Design, synthesis and pharmacological evaluation of N-benzyl-piperidinyl-aryl-acylhydrazone derivatives as donepezil hybrids: Discovery of novel multi-target anti-alzheimer prototype drug candidates

Article abstract of DOI:10.1016/j.ejmech.2018.01.066

A new series of sixteen multifunctional N-benzyl-piperidine-aryl-acylhydrazones hybrid derivatives was synthesized and evaluated for multi-target activities related to Alzheimer's disease (AD). The molecular hybridization approach was based on the combination, in a single molecule, of the pharmacophoric N-benzyl-piperidine subunit of donepezil, the substituted hydroxy-piperidine fragment of the AChE inhibitor LASSBio-767, and an acylhydrazone linker, a privileged structure present in a number of synthetic aryl- and aryl-acylhydrazone derivatives with significant AChE and anti-inflammatory activities. Among them, compounds 4c, 4d, 4g and 4j presented the best AChE inhibitory activities, but only compounds 4c and 4g exhibited concurrent anti-inflammatory activity in vitro and in vivo, against amyloid beta oligomer (AβO) induced neuroinflammation. Compound 4c also showed the best in vitro and in vivo neuroprotective effects against AβO-induced neurodegeneration. In addition, compound 4c showed a similar binding mode to donepezil in both acetylated and free forms of AChE enzyme in molecular docking studies and did not show relevant toxic effects on in vitro and in vivo assays, with good predicted ADME parameters in silico. Overall, all these results highlighted compound 4c as a promising and innovative multi-target drug prototype candidate for AD treatment.

Full text of DOI:10.1016/j.ejmech.2018.01.066

Accepted Manuscript
Design, synthesis and pharmacological evaluation of N-benzyl-piperidinyl-aryl-
acylhydrazone derivatives as donepezil hybrids: Discovery of novel multi-target anti-
alzheimer prototype drug candidates
Flávia Pereira Dias Viegas, Matheus de Freitas Silva, Miguel Divino da Rocha, Maísa
Rosa Castelli, Mariana Máximo Riquiel, Rafael Pereira Machado, Sarah Macedo
Vaz, Laís Medeiros Simões de Lima, Karla Cristine Mancini, Patrícia Cruz Marques
de Oliveira, Élida Parreira Morais, Vanessa Silva Gontijo, Fernanda Motta R. da
Silva, Dora D'Alincourt da Fonseca Peçanha, Newton Gonçalves Castro, Gilda A.
Neves, Alexandre Giusti-Paiva, Fabiana Cardoso Vilela, Lidiane Orlandi, Ihosvany
Camps, Márcia Paranho Veloso, Luis Felipe Leomil Coelho, Marisa Ionta, Guilherme
Álvaro Ferreira-Silva, Rodrigo Machado Pereira, Laurent E. Dardenne, Isabella Alvim
Guedes, Wellerson de Oliveira Carneiro Junior, Paula Maria Quaglio Bellozi, Antônio
Carlos Pinheiro de Oliveira, Fábio Furlan Ferreira, Letizia Pruccoli, Andrea Tarozzi,
Claudio Viegas, Jr.
PII:
S0223-5234(18)30079-5
10.1016/j.ejmech.2018.01.066
EJMECH 10143
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 12 December 2017
Revised Date: 18 January 2018
Accepted Date: 19 January 2018
Please cite this article as: Flá.Pereira. Dias Viegas, M. de Freitas Silva, M. Divino da Rocha, Maí.Rosa.
Castelli, Mariana.Má. Riquiel, R.P. Machado, S.M. Vaz, Laí.Medeiros. Simões de Lima, K.C. Mancini,
Patrí.Cruz. Marques de Oliveira, É.Parreira. Morais, V.S. Gontijo, F.M.R. da Silva, D. D'Alincourt
da Fonseca Peçanha, Newton.Gonç. Castro, G.A. Neves, A. Giusti-Paiva, F.C. Vilela, L. Orlandi,
I. Camps, Má.Paranho. Veloso, L.F. Leomil Coelho, M. Ionta, Guilherme.Á. Ferreira-Silva, R.M.
Pereira, L.E. Dardenne, I.A. Guedes, W. de Oliveira Carneiro Junior, P.M. Quaglio Bellozi, Antô.Carlos.
Pinheiro de Oliveira, Fá.Furlan. Ferreira, L. Pruccoli, A. Tarozzi, C. Viegas Jr., Design, synthesis and
pharmacological evaluation of N-benzyl-piperidinyl-aryl-acylhydrazone derivatives as donepezil hybrids:
Discovery of novel multi-target anti-alzheimer prototype drug candidates, European Journal of Medicinal
Chemistry (2018), doi: 10.1016/j.ejmech.2018.01.066.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT
Design, Synthesis and Pharmacological Evaluation of N-Benzyl-
Piperidinyl-Aryl-Acylhydrazone Derivatives as Donepezil Hybrids:
Discovery of Novel Multi-Target Anti-Alzheimer Prototype Drug
Candidates
1
1
1
Flávia Pereira Dias Viegas , Matheus de Freitas Silva , Miguel Divino da Rocha , Maísa
1
1
1
1
Rosa Castelli , Mariana Máximo Riquiel , Rafael Pereira Machado , Sarah Macedo Vaz ,
Laís Medeiros Simões de Lima , Karla Cristine Mancini , Patrícia Cruz Marques de
Oliveira , Élida Parreira Morais , Vanessa Silva Gontijo , Fernanda Motta R. da Silva ,
Dora D’Alincourt da Fonseca Peçanha , Newton Gonçalves Castro , Gilda A. Neves ,
Alexandre Giusti-Paiva , Fabiana Cardoso Vilela , Lidiane Orlandi , Ihosvany Camps ,
Márcia Paranho Veloso , Luis Felipe Leomil Coelho , Marisa Ionta , Guilherme Álvaro
Ferreira-Silva , Rodrigo Machado Pereira , Laurent E. Dardenne , Isabella Alvim
1
1
1
1
1
2
2
2
2
4
4
4
5
6
7
4
4
4
8
8
9
9
Guedes , Wellerson de Oliveira Carneiro Junior , Paula Maria Quaglio Bellozi , Antônio
9
10
11
11
Carlos Pinheiro de Oliveira , Fábio Furlan Ferreira , Letizia Pruccoli , Andrea Tarozzi
1
and Claudio Viegas Jr. *
1
Institute of Chemistry, Laboratory of Research on Medicinal Chemistry, Federal University of Alfenas, MG,
Brazil, 37133-840
2
Laboratory of Molecular Pharmacology, Institute of Biomedical Sciences, Federal University of Rio de
Janeiro, RJ, 21941-902, Brazil.
3
Institute of National Sciences, Federal University of Alfenas, MG, 37130-000, Brazil
4
Institute of Biomedical Sciences, Federal University of Minas Gerais, MG, 37130-000, Brazil
5
Institute of Exact Sciences, Federal University of Alfenas, MG, 37130-000, Brazil
6
Faculty of Pharmacy, Federal University of Alfenas, MG, 37130-000, Brazil
7
Laboratory of Vaccines, Institute of Biomedical Sciences, Federal University of Alfenas, MG, 37130-000,
Brazil.
8
National Laboratory of Computational Sciences, Petrópolis, RJ, 25651-075, Brazil
9
Institute of Biological Sciences, Federal University of Minas Gerais, MG, 31270-901, Brazil
1
0
1
Centre of Natural and Human Sciences, Federal University of ABC, Santo André, SP, 09210-580, Brazil
Department for Life Quality Studies, Alma Mater Studiorum-University of Bologna, Rimini, 47921, Italy
1
*
Corresponding author: Claudio Viegas Jr., Institute of Chemistry, Laboratory of Research
on Medicinal Chemistry, Avenida Jovino Fernandes Sales, 2600. Federal University of
Alfenas. 37130.000 Alfenas-MG, Brazil. Tel +55 (35) 3701-1880. e-mail:
cvjviegas@gmail.com
1

ACCEPTED MANUSCRIPT
ABSTRACT
A new series of sixteen multifunctional N-benzyl-piperidine-aryl-acylhydrazones hybrid
derivatives was synthesized and evaluated for multi-target activities related to Alzheimer’s
disease (AD). The molecular hybridization approach was based on the combination, in a
single molecule, of the pharmacophoric N-benzyl-piperidine subunit of donepezil, the
substituted hydroxy-piperidine fragment of the AChE inhibitor LASSBio-767, and an
acylhydrazone linker, a privileged structure present in a number of synthetic aryl- and aryl-
acylhydrazone derivatives with significant AChE and anti-inflammatory activities. Among
them, compounds 4c, 4d, 4g and 4j presented the best AChE inhibitory activities, but only
compounds 4c and 4g exhibited concurrent anti-inflammatory activity in vitro and in vivo,
against amyloid beta oligomer (AβO) induced neuroinflammation. Compound 4c also
showed the best in vitro and in vivo neuroprotective effects against AβO-induced
neurodegeneration. In addition, compound 4c showed a similar binding mode to donepezil
in both acetylated and free forms of AChE enzyme in molecular docking studies and did
not show relevant toxic effects on in vitro and in vivo assays, with good predicted ADME
parameters in silico. Overall, all these results highlighted compound 4c as a promising and
innovative multi-target drug prototype candidate for AD treatment.
Keywords: Alzheimer’s disease, multi-target directed ligands, N-benzyl-piperidine
derivatives, donepezil analogs, acetylcholinesterase inhibition, anti-inflammatory activity,
COX inhibition, amyloid beta oligomers, neuroprotection.
2

ACCEPTED MANUSCRIPT
1
. Introduction
Alzheimer's disease (AD) is a severe degenerative disorder of the central nervous
system (CNS) and is the most common type of dementia among the population above age
6
5. AD is characterized by progressive loss in cognition, decrease in motor and functional
capacity, impairment in behavioral and social autonomy and death.[1–5] Pharmacotherapy
of AD is only palliative and restricted to a few available drugs that act either through the
inhibition of acetylcholinesterase (AChE) enzyme or through a mild neuroprotective effect,
alleviating the symptoms and slightly delaying the progress of the disease. AD is
characterized by multiple physiological and biochemical disturbances involving
concurrently operating chemical mediators and numerous protein targets.[1,6–9] During the
last years, we have observed huge advances in the understanding of the complex multi
factorial pathophysiology of AD. Although its etiology is not completely understood, it is
well established that concomitant and interconnected multiple factors are decisive for the
onset, progression and severity of AD. It is well accepted that an idiopathic physiological
deregulation in some brain regions (e.g. hippocampus and cortex) is responsible for an
overproduction of Aβ-peptide (Aβ), cleaved from its precursor protein by β- and γ-
secretases. Aβ-fragments may form neurotoxic soluble oligomers (AβΟ) and, in turn,
insoluble fibrillar aggregates or senile plaques. Associated with amyloid neuropathology,
hyperphosphorylation of neuronal microtubule constitutive τ-protein occurs and leads to a
collapse in the structure of microtubules, disruption of axonal transport, and accumulation
of intracellular neurofibrillary tangles.[10–15] Acting together, these two biochemical
factors lead to cumulative changes in CNS homeostasis and function, eventually leading to
neuronal death.[9,13]
Several studies also suggest that Aβ deposits and neurofibrillary tangles provide
stimuli for neuroinflammation, which contribute to significantly exacerbate the AD
pathogenic processes.[16,17] Among the brain cells, the activated microglia cells have been
shown to be involved in the secretion of pro-inflammatory cytokines, such as interleukin-1
(
IL-1β), IL-6 and tumor necrosis factor alpha (TNF-α), thereby contributing towards the
progress of AD.[18,19] Further, studies on AD pathogenesis have demonstrated that
microglia produce Aβ, which is in itself pro-inflammatory and causes activation of several
3

ACCEPTED MANUSCRIPT
inflammatory componentes.[20] The pro-inflammatory cytokines act as potent stimuli in
brain inflammation through upregulation of cytosolic PLA2 and cyclooxygenase-1 (COX-
1
) and -2 (COX-2).[21] Upon inflammatory stimuli, COX-1, being constitutive in
microglia, is responsible for the primary inflammatory response by inducing the production
of prostaglandin, mainly prostaglandin E2. By contrast, COX-2 expression in neurons is
responsible, upon its induction, for a later and secondary inflammatory response.[21]
Therefore, drugs designed to treat the different neuroinflammatory conditions mediated by
COX-1 and COX-2 are needed.
Thus, in addition to cholinergic signaling, excitotoxic mechanisms and the
amyloid cascade, microglial cells and neuroinflammation have emerged as essential targets
in the polypharmacology-derived concept of multi-target directed drugs for AD,
representing a new paradigm in their design and discovery.[1,2,22–28]
Donepezil (1) is a potent AChE inhibitor with an IC50 of 5.7 nM and a 1250-fold
higher selectivity over butyrylcholinesterase (BuChE) and a long duration effect. To date,
donepezil is the reference drug for mild and moderate AD treatment and, in some cases, is
recommended to be used in association with the anti-excitotoxic drug memantine for the
most severe cases of the disease.[25,29–33]
In the search for novel bioactive chemical entities, the N-acylhydrazone moiety
deserves highlight as a privileged structure, capable to act as pharmacophore or auxophore
subunit in different pharmaceutic classes, with a variety of action profile, depending on the
other functionalities present in the molecular structure.[34,35] Recently, Silva and co-
workers synthetized a series of N-acylhydrazones, designed as AChE inhibitors. The
pharmacological evaluation evidenced that these derivatives showed significant AChE
inhibitory activity. Turnaturi and co-workers have also observed the antioxidant capacity of
some N-acylhydrazones with metal chelant ability[36,37]. As a part of an ongoing project
that aims at the discovery of innovative multi-target directed drugs, we describe herein the
discovery of novel donepezil-derived N-acyl-aryl-hydrazone ligands with multifunctional
profile as promising drug candidates for the treatment of AD. The design of the target-
molecules was based on molecular hybridization of the pharmacophoric subunit N-benzyl-
piperidine core (Fig. 1, a) present in structure of donepezil (1), with a common 3-O-
substituted-piperidine subunit (b) present in the structure of the AChE inhibitor LASSBio-
4

ACCEPTED MANUSCRIPT
7
67 (2), and the insertion of a N-acyl-arylhydrazone functionality (c, d) considered a
privileged structure present in a number of anti-inflammatory [35,38–42], AChE inhibitors
and antioxidant drug prototypes in the literature [36,37]. Our aim was to generate a novel
chemical library (Fig. 1), with an innovative scaffold to explore structure-activity
relationship of ligand candidates with a multitarget profile, that could act as AChE
inhibitors, anti-neuroinflammatory and neuroprotectants.
Fig. 1. Design of a new series (4) of multi-target directed ligands (MTDLs) based on the molecular
hybridization of the structures of donepezil (1), the AChE inhibitor LASSBio-767 (2) and a series
of anti-inflammatory N-aryl-acylhydrazone derivatives (3).
2
2
. Results and discussion
.1. Chemistry
4
-carboxybenzaldehyde (5) and rac-3-hydroxypiperidine (7) were used as starting
materials to prepare the key intermediate 4-((3-hydroxypiperidin-1-yl)-methyl)-
benzohydrazide 9 according to the synthetic approaches described in Scheme 1
(experimental details in the supporting information material). Initially, 4-
carboxybenzadehyde was reacted with SOCl /DMF, followed by treatment with anhydrous
2
MeOH to provide the ester derivative 6. Reductive amination reaction between ester 6 and
5

ACCEPTED MANUSCRIPT
2 3
rac-3-hydroxypiperidine (7) in the presence of ZnCl /NaBH CN led to the desired 3-
hydroxy-N-benzyl-piperidine ester 8 that was then converted to the key intermediate N-
acylhydrazone 9. In a final step, coupling reactions of 9 and the benzaldehydes 10a-l led to
the target molecules 4a-l in good yields. Additional 3-O-acetyl-piperidine derivatives 4m,
4
n and 4o were prepared directly by reaction of Ac O/4-DMAP with the correspondent 3-
2
hydroxy precursor. Compound 4h were obtained by catalytic hydrogenation of the para-
NO derivative 4d in quantitative yields. Compound 4n were obtained by reaction of
2
2 2
Ac O/4-DMAP and catalytic hydrogenation of the para-NO with 4d.
Scheme 1. Synthetic route for the target-compounds 4a-o
The purity of all compounds was determined as 94-98% by UPLC-UV. Considering
that the starting material was the racemic 3-hydroxypiperidine, the products were also
racemic, but the NMR and MS analysis of all compounds did not indicate the formation of
diastereoisomers. [34,43] We suppose that the coupling reaction between the hydrazide 9
and the functionalized benzaldehydes furnished only the most stable E relative
configuration at the hydrazone moiety. In order to verify this, the compounds were
subjected to X-ray powder diffraction analysis that confirmed the E relative configuration,
as can be seen in Fig. 2 for compounds 4c and 4g.
6

ACCEPTED MANUSCRIPT
Fig. 2. Crystallographic structures of compounds 4c (A) and 4g (B) obtained by X-Ray power
diffraction analysis.
2
.2. In Vitro Inhibition of AChE and BuChE
All fifteen N-benzyl-piperidine-acylhydrazone derivatives were first screened for
AChE inhibition.[44] The purified enzyme from Electrophorus electricus (EeAChE) was
chosen for these assays because of its high similarity to human AChE (100% identity of
residues lining the catalytic gorge[45] and relative low cost). Twelve compounds inhibited
EeAChE by more than 60% at 100 µM (Supporting Information, Fig. SI1) and the 3-
hydroxy-piperidine derivatives were clearly more active. Concentration-response curves for
all the twenty compounds confirmed acceptable potencies and full inhibition of AChE, with
the lowest IC50 of 2.7 µM, obtained for compound the para-nitrobenzyl derivative 4d
(Table 1). The different benzyl substitutions generally increased potency relative to the
unsubstituted compound 4a (IC50 30.0 µM). The differences in potency among the para-
halogenated derivatives were modest (3-fold maximum) and the potency rank was F > Br >
Cl, perhaps suggesting that factors other than size were more relevant to the interaction.
Seven 3-hydroxy piperidine derivatives were further evaluated for selectivity and
mechanism of AChE inhibition (Supporting Information, Fig. SI2 and Fig. SI3). Assays of
equine serum butyrylcholinesterase (eqBuChE) activity showed complete concentration-
dependent inhibition and yielded the corresponding IC (Table 1). All tested compounds
5
0
were 1.8 to 12.5 times more potent at inhibiting EeAChE than eqBuChE, with the para-
nitrobenzyl derivative 4d being the most selective. In the EeAChE assays of mechanism,
7

ACCEPTED MANUSCRIPT
none of the compounds showed a significant competitive behavior towards the substrate,
based on kinetic modeling of inhibition data as illustrated in Fig. 3 for compound 4g. The
simple (pure) noncompetitive inhibition model in which only Vmax is reduced without
differences in affinity for free and substrate-bound enzyme was consistently more likely
than mixed inhibition models, based on the corrected Akaike information criterion (AICc).
Accordingly, there were no marked differences between estimated K and IC (Table 1).
i
50
Only for compound 4l the probability of correctness of mixed inhibition was close to that of
pure noncompetitive inhibition.
Table 1: Potency, selectivity and mechanism of action of compounds 4a-o in
cholinesterase inhibition
EeAChE
eqBuChE
IC50b EeAChE
model P
ratio
Compounds
R1
R2
a
a
c
dc
IC50
(
µ
M)
IC50
(
µ
M)
ratio
K
i
(
µM)
4
a
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OH
OAc
OAc
OAc
H
Br
Cl
NO
OCH
30.01
11.89
25.39
2.71
13.19
14.73
8.65
21.46
≥100
10.89
21.71
10.39
4b
4
c
48.22
33.86
91.85
1.9
12.5
7.0
7.76
2.45
11.45
2.9
3.1
3.3
4d
2
4
4
e
f
3
Pyrrolidinyl
F
4g
39.04
4.5
13.41
3.3
4h
NH
2
4
4
i
j
Morpholinyl
Piperidinyl
Imidazole
58.14
39.62
59.30
5.3
1.8
5.7
9.70
21.10
8.42
1.8
2.5
1.1
4
k
4
l
SCH
Br
NH
OAc
3
4
m
≥100
≥100
≥100
4n
2
4o
e
donepezil
0.026
4.69
180
a
IC50 values are geometric means from 2-3 independent curves, each with 6 concentrations assayed in
b
50
50
c
triplicates. IC BuChE/IC AChE Kinetic inhibition constant for the average of three independent data sets
d
fitted to the most likely model of simple (pure) noncompetitive inhibition. Ratio of probabilities of correctness
of simple noncompetitive over mixed linear inhibition models, based on their AICc. Data from ref. [71].
8

ACCEPTED MANUSCRIPT
Fig. 3. Substrate competition assay of compound 4g for EeAChE. Data points are mean ±
SD from 3 independent experiments, each performed in triplicate. The curves represent the
best-fitting noncompetitive inhibition model, with estimated parameters (and 95% C.I.): Km
=
101.9 (87.7 to 116.2) µM and K
i
= 13.41 (11.87 to 14.96) µM. The inhibitor concentrations
(µM) are indicated to the right of the curves.
2
.3. Molecular docking study with human AChE
First molecular docking experiments with the inhibitor donepezil were performed
into the AChE binding site to validate the three states evaluated for the enzyme (i.e. free,
acetylated and Michaelis complex). The top ranked pose of docking experiments was
compared with the experimental conformation of donepezil complexed with AChE (PDB
code 4ey7). The docking simulation successfully found the experimental binding mode of
donepezil in both free and acetylated forms, while for the Michaelis complex the donepezil
molecule adopts a completely inverted binding mode with a significantly worse score.
These results validated the structures utilized in this work, whereas the unsuccessful
docking of donepezil with ACh present in the binding site might indicate that this inhibitor
does not favorably interact with the Michaelis complex. Compounds 4g and 4j,
representing halogen and heterocycle-substituted derivatives, respectively, occupy similar
regions in the AChE binding site when compared with donepezil in both free and acetylated
models, independent of the absolute configuration of C-3 carbon on the piperidine ring.
Furthermore, the docking scores for all compounds, including donepezil, are quite similar
for both free and acetylated models.
9

ACCEPTED MANUSCRIPT
Compound 4gR exhibited very similar binding modes for both the free and
acetylated forms of AChE, mainly diverging in the position of the 3-hydroxypiperidine
group and the intermolecular interaction performed by the hydroxyl group (Fig. 4A). In
both situations, compound 4gR interacts with Trp286 from the peripheral anionic site
(PAS) through stacking and interacts through a hydrogen bond with Tyr124 in the middle
of the enzyme gorge. In the free state, the protonated nitrogen from the 3-
hydroxipiperidinyl ring interacts with Trp86 through a weak cation-π interaction and the
hydroxyl group is hydrogen bonded with His447. In the acetylated enzyme, the positively
charged nitrogen from the 3-hydroxipiperidinyl ring interacts with Trp86 through a stronger
cation-π interaction while the hydroxyl group is hydrogen bonded to the W731.
Fig. 4. Docking results for 4gR in the (A) free (green) and acetylated (light blue) states, and in the
Michaelis complex (yellow). ACh is represented as salmon sticks and molecular waters as red
spheres.
The binding mode of 4gS is almost the same for 4gR, with a rotation of the 3-
hydroxypiperidine group between the free and acetylated models (Fig. 5A). The stacking
interaction between the phenyl ring and Trp286 is preserved and the hydrogen bond of the
N-acylhydrazone (NAH) group with Trp124 is kept. The protonated nitrogen atom of the 3-
hydroxypiperidine subunit is more weakly interacting through cation-π with Trp86 when
compared with the R configuration and the hydroxyl group of 4gS is hydrogen bonded to
1
0

ACCEPTED MANUSCRIPT
the catalytic Ser203 or to Glu202 residues in the free and acetylated models, respectively.
In the Michaelis complex (Fig. 5B) it is also possible to identify an additional binding of
the 4-fluoro-phenyl subunit with ACh through van der Waals interactions and a hydrogen
bond between the carbonyl group from NAH and the backbone NH of Arg296.
Fig. 5. Docking results for 4gS in the (A) free (green) and acetylated (light blue) states, and in the
Michaelis complex (B, yellow). ACh is represented as salmon sticks and molecular waters as red
spheres.
In the Michaelis complex, the predicted binding mode for compound 4j is
significantly different for the two configurations. In the R-isomer (Fig. 6A), the piperidinyl
group is oriented to the interior of the gorge interacting with ACh through van der Waals
interactions, the protonated nitrogen is interacting with Tyr341 through cation-π and the
hydroxyl group from the 3-hydroxypiperidine is interacting through a hydrogen bond with
Asp74. Conversely, the S-isomer exhibited an inverted conformation (Fig. 5B), with the
piperidine group oriented to the entrance of the gorge, the hydroxyl group hydrogen bonded
with the highly exposed Glu292, and the carbonyl from NAH hydrogen bonded with the
backbone NH of Arg296. The R configuration of compound 4j exhibited identical binding
modes in both free and acetylated forms of AChE with similar docking scores (Fig. 6A).
Such interaction is composed of cation-π of the protonated nitrogen of the piperidinyl group
with Trp86 (in a similar fashion with the ammonium group of ACh) and a hydrogen bond
1
1

ACCEPTED MANUSCRIPT
of the hydroxyl with W731. The N-acylhydrazone group interacts through a hydrogen bond
with Tyr124. At the entrance of the gorge, the phenyl ring makes a stacking interaction
with the side chain of Trp286, a major component of the PAS.
Fig. 6. Docking results for 4jR in the (A) free (green) and acetylated (light blue) states, and in the
Michaelis complex (B, yellow). ACh is represented as salmon sticks and water molecules as red
spheres.
Similarly, the S enantiomer of compound 4j exhibited almost the same binding
modes in both free and acetylated states of the enzyme, with a rotation of the 3-
hydroxypiperidine group at the bottom of the gorge (Fig. 7A). While the hydroxyl group is
hydrogen-bonded with the catalytic Ser203 in the free enzyme, it interacts through a
hydrogen bond with W737 in the acetylated model due to the steric hindrance caused by the
presence of the acetyl group attached to the Ser203. In both states of the enzyme, the
hydrogen atom of the charged nitrogen is pointed out to the Trp86, weakening the stacking
interaction. The N-acylhydrazone group interacts through a hydrogen bond with Tyr124,
located in the middle of the gorge, and the phenyl ring make a stacking interaction with the
Trp286 indole ring at the PAS.
1
2

ACCEPTED MANUSCRIPT
Fig. 7. Docking results for 4jS in the (A) free (green) and acetylated (light blue) states, and in the
Michaelis complex (yellow). ACh is represented as salmon sticks and water molecules as red
spheres.
The predicted binding modes of both 4jR (Fig. 6B) and 4jS (Fig. 7B) in the
Michaelis complex are significantly distinct from those predicted in the free and acetylated
models of AChE. The 4jR binding mode exhibits the 3-hydroxypiperidine located at the
bottom of the gorge, interacting with Tyr341 through cation-π interaction and with ACh
through nonpolar contacts, while the hydroxyl group interacts with Tyr124 and Asp74
through hydrogen bonds. The other piperidine group is oriented to the entrance of the
binding cavity, mostly exposed to the solvent. The binding mode predicted for 4jS exhibits
the 3-hydroxypiperidine group located at the entry of the gorge hydrogen bonded with the
Asn283 side chain. The other piperidine group is located at the bottom of the cavity,
located at hydrogen bond distance from the hydroxyl group of Tyr124 (3.13Å), and
interacts with ACh through nonpolar interactions.
2
.4.
In vivo Anti-inflammatory Activity
In parallel to the AChE inhibition studies, all compounds were evaluated for their
potential anti-inflammatory properties. For this purpose, classical animal models such as
mechanical allodynia test, formalin-induced hyperalgesia and carrageenan-induced paw
1
3

ACCEPTED MANUSCRIPT
edema assays were used as described in the literature.[46–49] Table 2 shows that eleven of
the fourteen compounds evaluated (4c, 4e-g, 4m, 4n, 4i, 4j, 4o and 4l) significantly reduced
the mechanical hyperalgesic index. The seven most active compounds 4c, 4e, 4g, 4i, 4j, 4n
and 4o were selected for testing in the formalin test. These compounds decreased the
licking time in the first (early) phase of the test, showing an analgesic effect. Compounds
4
c, 4e, 4g, 4i and 4n further reduced the licking time in the second phase of the formalin
test, and they were therefore selected for the carrageenan-induced edema test All five
compounds reduced edema volume, confirming an anti-inflammatory effect.
Table 2. Effect of the compounds 4b-o in mechanical allodynia, formalin and paw edema tests.
Formalin
Edema
Mechanical
Compound
Saline
ID50
hyperalgesic index
early phase
74.00±7.7
last phase
Volume (uL)
67.73±4.2
(
µmol/Kg)
324.1 ± 25.2
235.8 ± 55.1
168.3 ± 46.1***
320.1 ± 31.3
135.8±10.7
62.88±10.7
4
b
*
**
***
***
4
c
14.00±2.7
13.29±3.2
18.86±3.8
17.14±4.7
8.76
68.3
4.66
2.67
4
d
*
*
*
**
***
***
***
***
4
e
161.1 ± 29.1
206.4 ± 58.0
178.9 ± 28.6
244.4 ± 62.8
83.96 ± 26.1
64.56 ± 18.8
260.3 ± 54.5
206.9 ± 37.1
146.1 ± 27.2
353.1 ± 39.2
158.1 ± 19.9
15.86±6.3
24.17±8.4
25.71±2.0
*
4f
**
***
***
4
4
g
20.00±2.1
h
*
*
*
*
*
**
**
***
***
***
***
4
i
62.75±6.8
19.63±3.1
69.67±10.7
21.43±5.9
4
j
110.0±21.9
4
l
*
4m
**
**
**
***
4
n
o
47.75±13.1
69.75±13.9
11.43±4.0
4.20
2.12
4
*
**
***
***
Ind
65.43±6.8
22.75±6.5
3.375±1.6
14.29±2.9
***
***
Morph
7.500±1.8
Each value represents the mean (±S.E.M.) of eight animals. *p < 0.05, ** p < 0.01 and *** p < 0.001 when
compared with the vehicle group. Ind= indomethacin, Morph= morphine
14

ACCEPTED MANUSCRIPT
2
.5. In vitro and in vivo COX-1 and COX-2 Activity
Recent studies suggested that hydrazone derivatives display anti-inflammatory
effects through the inhibition of cyclooxygenases 1 (COX-1) and 2 (COX-2)[50–52]. We
selected compounds 4c and 4g, active in both AChE and anti-inflammatory assays, to
evaluate their inhibition of COX-1 and COX-2 activity. As reported in Fig. 8a, compounds
4
c, 4g and ibuprofen, used as positive control, inhibited both COX-1 (51%, 64% and 88%,
respectively) and COX-2 (65%, 53% and 94%, respectively) activity at 10 µM, with a COX
selectivity index (ratio of maximum inhibition on COX-2 and COX-1) of 1.27, 0.83 and
1
.07, respectively. Subsequently, we also evaluated the ability of 4c and 4g to decrease the
COX-1 and COX-2 protein levels of mice treated with the tested-compounds and
carrageenan, a pro-inflammatory polysaccharide. COX-1 and COX-2 levels were measured
in mice serum by using enzyme-linked immunosorbent assay. As shown in Fig.s 8b and 8c,
compounds 4c and 4g and indomethacin, used as positive control, significantly decreased
COX-1 (percentage of inhibition, 27%, 34% and 52%, respectively) and COX-2 (42%, 25%
and 40%, respectively).
COX-1
COX-2
a)
1
00
110
100
110
100
90
80
70
60
50
40
30
20
10
0
(
c)
(
(b)
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
9
8
7
6
5
4
3
2
1
0
0
0
0
0
0
0
0
0
0
*
*
*
*
*
*
Ibu
4c
4g
Fig. 8. Effect of compounds 4c and 4g on COX-1 and COX-2 enzyme activity and blood levels. (a)
In vitro peroxidative activity of human COX-1 and ovine COX-2 was determined by a
chemiluminescence assay in the presence of compounds (10 µM) or ibuprofen (Ibu), as a positive
control. Values are the mean ± SEM of two independent experiments. (b, c) Plasma COX-1 and
15

ACCEPTED MANUSCRIPT
COX-2 levels were evaluated in mice orally treated with the compounds (100 mM/kg, p.o) and then
injected with carrageenan (CG) in the paw. Indomethacin (IND) was a positive control. Values are
expressed as mean ± SEM (n=8) in terms of COX level percentage (*P<0.05 compared to the CG
group).
2
.6. Molecular docking study with COX-1 and COX-2
Considering the in vitro and in vivo anti-inflammatory results, we also studied the
binding mode and selectivity in silico of the target compounds toward COX-1 and COX-2.
According to the redocking results, the docking strategy was successfully applied to both
COX-1 and COX-2, providing predicted poses similar to the experimentally observed
binding mode of the reference ligands (i.e. celecoxib and rofecoxib) with RMSD values of
1
.029 Å and 0.400 Å, respectively.[53–55] According to affinity predictions, the
configuration R-R of the compounds 4c and 4g provide more favorable interaction with the
COX-1 enzyme (it is important to note that the R-R configurations have very similar
predicted affinities for COX-1), while S-S compounds interact better with COX-2 (Table 3
and Fig. 9).
Fig. 9. 2D structures of (R-R) and (S-S) configurations of 4c and 4g.
16

ACCEPTED MANUSCRIPT
Table 3. Binding affinities predicted by DockTScore for the top ranked pose of (R-R) and
S-S) compounds 4c, 4g and the reference ligands celeboxib and rofecoxib from docking with
(
DockThor.
a
a
Compound
R-R- 4c
S-S-4c
COX-1
COX-2
-10.427
-10.378
-10.326
-10.320
10.489
n.a.
-8.793
-10.129
-8.617
-10.080
n.a.
R-R-4g
S-S-4g
Celeboxib
Rofecoxib
-10.052
a
-1
Given in kcal mol .
In the computational study, both compounds 4c and 4g exhibited similar docking
scores and binding modes in both COX-1 and COX-2 (Table 3, Fig. 10 and 11), interacting
with the binding channel through diverse lipophilic and hydrophilic interactions. The
overall binding mode is characterized by the hydroxyl-piperidine group oriented to the
active site at the bottom of the channel, while halogen-phenyl ring is located at the entrance
of the binding cavity exposed to the solvent. Strikingly, the piperidine amine group is
located in a hydrophobic region without interacting with charged or polar atoms. We
believe that this interaction is thermodynamically unfavorable and might reflect on a
weakening of the affinity for both COX-1 and COX-2. Interestingly, S-S-4c interacts with
COX-2 through four hydrogen bonds: the hydroxyl group is hydrogen-bonded with the
Tyr385 side chain and the N-acylhydrazone is stabilized by three hydrogen bonds with
Leu319, Ser320 and His90. Furthermore, the hydroxyl group of 4c is close to the Ser530
residue, interacting through polar contacts (~ 3.8 Å). On the other hand, in COX-1, the R-R
configuration lost the interaction with Tyr385 due to the rotation of the hydroxy-piperidine
group.
17

ACCEPTED MANUSCRIPT
Fig. 10. Docking results of 4c (light blue) in the COX-1 (left) and COX-2 (right) catalytic channels.
Hydrogen bonds are represented as yellow dashes.
In COX-1, compound R-R-4g interacts through two hydrogen bonds: the N-
acylhydrazone is hydrogen bonded with Ser353 backbone and His90 side chain. In COX-2,
compound S-S-4g exhibits the NH from N-acylhydrazone interacting with Leu319 and
Ser320 backbones through two hydrogen bonds, while the hydroxyl group is oriented
toward the side chains of Ser530 and Tyr385 residues due to the rotation of the hydroxyl-
piperidine subunit (Fig. 10).
Fig. 11. Docking results of 4g (light green) in the COX-1 (left) and COX-2 (right) catalytic
channels. Hydrogen bonds are represented as yellow dashes.
2
.7. Toxicity profile in vitro and ADME prediction and in silico.
1
8

ACCEPTED MANUSCRIPT
Cytotoxicity and genotoxicity of the most promising compounds 4c and 4g was
evaluated in peripheral blood mononuclear (PBMC) cells, liver HepG2 and neuronal SH-
SY5Y cells. Initially, the IC50 (concentration of the compound resulting in 50% inhibition
of cell viability) values were determined after a 6 h treatment of PBMCs and HepG2 cells,
and a 24 h treatment of SH-SY5Y cells by MTT assay. As shown in Table 4, the
compounds were not toxic to any type of cells up to the highest tested concentration (100
µM for PBMCs and HepG2 cells; 80 µM for SH-SY5Y cells). Subsequently, we also
determined the genotoxicity to HepG2cells of low concentrations of compounds 4c and 4g
(20 µM or 40 µM), using the comet assay and micronucleus (MN) test. The electrophoretic
formation of nuclear DNA “tail” and the occurrence of micronuclei reflect DNA damage
and clastogenicity/aneugenicity, respectively [56,57]. Remarkably, the treatment of HepG2
cells with 4c and 4g did not induce any genotoxic effects (Table 4).
Table 4. In vitro cytotoxicity and genotoxicity profile of compound 4c and 4g
a
b
c
Cytotoxicity (IC , µM)
Genotoxicity
5
0
Compounds
HepG2
cells
SH-SY5Y
cells
Tail
moment
PBMCs
MNi
Control
10
10
14
12
26
23
24
24
4c
>100
>100
> 80
> 80
n.d.
n.d.
n.d.
> 80
> 80
n.d.
n.d.
n.d.
4
g
d
Donepezil
Doxorubicin
n.d.
e
f
n.d.
n.d.
35
45
e
Ultraviolet light
138
n.d.
a
b
Cytotoxicity was evaluated in PBMC, HepG2 and SH-SY5Y cells as reported in experimental section. IC50
values represent the concentration of the compound resulting in 50% inhibition of cell viability; the values are
c
the mean of two/three independent determinations. Genotoxicity was determined in HepG2 cells after
treatment with compounds (20 µM or 40 µM) in terms of comet tail moment (DNA damage parameter) and
d
e
micronuclei index (MNi, clastogenicity/aneugenicity parameter). n.d.: not determined; p<0.001 versus
f
untreated HepG2 cells (control); p<0.05 versus untreated HepG2 cells (control).
In addition, some relevant ADME parameters were also predicted in silico by using
the QikProp v. 3.5 (Schrödinger). Our data indicate that the most promising multifunctional
19

ACCEPTED MANUSCRIPT
compounds 4c and 4g are in accordance with the software reference parameters (data
shown in the Supporting Information, Table SI1), with adequate clogP and water solubility,
and good predicted serum albumin binding, human absorption and blood-brain barrier
transposition in comparison to donepezil, included as reference drug.
2
.8.
In vitro and in vivo neuroprotective activity
Among the different forms of Aβ aggregates, soluble Aβ oligomers (AβO) cause a
selective synaptic dysfunction and/or neuronal loss in cortex and hippocampus, two
stricken brain regions in AD[58,59]. AβOs display higher hydrophobicity and ability to
interact with and cross the membranes, leading to formation of ion channels, membrane
disruption and mitochondrial dysfunction with neuronal death[60]. In addition, AβOs
exacerbate responses from glial cells with the overproduction of inflammatory cytokines,
glutamatergic excitotoxicity, neuroinflammation and loss of synaptic function[61,62]. In
this regard, we evaluated the effects of compound 4c and 4g in in vitro and in vivo models
of neuroinflammation and neurodegeneration induced by either LPS or Aβ1-42 oligomers
(Aβ1-42O). First, we evaluated the ability of compounds 4c and 4g to counteract the release
of pro-inflammatory factors, such as TNFα and IL-1β, from human THP-1 cells activated
by LPS. The THP-1 cells are transformed monocytes that have a range of properties similar
to microglia, including release of superoxide anion, TNFα, IL-1β and PGE [63]. The
2
treatment of THP-1 cells for 24 h with 10 µM of either compound 4c or 4g significantly
inhibited the LPS-elicited release of both TNFα (98% and 69%, respectively) and IL-1β
(29% and 31%, respectively) (Fig. 12). Anti-neuroinflammatory effects were then
confirmed in vivo, in mice subjected to unilateral intrahippocampal injection Aβ1-42O. The
intraperitoneal administration of either compound 4c or 4g determined a significant
reduction of Iba-1 immunostaining, a marker of activated microglia, in the CA1 area of the
ipsilateral hippocampus, suggesting that both compounds reduced Aβ1-42O -evoked
neuroinflammation (Fig. 13a/b). Iba-1 immunostaining in the CA1 area of contralateral
hippocampus was not affected by any treatment, confirming the selective unilateral effect
of the intrahippocampal Aβ1-42O injection and the lack of systemic pro-inflammatory
effects of the compounds in mice (Fig. 13a/b).
20

ACCEPTED MANUSCRIPT
Fig. 12. Compounds 4c and 4g inhibit the release of TNFα and IL-1β induced by LPS in THP-1
cells, representative of human microglial cells. THP-1 cells were treated with compounds (10 µM)
and LPS (1 µg/mL) for 24 h. At the end of the treatment, the pro-inflammatory cytokine levels were
determined as described in the experimental section. The values are reported as mean ± SD of three
§
§§
*
***
independent experiments. p < 0.001 vs. untreated cells, p < 0.05 and p < 0.001 vs. cells treated
with LPS (ANOVA with Bonferroni post hoc test).
21

ACCEPTED MANUSCRIPT
Fig. 13. Compounds 4c and 4g inhibit microgliosis induced by unilateral intrahippocampal injection
of Aβ1-42O in mice. (a) The quantitative analysis of microgliosis in the CA1 area of the ipsilateral
and contralateral hippocampus by Iba1 immunostaining was performed 7 days after the
administration of compound (100 mg/kg, i.p.) and Aβ1-42O intrahippocampal injection in mice as
reported in the experimental section. (b) Representative Iba1 immunostaining of the hippocampus.
Scale bar 50 µm. Values are expressed as mean ± SEM (n = 10) of the pixel density of each image
§§
analyzed. p<0.01 vs. PBS + vehicle; *p < 0.05 and **p < 0.01 vs. Aβ + vehicle; ANOVA, with
Newman-Keuls post hoc test.
In parallel, we determined the ability of compounds 4c and 4g to counteract the
early and late neurotoxic events, in terms of MTT formazan exocytosis and necrosis elicited
by Aβ1-42O in neuronal SH-SY5Y cells. In particular, MTT formazan exocytosis induced
by Aβ1-42O reveals an early vacuole formation prodromic to neuronal death[64]. Ten
2
2

ACCEPTED MANUSCRIPT
micromolar of compound 4c significantly decreased both MTT formazan exocytosis and
neuronal death elicited by Aβ1-42O in neuronal SH-SY5Y cells. In contrast to 4c, compound
4
g showed only the ability to significantly counteract the late neuronal death elicited by
Aβ1-42 (Fig. 14). These results reveal the best in vitro neuroprotective activity of 4c,
suggesting its ability to interfere with intracellular signal transduction pathways by which
Aβ enhances early MTT formazan exocytosis.
Fig. 14. Effects of compound 4c and 4g on early and late neurotoxic events, in terms of MTT
formazan exocytosis and neuronal death elicited by Aβ1-42O in neuronal SH-SY5Y cells. The cells
were treated with the compounds (10 µM) and Aβ1-42O (10 µM) for increasing amounts time. At the
end of the treatment of 4 and 24 h, MTT formazan exocytosis and neuronal death were determined
using MTT and propidium iodide, respectively, as described in the experimental section. The values
§§
are reported as mean ± SD of three independent experiments. p<0.01 vs, untreated cells, **p<0.01
and ***p<0.001 vs. cells treated with Aβ1-42O, at ANOVA with Bonferroni post hoc test.
Subsequently, we evaluated the neuroprotective effects of compounds 4c and 4g
against neurodegeneration induced by unilateral intrahippocampal injection of Aβ1-42O in
mice. After the administration of the compounds we determined the neurodegeneration in
the CA1 area of the ipsilateral hippocampus using Fluoro-Jade C (FJC), a marker of
neuronal death. As shown in Fig. 15, there was a trend towards a reduction in FJC staining
by the compound 4c in CA1 area of the ipsilateral hippocampus highlighting that only
compound 4c counteracts the Aβ induced neuronal death. There was no significant change
in neuronal death in the CA1 area of the contralateral hippocampus among the groups
23

ACCEPTED MANUSCRIPT
injected with PBS, Aβ or Aβ treated with 4c. On the other hand, the compound 4g
increased the neurodegeneration in both hippocampi, suggesting that its lack of in vivo
neuroprotective effect could be ascribed to its in vivo intrinsic neurotoxic effects.
Fig. 15. Compounds 4c, but not 4g, counteract the neuronal death induced by unilateral
intrahippocampal injection of Aβ1-42O in mice. (a) The quantitative analysis of neuronal death in the
CA1 area of the ipsilateral and contralateral hippocampus by FJC staining was performed 7 days
after the administration of the compounds (100 mg/kg, i.p.) and Aβ1-42O intrahippocampal injection
in mice as reported in experimental section. (b) Representative FJC staining of hippocampus slices.
Scale bar 50 µm. Values are expressed as mean ± SEM (n= 10) of the pixel intensity of each
§
hippocampus slice analyzed. p<0.05 vs. PBS + vehicle; *p < 0.05 vs. AβO + vehicle; ANOVA,
with Newman-Keuls post hoc test.
24

ACCEPTED MANUSCRIPT
3
3
.
Experimental section
.1.
Chemistry.
Reagents were obtained from Sigma-Aldrich, Acros or Merck. The solvents were
prepared by standard procedures. Analytical thin-layer chromatography (TLC) was
performed using silica 60 F254 from Merck and spots were visualized with UV light.
Melting points were measured in capillary tubes on a Marte (PFM II) melting point
apparatus without correction. Nuclear magnetic resonance (NMR) spectroscopy was
performed on a Bruker AC-300 NMR (IS as TMS). Chemical shifts were reported in parts
per million (ppm, δ) downfield from tetramethylsilane. Proton coupling patterns were
described as singlet (s), doublet (d), triplet (t), quartet (q) and multiplet (m). High-
resolution mass spectra (HRMS) were obtained by electrospray ionization (ESI) using a
Agilent 6550 iFunnel Q-TOF LC/MS. Infrared (IR) spectra were obtained by Nicolet iS50
FTIR (Thermo Scientific USA) infrared spectrometer coupled to Pike Gladi ATR
-
1
Technologies. The absorption bands were reported in wave numbers (cm , ν). All spectra
are available in supporting information (Fig. SI4 – Fig. SI72).
3
.1.1. Methyl 4-formylbenzoate (6).
A solution of 4-formylbenzoic acid (5, 1g, 6.78mmol), catalytic DMF, SOCl
2
(2
mL, 27.26 mmol) in dichloromethane (10mL) was stirred and refluxed for 40 minutes, than
methanol in excess was added. The solution was stirred and refluxed for additional 1 hour.
Removal of the solvents produced a residue which was purified using column
chromatography, eluted with a mixture of EtOAc/EtOH (8:2) to afford compound 6 as a
1
white solid in 87% yield. H NMR (CD
3
OD, 300MHz):
δ
10.06 (s, 1H), 8.16 (d, J=8.0Hz,
191.6, 166.4,
ν 3300, 2500, 2835, 2735, 2997, 2954,
1
3
2
1
1
H), 7.97 (d, J= 10.0Hz, 2H), 3.93 (s, 3H). C NMR (CD OD, 75 MHz): δ
3
39.0, 134.6, 129.2, 128.7, 128.5, 128.4, 51.7. IR:
724, 1707, 1280, 1111.
3
.1.2. Methyl 4-((3-hydroxypiperidin-1-yl)methyl)benzoate (8).
To a mixture of methyl 4-formylbenzoate (6.2 g, 12.2 mmol) and piperidin-3-ol
(7.147 g, 14.6 mmol) in methanol (10 mL) was added another mixture of zinc chloride
(0.83 g, 6.1 mmol) and sodium cyanoborohydride (0.76 g, 12.2 mmol) in methanol (6 mL).
The resultant mixture was stirred at room temperature for 24h. The solvent was evaporated
25

ACCEPTED MANUSCRIPT
and water was added (40mL), followed by addition of a solution of NaOH 20% to alkalize
the mixture. Then, the mixture was extracted with CH Cl (3 × 20 mL). The organic layer
2
2
4
was separated and dried over brine and MgSO . Removal of the solvents produced a
residue which was purified using column chromatography, eluted with a mixture of
1
EtOAc/MeOH (9:1, v/v), to afford compound 8 as a white solid. 90% yield. H NMR
(
CD OD, 300MHz):
δ
7.95 (d, J= 8.0Hz, 2H), 7.43 (d, J= 7.4Hz, 2H), 3.88 (s, 3H), 3.70-
3
3
1
7
2
.62 (m, 1H), 3.57 (s, 2H), 2.86-2.82 (m, 1H), 2.67-2.64 (m, 1H), 2.06-2.02 (m, 1H), 1.91-
1
3
.86 (m, 2H), 1.76-1.68 (m, 1H), 1.58-1.49 (m, 1H), 1.24-1.19 (m, 1H). C NMR (CD
5 MHz): 167.0, 143.3, 129.1, 66.5, 62.1, 60.3, 53.0, 51.20, 32.4, 22.6. IR: 3335, 2937,
792, 2859, 2359, 2341, 1609, 1277, 1057.
.1.3. 4-((3-hydroxypiperidin-1-yl)methyl)benzohydrazide (9).
To a solution of methyl 4-((3-hydroxypiperidin-1-yl)methyl)benzoate (8, 0.5 g, 2
3
OD,
δ
ν
3
mmol) in ethanol (7 mL) was added hydrazine monohydrate (11.73mL, 60mmol). The
resulting mixture was refluxed and stirred for 3h. Removal of the solvents produced a
residue which was purified using chromatography column, eluted with a mixture of
1
EtOAc/MeOH (8:2, v/v) to afford the compound 9 as a white solid. 95% yield. H NMR
(
CD OD, 300MHz):
δ
7.75 (d, J= 7.8Hz, 2H), 7.42 (d, J= 7.4Hz, 2H), 3.67-3.63 (m, 1H),
3
3
1
1
2
.57 (s, 2H), 2.87-2.83 (m, 1 H), 2.69-2.65 (m, 1H), 2.08-2.01 (m, 1H), 1.96-1.90 (m, 2H),
1
3
.74-1.71 (m, 1H), 1.59-1.51 (m, 1H), 1.28-1.21 (m, 1H). C NMR (CD OD, 75 MHz):
δ
3
68.0, 141.3, 131.8, 129.2, 126.8, 66.4, 62.1, 60.2, 52.9, 32.4, 22.5. IR:
360, 2341, 1611, 1548, 1303.
ν 3296, 2939, 2815,
3
.1.4. General Method for the preparation of N-acylhydrazones
intermediates.
To solution of 4-((3-hydroxypiperidin-1-yl)methyl)benzohydrazide (9, 0.5 g, 2
mmol) in ethanol with catalytic amount of HCl was added the corresponding aldehyde (10,
.4 mmol). The mixture was stirred for 3h at room temperature. Removal of the solvent
2
produced a residue, which was purified using chromatography column with the appropriate
eluents systems.
3
.1.4.1.(E)-N'-benzylidene-4-((3-hydroxypiperidin-1-yl)methyl)benzohydrazide
(4a).
26

ACCEPTED MANUSCRIPT
1
Gray solid. 60% yield. Mp 152-153°C. H NMR (CD
3
OD, 300MHz):
δ
8.54 (s, 1H),
8
7
1
7
5
.12 (d, J= 8.12Hz, 2H), 7.98 (d, J= 7.98Hz, 2H), 7.72 (d, J= 7.73Hz, 2H), 7.58 (t, J=
.58Hz, 3H), 3.49 (s, 2H), 3.46-3.44 (m, 1H), 3.16-3.13 (m, 1H), 3.00 (m, 1H), 2.10 (m,
1
3
H), 2.04-2.01 (m, 2H), 1.84-1.75 (m, 1H), 1.60 (m, 1H), 1.42 (m, 1H). C NMR (CD OD,
3
5 MHz):
δ 165.1, 149.5, 132.5, 138.8, 134.1, 132.5, 165.1, 132.5, 30.2, 128.4, 65.0, 58.8,
2.9, 31.0, 21.0. IR:
ν
3331, 2939, 2817, 1644, 1522, 1283, 961, 831. HRMS (ESI) m/z
+
calcd. C H N O [M+H] : 338.1869, found 338.1868.
2
0
23
3
2
3
.1.4.2.(E)-N'-(4-bromobenzylidene)-4-((3-hydroxypiperidin-1-
yl)methyl)benzohydrazide (4b).
White solid. 52% yield. Mp 192-194ºC. H NMR (DMSO-d , 300MHz):
1
δ
11.88 (s,
6
1
1
1
H), 8.42 (s, 1H), 7.86 (d, J=7.9Hz, 2H), 7.67 (s, 4H), 7.43 (d, J=7.4Hz 2H), 3.59-3.50 (m,
H ), 3.59- 3.50 (m, 1H), 2.80- 2.75 (m, 1H), 2.65- 2.61 (m, 1H), 1.92-1.85 (m, 1H), 1.76-
7
13
.70 (m, 2H), 1.63- 1.59 (m, 1H), 1.48- 1.40 (m, 1H), 1.09- 1.03 (m, 1H). C NMR
163.5, 146.8, 143.2, 134.1, 132.3, 129.2, 128.0, 123.7, 66.0, 62.1,
(DMSO-d , 75 MHz):
δ
6
6
9
4
1.5, 53.4, 33.6, 23.6. IR:
ν
3226, 3066, 2931, 2791, 2359, 2341, 1644, 1544, 1286, 1066,
+
73, 958, 815, 671. HRMS (ESI) m/z calcd. C H BrN O [M+H] : 416.0974, found
20
22
3
2
16.0984.
.1.4.3.(E)-N'-(4-chlorobenzylidene)-4-((3-hydroxypiperidin-1-
yl)methyl)benzohydrazide (4c).
Pale White solid. 70% yield. Mp 177-179ºC. H NMR (DMSO-d , 300MHz):
3
1
δ
8.34
s, 1H), 7.95 (d, J=7.9Hz, 2H), 7.83 (d, J=7.8Hz, 2H), 7.56 (d, J=7.6Hz, 2H), 7.44 (d,
J=7.5Hz, 2H), 3.84 (m, 1H), 3.59 (s, 2H), 2.98- 2.94 (m, 1H), 2.86- 2.83 (m, 1H), 2.44-
6
(
13
2
.28 (m, 1H), 1.89- 1.86 (m, 2H), 1.65- 1.61 (m, 1H), 1.42- 1.28 (m, 2H). C NMR
(
DMSO-d , 75 MHz):
δ
163.5, 146.9, 132.4, 134.9, 133.8, 132.4, 128.1, 129.4, 66.1, 61.8,
6
6
1.0, 33.3, 23.2. IR:
ν
3196, 2940, 2361, 2341, 1651, 1279, 1087. HRMS (ESI) m/z calcd.
+
C H ClN O [M+H] : 372.1479, found 372.1409.
20
22
3
2
3
.1.4.4.(E)-4-((3-hydroxypiperidin-1-yl)methyl)-N'-(4-
nitrobenzylidene)benzohydrazide (4d).
Yellow solid. 71% yield. Mp 222-225ºC. H NMR (DMSO-d , 300MHz):
1
δ
8.71 (s,
6
1
1
H), 8.29 (d, J=8.3Hz, 2H), 7.96 (d, J=8.0Hz, 4H), 7.44 (d, J=7.4Hz, 2H), 2.80- 2.77 (m,
H), 2.76- 2.62 (m, 1H), 1.94- 1.87 (m, 1H), 1.80- 1.75 (m, 2H), 1.64- 1.59 (m, 1H), 1.48-
27