Design, synthesis and evaluation of novel feruloyl-donepezil hybrids as potential multitarget drugs for the treatment of Alzheimer's disease

Article abstract of DOI:10.1016/j.ejmech.2017.02.043

A novel series of feruloyl-donepezil hybrid compounds were designed, synthesized and evaluated as multitarget drug candidates for the treatment of Alzheimer's Disease (AD). In?vitro results revealed potent acetylcholinesterase (AChE) inhibitory activity f

Full text of DOI:10.1016/j.ejmech.2017.02.043

Accepted Manuscript
Design, synthesis and evaluation of novel feruloyl-donepezil hybrids as potential
multitarget drugs for the treatment of Alzheimer's disease
Kris Simone T. Dias, Cynthia T. de Paula, Thiago dos Santos, Isis N.O. Souza,
Marina S. Boni, Marcos J.R. Guimarães, Fernanda M.R. da Silva, Newton G. Castro,
Gilda Neves, Clarice C. Veloso, Márcio M. Coelho, Ivo Souza F. de Melo, Fabiana
C. Vilela, Alexandre Giusti-Paiva, Marcelo L. da Silva, Laurent E. Dardenne, Isabella
Guedes, Letizia Pruccoli, Fabiana Morroni, Andrea Tarozzi, Claudio Viegas, Jr.
PII:
S0223-5234(17)30116-2
10.1016/j.ejmech.2017.02.043
EJMECH 9239
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 4 August 2016
Revised Date: 15 February 2017
Accepted Date: 17 February 2017
Please cite this article as: K.S.T. Dias, C.T. de Paula, T. dos Santos, I.N.O. Souza, M.S. Boni, M.J.R.
Guimarães, F.M.R. da Silva, N.G. Castro, G. Neves, C.C. Veloso, M.M. Coelho, I.S.F. de Melo, F.C.
Vilela, A. Giusti-Paiva, M.L. da Silva, L.E. Dardenne, I. Guedes, L. Pruccoli, F. Morroni, A. Tarozzi, C.
Viegas Jr., Design, synthesis and evaluation of novel feruloyl-donepezil hybrids as potential multitarget
drugs for the treatment of Alzheimer's disease, European Journal of Medicinal Chemistry (2017), doi:
10.1016/j.ejmech.2017.02.043.
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 evaluation of novel feruloyl-donepezil hybrids
as potential multitarget drugs for the treatment of Alzheimer’s disease
a
a
a
b
Kris Simone T. Dias , Cynthia T. de Paula , Thiago dos Santos , Isis N. O. Souza ,
b
b
b
Marina S. Boni , Marcos J. R. Guimarães , Fernanda M. R. da Silva , Newton G.
b
b
c
d
Castro , Gilda Neves , Clarice C. Veloso , Márcio M. Coelho , Ivo Souza F. de Melo
d
e
e
f
,
Fabiana C. Vilela , Alexandre Giusti-Paiva , Marcelo L. da Silva , Laurent E.
g
g
h
i
h
Dardenne , Isabella Guedes , Letizia Pruccoli , Fabiana Morroni , Andrea Tarozzi
and Claudio Viegas Jr.
a
a
PeQuiM- Laboratory of Research in Medicinal Chemistry, Institute of Chemistry, Federal University of
Alfenas, 37130-000, Alfenas/MG, Brazil
b
Laboratory of Molecular Pharmacology, Institute of Biomedical Sciences, Federal University of Rio de
Janeiro, 21941-902, Rio de Janeiro/RJ, Brazil
c
Pharmaceutical Sciences Faculty, Federal University of Amazonas (UFAM), 69067-005, Manaus/AM,
Brazil
d
Department of Pharmaceutical Products, Pharmaceutical Faculty, Federal University of Minas Gerais,
31270-901, Belo Horizonte/MG, Brazil
e
Department of Physiological Sciences, Federal University of Alfenas, 37130-000, Alfenas/MG, Brazil
Physiotherapy Department, Federal University of Alfenas, 37130-000, Alfenas/MG, Brazil
National Laboratory of Scientific Computation, 25651-075, Petrópolis/RJ, Brazil
f
g
h
Department for Life Quality Studies, Alma Mater Studiorum-University of Bologna, 47921 Rimini,
Italy
i
Department of Pharmacy and Biotechnology, Alma Mater Studiorum-University of Bologna, 40126
Bologna, Italy
*Corresponding author: Claudio Viegas Jr.
Phone: +55 35 37011880
e-mail: cvjviegas@gmail.com

ACCEPTED MANUSCRIPT
ABSTRACT
A novel series of feruloyl-donepezil hybrid compounds were designed, synthesized and
evaluated as multitarget drug candidates for the treatment of Alzheimer’s Disease (AD).
In vitro results revealed potent acetylcholinesterase (AChE) inhibitory activity for some
of these compounds and all of them showed moderate antioxidant properties.
Compounds 12a, 12b and 12c were the most potent AChE inhibitors, highlighting 12a
with IC50 = 0.46 µM. In addition, these three most promising compounds exhibited
significant in vivo anti-inflammatory activity in the mice paw edema, pleurisy and
2
+
2+
formalin-induced hyperalgesy models, in vitro metal chelator activity for Cu and Fe ,
and neuroprotection of human neuronal cells against oxidative damage. Molecular
docking studies corroborated the in vitro inhibitory mode of interaction of these active
compounds on AChE. Based on these data, compound 12a was identified as a novel
promising drug prototype candidate for the treatment of AD with innovative structural
feature and multitarget effects.
Key words: feruloyl-donepezil hybrids, Alzheimer’s Disease, multitarget-directed
ligands

ACCEPTED MANUSCRIPT
1. INTRODUCTION
Alzheimer’s disease (AD) is a progressive neurodegenerative condition,
characterized by cognitive impairment and behavioral abnormalities [1]. In the late stage of
the disease, patients lose their personal independence, requiring daily familiar and nursing
care, progressing to death. AD is a multifactorial disease, with several interconnected
neuropathological events, probably as a result of the combination of risk factors, including
age, genetic and environmental factors [2]. It is estimated that more than 18 million people
are currently affected by AD worldwide, with an expected increase of patient numbers to
114 million by 2050 [3]. The pathophysiology of AD is not completely understood, but
much progress has been achieved in recent years in unveiling the mechanisms of
neurodegeneration, given the importance of this disease on the world scene. AD’s
pathogenesis involves imbalance in numerous biochemical pathways such as the
production and clearance of β-amyloid (Aβ) peptides, accumulation of
hyperphosphorylated tau protein [4–6], decrease in brain acetylcholine (ACh) levels [7,8],
2
+
2+
2+
dyshomeostasis of biometals such as Zn , Fe , Cu
[9,10], oxidative stress and
mitochondrial dysfunction [11], and neuroinflammation [12], leading to neuronal damage
and death.
Currently, most of the approved drugs for AD treatment are acetylcholinesterase
(
AChE) inhibitors (e.g. donepezil (1), tacrine (2), rivastigmine (3) and galantamine (4) Fig.
). Such drugs restore cognitive function and hence improve the quality of life of AD
patients [8].
1
Fig. 1. Chemical structures of AChE inhibitors currently used in the treatment of AD
It has now been accepted that complex multifactorial diseases, such as cancer,
diabetes, hypertension and neurodegenerative diseases, cannot be efficiently treated by the

ACCEPTED MANUSCRIPT
modulation of a single target. As a consequence, the multi-target-directed ligand (MTDL)
approach has emerged as a new paradigm in drug discovery. This innovative strategy
proposes the design of new scaffolds with two or more pharmacophoric subunits connected
in a single molecule that could, in turn, modulate the interaction and biological response of
multiple molecular targets simultaneously [13–15].
Considering the MTDLs concept, donepezil, the current first choice drug for AD
treatment, is among the most popular pharmacophore inspirations used for the design of
novel drug candidates, since it is a very potent, low toxic and well tolerated AChE
inhibitor [16–19]. Curcumin (5, Fig.2), an abundant natural polyphenol found in Curcuma
longa rhizomes, is also widely used as scaffold for the planning of new MTDLs for AD
[
20–24]. Curcumin has potent antioxidant and anti-inflammatory properties, playing an
important role in the decrease of oxidative damage, inflammation [25,26] and amyloid
accumulation [27], with an additional biometal chelating ability [28]. In vivo studies with
curcumin on transgenic animal models have disclosed its various beneficial effects for the
treatment of AD. These studies demonstrate that 5 is also capable of reverting the
formation of amyloid plaques, preventing Aβ-induced oxidative stress, and modulating the
expression of inflammatory mediators [29].
Based on the MTDL strategy, we report here the design, synthesis and
pharmacological evaluation of a new series of molecular hybrids feruloyl-donepezil based
on the combination of the pharmacophoric N-benzylpiperidine, subunit from donepezil (1)
responsible for its adequate recognition by AchE and the subunit feruloyl present in ferulic
acid (6) and curcumin (5) (Fig. 2). Variation of the substituent groups in the aromatic rings
moiety derived from donepezil and feruloyl aimed structure-activity relationship (SAR)
studies and the identification of potential effective drug candidates for AD treatment, with
an innovative mechanism of action.

ACCEPTED MANUSCRIPT
Fig. 2. Design strategy for a new series of feruloyl-donepezil hybrid compounds
This series of hybrid compounds were investigated for their in vitro direct and
indirect antioxidant, anticholinesterase, metal chelating and neuroprotective properties,
and in vivo anti-inflammatory effects. In addition, molecular docking studies were
performed to determine the binding mode of the best AChE inhibitors.
2
2
. RESULTS AND DISCUSSION
.1. Chemistry
The synthetic convergent route for accessing the desired compounds 12a-n is
shown in Scheme 1. 4-piperidinemethanol (6) was used as starting material for the
preparation of the key intermediates 9a-g, by a reductive amination reaction with a
3 2
series of functionalized benzaldehydes in the presence of NaBH CN/ZnCl . Thereafter,
ferulic acid (9a) and iso-ferulic acid (9b) were converted into their acyl-chloride
derivatives 10a-b by using thionyl chloride. In a final step, derivatives 10a and 10b

ACCEPTED MANUSCRIPT
were coupled to the N-benzylpiperidine key-intermediates 9a-g to yield the target-
compounds 12a-n.
3 2 2
Scheme 1. Reagents and conditions: (a) NaBH CN, ZnCl , MeOH, rt, 72 h, 70-87%; (b) SOCl ,
DMF, CH Cl , N , 1h, 98%; (c) CH Cl , N , 15 min, rt, 55-70%
2
2
2
2
2
2
2.2. In vitro AChE activity inhibition
Firstly, all feruloyl-donepezil hybrids compounds (12a-n) were screened for
their cholinesterase inhibition profile on Electrophorus electricus AChE (EeAChE), at
the fixed concentration of 30 ꢀM. As shown in Fig. 3, 12a, 12b and 12c were the most
potent AChE inhibitors. However, only compounds 12a and 12b reached 50%
inhibition, and their IC50 and selectivity versus equine serum butyrylcholinesterase
(eqBuChE) were determined next. The unsubstituted benzylpiperidine compound 12a
showed a submicromolar IC50 value for EeAChE (0.46 ꢀM) and marked selectivity,
being 53-fold less potent for eqBuChE (Fig. 4 and Table 1). Compound 12b was about
3
0-fold less potent than 12a for EeAChE and only inhibited eqBuChE by 4.3% at 30
M, so it was not further investigated. Curcumin (5) showed only a minor
anticholinesterase activity, which is unlikely to contribute to its biological effects in
ꢀ

ACCEPTED MANUSCRIPT
vivo. In the substrate competition assay for EeAChE, compound 12a reduced Vmax
without affecting the apparent K
inhibition, with a K of 1.04 ꢀM (Fig. 4). The reference compound donepezil (1) also
had no discernible effect on the apparent K , showing a single K of 38 nM. These
m
, therefore indicating a simple (pure) non-competitive
i
m
i
results suggested that the regiochemistry of the hydroxy and methoxy substituents on
the phenylpropanoid moiety is a strong determinant of AChE inhibitory activity, as
could be observed for the most active ferulic acid derivatives in comparison to all their
regioisomers derived from iso-ferulic acid. In addition, the higher potency of compound
12a suggests that substitution on the aromatic ring of the N-benzylpiperidine subunit
leads to a decrease in the AChE inhibitory activity, independent of its size or electron
donating or withdrawing ability.
Fig. 3. Screening of feruloyl-donepezil hybrids (12a-12n) for EeAChE inhibition. The
compounds were pre-incubated for 10 min with the enzyme and reaction was followed for 5 min
and activity (velocity) was expressed as percent of control (vehicle). Data are means and pooled
SD from 2-4 experiments, each performed in triplicate.

ACCEPTED MANUSCRIPT
Fig. 4. Inhibition curves and mechanism of inhibition of 12a. A. Concentration-dependent
inhibition of EeAChE and eqBuChE by 12a. Independent experiments are shown in different
symbols, which are mean ± SD of triplicates; fitted curve parameters are shown in Table 2. B.
Substrate competition data for EeAChE and best-fitting model (simple linear non-competitive
inhibition). The concentration of 12a (µM) is indicated to the right of each curve. Data are
pooled from two experiments, each performed in triplicate.
Table 1. Cholinesterase inhibitory activity, DPPH scavenging capacity, neurotoxic and direct
antioxidant effects in neuron cells of compounds 12a-n
Compounds
Cholinesterase
DPPH
Neuron
Antioxidant
effects
Neurotoxicity
a
d
inhibiton IC (µM)
scavenging
IC (µM)
50
50
b
EeAChE eqBuChE
EC50 (µM)
c
IC (µM)
5
0
f
g
1
1
1
1
1
1
1
1
1
1
1
1
1
2a
0.46 (3)
16.74 (3)
>30
24.97 (2)
49.41
46.66
45.38
73.77
78.09
> 100
> 100
> 100
> 100
> 100
20.47
> 100
12.45
16.87
25.94
n.d. (6%)
2b
2c
2d
2e
2f
-
-
-
-
-
-
-
-
-
-
-
-
n.d. (9%)
e
n.d.
n.d. (28%)
> 30
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
> 30
> 30
2g
2h
2i
> 30
> 30
> 30
2j
> 30
2k
2l
> 30
> 30
2m
> 30.00

ACCEPTED MANUSCRIPT
1
2n
> 30.00
-
> 100
-
-
-
-
Donepezil
Curcumin
Trolox
0.026 (2)
4.69 (2)
n.d.
132.13 (3) >300 (2)
-
-
-
-
-
-
-
-
04.86
35.54
> 100
-
11.82
-
-
-
-
ferulic acid
iso-ferulic acid
a
The IC50 are geometric means from the number of independent experiments indicated in parenthesis;
each was performed in triplicate.
b
Antioxidant activity are shown as EC50 values in µM concentrations. All compounds were analyzed in
triplicate and the results expressed as average
c
2 2
Concentration of compound resulting in 50% inhibition of ROS formation induced by H O (100 µM) in
neuronal SH-SY5Y cells. The values are mean of at least two independent experiments;
d
Concentration of compound resulting in 50% inhibition of neuronal viability after 24 h treatment;
n.d: IC50 not determined because less than 50% inhibition of ROS formation induced by H
e
2
2
O in
neuronal SH-SY5Y cells was observed at the highest concentration tested (40 µM);
f
n.d: IC50 not determined because less than 50% inhibition of neuronal viability was observed at the
highest concentration tested (40 µM);
g
Values in parenthesis are percent inhibition of neuronal viability observed at highest tested concentration
(
40 µM).
2.3. Molecular docking study with human AChE
The compound 12a exhibited similar binding mode and slightly similar affinity
scores in both the free and acetylated forms of the enzyme (Table 2 and Fig. 5), with
the scores proving to be slightly better than donepezil. Interestingly, the predicted
binding mode for this compound is quite similar to the experimental conformation
observed for the AChE-donepezil complex (PDB code 4ey7), retaining several key
interactions: (i) the phenyl and the piperidine groups are located in the same region,
with the phenyl group making stacking interactions with Trp86 and the piperidine
Nitrogen atom interacting with Tyr337 through cation-π and with a water molecule
(W931) through a hydrogen bond; (ii) hydrogen bond between the carbonyl of the ester
group and the NH main chain of Phe295 (carbonyl group of the indanone ring in
donepezil), (iii) stacking interaction between the 2-methoxyphenol ring and the indole
system of Trp286 located at the peripheral anionic site (PAS). Additionally, the
hydroxyl group of the phenol is located at hydrogen-bond distance from Tyr72 (3.06 Å).

ACCEPTED MANUSCRIPT
Fig. 5. Docking results for 12a in the free (light blue) and acetylated (yellow) states, superposed
with the experimental binding mode of donepezil (transparent green). W931 is represented as a
red sphere and the hydrogen bonds are highlighted as orange dashed lines.
The predicted binding modes of 12b in the free and acetylated enzyme are also
similar, exhibiting the nitrophenyl ring oriented to the catalytic triad (Table 2 and Fig.
6
). The main interactions found are: (i) hydrogen bond between the nitro group and the
hydroxyl from Tyr133, (ii) salt bridge between nitro and the side chain of Glu202, (iii)
stacking interactions between the nitrophenyl ring and Trp86 (not planar) and His447,
(iv) cation-π between the NH group in the protonated piperidine and Phe338 (bad
geometry), Tyr337 and Trp86 (long distance, > 6Å), (vi) hydrogen bond between the
carbonyl in the ester group with NH of the Phe295 main chain, (vii) stacking
interactions between the indanone moiety and Trp286 from PAS (good geometry).
Fig. 6. Docking results for 12b in the free (light blue) and acetylated (yellow) states, superposed
with the experimental binding mode of donepezil (transparent green). W931 is represented as a
red sphere and the hydrogen bonds are highlighted as orange dashed lines.

ACCEPTED MANUSCRIPT
The binding modes of 12c are significantly similar in both the free and
acetylated forms of AChE (Table 2 and Fig. 7), performing similar interactions to those
observed for donepezil with some geometry deviations: (i) stacking interactions
between the 4-bromine-phenyl ring and Trp86, with the bromine atom oriented to the
catalytic triad and surrounded by three structural waters (W731, W722 and W729, not
shown), (ii) cation-π between the NH group in the protonated piperidine and Tyr337
and Phe338, (iii) hydrogen bond between the carbonyl in the ester group and the NH of
the Phe295 main chain, (iv) stacking interactions with Trp286 from PAS (best
interaction in the acetylated form). The significant difference between the scores in the
free and acetylated forms (~2.6 kcal/mol) might be due to the best geometry of
interactions with Trp286 in the PAS (stacking interactions) and with Phe338 (cation-π
with the protonated NH oriented to the phenyl ring in the acetylated form while the
same group is pointing to the opposite site of the aromatic ring).
Fig. 7. Docking results for 12c in the (A) free (light blue) and acetylated (yellow) states,
superposed with the experimental binding mode of donepezil (transparent green). W931 is
represented as a red sphere and the hydrogen bonds are highlighted as orange dashed lines.
All three compounds exhibited a significant degradation of the docking scores in
the Michaelis complex. Compounds 12a and 12c, which exhibited the best docking
scores, retain the overall orientation in the binding site. On the other hand, compound
1
2b exhibited a totally inverted orientation when compared with the predicted binding
modes for the free and acetylated enzyme.

ACCEPTED MANUSCRIPT
Table 2. Docking scores of the compounds 12a, 12b, 12c and donepezil (used as reference)
provided by docking with Glide in the Extra Precision Mode.
Compound
2ª
2b
2c
Donepezil
Free AChE
-16.129
Acetylated AChE
-16.956
Michaelis Complex
-11.567
1
1
-15.586
-15.982
-9.961
1
-13.913
-15.469
-16.566
-15.862
-11.463
-11.757
2
.4. In vitro direct antioxidant activity evaluation
There is ample evidence that oxidative stress plays an important role in inducing
and activating multiple cell signaling pathways that contribute to production of toxic
substances and then promotes the development of AD [30]. Thereby, compounds 12a-n
were evaluated for their antioxidant activity by using the radical scavenging DPPH
assay in different concentrations (12.5, 25, 50, 100 µM). Ferulic acid, iso-ferulic acid
and trolox were used as standard compounds and the results are shown in Table 2. It
was found that almost all of the ferulic acid related derivatives were effective in
scavenging free radicals, in comparison to the iso-ferulic parent compounds.
Interestingly, ferulic acid exhibited more than 100-fold higher antioxidant potency
compared to iso-ferulic acid, which revealed that the higher antioxidant profile of
compounds 12a-e, 12k and 12m is probably derived from their ferulic acid pattern.
These results clearly demonstrate that, similarly to AChE inhibition, regiochemistry
plays a key role in the antioxidant activity of the target compounds. This result could be
explained by analogy with the effect of para-hydroxy substituted cinnamic acids, in
which the oxygen atom of the hydroxyl group is able to share a positive charge and
thereby increase radical stabilization through conjugation extension. However, if the
hydroxyl radicals assume a meta orientation, the oxygen atom is unable to share the
charge, probably influencing the ability for radical scavenging [31].
We then evaluated the neurotoxic and direct antioxidant effects of compounds
1
2a, 12b and 12c which showed the best abilities to inhibit the cholinesterase activity
and counteract the DPPH radical in neuronal SH-SY5Y cells. As reported in Table 1,

ACCEPTED MANUSCRIPT
treatment for 24 h of SH-SY5Y cells with compounds 12a and 12b (1.25-40 µM) did
not affect the neuronal viability using MTT assay, while the similar treatment with
compound 12c showed a slight decrease in neuronal viability (28%) at the highest
concentration (40
2a and 12b, but not 12c, to counteract the intracellular ROS formation induced by
in SH-SY5Y cells using DCFDA probe (Table 1). Both compounds 12a and 12b
showed concentration-dependent antioxidant effects with a significant decrease of ROS
formation at 10, 20 and 40 M (all p<0.001). Compound 12c significantly inhibited the
ROS formation only at 10 M (p<0.001), showing pro-oxidant effects at higher
µM) used. In parallel, we recorded the strong ability of compounds
1
2 2
H O
µ
µ
concentrations (Fig. 8). Antioxidant profiles similar to those of 12a and 12b were
recorded with ferulic acid (Table 1, Fig. 8). As expected, donepezil did not show any
direct antioxidant effects (data not shown). These results show the absence of cytotoxic
and pro-oxidant effects for compounds 12a and 12b and suggest that their ferulic acid
pattern effectively contributes to counteract the ROS formation in neuron SH-SY5Y
cells.
Fig. 8. Compounds 12a, 12b, 12c and ferulic acid counteract the ROS formation induced by
H O in neuronal SH-SY5Y cells. The cells were treated with various concentrations of
2
2

ACCEPTED MANUSCRIPT
compounds and H
2
O
2
(100 µM) for 30 min. At the end of the treatment, intracellular ROS
formation was determined using DCFDA probe (as described in the experimental section). The
arbitrary units of fluorescence (AUF) values are shown as mean SD of three independent
).
±
#
##
2 2
experiments ( p<0.001 vs sham; ***p<0.001 vs treated with H O
2
.5. Chelation of pro-oxidant metals
A considerable amount of literature data has shown that the dyshomeostasis of
2
+,
2+,
2+
3+
biometals such as Cu Zn Fe and Fe in CNS and their interactions with amyloid
protein precursor (APP) and Aβ could contribute to AD pathology [10], [9], [32]. In
particular, copper and iron are redox active and can generate ROS via the Fenton and
Haber–Weiss reaction, and promote neurotoxic Aβ oligomer formation. Therefore,
compounds with metal-chelating properties could represent an additional therapeutic
strategy for the treatment of AD. The ability of compounds 12a-c to chelate biometals
+2
+2
+3
2+
such as Cu , Fe , Fe and Zn was studied by UV−vis spectroscopy with
wavelengths ranging from 200 to 400 nm, according to the methodology described by
Chen and cols. [20]. Compound 12a selectively chelated copper and iron. The
absorption spectra of 12a at the concentration of 75 ꢀM alone or incubated in the
presence of CuSO
4 4 3 2
, FeSO , FeCl or ZnCl (150 ꢀM) for 30 min in ethanol are shown in
Fig. 9. Compound 12a displayed absorptions with ƛmax (log ε) at 328.20 nm (0.970),
2
+
2
99.00 nm (0.660) and 237.20 nm (0.517). When Cu was added to the solution of 12a,
it caused a significant shift of the band at 237.20 nm (0.517) to 248.00 nm (0.470),
indicating the formation of the corresponding complex via metal chelation. Similarly,
2+
after addition of Fe , a shift of the band at 237.20 nm (0.517) to 249.20 nm (0.430) was
also observed. In contrast, the addition of FeCl and ZnCl did not cause any significant
shift in the absorption bands, which confirms that compound 12a is capable of chelating
3
2
2
+
2+
3+
2+
only Cu and Fe , but not Fe and Zn . Compounds 12b and 12c showed the same
profile as the compound 12a (data not shown).

ACCEPTED MANUSCRIPT
3 4
Fig.9. UV-absorbance spectra of 12a alone or in the presence of ZnCl2, FeSO4, FeCl and CuSO .
We then determined the ability of compound 12a to counteract the ROS
2
+
2+
formation elicited by Fenton reaction with either Cu or Fe and H
2
2
O in neuronal SH-
SH5Y cells. In both the reactions, 5, 10 and 20 µM of compound 12a significantly
counteracted the ROS formation in neuronal SH-SH5Y cells (p<0.001) (Fig. 10).
Fig.10. Compound 12a counteracted the ROS formation induced by either FeSO
4 2 2
/H O or
CuSO /H O in neuronal SH-SY5Y cells. The cells were treated with various concentrations of
4
2
2
4 2 2 4 2 2
compound 12a and either FeSO /H O or CuSO /H O (25 µM/100 µM) for 30 min. At the end
of the treatment, intracellular ROS formation was determined using DCFDA probe (as
described in the experimental section). The arbitrary units of fluorescence (AUF) values are
#
##
shown as mean
±
SD of three independent experiments ( p<0.001 vs sham; ***p<0.001 vs
treated with either FeSO
4
/H or CuSO /H ).
2
O
2
4
2 2
O

ACCEPTED MANUSCRIPT
2
.6. In vitro indirect antioxidant activity evaluation
Recent studies report that molecules like curcumin which contain Michael
acceptor functionalities and phenolic hydroxyl groups with direct antioxidant properties
can also induce antioxidant and cytoprotective proteins at cellular level. In particular,
these indirect antioxidants activate the Keap1/Nrf2/ARE pathway resulting in
transcriptional induction of glutathione (GSH) and Phase II detoxification enzymes that
act catalytically (not being consumed and having long half-lives), and are unlikely to
evoke pro-oxidant effects [33].
On the basis of these considerations, we determined the intracellular GSH levels
after a 24 h treatment of neuronal SH-SH5Y cells with 20 µM compounds 12a, 12b and
1
2c using MCB probe. As reported in Fig. 12a, compounds 12a, 12b, but not 12c,
increased the GSH levels with a maximum of 1.4- and 1.2-fold increase, respectively.
Furthermore, compound 12a induced a time-dependent increase of GSH levels with a
maximum of 24-fold increase at 24 h (Fig. 12b). We then chose compound 12a to
evaluate its ability to activate Nrf2 binding to ARE at nuclear level, an essential event
for the induction of GSH synthetic enzyme genes, by TransAM ELISA kit. At the 20
µ
M concentration of 12a we recorded a strong time dependent increase of nuclear Nrf2
activity starting at 1 h. COS-7 cells transfected with Nrf2 were used as a positive
control (Fig. 12c). The activity of this positive control showed the reliability of this
assay (data not shown). Finally, to ascertain whether the increase of the total GSH level
observed in SH-SY5Y cells could really translate into antioxidant effects, we treated the
SH-SY5Y cells 24 h before adding the H
significantly prevented the ROS formation evoked by H
2d).
2
O
2
. Remarkably, 20
µM of compound 12a
2
O
2
in SH-SY5Y cells (Fig.
1

ACCEPTED MANUSCRIPT
Fig. 11. Compound 12a increases cytosolic GSH and nuclear Nrf2 levels as well as indirect
antioxidant effects in neuronal SH-SY5Y cells. The cells were treated with 20 µM of
compounds 12a, 12b, 12c for 24 h (a) or various increasing treatment time (b, c). At the end of
the treatment, cytosolic GSH and nuclear Nrf2 levels were determined using MCB probe and
immunoassay, respectively (as described in experimental section). (d) After the 24 h treatment
2 2
with 20 µM of compound 12a, the cells were also treated with H O (100 µM, 30 min) to
evaluate the intracellular ROS formation using DCFDA probe (as described in the experimental
section) in SH-SY5Y cells. The results are expressed as a percentage of control cells. The
values are reported as mean
±
SD of three independent experiments (*p<0.05, **p<0.01,
#
##
§§
*
**p<0.001 vs sham; p<0.05, p<0.01 vs untreated; p<0.01 vs treated with H
O
2 2
).
2
.7. In vitro neuroprotective activity evaluation
Several studies have suggested that Aβ oligomers cause a selective synaptic
dysfunction and/or neuronal loss in cortex and hippocampus, two stricken brain regions
in AD [4]. In one proposed mechanism, Aβ oligomers adhere to the plasma membrane
of neurons and cause lesions by a combination of radical species-initiated lipid
peroxidation and formation of ion-permeable pores which initiate a cascade of
pathological processes that end with neuronal death [34]. We evaluated the ability of

ACCEPTED MANUSCRIPT
compound 12a to counteract the ROS formation, early and late neurotoxic events
elicited by Aβ1-42 oligomers. Twenty micromolar of compound 12a decreased ROS
formation elicited by Aβ1-42 oligomers in neuronal SH-SY5Y cells (Fig. 12). In parallel,
compound 12a did not counteract early vacuole formation and enhancement of
exocytosis induced by Aβ1-42 oligomers in neuronal SH-SY5Y cells using MTT assay
(data not shown) but it showed interesting neuroprotective effects against the late
neuronal death elicited by Aβ1-42 oligomers.
Fig. 12. Compound 12a counteracts ROS formation and neuronal death induced by Aβ1-42
oligomers in neuronal SH-SY5Y cells. The cells were treated with 20 µM of compound 12a and
A
β
1-42 oligomers (20 µM) for either 4 h (a) or 24 h (b). At the end of the treatment, intracellular
ROS formation and neuronal death were determined using DCFDA and propidium iodide probe
as described in the experimental section). (c) Representative images of ROS formation. The
results are expressed as a percentage of control cells. The values are reported as mean SD of
(
±
#
##
three independent experiments ( p<0.001 vs sham; **p<0.01, ***p<0.001 vs treated with A
β
1-
oligomers).
4
2

ACCEPTED MANUSCRIPT
2
.8. In vivo COX-1, COX-2 and 5-LOX activity evaluation
A recent study suggested that the pharmacophoric subunit derived from
curcumin could display anti-inflammatory effects through the inhibition of
cycloxygenases 1 (COX-1) and 2 (COX-2), and 5-lypoxygenase (5-LOX) [35]. In this
context, we evaluated the ability of compounds 12a and 12c to inhibit the activity of
COX-1, COX-2 and 5-LOX of mice treated with the compounds and carrageenan, a
pro-inflammatory lipopolysaccharide. In particular, COX-1, COX-2 and 5-LOX
activities were measured in mice serum by using enzyme-linked immunosorbent assay.
As shown in Fig. 13, compounds 12a, 12c and indomethacin, a positive control,
significantly decreased the activity of COX-1 (inhibition of 40%, 30% and 50%,
respectively) and COX-2 (27%, 60% and 42%, respectively); by contrast, only
compound 12a was able to inhibit 5-LOX activity (30%). Face to these results,
compound 12a showed the widest inhibitory activity profile on COX-1, COX-2 and 5-
LOX.
Fig. 13. Effect of compounds 12a and 12c on COX-1, COX-2 and 5-LOX activity in mice
treated with carrageenan (CG). Mice were treated with compounds 12a, 12c (100 µmol/kg, p.o)
and indomethacin (IND, 100 µmol/kg, p.o.) 1 h before intraplantar injection of carrageenan (400
µg/paw). After 3 h, the COX-1, COX-2 and 5-LOX activity was determined as reported in the
Experimental section. Values are expressed as mean ± SEM (n=8) in terms of enzymatic
activity percentage (*P<0.05 compared to the carrageenan group; **P<0.01 compared to the
carrageenan group and the indomethacin group).

ACCEPTED MANUSCRIPT
2
.9. In vivo anti-inflammatory activity evaluation
It is currently accepted that inflammation is a hallmark of AD pathogenesis.
Therefore, suppression of the neuroinflammation process may be a potential therapeutic
approach against AD [36], [37], [38]. Thus, compounds 12a-c were additionally
evaluated for their in vivo anti-inflammatory effects. These compounds were firstly
assayed in the formalin-induced antinociceptive model at a dose of 100 µmol/kg p.o.
The formalin test consists of two distinct phases. In the first one (neurogenic phase), a
peripheral nociceptive effect due to formalin activation of tissue nociceptors is
observed. In a second phase (inflammatory phase), the nociceptive effect is observed as
a consequence of the release of several pro-inflammatory molecules (e.g. histamine,
serotonin and bradykinin) in an acute inflammatory response [39], [40]. According to
our results depicted in Fig. 14, only compounds 12a and 12c exhibited a significant
antinociceptive activity in comparison to the control and indomethacin. Compound 12a
showed an antinociceptive response in both first and second phases, while compound
1
2c was significantly active only in the inflammatory phase. In agreement with these
results, compounds 12a and 12c were capable of reducing the time of licking during the
inflammatory phase, suggesting that both have an anti-inflammatory effect, which could
be a result of inhibition of formation and/or liberation of inflammatory mediators or by
directly blocking their receptors [41]. In addition, compounds 12a and 12c did not alter
motor performance in the open field model, confirming that the reduced licking time
observed in the formalin test was not due to motor impairment (data not shown).

ACCEPTED MANUSCRIPT
Fig. 14. Effect of compounds 12a, 12b and 12c in a mice model of formalin-
induced inflammatory pain. Mice were pre-treated orally with vehicle, compounds 12a-c (100
µmol/kg), indomethacin (100 µmol/kg), morphine (39 µmol/kg) or vehicle (saline + DMSO
2
%) prior to formalin administration. The total time spent licking the hindpaw was measured in
the first and second phases after intraplantar injection of formalin. Values are expressed as mean
±
SEM (n=8) (*p<0.05, **p<0.01, and ***p<0.001 compared to control group).
Taking into account the effects of compounds 12a and 12c in the second phase
of the formalin test, they were then selected for further anti-inflammatory evaluation in
LPS-induced peritonitis, carrageenan-induced paw edema and Von Frey models.
Leukocyte migration to the inflammation site is a fundamental event in the onset of the
inflammatory response [42]. Considering that this cell migration inhibition is also one
of the characteristic effects of anti-inflammatory drugs, compounds 12a and 12c were
evaluated in the LPS-induced peritonitis assay. This acute model of inflammation is
used to evaluate leukocyte migration by the count of total leukocytes released during
acute inflammation response after intraperitoneal administration of a phlogistic agent
[
43]. Results depicted in Fig. 15 show that both compounds 12a and 12c significantly
inhibited leukocyte migration in comparison to the control group (vehicle + LPS) and
dexamethasone, reinforcing their anti-inflammatory properties.

ACCEPTED MANUSCRIPT
Fig. 15. Effect of compounds 12a and 12c in a mouse model of LPS-induced peritonitis. Mice
were pre-treated orally with compounds 12a and 12c (100 µmol/kg), dexamethasone (10
mg/kg, p.o.) or vehicle (saline + DMSO 2%) prior to LPS administration. The number of
leukocytes recruited in the peritoneal cavity was determined. Values are expressed as mean ±
*
**
**
SEM (n=8). #P<0.05 compared with the saline + vehicle group; P<0.01, P<0.01and
*
P<0.05 compared with the vehicle + LPS group.
In another approach, compounds 12a and 12c were additionally tested for their
ability to interfere in the carrageenan-induced mechanical hypernociception (Von Frey
test). In this model, it is evaluated the animal sensitivity to a mechanical stimulus after
induced-hyperalgesia by an intraplantar injection of carrageenan. The induction of
hypernociception by carrageenan occurs through activation of the cytokine cascade that
leads to the release of inflammatory mediators (e.g. prostaglandins), which cause the
activation of Aδ and C fiber sensory nerve endings, increasing the local flow and
vascular permeability by the release of substance P and neurokinin A [44], [45]. The
results shown in Fig. 16 A and B demonstrate that administration of carrageenan caused
rd
hyperalgesia with the peak of effect in the 3 hour after intraplantar administration.
rd
Compounds 12a and 12c exhibited a significant antinociceptive effect after the 3 hour
of treatment. Compound 12a showed a very similar effect to indomethacin used as drug
control, while 12c was shown to be active during all periods of the experiment with a
greater breadth of statistical significance compared to the indomethacin and vehicle
groups. As depicted in Fig. 16 B, both compounds exhibited their maximum anti-

ACCEPTED MANUSCRIPT
rd
inflammatory activity in the 3 hour after treatment, with a fast decrease in their
effectiveness after this time.
Fig. 16. Effect of compounds 12a and 12c on mechanical hyperalgesia in a mice model of
carrageenan-induced paw edema. Mice were pre-treated orally with compounds 12a and 12c
(
100 µmol/kg p.o) or indomethacin (100 µmol/kg p.o) prior to intraplantar injection of
carrageenan (100 g/paw). The results are presented as threshold for touch sensitivity (weight in
grams). Values are expressed as mean ± SEM (n=8). ### P<0.01 compared with the saline +
ꢀ
***
**
*
vehicle group;
group.
P<0.01, P<0.01 and P<0.05 compared with the vehicle + carrageenan
Aiming to confirm the anti-inflammatory effect of compounds 12a and 12c, we
also performed the carrageenan-induced paw edema assay. In this model, the
development of edema induced by carrageenan is a biphasic event. The early phase (1–2
h) is mainly mediated by histamine, serotonin and bradykinin, followed by a late phase
(3-6 h) sustained by the release of prostaglandins and NO produced by an inducible
isoform of cyclooxygenase (COX-2) and NO-synthase (iNOS) [46], [47]. In this assay,
compounds 12a and 12c promoted a significant decrease in the formation of

ACCEPTED MANUSCRIPT
carrageenan-induced paw edema in comparison to the vehicle group (Fig. 17). The
inhibitory profile of both compounds was similar to indomethacin, showing a maximum
rd
effect at the 3 hour, with a higher potency for compound 12a. The mechanism of
action of indomethacin is produced by blockade of cyclooxygenases and, in turn,
preventing the biosynthesis of prostaglandins. Taking all these results together, it is
possible to suggest that compounds 12a and 12c exert their anti-inflammatory effect by
this same mechanism of action.
Fig. 17. Effect of compounds 12a and 12c in a mouse model of carrageenan-induced paw
edema. Mice were pre-treated orally with compounds 12a and 12c (100 µmol/kg p.o) or
indomethacin (100 µmol/kg p.o) prior to intraplantar injection of carrageenan (Cg, 100 ꢀg/paw).
The results are presented as the paw volume (µL) variation in relation to the basal values.
Values are expressed as mean ± SEM (n=8). *P<0.05, **P<0.01, ***P<0.001 compared with
control group.
3
3
. EXPERIMENTAL SECTION
.1. Chemistry
All chemicals used were purchased from Sigma Aldrich. Solvents were purified
and dried by standard procedures. Reactions was monitored by thin layer
chromatography (TLC) on aluminum plates precoated with silica gel 60 F254 and the
spots were visualized using a UV lamp (λ = 254 nm). Chromatographic separations
were performed on silica gel (90-150 µm; Merck) columns by flash chromatography.
NMR spectra of target compounds measured on a Bruker AC-300 spectrometer
1
13
operating at 300 MHz for H and 75 MHz for C MHz and IR spectra were performed
on Thermo Scientific Nicolet iS50 FT-IR spectrometer.

ACCEPTED MANUSCRIPT
3
.1.1. General procedures for the preparation of compounds 9a-9g
A mixture of 4-piperidine methanol (1.2 mmol) and functionalized
benzaldehydes (1.0 mmol) in 10 mL of anhydrous methanol and molecular sieve 4A°
was stirred for 1 h. The solution of sodium cyanoborohydride (1.0 mmol) and zinc
chloride (0.5 mmol) in anhydrous methanol (10 mL) also was stirred for 1 hour in r.t.
After two hours the solutions were mixed and stirring was continued for 72 hours. The
reaction mixture was treated with 0.1 M NaOH (10 mL) and extracted with ethyl
acetate. The combined extracts were washed with brine, dried, and evaporated to
dryness. Purification of the crude products were purified by flash column
chromatography using DCM-MeOH (8:2) as eluents.to afford 9a-9g.
3
.1.1.1. 1-benzyl piperidine-4-methanol (9a)
-1
IR-ATR (cm ) 3340, 3084, 3026, 2915, 1600, 1494, 1451, 1256, 1040, 736, 697. NMR
1
H (
δ
,
CDCl3, 300 MHz): 7.23-7.31 (m, 4H, HAr), 3.50 (s, 2H, CH
.3 Hz, CH OH), 2.87-2.94 (m, 2H, CH CH N), 1.96 (t, =11.7 Hz, 2H, CH
.66-1.72 (m, 2H, CH CH N), 1.21-1.34 (m, 2H, CH CH N), 1.42-1.55 (m, 1H,
). NMR C ( CDCl3, 75 MHz): 137.74, 129.41, 128.13, 127.04, 67.34,
2
Ar), 3.45 (d, 2H,
J=
6
2
2
2
J
2
CH N),
2
1
2
2
2
2
13
CH
2
CHCH
2
δ
,
6
3.39, 53.34, 38.35, 28.59.
3
.1.1.2. 1-(4-nitrobenzyl) piperidine-4-methanol (9b)
-1
IR-ATR (cm ) 3306, 3103, 2945, 1604, 1595, 1511, 1443, 1347, 1252, 1036, 849, 803.
1
NMR H (
δ
,
CDCl3, 300 MHz): 8.17 (d, 2H,
Ar), 3.51 (d, 2H,
J
=9.0 Hz, HAr), 7.51 (d, 2H,
J=8.2 Hz,
HAr), 3.58 (s, 2H, CH
2
J
= 6.3 Hz, CH OH), 2.85-2.88 (m, 2H,
2
CH CH N), 1.98-2.07 (m, 2H, CH CH N), 1.71-1.74 (m, 2H, CH CH N), 1.27-1.37 (m,
2
2
2
2
2
2
1
3
2
1
H, CH
2
CH
2
N), 1.45-1.60 (m, 1H, CH
2
CHCH
2
). NMR C (
δ
,
CDCl3, 75 MHz): 147.08,
29.46, 123.47, 129.46, 146.79, 147.08, 67.82, 62.54, 53.60, 38.39, 28.78.

ACCEPTED MANUSCRIPT
3
.1.1.3. 1-(4-methoxybenzyl) piperidine-4-methanol (9c)
-1
IR-ATR (cm ) 3361, 3072, 2926, 1611, 1583, 1511, 1244, 1178, 1034, 851, 818. NMR
1
H (δ
,
CDCl3, 300 MHz): 6.84 (d, 2H,
J=8.7 Hz, HAr), 7.22 (d, 2H, J=8.7 Hz, HAr),
3
.44 (s, 2H, CH Ar), 3.45 (d, 2H,
J
= 6.4, CH OH), 3.79 (OCH ), 2.86-2.92 (m, 2H,
2
2
3
CH
2
CH
2
N), 1.93 (t,
2 2 2 2
J=11.7, 2H, CH CH N), 1.66-1.72 (m, 2H, CH CH N), 1.20-1.33
13
(
m, 2H, CH
2
CH N), 1.41-1.53 (m, 1H, CH
2
2
CHCH
2
). NMR C (
δ
,
CDCl3, 75 MHz):
1
59.24, 131.32, 126.91, 113.78, 66.60, 61.60, 52.51, 37.67, 27.60.
3
.1.1.4. 1-(4-chlorobenzyl) piperidine-4-methanol (9d)
-1
1
IR-ATR (cm ) 3326, 3056, 2919, 1596, 1576, 1489, 1253, 1039, 843, 802. NMR H (
δ
,
2
CDCl3, 300 MHz): 7.20-7.29 (m, 4H, HAr), 3.42 (s, 2H, CH Ar), 3.39 (d, 2H, J= 6.3
Hz, CH OH), 2.80-2.85 (m, 2H, CH CH N), 1.87-1.95 (m, 2H, CH CH N), 1.62-1.68
2
2
2
2
2
(m, 2H, CH
2
CH
2 2 2 2 2
N), 1.16-1.29 (m, 2H, CH CH N), 1.36-1.45 (m, 1H, CH CHCH ).
1
3
NMR C (δ CDCl3, 75 MHz): 136.30, 130.65, 132.70, 128.24, 67.18, 62.53, 53.24,
,
3
3
8.23, 28.52.
.1.1.5. 1-(4-bromobenzyl) piperidine-4-methanol (9e)
-1
IR-ATR (cm ) 3343, 3054, 2941, 1639, 1592, 1490, 1239, 1039, 1072, 844, 803. NMR
1
H (
δ
,
CDCl3, 300 MHz): 7.43 (d, 2H,
.45 (s, 2H, CH Ar), 3.47 (d, 2H, = 6.4 Hz, CH
.92-2.00 (m, 2H, CH CH N), 1.68-1.73 (m, 2H, CH
J
=8.5 Hz, HAr), 7.20 (d, 2H,
OH), 2.85-2.92 (m, 2H, CH
CH N), 1.21-1.36 (m, 2H,
J
=8.5 Hz, HAr),
3
1
2
J
2
2
CH N),
2
2
2
2
2
13
CH CH N), 1.42-1.56 (m, 1H, CH CHCH ). NMR C (δ, CDCl 75 MHz): 137.27,
2
2
2
2
3,
1
30.89, 131.27, 120.82, 67.76, 62.60, 53.34, 38.44, 28.59.
3
.1.1.6. 1-(4-morpholinobenzyl) piperidine-4-methanol (9f)

ACCEPTED MANUSCRIPT
-
1
1
IR-ATR (cm ) 3374, 3034, 1611, 1515, 1449, 1217, 1068, 1114, 924. NMR H (
CDCl3, 300 MHz): 7.22 (d, 2H, =8.4 Hz, HAr), 6.86 (d, 2H, =8.7 Hz, HAr), 3.86 (t,
=9.6 Hz, 2H, NCH CH O), 3.44 (s, 2H, CH Ar), 3.48 (d, 2H, = 6.3 Hz, CH OH), 3.15
t, J=9.6 Hz, 2H, NCH CH O), 2.88-2.94 (m, 2H, CH CH N), 1.90-2.00 (m, 2H,
CH CH N), 1.67-1.73 (m, 2H, CH CH N), 1.22-1.35 (m, 2H, CH CH N), 1.43-1.53 (m,
H, CH CHCH ). NMR de C (δ, CDCl 75 MHz): 131.30, 124.59, 115.23, 151.01,
δ
,
J
J
J
2
2
2
J
2
(
2
2
2
2
2
2
2
2
2
2
1
3
1
6
2
2
3,
6.48, 66.77, 61.19, 52.15, 48.90, 37.49, 27.20.
3
.1.1.7. 1-(1,3dioxolbenzyl-5-ylmethyl)-piperidine-4-methanol (9g)
-1
IR-ATR (cm ) 3369, 3068, 2943, 1502, 1491, 1441, 1209,1243, 1100, 1033, 872, 808.
1
NMR H (δ, CDCl 300 MHz): 6.86 (s, 1H, HAr), 6.74 (s, 2H, HAr), 5.93 (s, 2H,
3,
OCH
CH CH
H, CH CH N), 1.43-1.55 (m, 1H, CH CHCH ). NMR C (δ, CDCl 75 MHz): 131.72,
2
O), 3.43 (s, 2H, CH
2
Ar), 3.48 (d, 2H,
J
= 7.2 Hz, CH
2
OH), 2.88-2.94 (m, 2H,
2
2
N), 1.92-2.01 (m, 2H, CH
2
CH N), 1.68-1.73 (m, 2H, CH
2
2
CH N), 1.23-1.36 (m,
2
1
3
2
1
2
2
2
2
3,
47.59, 146.64, 122.49, 109.73, 107.85, 100.89, 67.73, 63.03, 53.19, 38.45, 28.59.
.1.2. General procedures for the preparation of compounds 12a-12n
To a mixture of ferulic acid or iso-ferulic acid (1 mmol) in dry dichloromethane
3
(5mL), thionyl chloride (10 mmol) and DMF (25 µL) were added, and the resulting
solution was stirred under argon atmosphere for 60 min at rt. After evaporation of the
solvent, the residue was dissolved in dry dichloromethane (10 mL) and slowly added to
the appropriate intermediate (9a-g) (1.5 mmol). The reaction mixture was stirred for 15
min at r.t. under argon atmosphere. The solvent was evaporated under vacuum.
Purification of the crude products was by flash column chromatography using DCM-
MeOH (9.5:0.5) as eluents to afford 12a-12n.
3
.1.2.1. (
E
)-(1-benzylpiperidin-4-yl)methyl 3-(4-hydroxy-3-methoxyphenyl)acrylate
(12a)
-1
IR-ATR (cm ): 3422, 3062, 2937, 1703, 1631, 1588, 1451, 1281, 1029, 1154, 736, 698.
1
NMR H (
δ
,
CDCl3, 300 MHz): 6.88 ( d, 1H,
J=8.1, HAr), 7.01 (d, 1H,
J
=1.8, HAr)
,

ACCEPTED MANUSCRIPT
7
.04 (dd, 1H,
J
=8.2 and 1.9, HAr)
,
7.30-7.31 (m, 5H, HAr), 7,60 (d, 1H,
J
=15.9,
2.93-2.97 (m,
1.71-1.76 (m, 3H, CH CH N and
3.53 (s, 2H, CH Ar) 3.89 (s, 3H, OCH ).
146.98 148.19 109.44 126.83 123.04 144.88
,
CH=CH)
,
6.27 (d, 1H,
J
=15.9, CH=CH)
,
4.06 (d, 2H,
CH N)
N)
2 ,
J=6.0, COOCH )
2
H, CH
2
CH
2
N), 1.97-2.05 (m, 2H, CH
1.40-1.49 (m, 2H, CH CH
CDCl3, 75 MHz): 114.95
2
2
,
2
2
CH
2
CHCH
2
)
,
2
2
,
2
,
3
1
3
NMR C (
δ
,
,
,
,
,
,
,
1
15.31 167.35 68.65 35.38 28.74 53.18 63.32 137.80 129.38 127.12 128.21 55.87.
,
,
,
,
,
,
,
,
,
,
,
+
HRMS (ESI) calculated for C23
H
4
27NO [M+H] : m/z 382.20183; found: 382.20205
3
.1.2.2. (E)-(1-(4-nitrobenzyl)piperidin-4-yl)methyl 3-(4-hydroxy-3-
methoxyphenyl)acrylate (12b)
-1
IR-ATR
(
cm ): 3373, 3072, 2992, 1698, 1631, 1588, 1509, 1343, 1256, 1149, 1028,
1
8
44, 741. NMR H (
δ
,
CDCl3, 300 MHz): 6.87 ( d, 1H,
J
=8.0, HAr), 7.00 (d, 1H,
J
J
J
=1.8,
=8.7,
=6.0,
HAr) 7.09 (dd, 1H,
J
= 8.1 and 1.8, HAr) 7.48 (d, 2H,
J
=8.7, HAr) 8.13 (d, 2H,
,
,
,
HAr) 7.58 (d, 1H,
COOCH 2.84-2.87 (m, 2H, CH
H, CH CH N and CH CHCH
CH Ar), 3.87 (s, 3H, OCH ). NMR C (
145.06 115.35 167.44
56.01. HRMS (ESI) calculated for C23
J
=15.9, CH=CH)
CH
1.31-1.45 (m, 2H, m, 2H, CH
,
6.26 (d, 1H,
J
=15.9, CH=CH)
CH
CH
CDCl3, 75 MHz): 115.05
68.69 35.41 28.98 53.50
,
4.05 (d, 2H,
2
)
,
2
2
N), 1.98-2.06 (m, 2H, CH
2
2
N), 1.70-1.75 (m,
N), 3.56 (s, 2H,
3
2
2
2
2
)
,
2
2
1
3
2
3
δ
,
,
146.62
,
148.27
,
1
1
4
09.57
,
,
126.94
,
123.15
,
,
,
,
,
,
,
,
62.58
,
147.19
,
+
23.56
129.64
,
147.07
,
H
26
N
2
O
6
[M+H] : m/z
27.18691; found: 427.18635
3
.1.2.3. (E)-(1-(4-bromobenzyl)piperidin-4-yl)methyl 3-(4-hydroxy-3-
methoxyphenyl)acrylate (12c)
-1
IR-ATR (cm ): 3428, 3063, 2937, 1699, 1630, 1588, 1509, 1253, 1115, 1030, 726.
1
NMR H (
δ
,
CDCl3, 300 MHz): 6.85 ( d, 1H,
= 8.3 and 2.0, HAr) 7.19 (d, 2H,
=15.9, CH=CH) 6.31 (d, 1H,
J
=7.5, HAr), 7.05 (d, 1H,
=8.4, HAr) 7.42 (d, 2H,
=15.9, CH=CH)
J
=2.0, HAr)
=8.3, HAr),
=6.0,
,
7
.09 (dd, 1H,
.62 (d, 1H,
J
,
J
,
J
7
J
,
J
,
4.05 (d, 2H,
J
COOCH ), 2.86-2.90 (m, 2H, CH CH N) 1.93-2.01 (m, 2H, CH CH N) 1.60-1.76 (m,
2
2
2
,
2
2
,
3
3
1
H, CH
.90 (s, 3H, OCH
27.32 122.58 144.60
2
CH
2
N and CH
). NMR C (
115.70 167.15
2
CHCH
2
)
,
1.31-1.47 (m, 2H, CH
CDCl3, 75 MHz): 113.02
69.91 35.37 28.88 53.14
2
CH
2
N), 3.44 (s, 2H, CH
149.15 151.06 109.54,
62.51 136.72 130.73
,
2 ,
Ar)
13
3
δ
,
,
,
,
,
,
,
,
,
,
,
,
,
,
,