

Clorgyline [22]

ꢁ500

e

267

end ¼ values not determined.

Values are expressed as mean ꢀ standard error of the mean of at least three different experiments in quadruplicate.

a

20-min pre-incubation.

30-min pre-incubation.

b

c

eqBuChE selectivity index ¼ IC₅₀(eqBuChE)/IC₅₀(EeAChE).

d

MAO B selectivity index ¼ IC₅₀(MAO B)/IC₅₀(MAO A).

DPH4 and DPH5 exhibiting similar antiacetylcholinesterase activ-

ity. In comparison to donepezil, DPH6 was signiﬁcantly less active

EeAChE inhibitor, but more potent at inhibiting eqBuChE (4.5-fold).

Moreover, DPH6 was 64-fold more potent inhibiting both cholin-

esterases than M30 (Chart 1), inactive on this enzyme.

Scheme 1. Synthesis of DPH1. Reagents and conditions: (a) i. 1-Benzyl-4-piperidone,

MeOH, TFA; ii. NaBH₃CN (31%); (b) 5-(chloromethyl) quinolin-8-ol (8), DCM, Et₃N

(33%).

Scheme 3. Synthesis of DPH4 and DPH5. Reagents and conditions: (a) Dieth-

yl(cyanomethyl)phosphonate, K₂CO₃, THF (99%); (b) I₂, Mg, MeOH (74%); (c) DIBALH,

THF, -78 ^ꢂC (27%); (d) i. Propargylamine, MeOH, TFA; ii. NaBH₃CN; (e) 5-(chloromethyl)

quinolin-8-ol (8), DCM, Et₃N.

Scheme 2. Synthesis of DPH2 and DPH3. Reagents and conditions: (a) i. Propargyl-

amine, MeOH, TFA; ii, NaBH₃CN; (b) 5-(chloromethyl) quinolin-8-ol (8), DCM, Et₃N.

546

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

showed moderate, almost equipotent inhibitory activity on both

enzyme isoforms. Once more, DPH6 revealed the most potent

proﬁle at inhibiting both monoamine oxidase isoforms [MAO A

(IC₅₀¼ 6.2 ꢀ 0.7

m

M; MAO B (IC₅₀¼ 10.2 ꢀ 0.9

mM)]. Compared to

M30 (Chart 1) [15], DPH6 was 166-fold and 179-fold less potent

inhibiting MAO A and MAO B, respectively. Yet, DPH6 was 138-fold

more active than donepezil, showing equipotency at inhibiting

MAO A and MAO B, respectively. The previously mentioned SAR

also applied: for similar linker length, a-aminonitriles were more

potent inhibitors than their corresponding amines versus both

isoenzymes, with the exception of DPH2 as MAO B inhibitor. From

doseeresponse curves (IC₅₀), DPH6 showed a non-selective MAO

inhibition in comparison to standard selective MAO B and MAO A

inhibitors l-deprenyl [14] and clorgyline [20], respectively.

To further characterize MAO inhibition by DPH6, reversibility

study of this inhibitor was addressed. Irreversible inhibition was

observed in both isoforms since enzyme activities were not

signiﬁcantly reverted after three consecutive centrifugations and

washings with phosphate buffer (Fig. 2) followed by a pre-

incubation with the inhibitor. Simultaneously, and for comparison

purpose, the same assay was performed with irreversible MAO A

inhibitor clorgyline and irreversible MAO B inhibitor l-deprenyl.

Next, the assessment of time-dependent inhibition of both MAO

isoforms with DPH6 was assessed by pre-incubating the enzyme

with the inhibitor at times ranging 0e360 min (Fig. 3A). Samples

with no compound were used to determine the maximum enzyme

activity. Enzyme inactivation was not achieved before a pre-

incubation of 180 min and 120 min for MAO A and MAO B,

respectively, revealing time-dependent inhibition in both isoforms.

These ﬁndings were conﬁrmed by determining a constant variation

on IC₅₀values as longer pre-incubations with DPH6 were per-

formed with both MAO isoforms (Fig. 3B, C and D).

Scheme 4. Synthesis of DPH6 and DPH7. Reagents and conditions: (a) Dieth-

yl(cyanomethyl) phosphonate, K₂CO₃, THF (33%); (b) I₂, Mg, MeOH (99%); (c) DIBALH,

THF, -78 ^ꢂC (28); (d) i. Propargylamine, MeOH, TFA; ii. NaBH₃CN; (e) 5-(chloromethyl)

quinolin-8-ol (8), DCM, Et₃N.

Scheme 5. Alternative synthesis of compound 21. Reagents and conditions: (a) Prop-

argylamine, TMSCN, MWI, 125 ^ꢂC, 10 min (60%)

To determine the type of AChE inhibition exerted by DPH6,

LineweavereBurk reciprocal plots were obtained (Fig. 1). Increasing

slopes (decreasing V_max) and intercepts (increasing K_m) with higher

inhibitory concentration were determined suggesting a mixed type

2.2.3. Assessment of metal-chelating properties

inhibition. Reversible inhibition constant (k_i) of 1.21 ꢀ 0.25

mM was

Metal-chelating properties of compound DPH6 were also

investigated towards biometals Cu(II), Fe(III) and Zn(II) by UVeVIS

spectrometry. The increase on brain levels of iron, zinc and

particularly copper is reported to actively contribute to the for-

mation of senile plaques by generating more reactive oxygen spe-

cies through the Ab_1-42-metal complex [21].

estimated from the slopes of double reciprocal plots versus DPH6

concentrations.

2.2.2. Monoamine oxidase inhibition

In order to assess their multipotent proﬁle, the inhibitory ca-

pacity of DPHs 1e7 as dual MAO A and MAO B inhibitors was

evaluated using [¹⁴C]-5HT and [¹⁴C]-phenylethylamine as sub-

strates, respectively [19]. As shown in Table 1, most of DPH de-

rivatives poorly inhibited MAO, except DPH 4 and DPH6 that

Upon the addition of varying concentrations of CuSO₄, Fe₂(SO₄)₃

or ZnSO₄, the maximum absorption detected at 240 nm with the

compound alone shifted to 257 nm, exhibiting the formation of

complexes DPH6-Cu(II), DPH6-Fe(III) and DPH6-Zn(II) (Fig. 4AeC).

Signiﬁcant variations on complexation times were observed

depending on the metal. No spectral differences were identiﬁed

when Cu(II) and was added after 5-min incubation whereas 1 h or

O/N incubations were required to detect the formation of DPH6-

Zn(II) and DPH6-Fe(III) complexes, respectively. These results

conﬁrm that DPH6 selectively complexes Cu(II) salts. The equations

obtained by the Job’s method provided a solution at a mole fraction

of 0.65 for compound DPH6 complexing Cu(II) and Zn(II) and a

mole fraction between 0.5 and 0.6 for the complex DPH6-Fe (III).

This data revealed a 2:1 stoichiometry for all complexes.

2.3. Molecular modeling of DPH6

2.3.1. Inhibition of EeAChE

A modeling study was carried out through docking simulations

for the purpose of gaining insights on the nature and spatial loca-

tion of the key interactions of the (R)- and (S)-enantiomers of DPH6

modulating the inhibitory activity of AChE and BuChE. As in our

previous studies, we have chosen the 3D structure of the enzyme

species (EeAChE and eqBuChE) used for the kinetic studies. The

kinetic data provide evidence that compound 6 displays a mixed

Fig. 1. Steady-state inhibition of EeAChE hydrolysis of acetylthiocholine (ASCh) by

DHP6 (0e5

concentrations (0.1e1

squares analysis of data.

m

M). LineweavereBurk reciprocal plots of initial velocity and substrate

m

M) are presented. Lines were derived from a weighted least-

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

547

Fig. 2. (A) Reversibility study of MAO A inhibition by 10

mM DPH6 and 50 nM clorgyline. (B) Reversibility study of MAO B inhibition by 20

mM DPH6 and 20 nM l-deprenyl. MAO A

and MAO B inhibition was performed following a 30-min pre-incubation. After three consecutive washes with buffer MAO activity was not signiﬁcantly reverted. Data expressed as

the mean ꢀ SEM of six independent experiments in triplicate. KruskaleWallis one-way analysis was used.

Fig. 3. (A) Study of time-dependent inhibition of MAO A and MAO B by DHP6 (10 mM and20 mM, respectively) over different pre-incubation times (0e360 min). (B and C) Dose-

response curves (IC₅₀) following different pre-incubation times (0e240 min) with DPH6 inhibiting MAO A and MAO B. (D) IC₅₀values determined following different pre-

incubations. Data are the mean ꢀ SEM of three independent experiments in triplicate.

548

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

Fig. 4. (A) UVeVIS spectrum of a solution containing 10

mM DPH6 and 0e4

m

M CuSO₄following 5-min incubation. (B) UVeVIS spectrum of a solution containing 10

mM DPH6 and

0e4 M Fe₂(SO₄)₃following O/N incubation. (C) UVeVIS spectrum of a solution containing 10

m

M DPH6 and 0e4 M ZnSO₄following 1-h incubation. (D) Determination of the

m

stoichiometry of complex DPH6-Cu(II), (E) complex DPH6-Fe(III) and (F) complex DPH6-Zn(II) following 5-min, O/N and 1-h incubation, respectively by using the Job’s method. All

solutions were prepared in distilled water and incubations performed at room temperature.

type inhibition and argue in favor of interactions of DPH6 with both

catalytic and peripheral binding sites of AChE. Molecular modeling

studies have been carried in order to validate this assumption.

Ligand docking studies were performed with Autodock Vina [22]

using a single catalytic sub-unit of EeAChE (PDB: 1C2B). This

docking procedure allows the docking of ligands on the entire

protein surface, without prior speciﬁcation of the binding site. As in

previous studies, the recognition process between (R)- and (S)-

enantiomers of DPH6 was theoretically investigated by ﬂexible

docking experiments. Flexible torsions in the ligands were assigned

and protein side chain ﬂexibility was incorporated allowing the

rearrangement of the side chains of eight residues, Trp286, Tyr124,

Tyr337, Tyr72, Asp74, Thr75, Trp86 and Tyr341. These residues

delineate the shape of the gorge entry and lining and their motion

may signiﬁcantly enlarge the gorge to facilitate bulky ligand access

to the catalytic site [23,24]. As shown in Figs. 5 and 6, it appears that

both enantiomers of DPH6 interact simultaneously with both cat-

appropriate length showing a strong correlation with the obser-

vations we have from LineweavereBurk plots.

2.3.1.1. Docking

studies

of

(R)-DPH6

with

EeAChE.

Computational docking studies of (R)-DPH6 with EeAChE yielded

four major binding modes at the enzyme binding site. In Fig. 7, the

four most favored binding modes are presented along with the ﬁrst

shell of residues surrounding (R)-DPH6.

Mode I and Mode II (Fig. 5A and B) placed the quinoline moiety

near the opening of the binding pocket. Mode III and Mode IV

(Fig. 5C and D) placed the quinoline deep into the binding pocket

next to the residues known to be involved in catalysis. The classical

catalytic triad is shaded in green in all four panels. A close exami-

nation of (R)-DPH6 in Mode I revealed that the interaction with the

AChE peripheral site involved a face to face

the indole ring of Trp286 and the quinoline moiety and a T-shaped

stacking between the phenyl ring of Tyr72 and the quinoline

moiety. Besides, in this complex, a bifurcated hydrogen bond was

pep stacking between

pep

alytic and peripheral site of EeAChE thanks to

a linker of

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

549

Fig. 5. Binding mode of inhibitor (R)-DPH6 at the active site of EeAChE. (A) Mode I, compound (R)-DPH6 is illustrated in green. (B) Mode II, compound (R)-DPH6 is illustrated in

blue. (C) Mode III, compound (R)-DPH6 is illustrated in red (D) Mode IV, compound (R)-DPH6 is illustrated in violet. Ligands are rendered as sticks and the side chains conformations

of the mobile residues are illustrated in the same color light as the ligand. Different subsites of the active site were colored: catalytic triad (CT) in green, oxyanion hole (OH) in pink,

anionic sub-site (AS) in orange, except Trp86, acyl binding pocket (ABP) in yellow, and peripheral anionic subsite (PAS) in blue. Black dashed lines are drawn among atoms involved

in hydrogen bond interactions. (For interpretation of the references to colour in this ﬁgure legend, the reader is referred to the web version of this article.)

formed between the nitrogen atom of the cyano group and Arg296-

NH and Phe295-OH. The benzyl moiety pointed towards the bot-

II, the benzyl and the quinoline moieties interacted with Trp86 of

the catalytic pocket and Trp286 of the peripheral site, respectively.

In the middle of the gorge, the nitrogen atom of the cyano group

forms a hydrogen bond with the hydroxyl group of the Phe295. At the

tom of the gorge and established edge-to-face

pep interactions

with Trp86 and face-to-face interactions with Tyr337. In Mode

pep

Fig. 6. Binding mode of inhibitor (S)-DPH6 at the active site of hAChE. (A) Mode I, compound (S)-DPH6 is illustrated in gray. (B) Mode II, compound (S)-DPH6 is illustrated in brown.

(C) Mode III, compound (S)-DPH6 is illustrated in pink. Ligands are rendered as sticks and the side chains conformations of the mobile residues are illustrated in the same light color

as the ligand. Different subsites of the active site were colored: catalytic triad (CT) in green, oxyanion hole (OH) in pink, anionic sub-site (AS) in orange, except Trp86, acyl binding

pocket (ABP) in yellow, and peripheral anionic subsite (PAS) in blue. Black dashed lines are drawn among atoms involved in hydrogen bond interactions. (For interpretation of the

references to colour in this ﬁgure legend, the reader is referred to the web version of this article.)

550

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

Fig. 7. Complex of compound (R)-DPH6 and (S)-DPH6 and eqBuChE homology built 3D-model. (A) Mode I, compound (R)-DPH6 is illustrated in orange and compound (S)-DPH6 in

green. (B) Mode II, compound (R)-DPH6 is illustrated in yellow and compound (S)-DPH6 in blue. The compounds are rendered as sticks. (For interpretation of the references to

colour in this ﬁgure legend, the reader is referred to the web version of this article.)

top of the gorge, the hydroxyl group of the quinoline moiety forms

a hydrogen bond with the carboxylate group of Asp74. As can be

seen in Fig. 5C (Mode III), (R)-DPH6 has several interactions along

the active-site gorge of EeAChE. At the top of the gorge, the benzyl

The benzene ring of (S)-DPH6 formed a pep stacking interaction

with Trp86. In Mode II (Fig. 6B), compound (S)-DPH6 can also adopt

a similar binding conformation to that in Mode I. The ligand had an

orientation along the active-site gorge, extending from the anionic

sub-site of the active site at the bottom, to the peripheral anionic

site at the top, via aromatic stacking interactions with Trp86,

Trp286 and Tyr72 residues (Mode II, Fig. 6B). Conversely, in this

orientation, compound (S)-DPH6 was not able to form hydrogen

bonds but a stronger stacking interaction between benzene ring

and Trp86. Mode III (Fig. 6C) also placed the ligand along the active-

ring and Trp286 indole ring formed a favorable T-shaped

pep

interaction. Near the bottom of the gorge the quinoline moiety

stacked against the Trp86 indole ring. In this region, the nitrogen

atom of the quinoline moiety was hydrogen bonded to the hydroxyl

group of Tyr133. Comparison of Mode IV (Fig. 5D) and Mode III

(Fig. 5C) revealed a broadly similar interaction but with two key

differences: a) a signiﬁcant movement of the Trp286 indole and

site gorge. The quinoline ring formed a

pep interaction with the

benzyl rings in order to establish a face-to-face

p

e

p

interaction,

indole ring of Trp86 and also established three additional hydrogen

bonds, which have an important contribution to the binding po-

tency of the ligand. The hydroxyl group of Tyr133 formed two

hydrogen bonds with the nitrogen and the hydroxyl group of the

quinoline moiety, the later group is also hydrogen bonded to the

carbonyl oxygen from the backbone of Gly120. On the other hand,

at the peripheral anionic site (PAS), the benzene ring of the ligand

and b) the rotation of the hydroxyl group of the quinoline moiety to

form a hydrogen bond with the carbonyl group of Gly120. In both

poses (Mode III and Mode IV), the three methylene units in the

spacer of (R)-DPH6 were long enough to allow a proper interaction

between (R)-DPH6 and both sites of the enzyme. The linker was

lodged in a narrow cavity described by Asp74, Tyr124, Phe297,

Tyr337, Phe338, and Tyr341. The protonated nitrogen was favorably

interacting with the residues of a kind of “electrostatic cage”

(Tyr124, Asp74, Tyr337 and Tyr341 side chains).

form edge-to-face

pep interaction with the indole ring of Trp286

and with the benzene ring of Tyr72. In comparison with the other

poses, the cyano residue is involved in the interactions with the

catalytic triad of the active site gorge. This group is able to bind by

mean of hydrogen bond with the Ser203 side chain.

2.3.1.2. Docking studies of (S)-DPH6 with EeAChE. Docking studies

of (S)-DPH6 with EeAChE yielded three major binding modes at the

enzyme binding-site. In Fig. 6 the three most favored binding

modes are presented along with the ﬁrst shell of residues sur-

rounding (S)-DPH6. Mode I and Mode II (Fig. 6A and B) placed the

quinoline moiety near the opening of the binding pocket. Mode III

(Fig. 6C) placed the quinoline deep into the binding pocket next to

the residues involved in catalysis. In Mode I (Fig. 6A), the main

stabilizing factors that keep stable the (S)-DPH6-AChE complex

On the basis of the similar orientation and the set of in-

teractions, which were identiﬁed between the (R) enantiomer

(Modes I-III, Fig. 5AeC) and (S) enantiomer (Modes I-III, Fig. 6AeC)

and the protein residues within the gorge, it can be hypothesized

that both enantiomers of DPH6 could simultaneously bind at both

the peripheral anionic site (PAS) and the catalytic active site (CAS)

of EeAChE.

were found to be the hydrophobic contacts,

p

ep

interactions and

2.3.2. Inhibition of eqBuChE

hydrogen bonding interactions. Compound (S)-DPH6 can simulta-

neously bind at both the peripheral anionic site (PAS) and the

catalytic active site (CAS) of EeAChE. (S)-DPH6 is able to bind in the

With respect to BuChE, in the absence of X-ray structure of

eqBuChE, a homology model was used. The modeling of the 3D

structure was performed by an automated homology-modeling

program (SWISS-MODEL) [25e27]. A putative three-dimensional

structure of eqBuChE has been created based on the crystal struc-

ture of human BuChE (pdb: 2PM8), as these two enzymes exhibited

89% sequence identity. In order to simulate the binding of both

enantiomers of compound DPH6 to eqBuChE, docking experiments

were performed as blind dockings following the same

PAS by face-to-face and edge-to-face pep interactions between the

quinoline moiety of the ligand and the Trp286 indole ring and the

Tyr72 phenyl ring, respectively. Other interactions like a bifurcated

hydrogen bond between the nitrogen atom of the cyano group and

Arg296-NH and Phe295-OH could also play an important role in

positioning and stabilizing the ligand inside the active site gorge.

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

551

computational protocol used for EeAChE. The best-ranked docking

solutions revealed that BuChE can effectively accommodate both

enantiomers of DPH6 inside the active site gorge and two major

binding modes can be proposed (Fig. 7). Modes I and II (Fig. 7A and

B) for both enantiomers placed the quinoline moiety into the

binding pocket next to the residues involved in catalysis. In this

orientation, the phenyl moiety interacts with Trp231 by means of

were considered as integral components of the protein structure

during the docking simulation. Docking of DPH6 was performed for

both enantiomers and a sole binding mode per enantiomer was

found. Fig. 8A and B illustrated the binding modes of (R)-DPH6 and

(S)-DPH6 enantiomers into the hMAO A binding cavity. Both en-

antiomers showed a relatively similar localization for the hydroxy-

quinoline and benzyl moieties, with the rest of the molecule easily

accommodated in the relatively large cavity of MAO A.

hydrophobic interactions and Trp82 allowed a

pep stacking

interaction with the quinoline ring of the ligand. In Mode I (Fig. 7A),

a close examination of the ﬁrst shell of residues surrounding (R)-

DPH6 and (S)-DPH6 revealed that in both enantiomers the hy-

droxyl group of the quinoline ring formed one hydrogen bond with

the carboxylate group of Glu197 and another one with the hydroxyl

group of the Tyr128. Moreover, the cyano group for the (R)-DPH6

enantiomer, formed a hydrogen bond with the NH₂of the side

chain of Asn83, this bond was bifurcated in the case of (S)-DPH6

enantiomer. In this orientation, hydrophobic interactions with the

catalytic triad residues, Ser198 and His438 were found. In mode II

(Fig. 7B), only one hydrogen bond was observable for the (R)

enantiomer, it was formed between the cyano group and the hy-

droxyl group of Thr120. The (S) enantiomer preserved the cyanide

bifurcated hydrogen bond already observed in Mode I, and the

hydroxyl group of the quinoline ring established a hydrogen bond

with the catalytic triad residue His438.

2.3.3.1. Inhibition of MAO A by (R)- and (S)-DPH6. Fig. 8A showed

that the optimal position for the (R)-DPH6 enantiomer placed the

quinoline ring in an “aromatic cage” formed by Tyr407, Tyr444 side

chains, as well as the isoalloxazine FAD ring. (R)-DPH6 forms pep

stacking interactions with Tyr407 and the carbonyl group of Gln215

residue. The hydroxy-quinoline moiety was involved in two

hydrogen bonds between the OH and the carbonyl oxygen of the

FAD and the 193w molecule. The benzyl group is located in a hy-

drophobic core delimited by residues Phe173, Phe208, Ileu325,

Ala111 Phe112 and Leu176. The cyano group is also able to form a

hydrogen bond with Cys323 side chain. Compound (S)-DPH6

showed a binding geometry very similar to that displayed by (R)-

DPH6 as for the hydroxy-quinoline and benzyl moieties (see

Fig. 8B), which displayed a set of intermolecular interactions

described before in the case of the (R)-enantiomer. The main dif-

ference with respect to the recognition of (R)-DPH6 was in the

spatial orientation of the propargyl, piperidine and cyano groups,

the later accommodated this time to establish hydrogen bond with

Thr336 residue. The docking studies rationalized the relevant

inhibitory activity of DPH6 towards MAO A, as due to the formation

of several favorable interactions with the catalytic site of the

enzyme.

Analysis of the intermolecular interactions indicated key resi-

dues responsible for ligand binding. The cyanide group is likely to

be an important feature for these derivatives to exhibit BuChE

inhibitory activity.

2.3.3. Inhibition of MAO A and MAO B

In order to explore the nature of the ligandereceptor in-

teractions, the ligand was docked to the active site of both MAO A

and MAO B isoforms using the program Autodock Vina [22]. We

have focused on DPH6, which showed the best both MAO A

The importance of the cyano group in properly positioning the

ligand by H-bond formation is pointed out. It is worth noting that

compounds lacking this CN, like in the case of DPH7, showed the

inhibitory activity drastically lowered.

(IC₅₀¼ 1.8 ꢀ 0.1

mM) and MAO B (IC₅₀¼ 1.6 ꢀ 0.25

mM) inhibitory

activities, with signiﬁcant EeAChE and eqBuChE inhibitory po-

tencies. Although rat MAO A (rMAO A) and rat MAO B (rMAO B) are

w90% identical in sequence with human enzymes, their functional

properties are similar but not identical to those of human enzymes.

Given that MAO inhibition assays were carried out on rat brain

mitochondria, docking experiments were carried out using the X-

ray structure of rMAO A (PDB ID: 1O5W) and the homology model

of rMAO B developed from a human MAO B (hMAO B) crystallo-

graphic structure (PDB ID: 1S3E) [28], as previously described for

eqBuChE. The recognition process between (R)- and (S)-enantio-

mers of DPH6 (chosen as reference compound) was theoretically

investigated by blind docking experiments, in accordance with a

protocol previously deﬁned by us and well validated [29]. The

enzymeeinhibitor interactions might allow a theoretical evaluation

of which enantiomer of the inhibitor could be better accommo-

dated into the catalytic site of MAO A and MAO B. Results from

several studies have shown that it must be the neutral amine that

reaches the active site of MAO A and MAO B to allow the chemistry

[30e33]. Therefore, the docking simulations were done using both

enantiomers of DPH6 as neutral species despite of at physiological

pH, most of the piperidine rings would be in the protonated,

positively charged form. In docking with MAO A, during each run,

the side chains of twenty-one residues (Tyr 69, Leu97, Gln99,

Ala111, Phe112, Tyr124, Trp128, Phe173, Leu176, Phe177, Ile180,

Asn181, Ile207, Phe208, Gln215, Cys323, Ileu325, Ileu335, Phe352,

Tyr407 and Tyr444) were allowed to relax with the ligand, while

the remainder of the enzyme was ﬁxed in 3-D space. Six water

molecules labeled as w72, w193, w11, w23, w15, and w53 in

accordance with the numbering reported for the hMAO B crystal-

lographic structure (PDB ID: 1S3E) located near the FAD cofactor

2.3.3.2. Inhibition of MAO B by (R)- and (S)-DPH6. To rationalize the

selectivity towards MAO A and MAO B of (R)- and (S)-DPH6, blind

docking studies of DPH6 into the MAO B were done. Up to date, a

reliable 3-D structure of rMAO B is not available and we used a 3D

homology model of rMAO B for the docking studies. The six

structural water molecules selected for rMAO A were also included

in the model. For MAO B, the inhibitor (R)-DPH6 crosses both

cavities, presenting the piperidine nucleus located between the

“entrance” and “catalytic” cavities, separated by the residues Ile199

and Tyr326. This complex was stabilized by hydrophobic contacts

of quinoline ring with Phe103, Pro104, Trp119, His90, Val316 and

Tyr115. Besides, the phenyl ring is hosted into the “aromatic cage”

framed by Tyr398, Phe343, Tyr435, and the FAD aromatic ring,

where it forms a number of

pep interactions also including

Gln206. No intermolecular H-bonds between ligand and enzyme

were observed (Fig. 9A). For MAO B, the quinoline system of the

inhibitor (S)-DPH6 was hosted in the entrance cavity made up by

lipophilic residues Phe103, Pro104, Trp119, His90, Ileu164, Val316

and Tyr115. The phenyl ring occupied the substrate cavity and was

in direct contact with the Tyr398, Phe343, Tyr435, and the FAD

aromatic ring. The compound established a H-bond with Tyr115 OH

hydrogen by its OH oxygen (Fig. 9B). The study conﬁrmed the

selectivity of DPH6 for MAO A isoform. Selectivity is likely due to

the orientation of the quinoline and phenyl moieties of DPH6 in

MAO A and in MAO B. For MAO B, the quinoline system was hosted

in the entrance cavity and for MAO A this system occupied the

substrate cavity. In this disposition the quinoline moiety interacted

directly with the FAD aromatic ring.

552

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

Fig. 8. Docking poses of inhibitor DPH6 into rMAO A. (A) (R)-DPH6 (purple sticks). (B) (S)-DPH6 (blue sticks). Amino acid residues of the binding site are color-coded. The ﬂavin

adenine dinucleotide cofactor (FAD) and the six water molecules are represented as an integral part of the MAO A structure model and are rendered as yellow sticks and red balls,

respectively. Green dashed lines are drawn among atoms involved in hydrogen bond interactions. (For interpretation of the references to colour in this ﬁgure legend, the reader is

referred to the web version of this article.)

Fig. 9. Docking pose of inhibitor DPH6 into rMAO B. (A) (R)-DPH6 (green sticks). (B) (S)-DPH6 (pink sticks). Amino acid residues of the binding site are color-coded. The ﬂavin

adenine dinucleotide cofactor (FAD) and the six water molecules are represented as an integral part of the MAO B structure model and are rendered as yellow sticks and red balls,

respectively. (For interpretation of the references to colour in this ﬁgure legend, the reader is referred to the web version of this article.)

2.3.4. Predicted binding afﬁnities

to ꢃ12.4 kcal/mol for AChE and ꢃ9.9 and ꢃ9.7 kcal/mol for BuChE.

Consistent with biological trends, both enantiomers were predicted

to have similar binding interactions with AChE and BuChE. The

weakest predicted binding afﬁnities were computed for the both

enantiomers interactions with the MAO A and MAO B enzymes.

Very similar binding afﬁnity values were also observed for both

enantiomers with MAO A and MAO B.

AutoDock Vina provides a computed binding afﬁnity for each

docking mode predicted. Binding afﬁnity data for both enantiomers

with the four enzymes are summarized in Table 2. The more

negative the value, the tighter the predicted bonding. For the (R)-

DPH6 interaction with the AChE, predicted binding afﬁnities were

very similar for the four modes (IeIV), ranging from ꢃ11.0

to ꢃ12.3 kcal/mol. Predicted binding afﬁnities for (R)-DPH6 with

BuChE (ꢃ10.4 and ꢃ10.7 kcal/mol) were slightly lower in range

than those for the AChE complex (Table 2). Data for (S)-DPH6 in-

teractions gave very similar predicted binding than that for (R)-

DPH6 interactions: predicted binding energies of ꢃ11.5

2.4. Theoretic ADMET analysis of DPH hybrids

ADMET properties are important conditions and major parts of

pharmacokinetics. Viable drugs should have proper ADMET prop-

erties to be approved as a drug in clinical tests. The drugs used for

neurological disorder treatment, such as AD, are generally CNS

acting drugs, so factors that are important to the success of CNS

drugs were analyzed. In particular, the new molecules should

present a good CNS penetration proﬁle and low toxic effects. Cur-

rent in silico ADMET predictions cannot fully replace well-

established in vitro cell-based approaches or in vivo assays, but

they can provide signiﬁcant insights [34]. Computer predictions

were performed with ADMET Predictor 6.53 [35] and ACD/Percepta

14.0.04 [36] software packages. According to the predictions

(Table 3), the lipophilicity increases with the hydrocarbon tether

Table 2

Predicted binding afﬁnities (kcal/mol).

DPH

Mode

EeAChE

EqBuChE

MAO A

MAO B

(R)-DPH6

I

ꢃ11.0

ꢃ12.3

ꢃ12.1

ꢃ11.9

ꢃ11.5

ꢃ12.4

ꢃ10.4

ꢃ10.7

ꢃ7.8

ꢃ8.9

II

III

IV

I

II

III

(S)-DPH6

ꢃ9.9

ꢃ9.7

ꢃ7.9

ꢃ8.8

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

553

Table 3

Calculated physicochemical properties for DPH1-7.^a,b

,c

,d

Molecular

weight

No. of H-bond

donors

No. of H-bond

acceptor

No. of rotatable

Bonds^a

logP (Moriguchi)^a

logP^a

TPSA

No. violations

Lipinski’s rule

LogBB^a

1

2

3

4

5

6

7

385.51

424.55

399.54

438.58

413.57

452.60

427.59

379.50

226.28

1

0

1

4

5

4

5

4

5

4

3

6

7

8

9

6

3

3.55

3.09

3.75

3.29

3.95

3.48

4.09

3.52

2.06

3.79

3.51

4.16

3.76

4.54

4.05

4.95

4.59

2.32

39.60

63.39

39.60

63.39

39.60

63.39

39.60

38.78

36.36

0

ꢃ0.12

ꢃ0.31

0.03

ꢃ0.14

0.20

0.07

0.36

0.88

0.17

Donezepil

M-30

,d

LogBB^b

logPS^eLog (PS*fu, brain)^ePeff (cm/s ꢅ 10⁴) Human intestinal In vitro Caco-2

MDCK (cm/s ꢅ 10⁷)^h% Plasma protein

Toxicity^a

absorption (%)^f

perm (nm/sec)^g

binding (in vitro)ⁱ

1

2

3

4

5

6

7

ꢃ0.02

ꢃ0.11

0.29

0.26

0.36

0.34

0.46

0.31

ꢃ0.27

ꢃ1.6

ꢃ1.5

ꢃ1.6

ꢃ1.4

ꢃ1.5

ꢃ1.4

ꢃ1.5

ꢃ2.8

ꢃ2.9

ꢃ3.0

ꢃ3.1

ꢃ3.2

ꢃ3.3

ꢃ3.4

ꢃ3.0

ꢃ2.1

4.47

3.36

4.99

3.75

5.59

4.18

6.24

6.69

2.5

96.79

96.83

96.84

96.91

96.87

96.99

96.90

97.95

95.86

45.66

49.28

52.26

51.96

54.36

54.37

54.08

55.51

42.90

350.7

348.2

325.9

321.9

313.0

306.1

301.9

306.4

429.3

72.68

80.31

76.37

78.44

77.11

79.97

77.40

84.62

69.02

hERG

Donezepil

M-30

a

ADME Predictor, v.6.5.

ACD/Percepta 14.0.

Moriguchi model (ref. [4]).

b

c

d

According to the classiﬁcation made by Ma et al. [9]: High absorption to CNS: logBB more than 0.3; Middle absorption to CNS: logBB 0.3 w ꢃ1.0; Low absorption to CNS:

logBB less than ꢃ1.0.

e

Other estimated parameters related to brain penetration were used to classify the compounds as CNS permeable or non-permeable: rate of brain penetration (LogPS) is the

rate of passive diffusion/permeability; brain/plasma equilibration rate (Log(PS*fu,brain)); fu,brain e fraction unbound in plasma.

f

Human intestinal absorption is the sum of bioavailability and absorption evaluated from ratio of excretion or cumulative excretion in urine, bile, and feces. A value

between 0 and 20% indicates poor absorption, 20ꢃ70% shows moderate absorption, and 70ꢃ100% indicates good absorption.

g

Caco-2 cells are derived from human colon adenocarcinoma and possess multiple drug transport pathways through the intestinal epithelium. A value <4 indicates low

permeability, 4ꢃ70 shows middle permeability, and >70 indicates high permeability.

h

The MDCK cell system may be used as a good tool for rapid permeability screening. A value <25 indicates low permeability, 25ꢃ500 shows middle permeability, and >500

indicates high permeability.

i

The percent of drug binds to plasma protein. A value <90% indicates weak binding, and >90% indicates strong binding to plasma proteins.

chain, and for the same value of n, the

a

-aminonitriles show the

available for diffusion or transport across cell membranes and

thereby ﬁnally interact with the target. Herein, the percent of drug

bound with plasma proteins was estimated and the compounds

were predicted to be weakly bind to plasma proteins.

expected slightly lower lipophilicity. All of the DPH show good

values (log P < 5 and mlog P < 4.15) [37], and as can be deduced, the

whole series of structures meets the Lipinski’s rule of ﬁve [38].

Moreover, a more severe rule for CNS drugs-like characteristics [39]

(MW ꢄ 450, HB donor ꢄ 3, HB acceptors ꢄ 7, log P ꢄ 5, PSA ꢄ 90,

and number of rotatable bonds ꢄ 8) is satisﬁed for all structures,

although the molecular weight of DPH6 is on the borderline.

However, drugs that penetrate CNS should have lower polar surface

areas (PSA) than other kinds of molecules, being the optimal range

Other ADMET predictions of the present DPH compounds show

satisfactory results. Thus, all of them show good intestinal ab-

sorption (HIA [44] of about 96% and effective permeability across

the intestinal membrane [45], Peff > 0.1), moderate apparent

permeability for in vitro MDCK cells (>25) [46] and middle

permeability for in vitro Caco-2 cells [47]. Regarding toxicity

properties, these compounds are predicted to show cardiac toxicity

(hERG potassium channel blockage), but, according to the models

used, they could lack carcinogenicity and hepatotoxicity [48e50].

In summary, these structures show moderate to good ADMET

properties; in particular those with increasing values of n in the

linker present proper drug-like properties and brain penetration

capacity for CNS activity.

2

ꢀ

60e70 A [40]. Therefore, only the a-aminonitriles DPH fall in the

suitable range of values.

The bloodebrain barrier (BBB) is a separation of circulating

blood and cerebrospinal ﬂuid in the central nervous system (CNS),

so an estimation of BBB penetration means predicting whether

compounds pass across the bloodebrain barrier. This is a crucial

pharmacokinetic property in drug design because CNS-active

compounds, such as those for AD, must pass across it. According

to the computed values [41], for the same n, the a-aminonitriles

present slightly lower values, whereas the values increase with the

chain length. In any case, the two models predict that all structures

should be moderate to good candidates, showing middle to high

absorption to CNS. In particular, the DPH3 and DPHs 5e7 show the

best results suggesting a brain penetration sufﬁcient for CNS ac-

tivity [42,43].

Generally, the degree to which any drug binds to plasma protein

inﬂuences not only the drug action but also its disposition and ef-

ﬁcacy. Usually, the drug that is unbound to plasma proteins will be

2.5. Toxicological evaluation of DPH6

Based on the predicted non hepatotoxicity (see above) of the

DHP compounds, next we analyzed our hit DHP6 hybrid in an

in vitro toxicity test in HepG2 cells, determining the cell viability

with MTT method [51], using donepezil as a reference molecule.

The results shown in Table 4 are very interesting as they prove that

DPH6 is by far less toxic than donepezil at high concentrations

(from 30 mM to 1000 mM), while at low concentrations (from 1 mM

554

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

Table 4

In vitro toxicity in HepG2 cells.^a

Compounds

Viability (%) HepG2 cells

1

m

M

3

mM

10

mM

30

mM

100

mM

300

m

M

1000 mM

ns

DPH6

Donepezil

98.4 ꢀ 1.64^ns

94.8 ꢀ 2.42 ^ns

94.4 ꢀ 1.91^ns

92.4 ꢀ 2.20

85.6 ꢀ 4.10*

71.44 ꢀ 3.33***

53.4 ꢀ 1.44***

51.5 ꢀ 3.39***

40.9 ꢀ 1.50***

98.6 ꢀ 0.62^ns

97.0 ꢀ 1.77 ^ns

92.4 ꢀ 1.05^*

83.5 ꢀ 2.15^***

80.0 ꢀ 2.36***

***P ꢄ 0.001, **P ꢄ 0.01, *P ꢄ 0.05 and ^nsnot signiﬁcant, with respect to control group. Comparisons between drugs and control group were performed by one-way ANOVA

followed by the NewmaneKeuls post-hoc test.

a

Cell viability was measured as MTT reduction and data were normalized as % control. Data are expressed as the means ꢀ s.e.m. of triplicate of four different cultures. All

compounds were assayed at increasing concentrations (1e1000 mM).

to 10

proﬁle.

m

M) both are non toxic showing a very similar cell viability

cognitive deﬁcits was reﬂected by a decreased latency in compar-

ison to saline (vehicle) group (***p < 0.001). Not surprisingly,

donepezil þ scop group showed a latency time similar to that

observed in the vehicle group. Likewise, scopolamine effect was

widely reversed in the DPH6 þ scop group (***p < 0.001). Char-

acteristically, not statistical difference in the latency time between

DPH6 þ scop and donepezil þ scop groups were observed and it

was similar to that exhibited by the vehicle.

When comparing individual groups between training and probe

sessions (Fig. 12), it is noteworthy that in the vehicle þ scop group

no signiﬁcant differences were observed in the latency time. In

striking contrast, during the probe session, handling groups: saline,

donepezil þ scop (1 mg/kg) and DPH6 þ scop (35 mg/kg) remained

in the brightly lit compartment for longer time than that measured

in the training session, signiﬁcantly increasing the latency time. The

retention impairment observed in the probe session after a single

dose of scopolamine (1 mg/kg), administered 30 min before

training, is in good agreement with previous reports [56]. Taken

these results together, we might conclude that both DPH6 (35 mg/

kg) and donepezil (1 mg/kgg) are able to induce similar effects on

cognition.

To sum up, hybrid DPH6 preserves all the anti-cholinesterasic

properties related to donepezil, but, as expected, shows an addi-

tional multipotent inhibitory activity in agreement with the in vitro

results previously determined such as MAO inhibition capacity,

biometal (Cu, Zn and Fe) chelating properties, favorable ADMET

data and a notorious lack of hepatotoxicity. Consequently, and

based on these in vitro pharmacological analyses, next we stepped

forward to preliminary in vivo tests in order to investigate the po-

tential biological proﬁle of selected derivative DPH6.

2.6. Relief of scopolamine-induced long-term memory deﬁcit in

mice by DHP6

To analyze the in vivo effect on contextual memory, DPH6 was

administered to scopolamine-treated mice that were subjected to

passive avoidance behavioral test. This task is based on contextual

memory in which the hippocampus plays an important role, with

the association between an environmental context that the animal

learns to avoid and an aversive stimulus (foot-shock). This behav-

ioral test has been extensively used to evaluate learning and

memory in rodent models of central nervous system disorders and

for screening of novel chemical entities on memory function

[53,54]. Scopolamine, a well-known non-selective muscarinic

antagonist that produces amnesic effects in both rodents and hu-

man, is frequently used to perform pharmacological assays for

studying the effects of enhancers on cognitive functions [52].

During the training day, no signiﬁcant differences in the latency

time between treated groups were found (Fig. 10), even though

scopolamine has been reported to inﬂuence motivational behavior

[55]. In the probe session (Fig. 11), scopolamine (1 mg/kg)-induced

3. Conclusion

DPHs 1e7, designed as hybrids from donepezil and M30, bearing

N-benzylpiperidine and a pro-chelator 8-hydroxyquinoline moi-

eties attached to a central N-propargylamine core, have been syn-

thesized and subjected to pharmacological evaluation. DPHs 1e7

were readily prepared in good yields, in short synthetic sequences,

from easily available precursors.

Fig. 11. Comparative effects on latency time of DPH6 and donepezil on scopolamine-

induced amnesia in the probe session. Bars represent the mean of latency time in

seconds (sec) corresponding to each of experimental handling group. SEM is repre-

sented by error bars. Analysis of variance (ANOVA) followed by Bonferroni’s post hoc

test shows signiﬁcant differences (***p < 0.001) between groups DPH6, donepezil or

vehicle with the scopolamine group. Values expressed as mean ꢀ SEM of 8e10 in-

dependent experiments.

Fig. 10. Comparative effect of DPH6, donepezil and scopolamine on latency time in the

training day. Bars represent the mean of latency time in seconds (sec) corresponding to

each of experimental handling group No signiﬁcant differences were found between

groups. Values expressed as mean ꢀ SEM of 8e10 independent experiments.

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

555

Fig. 12. Statistical differences within groups (training versus probe) determined by the Student’s t-test (**p < 0.01; ***p < 0.001).

The biochemical evaluation of these molecules showed DPHs

2e7 hybrids as non-selective ChE and MAO inhibitors with

moderate-to-low micromolar activity. Particularly interesting

resulted DPH6, an irreversible MAO A/B, mixed-type EeAChE

inhibitor able to complex biometals Cu(II) and Zn(II) and Fe(III).

Compared to related M30A [15] (Chart 1), the demethylated

derivative of M30 [15], or M30B [39] (Chart 1), DPH6 showed a

very similar MAO inhibitory proﬁle with biometal chelating

properties.

Computational docking studies of (R)- and (S)-enantiomers of

DPH6 in AChE and BuChE yielded several binding modes for each

enantiomer, but we have not observed signiﬁcant differences be-

tween the binding energies of the different binding patterns for

each enantiomer. For both MAO A and MAO B, only one binding

mode has been proposed based on the most stable complex formed

between the (R)- and (S)-enantiomers of DPH6 and the enzymes,

showing also very similar binding energies. In addition, we have

discovered an essential role for the cyano group in properly posi-

tioning the ligand by hydrogen-bond formation with the amino

acid residues of the binding sites of AChE, BuChE and MAO A. It is

worth noting that compounds lacking the cyano motif, like in the

case of DPH7, showed drastically lowered inhibitory activity.

side effects that include bradycardia, nausea, diarrhea, loss of

appetite and stomach pain [44].

In addition, and according to the theoretic ADMET analysis, DPH

compounds showed proper drug-like properties and brain pene-

tration capacity for CNS activity. In particular, DPH6 is lees toxic tan

donepezil at high concentrations in an in vitro model of toxicity in

HepG2 cells.

Finally, in terms of translational science, it is remarkable the

in vivo effect of DPH6 as enhancer on cognitive functions. Scopol-

amine induced memory deﬁcits is greatly dependent on the

cholinergic system [57]. We have found that donepezil is much

more powerful than DPH6 as cholinesterase inhibitor. This result is

consistent with our own results on ASS234 [8,9] (more potent ChEI

than DPH6) which signiﬁcantly lowers scopolamine-induced

learning deﬁcits in healthy adult mice, suggesting that product

ASS234 works as cognitive enhancer, as efﬁciently as donepezil at

lower dosage, likely useful for the treatment of AD. Given the latter,

the well-balanced anti-cholinesterasic and MAO inhibition proﬁle

in addition to the other attractive pharmacological properties of

DPH6 described here, and that DPH6 has effect as enhancer on

cognitive functions, prompt us to propose DPH6 as the ﬁrst racemic

a-aminonitrile identiﬁed so far as a multifunctional chelator for

Finally, it is worth pointing out that

a

-aminonitriles have been

biometals, able to interact in two key enzymatic systems implicated

in AD, and a new lead compound that deserves further investiga-

tion for the potential treatment of this disease.

To sum up, the well-balanced anti-cholinesterasic and MAO

inhibition proﬁle, in addition to the other attractive pharmacolog-

ical properties described here, prompt us to propose DPH6 as the

seldom investigated as ChE inhibitors [40], and to the best of our

knowledge, they have never been analyzed as MAO inhibitors [41].

However, some works have described nitriles as MAOIs [42], and

Tipton and collaborators have reported that cyanide potentiates the

inhibition of MAO by the irreversible inhibitor, phenelzine and

pheniprazine [43]. One of the reviewers has kindly addressed our

attention to the fact that “historically, prescribing MAO inhibitors

have been reserved as a lasteline of treatment because of poten-

tially lethal effects due to drug interactions with a wide variety of

therapeutic agents”. Also, donepezil produces a wide assortment of

ﬁrst racemic

a-aminonitrile that has been identiﬁed as a multi-

functional chelator for biometals, able to interact in two key

enzymatic systems implicated in AD, and a new lead compound

that deserves further investigation for the potential treatment of

this disease.

556

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

4. Experimental section

was not detected); ¹³C NMR (100 MHz, CDCl₃)

d 29.6, 38.0, 51.3,

53.0, 58.5, 63.0, 73.1, 80.0, 108.8, 121.4, 124.9, 127.1, 127.7, 128.2 (2C),

129.2, 129.3, 133.9, 138.7, 147.5, 151.8. Anal. Calcd. for C₂₅H₂₇N₃O: C,

77.89; H, 7.06; N, 10.90. Found: C, 77.88; H, 7.10; N, 10.89.

4.1. Chemistry

4.1.1. General methods

Melting points were determined on a Kofﬂer apparatus, and are

uncorrected. ¹H NMR and ¹³C NMR spectra were recorded in CDCl₃

or DMSO-d₆at 300, 400 or 500 MHz and at 75, 100 or 125 MHz,

respectively, using solvent peaks [CDCl₃: 7.27 (D), 77.2 (C) ppm; and

DMSO-d₆: 2.49 (D), 40 (C) ppm] as internal reference. The assign-

ment of chemical shifts WAs based on standard NMR experiments

(¹H, ¹³C, DEPT, COSY, gHSQC, gHMBC). Mass spectra were recorded

on a GC/MS spectrometer with an API-ES ionization source.

Elemental analyses were performed at CNQO (CSIC, Spain). TLC

were performed on silica F254 and detection by UV light at 254 nm

or by charring with either ninhydrin, anisaldehyde or phospho-

molybdic-H₂SO₄dyeing reagents. Anhydrous solvents were used in

all experiments. Column chromatography was performed on silica

gel 60 (230 mesh). All known compounds have been synthesized as

reported. All compounds were ꢁ95% purity as determined by ex-

amination of their combustion analyses.

4.1.4. 2-(1-Benzylpiperidin-4-yl)-2-(((8-hydroxyquinolin-5-yl)

methyl)(prop-2-yny)amino)acetonitrile (DPH2)

To a cooled solution of 1-benzyl-4-piperidinecarboxaldehyde

[58] (407 mg, 2.0 mmol) and propargylamine (165 mg, 3.0 mmol)

in MeOH (6 mL) at 0 ^ꢂC, a small amount of CF₃CO₂H (5 drops) was

added. After being stirred for 1 h, NaBH₃CN (189 mg, 2.9 mmol) was

added to the solution portionwise. The mixture was stirred at rt

overnight and then quenched with aqueous saturated NaHCO₃. The

mixture was concentrated in vacuo and then extracted with AcOEt.

The combined organic layers were dried over MgSO₄and concen-

trated under reduced pressure to give a mixture that was puriﬁed

by column chromatography (SiO₂, eluting solvent was changed

from hexane/AcOEt ¼ 5:1 to AcOEt and then AcOEt/MeOH ¼ 4:1 v/

v) to give N-[(1-benzylpiperidin-4-yl)methyl]prop-2-yn-1-amine

(10) (232 mg, 50.2%), as an orange oil [R_f¼ 0.23 (AcOEt/

MeOH ¼ 4:1); IR (ﬁlm)

n

3287, 3028, 2918, 2801, 2758, 1495, 1454,

1.15e1.55

1366, 1342, 1121, 1078 cm^ꢃ1; ¹H NMR (400 MHz, CDCl₃)

d

4.1.2. 5-(Chloromethyl) quinolin-8-ol (8) [17]

(m, 3H), 1.72 (d, J ¼ 12.5 Hz, 2H), 1.95e2.11 (m, 2H), 2.20 (t,

J ¼ 2.3 Hz, 1H), 2.57 (d, J ¼ 6.6 Hz, 2H), 2.94 (d, J ¼ 11.7 Hz, 2H),

3.33e3.65 (m, 4H), 7.19e7.40 (m, 5H) (the NH signal was not

d 30.1, 35.8, 38.4, 53.4, 54.5,

63.1, 71.4, 82.2, 127.3, 128.3, 129.4, 137.3], and (1-benzylpiperidin-4-

To a cooled solution of 8-hydroxyquinoline (14.6 g,100 mmol) in

conc. HCl (44 mL) at 0 ^ꢂC, a 37% aqueous formaldehyde solution

(20 mL) was added. Then HCl (g) was bubbled through the solution

with stirring for 2 h. The mixture was allowed to warm to rt with

further stirring for 6 h and without stirring for 2 h more. The

product was ﬁltered and the solid was rinsed with conc. HCl, giving

product 8 [17] (19.9 g, 77.5%).

observed); ¹³C NMR (100 MHz, CDCl₃)

yl)(prop-2-yn-1-ylamino)acetonitrile (11) (165 mg, 30.9%) as an

orange oil [R_f¼ 0.78 (AcOEt/MeOH ¼ 4:1); IR (ﬁlm)

2920, 2803, 2760, 1732, 1452, 1368, 1246, 1223, 1072 cm^ꢃ1; ¹H NMR

(400 MHz, CDCl₃) 1.37e1.72 (m, 4H), 1.76e1.88 (m, 2H), 1.92e2.02

(m, 2H), 2.26e2.33 (m, 1H), 2.87e2.98 (m, 2H), 3.44e3.69 (m, 5H),

7.17e7.44 (m, 5H); ¹³C NMR (100 MHz, CDCl₃)

28.4, 29.0, 36.8,

n 3296, 2940,

d

4.1.3. 5-(((1-Benzylpiperidin-4-yl)(prop-2-ynyl)amino)methyl)

quinolin-8-ol (DPH1)

d

To a solution of commercial 1-benzyl-4-piperidone (1.3 g,

7.0 mmol) and propargylamine (580 mg, 10.5 mmol) in MeOH

(22 mL) at 0 ^ꢂC, was added a small amount of CF₃CO₂H (5 drops).

After being stirred for 1 h, NaBH₃CN (1.3 g,19.7 mmol) was added to

the solution. The mixture was stirred at 0 ^ꢂC overnight and then

quenched with aqueous saturated NaHCO₃. The mixture was

concentrated in vacuo and extracted with AcOEt. The combined

organic layers were dried over MgSO₄and concentrated under

reduced pressure. The crude product was puriﬁed by column

chromatography (SiO₂, AcOEt/MeOH ¼ 4:1 v/v) to give 1-benzyl-N-

(prop-2-ynyl)aminopiperidine (9) (500 mg, 31.2%) as an orange oil

39.0, 53.0, 54.5, 63.0, 72.8, 80.1, 118.7, 127.1, 128.2, 129.1, 138.3].

To a solution of compound 11 (70 mg, 0.29 mmol) and Et₃N

(0.085 mL, 0.58 mmol) in CH₂Cl₂(2 mL), chloride 8 (70 mg,

0.25 mmol) was added. After stirring for 30 min at rt, brine was

added to the mixture. The product was extracted with CH₂Cl₂, the

extract was dried over Mg₂SO₄and then concentrated in vacuo. The

crude product was puriﬁed by column chromatography (hexane/

AcOEtby increasing the gradient from 5:1 to 1:1 v/v), to give

compound DPH2 (99 mg, 80.6%) as a white solid: IR (KBr)

3300, 3026, 2949, 2814, 1580, 1504, 1476, 1454, 1424, 1371, 1269,

1229, 1193, 1150, 1072 cm^ꢃ1; ¹H NMR (600 MHz, CDCl₃)

0.68e0.70

n 3443,

d

[R_f¼ 0.19 (AcOEt/MeOH ¼ 4:1); ¹H NMR (400 MHz, CDCl₃)

d

1.34e

(m, 1H), 1.17e1.22 (m, 1H), 1.65e1.95 (m, 5H), 2.41 (s, 1H), 2.65 (s,

1H), 2.81e2.83 (m, 1H), 3.23e3.29 (m, 3H), 3.34 (s, 1H), 3.45e3.48

(m, 1H), 3.68 (d, J ¼ 13.2 Hz, 1H), 4.63 (d, J ¼ 13.8 Hz, 1H), 7.1e7.20

(m, 1H), 7.21e7.28 (m, 5H), 7.40e7.44 (m, 2H), 8.60 (d, J ¼ 8.4 Hz,

1.49 (m, 2H), 1.76e1.88 (m, 2H), 2.00e2.15 (m, 3H), 2.17e2.22 (t,

J ¼ 2.3 Hz, 1H), 2.65e2.76 (m, 1H), 2.81e2.90 (m, 2H), 3.41e3.46 (d,

J ¼ 2.3 Hz, 2H), 3.52 (s, 2H), 7.20e7.35 (m, 5H); ¹³C NMR (100 MHz,

CDCl₃)

d

31.8, 34.9, 51.8, 62.7, 62.8, 71.1, 82.0, 126.8, 128.0, 128.9,

1H), 8.76 (d, J ¼ 4.2 Hz, 1H) (the signal for OH was not detected); ¹³

C

138.0]. To a solution of compound 9 (475 mg, 2.1 mmol) and

quinoline 8 (387 mg, 1.5 mmol) in CH₂Cl₂(12 mL), was added Et₃N

(0.84 mL, 6.0 mmol) at rt. After being stirred overnight, the mixture

was quenched with aqueous saturated NaHCO₃. The product was

extracted with CH₂Cl₂. The organic layer was dried over Mg₂SO₄

and concentrated in vacuo. The product was puriﬁed by column

chromatography (SiO₂, hexane/AcOE by increasing the gradient

from 5:1 to 3:1 v/v). Further puriﬁcation was achieved by recrys-

tallization (hexane/AcOEt ¼ 1:1) to give compound DPH1 (196 mg,

NMR (150 MHz, CDCl₃) d 29.4, 30.7, 36.0, 40.0, 52.5, 54.2, 57.7, 63.1,

74.0, 78.8, 84.7, 108.9, 111.8, 116.0, 121.7, 122.2, 127.2, 127.5, 128.2,

129.3, 130.0, 134.0, 138.8, 148.9, 152.6. Anal. Calcd for C₂₇H₂₈N₄O: C,

76.39; H, 6.65; N, 13.20. Found: C, 76.45; H, 6.75; N, 13.17.

4.1.5. 5-((((1-Benzylpiperidin-4-yl)methyl)(prop-2-ynyl)amino)

methyl)quinolin-8-ol (DPH3)

To a stirring solution of compound 10 (638 mg, 2.6 mmol) and

chloride 8 (694 mg, 2.7 mmol) in CH₂Cl₂(12 mL), Et₃N (1.09 mL,

5.2 mmol) was added at rt. After being stirred overnight, the

mixture was quenched with water and the product was extracted

with CH₂Cl₂, dried over Mg₂SO₄, concentrated in vacuo, and puri-

ﬁed by column chromatography (SiO₂, hexane/AcOEt by increasing

the gradient from 5:1 to 1:1 v/v) to give amine DPH3 (572 mg, 55%)

33.8%) as a white solid: mp 145.5 ^ꢂC; IR (KBr)

n

3289, 2958, 2914,

2805, 2754, 1582, 1510, 1422, 1360, 1281, 1229, 1186, 1152, 1107,

1069, 1001 cm^ꢃ^{1 1}H NMR (400 MHz, CDCl₃)

1.60e1.76 (m, 2H),

;

d

1.90e2.02 (m, 4H), 2.17 (t, J ¼ 2.3 Hz, 1H), 2.48e2.60 (m, 1H), 2.91

(d, J ¼ 11.5 Hz, 2H), 3.16 (d, J ¼ 2.3 Hz, 2H), 3.46 (s, 2H), 4.00 (s, 2H),

7.00 (d, J ¼ 7.6 Hz, 1H), 7.15e7.29 (m, 5H), 7.31e7.40 (m, 2H), 8.60

(dd, J ¼ 8.5, 1.6 Hz, 1H), 8.69 (dd, J ¼ 4.3, 1.5 Hz, 1H) (the OH signal

as a light brown solid: mp 128.5 ^ꢂC; IR (KBr)

2914, 2754, 2365, 1734, 1580, 1506, 1478, 1422, 1366, 1279, 1229,

n 3320, 3248, 2946,

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

557

1192, 1128, 1065 cm^ꢃ1

;

¹H NMR (400 MHz, CDCl₃)

d

0.97e1.14 (m,

(t, J ¼ 7.2 Hz, 2H), 2.84 (d, J ¼ 11.5 Hz, 2H), 3.30e3.55 (m, 4H), 7.12e

7.33 (m, 5H) (the NH signal was not observed); ¹³C NMR (100 MHz,

CDCl₃) d 32.0, 33.3, 36.3, 38.0, 45.9, 53.5, 63.1, 71.2, 82.0, 127.0, 128.0

(2C), 129.2 (2C), 137.5], and 3-(1-benzylpiperidin-4-yl)-2-(prop-2-

yn-1-ylamino)propanenitrile (16) (190 mg, 40.2%) as an orange

2H), 1.32e1.48 (m, 1H), 1.60 (d, J ¼ 12.5 Hz, 2H), 1.84 (t, J ¼ 11.0 Hz,

2H), 2.19 (s, 1H), 2.36 (d, J ¼ 7.2 Hz, 2H), 2.76 (d, J ¼ 11.1 Hz, 2H), 3.15

(d, J ¼ 1.9 Hz, 2H), 3.39 (s, 2H), 3.83 (s, 2H), 6.98 (d, J ¼ 7.8 Hz, 1H),

7.12e7.37 (m, 7H), 8.53 (dd, J ¼ 8.5, 1.4 Hz, 1H), 8.68 (dd, J ¼ 4.2,

1.4 Hz,1H) (the signal for OH was not detected); ¹³C NMR (100 MHz,

oil [R_f¼ 0.56 (MeOH/AcOEt ¼ 1:4); IR (ﬁlm)

2782, 1493, 1452, 1388, 1343, 1125, 1074 cm^ꢃ1; ¹H NMR (400 MHz,

CDCl₃)

n 3295, 2924, 2905,

CDCl₃) d 30.8, 33.5, 40.8, 53.6, 56.6, 59.1, 63.5, 73.4, 78.4,108.7,121.3,

124.9, 126.9, 127.8, 128.1, 129.1, 129.2, 134.1, 138.5, 138.7, 147.5, 151.9.

Anal. Calcd for C₂₆H₂₉N₃O: C, 78.16; H, 7.32; N, 10.52. Found; C,

78.05; H, 7.36; N, 10.40.

d

1.15e1.38 (m, 2H), 1.43e1.73 (m, 5H), 1.92 (t, J ¼ 11.5 Hz,

2H), 2.21 (t, J ¼ 2.3 Hz, 1H), 2.82 (d, J ¼ 10.7 Hz, 2H), 3.35e3.64 (m,

4H), 3.66e3.78 (m, 1H), 7.13e7.33 (m, 5H) (the NH signal was not

observed); ¹³C NMR (100 MHz, CDCl₃)

d 31.7, 31.8, 32.2, 36.4, 39.9,

4.1.6. (1-Benzylpiperidin-4-ylidene)acetonitrile (12) [59]

46.9, 53.2, 63.2, 72.7, 79.8 (2C), 119.6, 126.9, 128.0 (2C), 129.0 (2C),

138.1].

A

solution of diethyl (cyanomethyl)phosphonate (2.13 g,

12 mmol) [prepared by heating triethylphosphite (1.0 equiv) and

chloroacetonitrile (1.0 equiv) at 150 ^ꢂC for 3.5 h; the crude product

was directly used in the next reaction] and K₂CO₃(1.39 g, 10 mmol)

in dry THF (5 mL) was stirred for 15 min at rt. Then the mixture was

heated to reﬂux for 20 min. After cooling down to rt, 1-benzyl-4-

piperidone (1.90 g, 10 mmol) was added dropwise to this solu-

tion. Then the mixture was heated at reﬂux for 12 h. After cooling

down to rt, 10% K₂CO₃aqueous solution was added. The reaction

mixture was extracted with AcOEt, and the organic layers were

dried over MgSO₄and concentrated in vacuo to give the crude

product 12 (2.7 g, >99.0%), which was directly used in next step.

To a cooled mixture of 16 (84 mg, 0.3 mmol) and Et₃N (0.093 mL,

0.67 mmol) in CH₂Cl₂(6 mL), chloride 8 (77 mg, 0.3 mmol) was

added at 0 ^ꢂC. After being stirred at rt overnight, the reaction was

quenched with aqueous 5% NaHCO₃. The reaction mixture was

extracted with CH₂Cl₂, and extracts were dried with Mg₂SO₄and

concentrated in vacuo. The crude product was puriﬁed by column

chromatography (hexane: AcOEt by increasing the gradient from

5:1 to 1:1 v/v) to give compound DPH4 (91.4 mg, 69.5%) as an or-

ange oil: IR (ﬁlm)

1233, 1198, 1148, 1072, 1028 cm^ꢃ1

0.81e0.93 (m,1H), 0.93e1.07 (m,1H),1.12e1.32 (m, 2H),1.38e1.62

n

3298, 2934, 2807, 1580, 1505, 1476, 1370, 1273,

;

¹H NMR (400 MHz, CDCl₃)

d

(m, 3H), 1.69 (ddd, J ¼ 14.1, 8.8, 5.5 Hz, 1H), 1.76e1.89 (m, 1H), 2.39e

2.47 (m, 1H), 2.58 (d, J ¼ 11.3 Hz, 1H), 2.80 (d, J ¼ 11.5 Hz, 1H), 3.21e

3.32 (m, 1H), 3.36e3.53 (m, 3H), 3.65e3.75 (m, 2H), 4.60 (d,

J ¼ 13.1 Hz, 1H), 7.06e7.11 (m, 1H), 7.21e7.33 (m, 5H), 7.40e7.47 (m,

2H), 8.58e8.65 (m, 1H), 8.75e8.81 (m, 1H) (the signal for OH was

4.1.7. (1-Benzylpiperidin-4-yl)acetonitrile (13) [60]

To a solution of nitrile 12 (2.7 g, 10 mmol) in MeOH (100 mL), Mg

(4.6 g, 191 mmol) and inﬁnitesimal quantity of I₂was added. The

mixture was stirred until it became gray gel. After conc. HCl was

added, the mixture became clear solution. Then it was treated with

10 N NaOH to alkaline. The precipitate was ﬁltered and washed

with large amount of EtOAc. The ﬁltrate was extracted with AcOEt,

and the combined organic layers were dried over MgSO₄and

concentrated in vacuo to give the product 13 (1.59 g, 74.5%).

not detected); ¹³C NMR (100 MHz, CDCl₃)

d 30.6, 31.5, 37.2, 39.8,

49.6, 52.7, 53.0, 53.6, 62.8, 74.1, 78.6, 108.9, 117.0, 121.6, 122.3, 127.2,

127.5, 128.1, 129.3, 129.8, 133.6, 138.6, 147.7, 152.5. HRMS. Calcd. for

C

₂₈H₃₁N₄O (M þ H)^þ: 439.2428. Found: 439.2532 (M þ H)^þ.

4.1.10. 5-(((2-(1-Benzylpiperidin-4-yl)ethyl)(prop-2-ynyl)amino)

methyl)quinolin-8-ol (DPH5)

4.1.8. (1-Benzylpiperidin-4-yl)acetaldehyde (14) [61]

To an oven-dried and argon-purged ﬂask were added the nitrile

13 (1.29 g, 6.0 mmol) and THF (13 mL). The mixture was cooled

to ꢃ78 ^ꢂC, and DIBAL-H (6.22 mL, 1 mmol/mL) was added to the

reaction via syringe. The reaction was stirred at ꢃ78 ^ꢂC for 1 h, and

then quenched with aqueous saturated NaHCO₃. The precipitate

was ﬁltered and washed with large amount of EtOAc. The ﬁltrate

was extracted with EtOAc. The combined organic layers were

washed with brine and dried over MgSO₄. After concentrated in

vacuum, the crude product was puriﬁed by chromatography (SiO₂,

CH₂Cl₂/MeOH ¼ 20:1 v/v) to give product 10 (365 mg, 27.8%).

To a cooled mixture of 15 (46 mg, 0.18 mmol) and Et₃N

(0.053 mL, 0.38 mmol) in CH₂Cl₂(2.5 mL) at 0 ^ꢂC, chloride 8 (47 mg,

0.18 mmol) was added. After being stirred at rt overnight, the re-

action was quenched with aqueous 5% NaHCO₃. The product was

extracted with CH₂Cl₂, and extracts were dried over Mg₂SO₄and

then concentrated in vacuo. The crude product was puriﬁed by

column chromatography (hexane/AcOE tby increasing the gradient

from 5:1 to 1:1 v/v) to give compound DPH5 (40 mg, 53.5%) as a

yellow oil: IR (ﬁlm)

1478, 1420, 1375, 1350, 1279, 1231, 1194, 1148, 1121 cm^ꢃ1

(400 MHz, CDCl₃)

2H), 2.67 (t, J ¼ 2.40 Hz, 1H), 2.58 (t, J ¼ 7.20 Hz, 2H), 2.83 (d,

J ¼ 11.20 Hz, 2H), 3.26 (d, J ¼ 2.40 Hz, 2H), 3.48 (s, 2H), 3.92 (s, 2H),

7.08 (d, J ¼ 8.00 Hz, 1H), 7.30e7.44 (m, 7H), 8.61e8.61 (m, 1H),

8.76e8.77 (m, 1H) (the signal for OH was not detected);¹³C NMR

n

3254, 2945, 2915, 2805, 2758, 1578, 1508,

;

¹H NMR

d

1.24 (s, 3H), 1.44 (d, J ¼ 33.20 Hz, 4H), 1.82 (m,

4.1.9. 3-(1-Benzylpiperidin-4-yl)-2-(((8-hydroxyquinolin-5-yl)

methyl)(prop-2-ynyl)amino)propanenitrile (DPH4)

A solution of 14 (365 mg, 1.68 mmol) and propargylamine

(187 mg, 3.4 mmol) in MeOH (10 mL) was stirred at 0 ^ꢂC for 1 h;

then NaBH₃CN (214 mg, 3.4 mmol) was added. The mixture was

stirred at rt overnight. Water was added to the mixture and MeOH

was removed under reduced pressure. Then aqueous saturated

NaHCO₃was added, and the mixture was extracted with AcOEt. The

separated organic layers were dried over MgSO₄and concentrated

in vacuo to give the crude product. Further puriﬁcation was ach-

ieved by column chromatography (SiO₂, eluting solvent was

changed with gradient from hexane/AcOEt 5:1 to 1:5v/v and then

change from AcOEt to AcOEt/MeOH 4:1 v/v) to give N-[2-(1-

benzylpiperidin-4-yl)ethyl]prop-2-yn-1-amine (15) (168 mg,

(100 MHz, CDCl₃) d 32.2, 33.5, 33.9, 41.0, 50.3, 53.7, 55.9, 63.3, 72.3,

78.3, 108.7, 121.4, 124.9, 127.0, 127.9, 128.1, 129.2, 129.3, 134.0, 138.7,

147.5, 151.8. HRMS. Calcd. for C₂₇H₃₂N₃O (M þ H)^þ: 414.2545.

Found: 414.2570 (M þ H)^þ.

4.1.11. (2E)-3-(1-benzylpiperidin-4-yl)prop-2-enenitrile (17) [62]

A

mixture of diethyl (cyanomethyl)phosphonate (2.2 g,

12 mmol) and K₂CO₃(1.4 g, 10 mmol) in dry THF (100 mL) was

stirred at rt for 15 min, and then heated at reﬂux for 20 min. After

cooling down to rt, 1-benzyl-4-piperidinecarboxaldehyde [58]

(2.0 g, 10 mmol) was added. The mixture was heated to reﬂux for

3 h. Aqueous 10% K₂CO₃water (100 mL) was added after cooling

down to rt. The product was extracted with AcOEt and dried over

MgSO₄. After concentrated in vacuum, the crude product was

38.9%), as an orange oil [R_f¼ 0.11 (MeOH/AcOEt ¼ 1:4); IR (ﬁlm)

3302, 2922, 2945, 2801, 2758, 1493, 1454, 1366, 1148, 1121, 1078,

1028 cm^ꢃ1; ¹H NMR (400 MHz, CDCl₃)

1.15e1.45 (m, 5H), 1.61 (d,

J ¼ 10.5 Hz, 2H), 1.94 (t, J ¼ 11.1 Hz, 2H), 2.13 (t, J ¼ 2.1 Hz, 1H), 2.63

n

d

558

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

puriﬁed by column chromatography (SiO₂, hexane/AcOEt from 5:1

to 1:1 v/v) to give nitrile 17 (0.74 g, 32.6%).

(ﬁlm)

n

3295, 2931, 2812, 1578, 1505, 1476, 1371, 1271, 1203, 1196,

0.99e0.77

1148, 1123, 1072, 1028 cm^ꢃ1; ¹H NMR (400 MHz, CDCl₃)

d

(m, 1H), 1.40e1.38 (m, 6H), 1.53e1.40 (m, 1H), 1.84e1.60 (m, 5H),

2.42 (t, J ¼ 2.4 Hz, 1H), 2.77 (ddt, J ¼ 12.0, 8.3, 2.1 Hz, 1H), 3.28 (dd,

J ¼ 16.8, 2.4 Hz, 1H), 3.45 (s, 2H), 3.58 (dd, J ¼ 16.8, 2.4 Hz, 1H), 3.75

(d, J ¼ 13.1 Hz, 1H), 4.60 (d, J ¼ 13.1 Hz, 1H), 7.11 (d, J ¼ 7.6 Hz, 1H),

7.32e7.22 (m, 5H), 7.57e7.39 (m, 2H), 8.64 (dd, J ¼ 8.5, 1.6 Hz, 1H),

8.78e8.80 (m, 1H) (the signal for OH was not detected); ¹³C NMR

4.1.12. 3-(1-Benzylpiperidin-4-yl)propanenitrile (18) [63]

To a solution of 17 (1.74 g, 7.70 mmol) in MeOH (33 mL) at rt,

turning of Mg (3.70 g, 154 mmol) and inﬁnitesimal quantity of I₂

was added to the mixture. The mixture was stirred until it became

gray gel. After conc. HCl was added, the mixture became clear so-

lution. Then it was treated with 10 N NaOH to alkaline. The pre-

cipitates were ﬁltered and washed with large amount of EtOAc. The

ﬁltrate was extracted with AcOEt, and the combined organic layers

were dried over MgSO₄and concentrated in vacuo to give the crude

product 18 (1.76 g, 99.9%).

(100 MHz, CDCl₃)

d 28.1, 31.5, 31.9, 32.0, 34.4, 39.8, 52.3, 53.4, 53.5,

53.7, 63.3, 74.2, 78.6, 108.9, 117.2, 121.8, 122.5, 127.1, 127.6, 128.2,

129.3, 129.9, 133.7, 138.8, 147.9, 152.6. HRMS. Calcd for C₂₉H₃₃N₄O

(M þ H)^þ: 453.2654. Found: 452.2526 (M þ H)^þ.

4.1.15. 5-(((3-(1-Benzylpiperidin-4-yl)propyl)(prop-2-ynyl)amino)

methyl)quinolin-8-ol (DPH7)

4.1.13. 3-(1-Benzylpiperidin-4-yl)propanal (19) [64]

To an oven-dried and argon-purged ﬂask were added the nitrile

18 (1.00 g, 4.4 mmol) and THF (10 mL). The reaction was cooled

to ꢃ78 ^ꢂC, and DIBAL-H (4.54 mL, 1 mmol/mL) was added to the

reaction via syringe. The mixture was stirred at ꢃ78 ^ꢂC for 1 h, and

then quenched with aqueous saturated NaHCO₃. The precipitates

were ﬁltered and washed with large amount of EtOAc. The ﬁltrate

was extracted with EtOAc. The combined organic layers were

washed with brine and dried over MgSO₄. After concentrated in

vacuum, the crude product was puriﬁed by chromatography (SiO₂,

CH₂Cl₂/MeOH ¼ 20:1 v/v), to give aldehyde 19 (427 mg, 42.7%).

To a cooled mixture of amine 20 (189 mg, 0.70 mmol) and Et₃N

(0.51 mL, 2.5 mmol) in CH₂Cl₂(15 mL) at 0 ^ꢂC, chloride 8 (186 mg,

0.73 mmol) was added. After being stirred at rt overnight, the re-

action was quenched with aqueous 5% NaHCO₃. The product was

extracted with CH₂Cl₂, and extracts were dried over Mg₂SO₄and

then concentrated in vacuo. The crude products were puriﬁed by

column chromatography (SiO₂, hexane/AcOEt by increasing the

gradient from 5:1 to 1:1 v/v) to give compound DPH7 (162 mg,

54.3%) as a yellow oil: IR (ﬁlm)

1505, 1476, 1425, 1373, 1289, 1238, 1198, 1047 cm^ꢃ1

(300 MHz, CDCl₃) 0.99e1.33 (m, 5H), 1.42e1.66 (m, 4H), 1.86 (t,

n

3381, 3293, 2911, 2801, 1738, 1580,

;

¹H NMR

d

4.1.14. 4-(1-Benzylpiperidin-4-yl)-2-(((8-hydroxyquinolin-5-yl)

methyl)(prop-2-yny)amino)butanenitrile (DPH6)

J ¼ 11.3 Hz, 2H), 2.29 (t, J ¼ 2.1 Hz, 1H), 2.56 (t, J ¼ 7.1 Hz, 2H), 2.86

(d, J ¼ 11.4 Hz, 2H), 3.26 (d, J ¼ 2.2 Hz, 2H), 3.48 (s, 2H), 3.93 (s, 2H),

7.09 (d, J ¼ 7.6 Hz, 1H), 7.20e7.37 (m, 5H), 7.37e7.49 (m, 2H), 8.65

(dd, J ¼ 8.5,1.3 Hz,1H), 8.77 (dd, J ¼ 4.1,1.3 Hz,1H) (the signal for OH

A solution of aldehyde 19 (427 mg, 1.9 mmol) and propargyl-

amine (204 mg, 3.7 mmol) in MeOH (6 mL) was stirred at 0 ^ꢂC for

1 h, then NaBH₃CN (1.3 g, 2.0 mmol) was added. The mixture was

stirred at rt overnight. Water was added to the mixture and MeOH

was removed under reduced pressure. Then, aqueous saturated

NaHCO₃was added, and the mixture was extracted with AcOEt. The

separated organic layers were dried over MgSO₄and concentrated

in vacuo to give the crude products. Further puriﬁcation was ach-

ieved by column chromatography (SiO₂, eluting solvent was

changed with gradient from hexane/AcOEt 5:1 to 1:5v/v and then

change from AcOEt to AcOEt/MeOH 4:1 v/v) to give N-[3-(1-

benzylpiperidin-4-yl)propyl]prop-2-yn-1-amine (20) (221 mg,

was not detected); ¹³C NMR (100 MHz, CDCl₃)

d 24.4, 32.2, 34.0,

35.3, 40.9, 53.2, 53.8 (2C), 55.8, 63.4, 73.4, 78.4, 108.9, 121.4, 124.9,

127.0, 127.9, 128.2, 129.2, 129.3, 134.1, 138.7, 147.6, 152.0. Anal. Calcd

for C₂₈H₃₃N₃O: C, 78.65; H, 7.78; N, 9.83. Found; C, 76.90; H, 8.03; N

9.98.

4.1.16. 4-(1-Benzylpiperidin-4-yl)-2-(prop-2-yn-1-ylamino)

butanenitrile (21)

3-(1-Benzylpiperidin-4-yl)propanal (19) (136 mg, 0.59 mmol),

prop-2-yn-1-amine (47

mL, 0.74 mmol, 1.25 equiv) and trime-

44.2%), as a yellow oil [R_f¼ 0.19 (MeOH/AcOEt ¼ 1:4); IR (ﬁlm)

3304, 3028, 2922, 2847, 2794, 2758, 1495, 1452, 1386, 1343, 1119,

1028 cm^ꢃ^{1 1}H NMR (600 MHz, CDCl₃)

1.10e1.26 (m, 5H), 1.33e

n

thylsilyl cyanide (TMSCN) (0.14 mL, 1.1 mmol, 1.85 equiv) were

mixed and submitted to microwave irradiation at 125 ^ꢂC for 10 min.

The crude reaction was puriﬁed by column chromatography (SiO₂,

hexane/AcOEt from 9:1 to 1:1 v/v) to give compound 21 (104 mg,

60%) as a yellow oil, that showed identical spectroscopic data to the

compound isolated in the Mannich reductive amination of alde-

;

d

1.46 (m, 2H),1.58 (d, J ¼ 9.9 Hz, 2H),1.87 (br s, 2H), 2.13 (t, J ¼ 2.3 Hz,

1H), 2.31e2.39 (m, 1H), 2.59 (t, J ¼ 7.1 Hz, 1H), 2.81 (d, J ¼ 10.8 Hz,

2H), 3.35 (d, J ¼ 2.3 Hz, 2H), 3.43 (s, 2H), 7.13e7.29 (m, 5H) (the NH

signal was not observed); ¹³C NMR (150 MHz, CDCl₃)

d

24.6, 27.0,

hyde 19. a-Aminonitrile 21 was also transformed into target com-

32.2, 32.3, 34.1, 35.6, 38.0, 48.8, 53.8, 63.4, 71.1, 82.2, 126.7, 128.0,

129.1, 138.4], and 4-(1-benzylpiperidin-4-yl)-2-(prop-2-yn-1-

ylamino)butanenitrile (21) (234 mg, 42.7%), as an yellow oil

pound DPH6 (see above).

4.2. Pharmacological evaluation

[R_f¼ 0.59 (MeOH/AcOEt ¼ 1:4); IR (ﬁlm)

2780, 1493, 1452, 1368, 1343, 1312, 1260, 1146, 1119, 1074; ¹H NMR

(400 MHz, CDCl₃) 1.18e1.69 (m, 8H), 1.70e1.83 (m, 2H), 1.85e2.02

n 3296, 2911, 2845, 2801,

4.2.1. Cholinesterase inhibition

d

Cholinesterase activities were assessed following a spec-

(m, 2H), 2.23e2.37 (m, 1H), 2.87 (d, J ¼ 10.5 Hz, 2H), 3.29e3.76 (m,

trophometric method [18], using puriﬁed AChE from Electrophorus

electricus (type VeS) and puriﬁed BuChE from equine serum

(lyophilized powder) (SigmaeAldrich, Madrid, Spain). Enzymatic

reactions took place in 96-well plates in solutions containing 0.1 M

phosphate buffer (pH 8.0), 0.035 U/mL AChE or 0.05 U/mL BuChE

and 0.35 mM of 5,5⁰-dithiobis-2-nitrobenzoic acid (DTNB, Sigmae

Aldrich, Madrid, Spain). Inhibition curves were plotted by pre-

incubating this mixture with serial dilutions of each compound

for 20 min. The activity in absence of compounds was used to

determine the 100% of enzyme activity. After pre-incubation times,

4H), 7.16e7.38 (m, 5H) (the NH signal was not observed); ¹³C NMR

(100 MHz, CDCl₃)

d 31.0, 31.6, 32.1, 32.2, 32.3, 35.4, 36.6, 49.5, 53.7,

63.4, 72.8, 80.0, 119.4, 126.7, 127.9, 128.9, 138.1].

To a mixture of amine 21 (89 mg, 0.3 mmol) and Et₃N (0.09 mL,

0.7 mmol) in CH₂Cl₂(6 mL) at 0 ^ꢂC, chloride 8 (77 mg, 0.3 mmol)

was added. After being stirred at rt overnight, the reaction was

quenched with aqueous 5% NaHCO₃and the product was extracted

with CH₂Cl₂. The extracts were dried Mg₂SO₄, and concentrated in

vacuo. The crude products were puriﬁed by column chromatog-

raphy (SiO₂, hexane/AcOEt by increasing the gradient from 5:1 to

1:1 v/v), to give compound DPH6 (89 mg, 65.4%) as a yellow oil: IR

50

ml of substrate were added to a ﬁnal concentration of 0.35 mM

acetylthiocholine iodide or 0.5 mM butyrylthiocholine iodide

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

559

(SigmaeAldrich). Enzymatic reactions were followed for 5 min

with AChE or 25 min with BuChE. Changes in absorbance at l₄₀₅in a

spectrophotometic plate reader (FluoStar OPTIMA, BMG Labtech)

were detected and IC₅₀values calculated at the compound con-

centration inhibiting 50% of enzymatic activity by using the

GraphPad ‘PRISM’ software (version 3.0). Data were expressed as

mean ꢀ SEM of at least three different experiments performed in

triplicate.

biometals CuSO₄, ZnSO₄or Fe₂(SO₄) ₃were prepared at a ﬁnal sum

of concentrations of both species of 10 M, varying the proportions

of both components between 0 and 100%. Absorbance at 257 nm

m

was plotted versus the mole fraction of DPH6 for each metal.

4.2.4. Toxicity in human hepatoma cell line HepG2: culture of

HepG2 liver cells and treatment

Human hepatoma cell line HepG2 was cultured in Eagle’s min-

imum essential medium (EMEM) supplemented with 15 non-

4.2.1.1. Kinetic studies of AChE inhibition by DHP6. To estimate the

mechanism of action of DPH6 inhibiting AChE, reciprocal plots of 1/

V versus 1/[S] were determined at different acetylthiocholine

essential amino acids,

inactivated fetal bovine serum (FBS), 100 units/mL penicillin, and

100 g/mL streptomycin (reagents from Invitrogen, Madrid, Spain).

1 mM sodium pyruvate, 10% heat-

m

(ASCh) concentrations (0.1e5

weighted least-squares analysis. Slopes of reciprocal plots were

plotted against concentration of DPH6 (0e5 M) for AChE to

m

M). The plots were assessed by a

Cells were seeded into ﬂasks containing supplemented medium

and maintained at 37 ^ꢂC in a humidiﬁed atmosphere of 5% CO₂and

95% air. For assays, HepG2 cells were subcultured in 96-well plates

at seeding density of 8 ꢅ 10⁴cells per well. When HepG2 cells

reached 80% conﬂuence, the medium was replaced with fresh

m

determine the reversible inhibition constant (K_i).

4.2.2. Monoamine oxidase inhibition

medium containing 1e1000 mM compounds or 0.1% DMSO as a

To assess the inhibition of MAO A and MAO B by derivatives

DHPs 1e7, ¹⁴C-labeled substrates were used (Perkin Elmer, USA).

Rat liver homogenates were used as source of MAO [19]. MAO A

vehicle control.

4.2.5. MTT assay and cell viability

activity was determined using 100

hydroxytryptamine) (5-HT) whereas MAO B activity was deter-

mined using 20

M (2.5 mCi/mmol) [¹⁴C]-phenylethylamine (PEA).

m

M (0.5 mCi/mmol) [¹⁴-C]-(5-

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)

m

Inhibition curves were plotted by pre-incubating the enzyme with

several concentrations of each compound for 30 min in 50 mM

phosphate buffer (pH 7.4). Inhibitor-free samples were used to

determine the 100% of enzyme activity. At the end of each pre-

(Sigma Aldrich), as described previously [51]. Brieﬂy, 50 ml of the

MTT labeling reagent, at a ﬁnal concentration of 0.5 mg/ml, was

added to each well at the end of the incubation time and the plate

placed in a humidiﬁed incubator at 37 ^ꢂC with 5% CO₂and 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

EMEM were used as 100% viability.

incubation, substrate (25

ml) was added in a ﬁnal volume of

225 l and reactions allowed for 20 min (MAO A) or 4 min (MAO B).

m

Reactions were stopped by adding 2 M citric acid and radiolabelled-

aldehyde products extracted into toluene/ethylacetate (1:1, v/v)

containing 0.6% (w/v) 2,5-diphenyloxazole prior to liquid scintil-

lation counting (Tri-Carb 2810TR). Inhibition curves were plotted as

previously mentioned. Total protein was measure by the method of

Bradford (1976) using bovine-serum albumin as standard.

4.2.6. In vivo analysis: relief of scopolamine-induced long-term

memory deﬁcit in mice by DHP6

4.2.6.1. Animals. Thirty-six male C57BL/6J mice (Harlan), weighing

25 g served as subjects in order to obtain our results. Animals were

housed under controlled light (with a 12-h light/12-h dark cycle,

lights on at 7:00 a.m.). Procedures were carried out following the

European Communities Council Directive (86/609/EEC) on animal

experiments.

4.2.2.1. Reversibility and time-dependence of MAO by DHP6.

The study of reversibility inhibition exerted by DPH6 towards MAO

A and MAO B were determined by incubating the enzyme in the

presence and in the absence of the inhibitor before and after three

consecutive washings with buffer. MAO A and MAO B samples were

pre-incubated for 30 min at 37 ^ꢂC with 10

mM DPH6 and 20

mM

4.2.6.2. Drugs and treatments. Scopolamine hydrobromide, 1 mg/

kg (Sigma, St. Louis, MO, USA), donepezil, 1 mg/kg (Tocris Biosci-

ence, R&D Systems Inc., Minneapolis, USA), and DPH6 (35 mg/kg)

were given in 10 ml/kg of saline solution (0.9% NaCl) by intraperi-

toneal via. Animals were divided into four experimental handling

groups: mice administered with i) saline (vehicle, n ¼ 9); ii)

scopolamine (vehicle þ scop, n ¼ 9); iii) donepezil plus scopol-

amine (donepezil þ scop, n ¼ 10); and iv) DPH6 plus scopolamine

(DPH6 þ scop, n ¼ 8).

DPH6, respectively. A sample of 50 nM clorgyline and 20 nM l-

deprenyl was also used as control of irreversible MAO A and MAO B

inhibition. Samples were washed with 50 mM phosphate buffer

(pH 7.4) and centrifuged at 25,000 g for 10 min at 4 ^ꢂC consecutively

three times. Total protein was measured by the Bradford method

and MAO activity determined as described above. Time-

dependence inhibition of DPH6 towards MAO A and MAO B was

evaluated by pre-incubating varying concentrations of the inhibitor

with the enzyme at different times (0e180 min). Dose-response

curves (IC₅₀) values were accordingly determined for each time as

described above.

4.2.6.3. Passive avoidance task. The test was performed using the

Ugo Basile (Comerio, Italy) apparatus. Basically, the device consists

of two compartments (10 ꢅ 13 ꢅ 15 cm) connected by a sliding

door. One compartment is brightly lit (10 W) and, on the contrary,

the other one is dark and equipped with an electriﬁed grid ﬂoor.

Rodents tend to prefer dark environments and will immediately

enter the darkened compartment. The day before of the experi-

ment, mice are placed in the experimental room for 1 h. The

experiment is performed in two consecutive days. The ﬁrst day,

scopolamine or saline is administered 30 min before the training

session and saline, donepezil or DPH6 are administered 90 min

prior to the administration of scopolamine or saline. During the

training session, each mouse is individually placed in the

4.2.3. Determination of metal-chelating properties

Complexing studies were performed in distilled water at room

temperature using a UVeVIS spectrophotometer (Lambda25, Per-

kinElmer). Spectrums (220e300 nm) of DPH6 alone (10 mM) and in

presence of varying concentrations of CuSO₄(no changes observed

after 5-min incubation), Fe₂(SO)₃(no changes observed following

O/N incubation) and ZnSO₄(no changes observed after 1-h incu-

bation) were recorded in 1 cm quartz cells. The stoichiometry of the

complexes DPH6-Cu(II)/Zn(II)/Fe(III) was determined by the Job’s

method [65]. Series of different solutions containing DPH6 and

560

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

illuminated compartment with the sliding door closed. 30 sec later,

the door is open so that the mouse can move freely to each room.

Once the mouse gets into the dark compartment, the sliding door

closes automatically and the animal receives an electric foot shock

(0.5 mA, 1 s). Then, mice are returned their home cage and 24 h

later, in the second day, the probe session takes place. In the second

day, each individual is placed again into illuminated box. Then the

sliding door opens for 5 s with the electric foot-shock switched off

in the dark room. The latency in seconds taken by a mouse to enter

the dark compartment after door opening during the training and

the probe session was automatically determined by the computer

device. A cut-off time of 5 min was deﬁned.

putative three-dimensional structure of eqBuChE has been created

based on the crystal structure of hBuChE (PDB ID: 2PM8), these two

enzyme exhibited 89% sequence identity. Proper bonds, bond or-

ders, hybridization and charges were assigned using protein model

tool in Discovery Studio, version 2.1, software package. CHARMm

force ﬁeld was applied using the receptoreligand interactions tool

in Discovery Studio, version 2.1, software package. Docking calcu-

lations were performed following the same protocol described

before for EeAChE. All dockings were performed as blinds dockings

where a cube of 75 Ǻ with grid points separated 1 Ǻ, was positioned

at the middle of the protein (x ¼ 29.885; y ¼ ꢃ54.992; z ¼ 58.141).

Default parameters were used except num_modes, which was set

to 40. The lowest docking-energy conformation was considered as

the most stable orientation. Finally, the docking results generated

were directly loaded into Discovery Studio, version 2.1.

4.2.6.4. Statistical analysis. Signiﬁcance was determined using a

one-way analysis of variance (ANOVA), followed by Bonferroni’s

post hoc test. Statistical analysis of differences within a group be-

tween training and probe sessions were determined by using the

Student’s t- test. These statistical analyses have been carried out by

using GraphPad Prism 5 computer software. Results are expressed

as the average of latencies ꢀS.E.M. Differences are considered sta-

tistically signiﬁcant at p < 0.05.

4.2.7.4. Molecular docking of compounds (R)-DPH6 and (R)-DPH7

into rat MAO A/B. Compounds (R)-DPH6 and (R)-DPH7 were

assembled as non-protonated amine within Discovery Studio,

version 2.1, software package, following the procedure described

before for cholinesterases. The crystal structure of rat MAO A in

complex with its irreversible inhibitor MLG-709 was obtained from

the Protein Data Bank (PDB ID 1O5W). The rat MAO B model has

been retrieved from the SWISS-MODEL Repository. A putative

three-dimensional structure of rat MAO B has been created based

on the crystal structure of hMAO B (PDB ID: 1S3E), these two en-

zymes exhibited 89% sequence identity. For docking studies initial

proteins were prepared. First, in the PDB crystallographic structure

1O5W (rat MAO A), any co-crystallized solvent and the ligand were

removed; it is not necessary in the PDB MAO B model. Then, proper

bonds, bond orders, hybridization and charges were assigned using

protein model tool in Discovery Studio, version 2.1, software

package. CHARMm force ﬁeld was applied using the receptore

ligand interactions tool in Discovery Studio, version 2.1, software

package. Six water molecules located around the FAD cofactor were

considered in the docking experiments because of their well-

known role into the MAO’s inhibition. Finally, atoms of the FAD

cofactor were deﬁned in their oxidized state. Docking calculations

were performed following the same protocol described before for

EeAChE. In docking with MAO A, Tyr 69, Leu97, Gln99, Ala111,

Phe112, Tyr124, Trp128, Phe173, Leu176, Phe177, Ile180, Asn181,

Ile207, Phe208, Gln215, Cys323, Ileu325, Ileu335, Phe352, Tyr407

and Tyr444 receptor residues were selected to keep ﬂexible during

docking simulation using the AutoTors module. All dockings were

performed as blind dockings where a cube of 40 Ǻ with grid points

separated 1 Ǻ, was centered on the FAD N5. Default parameters

were used except num_modes, which was set to 40. According to

Vina best scored poses, the most stable complex conﬁgurations

were considered. At the end of the docking, the best poses were

analyzed using Discovery Studio.

4.2.7. Molecular modeling

4.2.7.1. Molecular docking into AChE and BuChE. (R)-DPH6 and (S)-

DPH6 were assembled as hydrochlorides and free bases within

Discovery Studio, version 2.1, software package, using standard

bond lengths and bond angles. With the CHARMm force ﬁeld [66]

and partial atomic charges, the molecular geometries of (R)-DPH6

and (S)-DPH6 were energy-minimized using the adopted-based

NewtoneRaphson algorithm. Structures were considered fully

optimized when the energy changes between iterations were less

than 0.01 kcal/mol [67].

4.2.7.2. Molecular docking of (R)-DPH6 and (S)-DPH6 into EeAChE.

The coordinates of E. electricus AChE (PDB ID: 1C2B), were obtained

from the Protein Data Bank (PDB). For docking studies, initial pro-

tein was prepared by removing all water molecules, heteroatoms,

any co-crystallized solvent and the ligand. Proper bonds, bond or-

ders, hybridization and charges were assigned using protein model

tool in Discovery Studio, version 2.1, software package. CHARMm

force ﬁeld was applied using the receptoreligand interactions tool

in Discovery Studio, version 2.1, software package. Docking calcu-

lations were performed with the program Autodock Vina

[22]. AutoDockTools (ADT; version 1.5.4) was used to add hydro-

gens and partial charges for proteins and ligands using Gasteiger

charges. Flexible torsions in the ligands were assigned with the

AutoTors module, and the acyclic dihedral angles were allowed to

rotate freely. Trp286, Tyr124, Tyr337, Tyr72, Asp74, Thr75, Trp86,

and Tyr341 receptor residues were selected to keep ﬂexible during

docking simulation using the AutoTors module. Because VINA uses

rectangular boxes for the binding site, the box center was deﬁned

and the docking box was displayed using ADT. For E. electricus AChE

(PDB ID: 1C2B) the docking procedure was applied to whole protein

target, without imposing the binding site (“blind docking”). A grid

box of 60 ꢅ 60 ꢅ 72 with grid points separated 1 Ǻ, was positioned

at the middle of the protein (x ¼ 21.5911; y ¼ 87.752; z ¼ 23.591).

Default parameters were used except num_modes, which was set

to 40. The AutoDock Vina docking procedure used was previously

validated [23].

4.3. Statistical analysis

Signiﬁcance was determined using a one-way analysis of vari-

ance (ANOVA), followed by Bonferroni’s post hoc test. Signiﬁcant

differences within group were determined with the Student’s t-

test. Results are expressed as the average of latencies ꢀ S.E.M.

Differences are considered statistically signiﬁcant at p < 0.05.

4.2.7.3. Molecular docking of inhibitors (R)-DPH6 and (S)-DPH6 into

eqBuChE. The horse BuChE model has been retrieved from the

SWISS-MODEL Repository. This is a database of annotated three-

dimensional comparative protein structure models generated by

the fully automated homology-modeling pipeline SWISS-MODEL. A

Acknowledgments

Supported by MINECO (SAF2012-33304 to JMC; to SAF2010-

15173 RMM), MICINN for a grant (CTQ2009-10478 to ES), and EU

(COST Action CM1103). Oscar M. Bautista-Aguilerta thanks MINECO

L. Wang et al. / European Journal of Medicinal Chemistry 80 (2014) 543e561

561

[29] A. Samadi, C. de los Ríos, I. Bolea, M. Chioua, I. Iriepa, I. Moraleda, M. Bartolini,

V. Andrisano, E. Gálvez, C. Valderas, M. Unzeta, J.L. Marco-Contelles, European

Journal of Medicinal Chemistry 52 (2012) 251e262.

(Spain) for a FPI fellowship. G.E. and L.W have equally contributed

to this work.

[30] D.E. Edmonson, A. Mattevi, C. Binda, M. Li, F. Hubálek, Current Medicinal

Chemistry 11 (2004) 1983e1993.

[31] T.Z.E. Jones, D. Balsa, M. Unzeta, R.R. Ramsay, Journal of Neural Transmission

114 (2007) 707e712.

Appendix A. Supplementary data

[32] J. Wang, D.E. Edmonson, Biochemistry 50 (2011) 7710e7717.

[33] R.A. Scerrer, A.J. Le, Molecular Informatics 30 (2011) 386.

[34] F. Cheng, W. Li, G. Liu, Y. Tang, Current Topics in Medicinal Chemistry 13

(2013) 1273e1289.

Supplementary data related to this article can be found at http://

dx.doi.org/10.1016/j.ejmech.2014.04.078.

[35] AMET Predictor, v.6.5, Simulations Plus, Inc., Lancaster, CA, 2013.

[36] ACD/Percepta 14.0.0, Advanced Chemistry Development, 2013.

[37] I. Moriguchi, S. Hirono, Q. Liu, I. Nakagome, Y. Matsushita, Chemical & Phar-

maceutical Bulletin 40 (1992) 127e130.

[38] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Advanced Drug Delivery

Reviews 46 (2001) 3e26.

[39] H. Pajouhesh, G.R. Lenz, Drugs NeuroRx 2 (2005) 541e553.

[40] J. Kelder, P.D.J. Grootenhuis, D.M. Bayada, L.P.C. Delbressine, J.-P. Ploemen,

Pharmaceutical Research 16 (1999) 1514e1519.

[41] P.A. Nielsen, O. Andersson, S.H. Hansen, K.B. Simonsen, G. Andersson, Drug

Discovery Today 16 (2011) 472e475.

References

[1] M. Goedert, M.G. Spillantini, Science 314 (2006) 777e781.

[2] A. Castro, A. Martínez, Current Pharmaceutical Design 12 (2006) 4377e4387.

[3] A. Gella, N. Durany, Cell Adhesion & Migration 3 (2009) 88e93.

[4] C. Geula, M.M. Mesulam, Neuroscience 33 (1989) 469e481.

[5] E. Scarpini, P. Scheltens, H. Feldman, Lancet Neurology 2 (2003) 539e547.

[6] V.N. Talesa, Mechanisms of Ageing and Development 122 (2001) 1961e1969.

[7] A. Cavalli, M.L. Bolognesi, A. Minarini, M. Rosini, V. Tumiatti, M. Recanatini,

C. Melchiorre, Journal of Medicinal Chemistry 51 (2008) 347e372.

[8] I. Bolea, J. Juárez-Jiménez, C. de los Ríos, M. Chioua, R. Pouplana, F.J. Luque,

M. Unzeta, J. Marco-Contelles, A. Samadi, Journal of Medicinal Chemistry 54

(2011) 8251e8270.

[42] X. Ma, C. Chen, J. Yang, Acta Pharmacologica Sinica 26 (2005) 500e512.

[43] F. Cheng, W.L. Li, Y. Zhou, J. Shen, Z. Wu, G. Liu, P.W. Lee, Y. Tang, Journal of

Chemical Information and Modeling 52 (2012) 3099e3105.

[44] S. Yee, Pharmaceutical Research 14 (1997) 763e766.

[9] I. Bolea, A. Gella, L. Monjas, C. Perez, M.I. Rodríguez-Franco, J. Marco-Contelles,

A. Samadi, M. Unzeta, Currrent Alzhimer Research 9 (2013) 797e808.

[10] J. Mitoma, A. Ito, Journal of Biochemistry 111 (1992) 20e24.

[11] T.P. Singer, Chemistry and Biochemistry of Flavoenzymes, F. Müller, (Ed.), CRC

Press, Boca Ratón, FL, USA, vol. III, pp. 437ꢃ470.

[45] A. Sjöberg, M. Lutz, C. Tannergren, C. Wingolf, A. Borde, A.I. Ungell, European

Journal of Pharmaceutical Sciences 48 (2013) 166e180.

[46] J.D. Irvine, L. Takahashi, K. Lockhart, J. Cheong, J.W. Tolan, H.E. Selick,

J.R. Grove, Journal of Pharmaceutical Sciences 88 (1999) 28e33.

[47] S. Yamashita, T. Furubayashi, M. Kataoka, T. Sakane, H. Sezaki, H. Tokuda,

European Journal of Pharmaceutical Sciences 10 (2000) 195e204.

[48] L.S. Gold, N.B. Manley, T.H. Slone, L. Rohrbach, Environmental Health Per-

spectives 107 (Suppl. 4) (1999) 527e600.

[12] S.Y. Son, J. Ma, Y. Kondou, M. Yoshimura, E. Yamashita, T. Tsukihara, Pro-

ceedings of the National Academy of Sciences of the United States of America

105 (2008) 5739e5744.

[13] C. Binda, M. Li, F. Hubalek, N. Restelli, D.E. Edmondson, A. Mattevi, Proceedings

of the National Academy of Sciences of the United States of America 100

(2003) 9750e9755.

[14] V. Pérez, J.L. Marco, E. Fernández-Alvarez, M. Unzeta, British Journal of

Pharmacology 127 (1999) 869e876.

[15] Y. Avramovich-Tirosh, T. Amit, O. Bar-Am, H. Zheng, M. Fridkin,

M.B.H. Youdim, Journal of Neurochemistry 100 (2007) 490e502.

[16] H. Zheng, M.B.H. Youdim, M. Fridkin, Journal of Medicinal Chemistry 52

(2009) 4095e4098.

[49] E.J. Matthews, N.L. Kruhlak, R.D. Benz, J.F. Contreras, Current Drug Discovery

Technologies 1 (2004) 61e76.

[50] L.G. Valerio Jr., K.B. Arvidson, R.F. Chanderbhan, J.F. Contrera, Toxicology and

Applied Pharmacology 222 (2007) 1e16.

[51] F. Denizot, R. Lang, Journal of Immunological Methods 89 (1986) 271e277.

[52] R.A. Lenz, J.D. Baker, C. Locke, L.E. Rueter, E.G. Mohler, K. Wesnes, W. biSaab,

M.D. Saltarelli, Psychopharmacology (Berl) 220 (2012) 97e107.

[53] P.J. Baarendse, G.G. Van, R.F. Jansen, A.W. Pieneman, S.O. Ogren, M. Verhage,

O. Stiedl, Hippocampus 18 (2008) 11e19.

[17] V. Moret, Y. Laras, T. Cresteil, G. Aubert, D.Q. Ping, C. Di, M. Barthélémy-

Requin, C. Béclin, V. Peyrot, D. Allegro, A. Rolland, F. De Angelis, E. Gatti,

Ph. Pierre, L. Pasquini, E. Petrucci, U. Testa, J.L. Kraus, European Journal of

Medicinal Chemistry 44 (2009) 558e567.

[54] G.R. Dawson, G. Bentley, F. Draper, W. Rycroft, S.D. Iversen, P.G. Pagella,

Pharmacology Biochemistry and Behavior 39 (1991) 865e871.

[55] M. Grant, Journal of Comparative and Physiological Psychology 86 (1974)

853e857.

[18] G.L. Ellman, K.D. Courtney, V. Andres, R.M. Featherstone, Biochemical Phar-

macology 7 (1961) 88e95.

[56] R. Gupta, L.K. Gupta, P.K. Mediratta, S.K. Bhattacharya, Pharmacological Re-

ports: PR 64 (2012) 438e444.

[19] C.J. Fowler, K.F. Tipton, Biochemical Pharmacology 30 (1981) 3329e3332.

[20] A. Ozaita, G. Olmos, M.A. Boronat, J.M. Lizcano, M. Unzeta, J.A. García-Sevilla,

British Journal of Pharmacology 121 (1997) 901e912.

[57] I. Izquierdo, Trends in Pharmacological Sciences 10 (1989) 175e177.

[58] H. Sugimoto, Y. Iimura, Y. Yamanishi, K. Yamatsu, Bioorganic & Medicinal

Chemistry Letters 2 (1992) 871e876.

[21] X. Huang, M.P. Cuanjungco, C.S. Atwood, M.A. Hartshorn, J.D. Tyndall,

G.R. Hanson, K.C. Stokes, M. Leopold, G. Multhaup, L.E. Goldstein, R.C. Scarpa,

A.J. Saunders, J. Lim, R.D. Moir, C. Glabe, E.F. Bowden, C.L. Masters, D.P. Fairlie,

R.E. Tanzi, A.I. Bush, The Journal of Biological Chemistry 274 (1999) 37111e

37116.

[22] O. Trott, A.J. Olson, Journal of Computational Chemistry 31 (2010) 455e461.

[23] M. Bartolini, M. Pistolozzi, V. Andrisano, J. Egea, M.G. López, I. Iriepa,

I. Moraleda, E. Gálvez, J.L. Marco-Contelles, A. Samadi, ChemMedChem 6

(2011) 1990e1997.

[59] J.M. Contreras, Y.M. Rival, S. Chayer, J.-J. Bourguignon, C.G. Wermuth, Journal

of Medicinal Chemistry 42 (1999) 730e741.

[60] D. Shao, C. Zou, C. Luo, X. Tang, Y. Li, Bioorganic & Medicinal Chemistry Letters

14 (2004) 4639e4642.

[61] M.I. Rodríguez-Franco, M.I. Fernández-Bachiller, C. Pérez, A. Castro,

A. Martínez, Bioorganic & Medicinal Chemistry 13 (2005) 6795e6802.

[62] H. Takasugi, A. Kuno, M. Ohkubo, WO 9313083, 1993.

[63] H. Sugimoto, Y. Tsuchiya, H. Sugumi, K. Higurashi, N. Karibe, Y. Iimura,

A. Sasaki, Y. Kawakami, T. Nakamura, S. Araki, Y. Yamanishi, K. Yamatsu,

Journal of Medicinal Chemistry 33 (1990) 1880e1887.

[64] R. Kitbunnadaj, O.P. Zuiderveld, I.J.P. De Esch, R.C. Vollinga, R. Bakker, M. Lutz,

A.L. Spek, E. Cavoy, M.-F. Deltent, W.M.P.B. Menge, H. Timmerman, R. Leurs,

Journal of Medicinal Chemistry 46 (2003) 5445e5457.

[24] C. Martins, M.C. Carreiras, R. Léon, C. de los Ríos, M. Bartolini, V. Andrisano,

I. Iriepa, I. Moraleda, E. Gálvez, M.G. García, J. Egea, A. Samadi, M. Chioua,

́

J.L. Marco-Contelles, European Journal of Medicinal Chemistry 46 (2011)

6119e6130.

[25] K. Arnold, L. Bordoli, J. Kopp, T. Schwede, Bioinformatics 22 (2006) 195e201.

[26] F. Kiefer, K. Arnold, M. Künzli, L. Bordoli, T. Schwede, Nucleic Acids Research

37 (2009) D387eD392.

[27] M.C. Peitsch, Bio/Technology 13 (1995) 658e660.

[28] C. Binda, P. Newton-Vinson, F. Hubalek, N. Restelli, D.E. Edmondson,

A. Mattevi, Nature Structural & Molecular Biology 2 (2002) 22e26.

[65] C.Y. Huang, Methods in Enzymology 87 (1982) 509e521.

[66] B.R. Brooks, R.E. Bruccoleri, B.D. Olafson, D.J. States, S. Swaminathan,

M. Karplus, Journal of Computational Chemistry 4 (1983) 187e217.

[67] A. Morreale, F. Maseras, I. Iriepa, E. Gálvez, J. Mol, Graphics Modell 2 (2002)

111e118.

Article Doi

Donepezil + propargylamine + 8-hydroxyquinoline hybrids as new multifunctional metal-chelators, ChE and MAO inhibitors for the potential treatment of Alzheimer's disease

DOI: 10.1016/j.ejmech.2014.04.078

Source and publish data:

Authors:

Article abstract of DOI:10.1016/j.ejmech.2014.04.078

Full text of DOI:10.1016/j.ejmech.2014.04.078

Products guided by the article

R&D Labs maybe for 1019508-30-5

Relevant to this article

Hot Product