Novel donepezil-like N-benzylpyridinium salt derivatives as AChE inhibitors and their corresponding dihydropyridine “bio-oxidizable” prodrugs: Synthesis, biological evaluation and structure-activity relationship

Article abstract of DOI:10.1016/j.ejmech.2017.12.084

As an extension of our previous work on donepezil-based “bio-oxidizable” prodrug approach, two new classes of N-benzylpyridinium donepezil analogues in tetralone B2 and acetophenone B3 series and a new set of indanone derivatives B1 were investigated along with the corresponding dihydropyridine prodrugs A1-3. A total of fifty one N-benzylpyridinium quaternary donepezil analogues B1-3 and twenty two prodrugs A1-3 were synthesized and evaluated for their inhibitory activities against hAChE and eqBuChE. While most prodrugs A1-3 were demonstrated to be inactive against AChE (IC₅₀ > 10 μM), a large number of the corresponding N-benzylpyridinium salt B1-3 exhibited appealing three-to-one-digit nanomolar hAChE inhibitory activities and even reaching subnanomolar activity (IC₅₀ = 0.36 nM). In addition, in silico docking studies were conducted for several compounds to explain the more relevant in vitro results. Lastly, the influence of the two stereogenic centers in prodrugs A was also evaluated, highlighting not only marked differences in residual AChE inhibitory activity of the four separated isomers of prodrug 23h (IC₅₀ ranging from 173 nM to 10 μM) but also significant variations of the oxidation rate between two separated diastereoisomers of prodrug 24a. This work provides useful information in the search of a preclinical candidate to conduct further development of this attractive “bio-oxidizable” prodrug strategy.

Full text of DOI:10.1016/j.ejmech.2017.12.084

Accepted Manuscript
Novel donepezil-like N-benzylpyridinium salt derivatives as AChE inhibitors and
their corresponding dihydropyridine “bio-oxidizable” prodrugs: Synthesis, biological
evaluation and structure-activity relationship
Rabah Azzouz, Ludovic Peauger, Vincent Gembus, Mihaela-Liliana Ţînţaş, Jana
Sopková-de Oliveira Santos, Cyril Papamicaël, Vincent Levacher
PII:
S0223-5234(17)31115-7
10.1016/j.ejmech.2017.12.084
EJMECH 10062
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 20 October 2017
Revised Date: 18 December 2017
Accepted Date: 23 December 2017
Please cite this article as: R. Azzouz, L. Peauger, V. Gembus, M.-L. Ţînţaş, J. Sopková-de Oliveira
Santos, C. Papamicaël, V. Levacher, Novel donepezil-like N-benzylpyridinium salt derivatives as AChE
inhibitors and their corresponding dihydropyridine “bio-oxidizable” prodrugs: Synthesis, biological
evaluation and structure-activity relationship, European Journal of Medicinal Chemistry (2018), doi:
10.1016/j.ejmech.2017.12.084.
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
Novel donepezil-like N-benzylpyridinium salt derivatives as AChE inhibitors
and their corresponding dihydropyridine “bio-oxidizable” prodrugs:
synthesis, biological evaluation and structure-activity relationship.
a
a
,a
b
Rabah Azzouz , Ludovic Peauger , Vincent Gembus* , Mihaela-Liliana Ţînţaş , Jana Sopková-
c
b
,b
de Oliveira Santos , Cyril Papamicaël , Vincent Levacher*
a
VFP Therapies, 15 rue François Couperin, 76000 Rouen, France
b
Normandie Univ, COBRA, UMR 6014 et FR 3038; Univ Rouen; INSA Rouen; CNRS, IRCOF, 1
rue Tesnière, 76821 Mont Saint Aignan Cedex, France
c
Centre d’Etudes et de Recherche sur le Médicament de Normandie, UFR des Sciences
Pharmaceutiques; Université de Caen, 1 rue Vaubénard, 14032 Caen Cedex, France
*
Corresponding authors : Vincent Levacher, tel/fax : +33235522485/+33235522462, e-mail:
vincent.levacher@insa-rouen.fr ; Vincent Gembus, tel/fax : +33235522465/+33235522462, e-
mail: vgembus@vfp-therapies.com
ABSTRACT
As an extension of our previous work on donepezil-based “bio-oxidizable” prodrug approach,
two new classes of N-benzylpyridinium donepezil analogues in tetralone B2 and acetophenone
B3 series and a new set of indanone derivatives B1 were investigated along with the
corresponding dihydropyridine prodrugs A1-3. A total of fifty one N-benzylpyridinium
quaternary donepezil analogues B1-3 and twenty two prodrugs A1-3 were synthesized and
evaluated for their inhibitory activities against hAChE and eqBuChE. While most prodrugs A1-3
were demonstrated to be inactive against AChE (IC50 >10 µM), a large number of the
corresponding N-benzylpyridinium salt B1-3 exhibited appealing three-to-one-digit nanomolar
1

ACCEPTED MANUSCRIPT
hAChE inhibitory activities and even reaching subnanomolar activity (IC50= 0.36 nM). In
addition, in silico docking studies were conducted for several compounds to explain the more
relevant in vitro results. Lastly, the influence of the two stereogenic centers in prodrugs A was
also evaluated, highlighting not only marked differences in residual AChE inhibitory activity of
the four separated isomers of prodrug 23h (IC50 ranging from 173 nM to 10µM) but also
significant variations of the oxidation rate between two separated diastereoisomers of prodrug
2
4a. This work provides useful information in the search of a preclinical candidate to conduct
further development of this attractive “bio-oxidizable” prodrug strategy.
Keywords: AChEIs, Alzheimer, “bio- oxidizable” prodrug, donepezil analogues
1
. Introduction
Alzheimer’s disease (AD) is the most common form of dementia occurring among ageing
people. There is an estimated 40 million people affected by AD worldwide, making this disease a
major public health issue [1, 2]. AD is a slow and progressive degenerative brain disease that is
characterized by impairments of memory, language, speech and awareness. The main
pathological hallmarks of AD are senile plaques stemming from beta-amyloid aggregates,
neurofibrillary tangles arising from abnormal hyperphosphorylation of tau proteins and the loss
of cholinergic neurotransmission [3-8]. All attempts to develop curative treatments, inter alia, by
seeking to limit the formation of senile plaques [9-11] and neurofibrillary tangles [11] have been
so far unsuccessful. Only acetylcholinesterase (AChE) inhibitors such as donepezil, rivastigmine
and galantamine show a certain effectiveness in restoring the central cholinergic function in AD
2

ACCEPTED MANUSCRIPT
patients during early stages of the disease and are the only marketed drugs together with
memantine for the palliative treatment of AD [12-16]. However, the effectiveness of AChE
inhibitors for patients with moderate- to late-stage AD is seriously hampered by the emergence
of severe adverse effects that result from peripheral anticholinesterase activity and triggered by
the need to gradually increase the dose over time [17, 18]. It is worth noting that new generation
of dual binding site AChE inhibitors targeting both catalytic active site (CAS) and peripheral
anionic site (PAS) are highly sought after, as they can not only improve cognition by restoring
cholinergic activity but also shrink the formation of senile plaques by inhibiting AChE-induced
Aβ aggregation [19-28]. In recent years, much attention has been paid to N-benzylpyridinium
salts bearing various heterocycles at C-4 position of the pyridine core as new dual binding site
AChE inhibitors analogues of donepezil [29-34]. While a number of these donepezil analogues
proved to be potent inhibitors of AChE, their therapeutic potential in AD is seriously impeded by
the presence of a permanent positive charge which strongly reduces the probability for high
blood brain barrier (BBB) permeability. With the aim to improve the symptomatic treatments of
AD by reducing side effects of AChE inhibitors, our group develops new redox chemical
delivery systems [35-37], initially reported by N. Bodor [38-45], to transport AChE inhibitors
into the brain. More recently, we designed a new class of donepezil-like compounds as “bio-
oxidizable” prodrugs of central AChE inhibitors [46]. The design of this prodrug strategy is
based on the redox properties of indanone-derived N-benzyl 1,4-dihydropyridines A1 able to
cross BBB which would be then converted into dual binding site AChEIs B1 via an oxidative
activation step occurring into the brain. Further in vitro experiments showed that prodrugs A1
are stable in human plasma, while their oxidation into drugs B1 occurs within few minutes in the
2
presence of mice’s brain homogenate or vitamin B [46]. These in vitro experiments thus
3

ACCEPTED MANUSCRIPT
foreshadow a rapid activation of the prodrug A once in the brain. Gratifyingly, whereas
indanone-derived N-benzylpyridinium salts B1 proved to be highly potent dual binding site
hAChE inhibitors (IC up to 3 nM), most of the corresponding 1,4-dihydropyridine prodrugs A1
5
0
showed to be inactive (IC > 10 µM). As evidenced by docking studies, the non-activity of
50
prodrugs A1 against AChE would result from the absence of positive charge at the nitrogen, not
basic enough to be protonated and thus preventing any cation-π interactions with aromatic
residues within the active site gorge of AChE. Furthermore, Pampa-BBB assay showed good
prediction of BBB penetration for all prodrugs A1 and brain trapping of the corresponding
permanently charged pyridinium salts B1. To gain further insight into this promising “bio-
oxidizable” prodrug system, we wish to extend the AChE inhibitors SAR study to newly
synthetized indanone-derived N-benzylpyridinium salts B1 together with two new classes of N-
benzylpyridinium quaternary donepezil analogues in tetralone and acetophenone series B2 and
B3. This work includes (1) the preparation of both aryl ketone-derived N-benzylpyridinium salts
B1-3 and the corresponding 1,4-dihydropyridine derivatives A1-3; (2) their in vitro evaluation
against both hAChE and eqBChE. The SAR analysis will focus on (1) the substitution pattern at
aryl ketone moieties interacting with the PAS region; (2) the nature of the N-substituent at the
pyridine ring located in the CAS and (3) the nature of the electron withdrawing group (EWG) at
C-3 position of the central pyridine ring lodged within the active site gorge. Although the main
role of this EWG is to ensure a good stability of the 1,4-dihydropyridine prodrug A, it also
provides the opportunity to carry out a SAR study of the central part of donepezil analogues, still
largely unexplored in the literature. Last but not least, we also investigated the effect of the two
stereogenic centers in dihydropyridine A both on undesirable hAChE inhibitory activity of the
prodrug A and on its conversion rate into the corresponding pyridinium salt B.
4

ACCEPTED MANUSCRIPT
Figure 1. (a) rhAChE active site bound donepezil ; (b) Previous work: design of "bio-oxidizable" prodrugs of
central AChE inhibitors derived from donepezil [46] ; (c) This work: expanding the scope of evaluation to new
indanone, tetralone and acetophenone derivatives.
2
2
. Results and discussion
.1. Chemistry
The key intermediates 4a-m were straightforwardly prepared by a two-step sequence
involving an aldol condensation between the commercially available aryl ketones 1a-m and 3-
bromo-4-pyridinecarboxaldehyde 2 under basic or acidic conditions affording compounds 3a-
m in moderate to excellent yields (26-99% yields). Selective hydrogenation of the alkene
under mild conditions furnished the desired 3-bromopyridine derivatives 4a-m in 58-99%
yields (Scheme 1).
5

ACCEPTED MANUSCRIPT
Scheme 1. Reagents and conditions: (a) K CO , EtOH, reflux for 3a (64%); or NaOH, EtOH, rt for 3c-d (57-99%);
2
3
or PTSA, toluene, reflux, Dean-Stark for 3b,e-m (26-79%); (b) H , Pt/C, EtOH, 20°C (58-99%).
2
From these key intermediates 4a-m, various Pd-catalyzed cross-coupling reactions were
achieved to set up a variety of EWGs at C-3 of the pyridine core (Scheme 2). Thus, Pd-catalyzed
carbonylation of compounds 4a-c in the presence of a carbon monoxide surrogate; namely 2,4,6-
trichlorophenyl formate, provided the corresponding activated 2,4,6-trichlorophenol esters 5a-c
(51-59% yields). The later could be readily converted into an assortment of carboxylic acid
derivatives by reacting various nucleophiles; namely methanol to afford methyl esters 6a-b (75,
8
7
4% yields) and various amines such as NH
3
or methylamine providing carboxamides 7a-c (58-
8% yields) and 8a-b (96-99% yields) respectively. Alternatively, Pd-catalyzed cross-coupling
reactions of compounds 4a-m with 1-(ethoxyvinyl)tri(n-butyl)stannane furnished the
corresponding 3-acetyl pyridine derivatives 9a-m in 58-93% yields. The conversion of 3-
bromopyridine derivatives 4a-b into the corresponding 3-carbonitrile pyridine derivatives 10a-b
was accomplished by a Pd-catalyzed cyanation reaction in 37-83% yields. Lastly, 3-
bromopyridine derivative 4b could be transformed into the trifluoromethylpyridine derivatives
6

ACCEPTED MANUSCRIPT
1
2 by a two-step reaction sequence involving halogen exchange reaction to furnish the more
reactive 3-iodopyridine derivative 12 (75% yield) which was then subjected to Pd-catalyzed
trifluoromethylation reaction (38% yield).
Scheme 2. Reagents and conditions: (a) Xantphos, Pd(OAc) , 2,4,6-trichlorophenyl-formate, n-Bu N, toluene,
2
3
Schlenk tube, 110°C, overnight (51-59%); (b) MeOH, NEt , Schlenk tube, 70°C, 20h for 6 (75-84%) or NH in
3
3
dioxan, Schlenk tube, 70°C, 20h for 7 (58-78%) or MeNH , NEt , THF, Schlenk tube, 70°C, 24h for 8 (96-99%); (c)
2
3
Pd(dba) , PPh , 1-(ethoxyvinyl)tri(n-butyl)stannane, toluene, reflux then aq. HCl, rt (58-93%); (d) Zn(CN) ,
2
3
2
Pd(PPh ) , DMF, microwaves, 170°C, 5 min to 1h for 10a (37%) or CuCl , MSTFA, toluene for 10b from 7b
3
4
2
(
83%); (e) CuI, NaI, DMEDA, dioxan, 110°C, 20h (75%); (f) PhenCuCF , DMF, Schlenk tube, 80°C, 24h (38%).
3
All compounds 4-12 were subsequently quaternized affording the corresponding indanone-,
tetralone – and acetophone-derived N-alkylpyridinium salts 13-21 in moderate to good yields
(32-99% yields). The regioselective reduction of these putative AChE inhibitors 13-21 was
performed by using either N-benzyl-1,4-dihydronicotinamide (BNAH) or sodium dithionite
7

ACCEPTED MANUSCRIPT
giving rise to the corresponding 1,4-dihydropyridine prodrugs 22-24 in 11 to 70% yields
(Scheme 3).
Scheme 3. Reagents and conditions: (a) RCH Br or RCH I, CH Cl or ACN, reflux (32-99%); (b) BNAH, CH Cl ,
2
2
2
2
2
2
rt or Na S O , NaHCO , dioxan/water, 50°C (11-70%).
2
2
4
3
In addition, we were also interested in investigating whether a longer linker connecting both the
central pyridine core and the aryl ketone moiety could have an influence on AChE inhibitory
activity. To this end, both acetophenone- and indanone-derived N-benzylpyridinium salts 33 and
1 were prepared as outlined in schemes 4 and 5 respectively. In short, the synthesis of 33 was
4
undertaken starting from acetophenone derivative 26, itself obtained in 90% yield by Friedel-
Craft acylation of veratrole 25 with 4-chlorobutyryl chloride. As a result of a standard
chloride/iodide exchange reaction, the iodo-derivative 27 was isolated in 97% yield before being
reacted with triphenylphosphine to give phosphonium salt 28 (97% yield). Reaction between 28
and 3-bromo-4-pyridinecarboxaldehyde was accomplished under mild basic conditions (K CO ,
2
3
1
8-crown-6) affording the Witting reaction product 29 in 57% yield. After hydrogenation of
alkene 29, proceeding smoothly under hydrogen atmosphere over Pd/C, the resulting reduced
product 30 was subjected to Pd-catalyzed carbonylation in the presence of 2,4,6-trichlorophenyl-
formate as monoxide carbon source to give the activated ester 31 in 72% yield which was
8

ACCEPTED MANUSCRIPT
subsequently transformed into carboxamide 32 in 72% yield upon treatment with ammonia in
dioxan. A last step of N-benzylation with 3-chlorobenzyl bromide furnished the required
acetophenone-derived N-benzylpyridinium 33 in 87% yield (Scheme 4).
3 2 2
Scheme 4. Reagents and conditions: (a) 4-chlorobutyryl chloride, AlCl , CH Cl , 0°C to rt (90%); (b) NaI, acetone,
reflux (97%); (c) PPh , toluene, reflux (97%); (d) 3-bromo-4-pyridinecarboxaldehyde, K
3
2
3 2 2
CO , 18-crown-6, CH Cl ,
5
5°C (sealed tube) (57%); (e) H , Pt/C, EtOH, 1 atm., rt (58%); (f) XantPhos, Pd(OAc) , 2,4,6-trichlorophenyl-
2
2
formate, n-Bu N, toluene, Schlenk tube, 110°C, overnight (72%); (g) NH in dioxan, Schlenk tube, 70°C (78%); (h)
3
3
3
-chlorobenzyl bromide, CH Cl , Schlenk tube, 55°C (87%).
2
2
The last target compound 41 was prepared through an alkylation decarboxylation sequence from
the β-ketoester 34 and the alkylating agent 37 affording the indanone-derived pyridine 39 in 72%
over the two steps. The pyridine 3-carboxamide derivative 40 was obtained through a two-step
sequence already described above (65% overall yield); namely a Pd-catalyzed carbonylation
reaction with 2,4,6-trichlorophenyl-formate followed by amidation of the resulting activated
ester. Lastly, the resulting carboxamide 40 undergoes smooth quaternization with 3-chlorobenzyl
bromide to produce in 85% yield the target indanone-derived N-benzylpyridinium 41 (Scheme
5
).
9

ACCEPTED MANUSCRIPT
Scheme 5. Reagents and conditions: (a) Mesyl chloride, NEt
68%); (c) K CO , DMF, 50°C (80%); (d) 30% aq. NaOH, THF, reflux (90%); (e) Xantphos, Pd(OAc)
trichlorophenyl-formate, n-Bu N, toluene, Schlenk tube, 110°C, overnight then NH in dioxane, Schlenk tube, 70°C
65%); (f) 3-chlorobenzyl bromide, CH Cl , Schlenk tube, 55°C (85%).
3
, CH
2
Cl
2
, 0°C to rt (99%); (b) NaI, acetone, reflux
(
2
3
2
, 2,4,6-
3
3
(
2
2
2
.2. In vitro inhibition of AChE and BuChE, propidium displacement assays
The three classes of aryl ketone-derived N-benzylpyridinium salts B1-3 and the corresponding
,4-dihydropyridine prodrugs A1-3 were evaluated against hAChE and eqBChE using the
1
method of Ellman et al. [47]. Donepezil and tacrine were used as standards. It is worth noting
that unlike isoenzyme BChE, the level of AChE decreased significantly at late stage of AD [48-
5
3], hence making BChE an attractive target to tackle the cholinergic deficit at an advanced stage
of the disease. The dual binding site mode of these AChE inhibitors was assessed by their
capacity to displace propidium iodide from the PAS of electric eel AChE [54-59].
2
.2.1. By indanone-derived N-benzylpyridinium salts (series B1)
To gain an overall view, a selection of some key highlights of our previous work [46] obtained in
this series B1 was included in Table 1 (compounds 14a-f, 15a, 16a, 19a,b), so that the
performances of the newly synthetized indanone-derived N-benzylpyridinium salts 13a, 14g-j,
10

ACCEPTED MANUSCRIPT
1
7a, 18, 19c-g, 21a and 41, can be fully analyzed. First of all, to further expand the range of
EWGs installed at C-3, carboxylic acid 17a was assessed and showed a good anti-hAChE
activity (IC = 18 nM), akin to that previously observed with acetyl 14a (IC = 3.27 nM), ester
5
0
50
1
6a (IC = 5.3 nM) and primary amide 19b (IC = 41 nM) derivatives, each of them including an
50 50
identical substitution pattern at both indanone and N-benzyl moieties. Interestingly, the
introduction of a more sterically hindered 2,4,6-trichlorophenol ester at C-3 of the pyridinium
salt 18 does not appear to seriously affect the hAChE inhibitory activity (IC50= 69.3 nM).
Remarkably, derivative 13a wherein a bromine atom is attached at the C-3 position of the
pyridine ring displayed a sub-nanomolar IC50 (0.36 nM). The substitution pattern at the indanone
moiety was then examined more closely with compounds 14b-j (EWG= Ac), confirming the
general trend already observed with donepezil [60-63]; namely that compounds bearing electron
donating groups at the indanone moiety display good affinities for hAChE with a one- to three-
digit nanomolar IC50 values (2.7 nM to 116.5 nM), while the introduction of an electron
1
withdrawing group in 14g (R = F) lead to a detrimental effect on the anti-AChE activity (IC >
5
0
1
4
0µM). All attempts to improve the hAChE inhibitory activity from 19b (EWG= CONH
2
; IC50=
1 nM) by varying the N-benzyl appendage in compounds 19c-g failed; only IC values in the
5
0
micromolar range were obtained. It is worth noting that changing the N-3-chlorobenzyl moiety in
9b (IC = 41 nM) by a phenethyl group caused a dramatically drop of the inhibitory potency of
1
5
0
the resulting pyridinium salt 19e (IC50 = 896 nM). Similarly, using a larger linker to connect both
indanone and pyridine moieties in target 41 is extremely deleterious to the inhibitory activity
(
IC50 > 10 µM). As already reported with donepezil analogues [46], only low affinity for
eqBuChE was observed for all the newly synthesized compounds 13a, 14g-j, 17a, 19c-g, 21a
11

ACCEPTED MANUSCRIPT
and 41 (IC50 values > 10µM) with the exception of the 2,4,6-trichlorophenol ester 18 which
inhibits eqBChE in the sub-micromolar range (IC = 0.373 µM).
5
0
Table 1. Inhibition of hAChE and eqBChE by indanone-derived N-benzylpyridinium salts
IC50 (nM) ± SD
Propidium
2
1
c
iodide
compd
n
EWG
R
R
R
SI
a
b
displacement
hAChE
eqBChE
d
(%)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
4
3a
4a
4b
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
Br
OCH
OCH
OCH
H
3
3
2
OCH
OCH
-
3
Ph
0.36±0.04
8.8±0.2
60±3
1790±50
5930±440
1615±45
2990±240
262±15
4972
15±1
14±1
19±2
23±3
18±2
20±2
20±3
4
6
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
CN
2-MePh
2-MePh
2-MePh
2-MePh
2-MePh
2-MePh
3-Cl-Ph
2-Me-Ph
3-Cl-Ph
3-Cl-Ph
Ph
674
3
3
3
3
3
3
3
3
3
3
3
4
6
O
26.9
4
6
4c
OCH
3
30±1
133
4
6
4d
CH
3
CH
CH
Cl
3
3
27.7±3.7
91±9
6.95
4
6
4e
H
3815±375
3630±90
41.9
4
6
4f
4g
4h
4i
H
262±17
>10µM
37.1±2.6
2.7±0.1
116.5±1.1
2.9±0.1
5.3±0.02
18.1±3
69.3±2
678±42
41±20.9
13.9
e
e
H
F
nd
-
nd
H
OCH
OCH
OH
CF
4147±218
>10µM
112
19±2
2
3
e
OH
3
-
nd
e
4j
H
>10µM
-
nd
4
6
e
5a
6a
7a
8
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
3
3
3
3
3
3
3
3
3
3
3
3
3
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
3
3
3
3
3
3
3
3
3
3
3
3
3
nd
-
25±2
4
6
e
COOCH
COOH
3-Cl-Ph
3-Cl-Ph
2814±821
531
nd
3
e
e
nd
-
nd
f
COOY
2-CH
OOC-Ph
373.2±30
>10µM
5.38
14±1
16±1
19±2
3
4
6
9a
9b
9c
9d
9e
9f
CONH
CONH
CONH
CONH
CONH
CONH
CONH
2
2
2
2
2
2
2
2-MePh
3-ClPh
-
-
-
-
-
-
-
-
-
4
6
>10µM
e
3-HOPh
>10µM
>10µM
5361±199
nd
e
4-(NH
CH Ph
2-(6-CH
SO
)Ph
>10µM
>10µM
>10µM
>10µM
4765±425
3264
nd
2
2
e
2
896±21
nd
e
)Py
1861±730
9267±738
nd
3
e
9g
2-benzoxazole
Ph
nd
4
6
e
1a
CONHMe OCH
CONH OCH
>10µM
>10µM
nd
e
1
2
3-ClPh
nd
12

ACCEPTED MANUSCRIPT
2
3
donepezil
6.4±0.4 or
3365
13±1
6
4
1
1
6
5
tacrine
45.1±7
3.2±0.2
a
b
The 50% inhibitory concentration (means ± SD of three experiments) of AChE from human erythrocytes. The 50%
c
inhibitory concentration (means ± SD of three experiments) of BChE from equine serum. Selectivity Index = IC50
(
donepezil as control. Not determined. Y = 2,4,6-trichlorophenyl
d
eqBChE)/IC50 (hAChE). Propidium iodide displacement conducted on EeAChE by the tested compound and
e
f
2
.2.2. By tetralone-derived N-benzylpyridinium salts (series B2)
The AChE inhibitory activity of tetralone-derived N-benzylpyridinium salts follow the same
general trend as that observed in indanone series. Thus, targets 14a-q (EWG=Ac) bearing
methoxy electron donating groups on the tetralone moiety were found to be highly potent against
hAChE (IC = 10-71 nM), although targets 14r-u (EWG=Ac) furnished somewhat lower
5
0
inhibitory activities (IC = 110-568 nM), thus highlighting the marked influence exerted by the
5
0
N-benzyl appendage over the AChE inhibitory potency. It is worth mentioning that targets in
tetralone series displayed significantly higher inhibitory activity towards the isoenzyme eqBChE
(
IC50= 0.256-9.1 µM), compared to those obtained in indanone series with IC50 generally above
0 µM (Table 1). As might be anticipated, N-propyl pyridinium salt 14v showed no substantial
1
inhibitory activity against hAChE (IC50 > 10 µM). This result is in line with the putative dual
binding mode of the class of donepezil analogues in which the N-benzyl moiety forms a π-π
stacking interaction with the aromatic ring of the W86 residue of hAChE (see docking study).
More surprisingly, N-methylene-c-hexyl pyridinium salt 14w revealed a submicromolar IC50
value with both hAChE (IC50= 0.210 µM) and eqBChE (IC50= 0.341 µM). Various EWGs at C-3
of the pyridinium salt were also investigated in this tetralone series, including bromine, primary
and secondary amides as well as a trifluoromethyl group. Firstly, the bromine derivative 13b
displayed a good inhibitory effect against hAChE (IC =16.5 nM). Whereas the two primary
5
0
13

ACCEPTED MANUSCRIPT
amides 19h and 19i displayed noticeable inhibitory activities against hAChE (IC50= 0.893 and
0
.379 µM respectively), no significant activity was observed with the secondary amide 21b (IC50
10µM). It must also be mentioned that, trifluoromethyl derivative 20 disclosed an interesting
activity against hAChE (IC =0.250 µM). Last but not least, all tested AChE inhibitors in
>
5
0
indanone and tetralone series (Table 1, 2) were able to displace propidium (16-27%) and tuned
out to be slightly more efficient than the standard drug donepezil (13%).
Table 2. Inhibition of hAChE and eqBChE by tetralone-derived N-benzylpyridinium salts 13b,
1
4k-w; 19h-i, 20, 21b
IC50 (nM) ± SD
Propidium
iodide
2
1
c
compd
n
EWG
R
R
R
SI
a
b
displacement
hAChE
eqBChE
d
(%)
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
2
2
3b
4k
4l
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
Br
H
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3-MePh
16.5±3.0
21±2.8
665±30
nd
40
-
19±1
nd
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
3
3
3
3
3
3
3
3
3
3
3
3
3
OCH
OCH
H
3
3
3-ClPh
Ph
24±3.9
nd
-
24±1
20±1
22±2
21±1
23±1
24±1
27±2
25±2
15±1
19±2
4m
4n
4o
4p
4q
4r
4s
3-ClPh
2-ClPh
2-FPh
3-FPh
2-MePh
3-MePh
71.0±4.2
29.5±2.1
43.0±4.2
55.5±374.9
10.0±1.3
110.5±12
307.5±10.5
568±13
496±50
256±50
489±79
412±16
433±65
596±29
6.9
8.6
11
7.5
43.3
5.3
18
10.7
7.4
-
H
H
H
H
H
H
3-CF
Ph
2170±120
6090±50
3
4t
H
4-ClPh
4-MePh
n-ethyl
c-pentyl
2-MePh
2-ClPh
2-MePh
2-MePh
4u
4v
4w
9h
9i
H
556±9
4140±420
>10µM
e
H
>10µM
nd
H
210±0
341±14
1.6
10.2
6.9
9.2
-
14±2
e
CONH
2
2
H
893.8±109
378.8±30
242.4±4.2
>10µM
9115±238
2637.1±13.1
2239±113.9
2505±95
nd
e
CONH
H
nd
e
0
CF
3
H
nd
e
1b
CONHMe
H
nd
14

ACCEPTED MANUSCRIPT
64
23
donepezil
6.4±0.4 or 11
3365
-
13±1
65
tacrine
45.1±7
3.2±0.2
a
b
The 50% inhibitory concentration (means ± SD of three experiments) of AChE from human erythrocytes. The 50%
c
inhibitory concentration (means ± SD of three experiments) of BChE from equine serum. Selectivity Index = IC
(
donepezil as control. Not determined.
5
0
d
eqBChE)/IC50 (hAChE). Propidium iodide displacement conducted on EeAChE by the tested compound and
e
2
.2.3. By acetophenone-derived N-benzylpyridinium salts (series B3)
To complete this SAR map, we were also interested in assessing the hAChE inhibitory activity of
acyclic aryl ketone-derived N-benzylpyridinium salts. Given that the presence of methoxy groups
was found to be a structural requirement for high hAChE inhibitory potency in indanone series,
we made the decision, as we did in tetralone series, to restrict our screening to electron rich
aromatics; namely 3,4-dimethoxyacetophenone derivatives. A brief survey of table 3 revealed
that the nature of both the EWG at C-3 and the N-benzyl appendage has a striking effect on the
activity. Whereas acetyl derivative 14x (IC =0.143 µM) and ester derivative 16b (IC =0.171
5
0
50
µM) inhibit hAChE with sub-micromolar affinity, carboxylic acid 17b and both primary amides
9j and 33 displayed IC values > 10 µM. More interestingly, a two-digit nanomolar hAChE
1
5
0
inhibitory activity could be obtained with the more potent hAChE inhibitors 14a (IC50= 22.2 nM)
and 15e (IC =11 nM) characterized by the presence of a nitrile at C-3 along with a N-benzyl
5
0
moiety substituted at the meta-position either by a chlorine or fluorine atom.
Table 3. Inhibition of hAChE and eqBChE by acetophenone-derived N-benzylpyridinium salts
IC50 (nM) ± SD
Propidium
iodide
c
compd
n
0
EWG
R2
R1
R
SI
a
b
displacement
hAChE
eqBChE
d
(
%)
1
4x
COCH
3
OCH
3
OCH
3
3-ClPh
143.5±3
3198±75
22.2
nd
15

ACCEPTED MANUSCRIPT
e
e
e
e
1
1
1
1
1
1
1
3
5b
5c
5d
5e
6b
7b
9j
0
0
0
0
0
0
0
0
CN
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
3
3
3
3
3
3
3
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
3
3
3
3
3
3
3
3-ClPh
4-ClPh
22.2±0.2
>10µM
4242±130
191
nd
nd
nd
nd
nd
nd
nd
nd
CN
>10µM
-
-
-
-
-
-
-
e
CN
2-CF
Ph
275±11
nd
3
CN
3-FPh
11±1.2
>10µM
2035±55
>10µM
>10µM
>10µM
e
e
e
e
COOCH
COOH
3
3-Cl-Ph
3-Cl-Ph
3-ClPh
3-ClPh
171±7.1
>10µM
CONH
2
2
>10µM
3
CONH
3
3
>10µM
64
23
donepezil
6.4±0.4 or 11
3365
13±1
65
tacrine
45.1±7
3.2±0.2
a
b
The 50% inhibitory concentration (means ± SD of three experiments) of AChE from human erythrocytes. The 50%
c
inhibitory concentration (means ± SD of three experiments) of BChE from equine serum. Selectivity Index = IC
(
donepezil as control. Not determined.
5
0
d
eqBChE)/IC50 (hAChE). Propidium iodide displacement conducted on EeAChE by the tested compound and
e
2
.2.4. In vitro evaluation of selected “bio-oxidizable”prodrugs (series A1-A3)
As expected, a large majority of selected indanone-, tetralone-, and acetophenone-derived N-
benzyl 1,4-dihydropyridine prodrugs 22-24 were observed to be inactive against hAChE (IC50
>
10µM). Even if one can notice that some prodrugs 22a, 23a,d,j and 24a exhibited a three-digit
nanomolar activity (IC = 415-923 nM), their corresponding pyridinium salts revealed to be 10-
5
0
to 100-fold more potent (see Table 1-3). These findings therefore suggest that the interaction
between the positive charge in previously assessed pyridinium salts B and aromatic residues
within the active site gorge of AChE is largely responsible for the high hAChE inhibitory
activity observed. Futhermore, PAMPA-BBB assays were performed at 100µM in a buffer pH
7
.4 to evaluate the ability of some dihydropyridines to cross the BBB by passive diffusion. All
-
6
-1
tested prodrugs A1-3 displayed high permeability values with Pe >14x10 cm.s suggesting a
likely ability to reach the brain (see supplementary material).
16

ACCEPTED MANUSCRIPT
Table 4. Inhibition of hAChE and eqBChE by Selected Prodrugs 22-24
IC50
(nM)±SD
IC50
(nM)±SD
BBB
permeation
prediction
1
2
compd
n
EWG
R
R
Ar
c
a
b
hAChE
eqBChE
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2a
2b
2c
2d
0
0
0
0
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
CN
OCH
OCH
OCH
OCH
OCH
OCH
OCH
3
3
3
3
3
2
3
OCH
OCH
OCH
OCH
OCH
-
3
3
3
3
3
3-ClPh
3-ClPh
3-ClPh
3-ClPh
2-MePh
2-MePh
2-MePh
2-MePh
2-MePh
2-MePh
2-MePh
3-ClPh
3-ClPh
Ph
923±185
>10µM
>10µM
CNS+
d
CO
COCH
CONH
Me
>10µM
>10µM
>10µM
428±42
nd
2
3
>10µM
CNS+
CNS+
CNS+
CNS+
CNS+
2
>10µM
4
6
3a
3b
COCH
COCH
COCH
COCH
COCH
COCH
3
3
>10µM
4
6
O
>10µM
>10µM
653±39
>10µM
>10µM
>10µM
>10µM
>10µM
415±11
>10µM
>10µM
>10µM
>10µM
>10µM
>10µM
>10µM
>10µM
6.4±0.4
4615±65
5435±15
1255±55
4820±150
4
6
3c
3
3
3
H
4
6
d
3d
CH
CH
Cl
3
3
CH
H
3
nd
4
6
d
3e
nd
4
6
d
3f
3
H
>10µM
>10µM
nd
4
6
d
3g
3h
3i
CONH
CONH
2
2
OCH
OCH
F
3
3
OCH
OCH
H
3
nd
4
6
3
>10µM
CNS+
d
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
COCH
>10µM
nd
3
3
3
3
3
3
3
3
3
4a
4b
4c
4d
4e
4f
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
OCH
3
3
3
3
3
3
3
3
3
OCH
3
3970±230
6010±300
4400±160
5775±35
4020±90
CNS+
d
H
2-FPh
nd
d
H
3-FPh
nd
H
2-ClPh
3-MePh
CNS+
d
H
nd
d
H
3-CF
Ph
>10µM
>10µM
>10µM
>10µM
nd
3
d
4g
4h
4i
H
4-ClPh
4-MePh
2-MePh
nd
d
H
nd
CONH
H
CNS+
2
23
donepezil
3365
64
or 11
6
5
tacrine
45.1±7
3.2±0.2
a
b
The 50% inhibitory concentration (means ± SD of three experiments) of AChE from human erythrocytes. The
c
5
0% inhibitory concentration (means ± SD of three experiments) of BChE from equine serum. Classification of
CNS positive (CNS+) and CNS negative (CNS-) compounds using PAMPA-BBB assay (see supplementary
d
material). Not determined.
17

ACCEPTED MANUSCRIPT
2
.3. Docking
To gain insight into binding interactions of AChEI in the hydrolytic active site, we carried out
molecular docking studies of compounds 13a, 18, 19j and 41. These studies were performed into
a human AChE structure (PDB ED: 4EY7)[66] what protonation schema was checked up by
proPKA software (Glu202 protonate, His447 protonate at δ position…) and a water molecule
interacting with protonated piperidine ring of donepezil was conserved during the docking
(residue number 931). Both R- and S-enantiomers of each compounds were docked
independently in the active site using the GOLD program and applying ChemPLP scoring
function.
Both enantiomers of 13a took a similar position close to that of donepezil in the crystal structure
of hAChE (Figure 2A and 2B). The docking results suggest that, as donepezil, the indanone
moiety of 13a is situated in W286 proximity of the peripheric binding site (PAS) well positioned
for π-stacking. N-benzyl group stacks above W86 of anionic hole and the indanone carbonyl
group establishes an H-bond with Phe295 backbone NH. As observed in donepezil, the cyclic
nitrogen atom of 13a takes a position suitable for an electrostatic interaction with the water
molecule in the proximity of Tyr337 and Tyr341. According to the docking results, all key
interactions are preserved by 13a. Furthermore the bromide substituent is situated in the cavity in
HIS447 proximity. The bromide atom is situated at a 3.8Ǻ distance of Tyr 124 OH group and at
a 3.7Ǻ of backbone NH Gly 122.
+
The docking study showed that the compound 19j and 41 loos the NH interaction with water
9
31 as well as the hydrogen bond through of bicycle C = O (even if the distance was correct for
18

ACCEPTED MANUSCRIPT
an H bond for compound 19j the orientation of NH partner and C = O was not in linearity)
(
Figure 1C and D). These observations may explain the loos of the inhibition activity of these
two compounds observed experimentally.
Figure 2. Compound 13a (B), 19j (C) end 41 (D) positioned in AChE binding sites using the docking studies,
compared to the donepezil position from the X-ray structure (A). The compound and the selected side chains of the
19

ACCEPTED MANUSCRIPT
binding site residues are in stick and the protein in ribbon representation. This figure was made with PYMOL
(
DeLano Scientific, 2002, San Carlo, USA).
In the case of the bulkiest compound 18, the docking study proposed a wide variety of solutions.
Among them two poses seemed interesting to us (Figure 3), one with the compound 18
positioned high in the groove and so covering a large area of the peripheric binding site (Figure
3
B) and second for which the compound 18 occupied the locality close to donepezil (Figure 3A).
As the propidium displacement test has given for compound 18 the same value (Table 1) as that
for the donepezil the second position seems most likely.
Figure 3. Two proposed poses for compound 18 positioned in AChE binding sites using the docking studies. The
compound and the selected side chains of the binding site residues are in stick and the protein in ribbon
representation. This figure was made with PYMOL (DeLano Scientific, 2002, San Carlo, USA).
2
.4. Influence of the stereochemistry in prodrugs A on both hAChE inhibitory activity
and their conversion rate into the corresponding pyridinium salts B
Although both enantiomers of donepezil are active and were found to display similar
pharmacological profiles [67], we were interested in looking more closely at the influence of an
20

ACCEPTED MANUSCRIPT
additional stereogenic center at C-4 position of the pyridine ring in our “bio-oxidizable” prodrug
system. Indeed, this supplementary stereogenic center may possibly have an influence not only
on the undesired side
ultimately on the overall pharmacological profile. We decided to focus our study on compound
4a which is one of the few compounds among the synthesized prodrugs exhibiting such activity
hAChE inhibitory activity but also on the oxidation rate of the prodrug, and
2
against AChE. To this end, the four enantiomers of the prodrug 24a were separated by
supercritical fluid chromatography (SFC) affording four enantiopure fractions of 24a-A, -B, -C
1
and -D in 76% yield (Table 5). We could determine from the H NMR spectra and the circular
dichroism (CD) responses that 24a-A pairs with 24a-C, and 24a-B with 24a-D. As can be seen
from Table 5, each enantiomer exhibits a quite distinctive inhibitory activity, the more active
enantiomer 24a-C (IC = 173 nM) being at least 100-fold more potent than 24a-C (IC > 10
5
0
50
µM). This clearly points out the importance of controlling the absolute configuration at both
stereogenic centers in these donepezil-based prodrugs of AChE inhibitors.
Table 5. Inhibitory activity of the four enantiomers of prodrug 24a toward hAChE
IC50 (nM)±SD
IC50 (nM)±SD
Circular dichroism
response at 250 nm
compd
Mass (mg)
a
b
hAChE
eqBChE
2
2
2
2
4a
132
19
415
3970
c
d
4a-A
4a-B
4a-C
4a-D
+
-
36%
45%
20
286±15
173±3
1830±250
1670±250
29
+
-
2
33
924±51
9070±39310
a
b
The 50% inhibitory concentration (means ± SD of three experiments) of AChE from human erythrocytes. The
c
5
0% inhibitory concentration (means ± SD of three experiments) of BChE from equine serum. % inhibition of
d
hAChE at 10µM. % inhibition of eqBChE at 10µM.
21

ACCEPTED MANUSCRIPT
The in vivo oxidation rate of the prodrug is a key issue for the success of this approach; the
prodrug must be stable enough at the periphery to prevent any peripheral side effect, whilst once
in the brain, oxidation should take place to entail central cholinergic activation. To complete this
study, we also investigated the influence of the stereochemistry on the redox activation step from
the selected prodrug 23h which displays favorable solubility in aqueous media. The
diastereoselective separation of 23h was accomplished from preparative HPLC by means of a C-
1
8 column to furnish the two diastereoisomers 23h-A (IC50= 5.3
µM) and 23h-B (IC50= 26.9
µM) with a satisfactory chromatographic yield. The oxidation rate of both diastereoisomers was
assessed by incubation in a 2% riboflavin solution in PBS. Riboflavin is a known oxidant of 1,4-
dihydropyridines [68-71], making it a potential oxidant that might be involved in the enzymatic
activation step of the prodrug. Pleasingly, whereas both diastereoisomers 23h-A and 23h-B
proved to be stable for several hours in PBS, suggesting a good stability at the periphery, their
oxidation rate is significantly accelerated in the presence of riboflavin to give rise almost
exclusively the corresponding drug 19b (see supplementary materials). Furthermore, the
substantial difference observed in the oxidation rate between 23h-A (t = 24.6 min) and 23h-B
1
/2
(t1/2 = 17.7 min) suggests that the pharmacological profile may be affected by the stereochemistry
of the prodrugs.
Table 6. Half-life of prodrugs 23h-A and 23h-B (diastereoisomers) incubated in a 2% riboflavin
solution
IC50 (nM)±SD
IC50 (nM)±SD
compd t1/2 in 2% Riboflavine in PBS (min)
a
b
hAChE
eqBChE
c
d
2
2
3h
--
29%
6%
3h-A
24.6
5339±904
14295±120
22

ACCEPTED MANUSCRIPT
2
3h-B
17.7
26950±1725
19925±3019
a
b
The 50% inhibitory concentration (means ± SD of three experiments) of AChE from human erythrocytes. The
c
5
0% inhibitory concentration (means ± SD of three experiments) of BChE from equine serum. % inhibition of
d
hAChE at 10µM. % inhibition of eqBChE at 10µM.
3
. Conclusion
Current symptomatic treatments of mild to moderate AD involve the administration of AChE
inhibitors to alleviate the cholinergic deficit associated with the disease [72-76]. Although
effective in the early stages of the disease, these treatments are not curative and have severe side-
effects leading to treatment discontinuation at an advanced stage of the disease [77,78]. Although
the discovery of a curative treatment for AD remains the Holy Grail, the development of new
central AChE inhibitors without side effects would provide a major breakthrough in the
symptomatic treatments of the disease. To this end, as part of our research program devoted to
the design of donepezil-based “bio-oxidizable” prodrugs of AChE inhibitors [35-37, 46], a new
set of prodrug systems in indanone, tetralone and acetophenone series were synthesized and
evaluated in vitro as new potential central AChE inhibitors. Generally speaking, most of the
newly prepared aryl ketone-derived
inactive against AChE (IC50 > 10 M), in contrast to their corresponding aryl ketone-derived
benzylpyridinium salts B1-3 which showed in numerous cases high AChE inhibitory activity
B1: IC50 up to 0.36 nM; B2: IC50 up to 10.0 nM; B3: IC50 up to 11.0 nM). These findings
N-benzyl 1,4-dihydropyridine A1-3 were found to be
h
µ
N-
h
(
support the assumption made in our preliminary work [46] whereby the presence of a positive
charge at the nitrogen heterocycle atom in donepezil-like compounds is a necessary, albeit not
sufficient, condition for high
concept of donepezil-based “bio-oxidizable” prodrug of AChE inhibitors, initially developed by
us in indanone series [46], to tetralone- and acetophenone-derived -benzyl 1,4-dihydropyridines
hAChE inhibitory activity. Furthermore, these results generalize the
N
23

ACCEPTED MANUSCRIPT
A2-3. With this large panel of aryl ketone-derived
N
-benzyl 1,4-dihydropyridines in hands,
further in vitro studies are currently in progress to identify the first set of prodrug candidates
eligible for preliminary in vivo studies. In the course of this work, we also highlighted the strong
incidence of the stereogenic centers present in prodrugs A on both the anti-AChE activity and on
the oxidation rate of the prodrug. These findings prefigure a different pharmacological profile for
each enantiomer of the prodrug and, consequently, the need to develop diastereo- and
enantioselective syntheses of this class of compounds. Asymmetric synthesis of these
encouraging donepezil-based “bio-oxidizable” prodrugs of AChE inhibitors are currently under
investigation in our laboratory and will be reported in due course.
4
. Experimental section
4
.1. General
All commercial reagents were used without further purification. The solvents were dried with
appropriate desiccants and distilled prior to use or were obtained anhydrous from commercial
suppliers. Silica gel (60, 230–400 mesh or 70–230 mesh) was used for column chromatography.
Reactions were monitored by thin layer chromatography on silica gel precoated aluminum plates.
1
13
UV light at 254 nm or KMnO stains were used to visualize TLC plates. H, C NMR spectra
4
were recorded using a spectrometer operating at 300 and 75 MHz respectively. Abbreviations
used for peak multiplicities are s: singlet, d: doublet, t: triplet, q: quadruplet dd = doublet of
doublet, br = broad and m: multiplet. Coupling constants
given in ppm and calibrated with DMSO- 6
J
are in Hz and chemical shifts are
1
d
or CDCl (residual solvent signals). H NMR spectra
3
1
3
obtained in CDCl were referenced to 7.26 ppm. C NMR spectra obtained in CDCl were
3
3
referenced to 77.16 ppm and in DMSO- 6
d
were referenced to 39.52 ppm. Elemental analyses
were performed by the microanalysis service of the University of Rouen and were recorded with
24

ACCEPTED MANUSCRIPT
a Thermo Scientific FLASH 2000 analyzer. Compounds 3b, 3e-i, 4b, 4e-i, 6b, 7b, 8a, 9b, 9e-i,
1
0b, 14a-f, 15a, 16a, 19a-b, 21a and 23a-h were prepared according the procedure described in
J. Med. Chem. 2017, 60, 5909.[46]
4
.2. Chemistry
4
.2.1. (E)-3-(3-bromopyridin-4-yl)-1-(3,4-dimethoxyphenyl)prop-2-en-1-one (3a)
To a solution of 3,4-dimethoxyacetophenone 1a (7.21 g, 40.0 mmol) and 3-bromo-4-
pyridinecarboxaldehyde 2 (7.44 g, 40.0 mmol) in ethanol (160 mL), was added K CO (12.72 g,
2
3
1
20.0 mmol) in one portion and the resulting mixture was stirred at reflux for 28 hours. After
cooling at room temperature, the mixture was concentrated to dryness, diluted with
dichloromethane, washed twice with water and brine. The organic layer was then dried over
anhydrous MgSO , filtered and concentrated under vacuum. The crude solid was triturated with
4
diethyl ether, and the resulting solid was filtered and dried to afford the compound 3a as a yellow
1
solid. (Yield 8.96 g, 64%). H NMR (300 MHz, CDCl ,
δ
): 3.97 (s, 3H), 3.98 (s, 3H), 6.94 (d,
3
1
H, J = 8.4 Hz), 7.54-7.61 (m, 3H), 7.67 (dd, 1H, J = 8.4 Hz, J = 2.1 Hz), 7.95 (d, 1H, J = 15.6
1
3
Hz), 8.56 (d, 1H, J = 5.1 Hz), 8.80 (s, 1H). C NMR (75 MHz, CDCl
3
, δ): 56.1, 56.2, 110.0,
1
10.7, 121.5, 122.8, 123.6, 128.1, 130.4, 139.2, 142.5, 148.6, 149.4, 153.0, 153.8, 187.7. HRMS
+
+
(
ESI ): calcd for [M+H] C H NO Br m/z 348,0235; found 348.0250.
16 15 3
4
.2.2. (E)-2-((3-bromopyridin-4-yl)methylene)-6-methoxy-3,4-dihydronaphthalen-1(2H)-one (3c)
25

ACCEPTED MANUSCRIPT
At 35°C, to a solution of NaOH (0.39 g, 9.7 mmol) in EtOH (12 mL) were added in one portion
a mixture of 6,7-dimethoxy-1-tetralone 1c (1.0 g, 4.85 mmol) and 3-bromo-4-
pyridinecarboxaldehyde 2 (0.9 g, 4.85 mmol). The reaction mixture was then stirred for 4 hours
at the same temperature. The resulting precipitate was filtered off then rinsed twice with cold
1
diethyl ether to afford the title compound 3c as a pale yellow solid. (Yield 1.80 g, 99%). H
NMR (300 MHz, CDCl
3
,
δ
): 2.89 (br s, 4H); 3.92 (s, 6H), 6.65 (s, 1H), 7.17 (d, 1H,
J
= 5.1 Hz),
): 27.7, 28.6,
6.1, 56.2, 109.6, 110.0, 122.6, 124.5, 126.1, 131.6, 139.5, 144.5, 148.0, 148.5, 152.3, 154.1,
1
3
7
5
1
.60 (s, 2H), 8.51 (d, 1H, J = 5.1 Hz), 8.75 (s, 1H). C NMR (75 MHz, CDCl , δ
3
+
+
3
85.7. HRMS (ESI ): calcd for [M+H] C18H17NO Br m/z 374,0392; found 374.0392.
4
.2.3. (E)-2-((3-bromopyridin-4-yl)methylene)-6,7-dimethoxy-3,4-dihydronaphthalen-1(2H)-one
(3d)
The title compound 3d was prepared according to the procedure for the synthesis of 3c with 6-
methoxy-1-tetralone 1d (3.0 g, 17.02 mmol) and 3-bromo-4-pyridinecarboxaldehyde 2 (3.17 g,
1
1
7.02 mmol) as reactants and obtained as an yellow solid. (Yield 3.35 g, 57%). H NMR (300
MHz, CDCl3,
δ
): 2.89 (br s, 4H); 3.84 (s, 3H), 6.67 (s, 1H), 7.17 (dd, 1H,
= 4.8 Hz), 7.61 (s, 1H), 8.09 (d, 1H, = 8.7 Hz), 8.51 (d, 1H,
.74 (s, 1H). C NMR (75 MHz, CDCl ): 27.4, 29.2, 55.9, 112.4, 113.7, 122.6, 124.5, 126.4,
31.0, 131.6, 139.6, 144.4, 145.9, 147.9, 152.3, 164.0, 185.6. HRMS (ESI ): calcd for [M+H]
J
= 8.7 Hz,
J = 1.8
Hz), 7.17 (d, 1H,
J
J
J
= 4.8 Hz),
1
3
8
1
3
, δ
+
+
C H NO Br m/z 344,0286; found 344.0291.
17
15
2
4
.2.4. General procedure A for the synthesis of compounds 3j-m.
26

ACCEPTED MANUSCRIPT
4
.2.4.1. (2E)-2-[(3-Bromo-4-pyridyl)methylene]-5-fluoro-indan-1-one (3j)
To a solution of 5-fluoro-1-indanone 1j (750.7 mg, 5.0 mmol) and 3-bromo-4-
pyridinecarboxaldehyde 2 (0.93 g, 5.0 mmol) in toluene (50 mL) was added -toluenesulfonic
p
acid (1.42 g, 7.5 mmol). After heated at reflux using a Dean-Stark apparatus for 4 hours, the
mixture was cooled to room temperature and the solvent was removed in vacuum, then 5%
sodium bicarbonate aqueous solution was added until reach pH = 8. After extraction with
dichloromethane (4x), the organic layer was dried over magnesium sulfate and concentrated to
dryness. The residue was taken up with EtOAc and the solid was filtered, rinsed with EtOAc to
1
afford compound 3j as a grey solid (1.59 g, 53%). H NMR (300 MHz, CDCl ,
δ
): 3.96 (s, 2H),
= 2.1 Hz), 7.95 (dd, 1H, = 8.4,
= 5.1 Hz), 8.82 (s, 1H). C NMR (75 MHz, CDCl , ): 31.6, 113.0,
13.3, 116.4, 116.7, 123.5, 127.3, 129.6, 134.1, 140.1, 142.7, 148.4, 152.2, 152.3, 153.0, 165.7,
3
7
5
1
1
.12-7.20 (m, 2H), 7.48 (d, 1H,
.4 Hz), 8.60 (d, 1H,
J
= 5.1 Hz), 7.79 (t, 1H,
J
J
J =
1
3
J
δ
3
+
+
69.1, 191.4. HRMS (ESI ): calcd for [M+H] C H NOBrF m/z 317.9924; found 317.9923.
1
5
10
4
.2.4.2. (2E)-2-[(3-bromo-4-pyridyl)methylene]-5-hydroxy-indan-1-one (3k)
The title compound 3k was prepared according to the general procedure A with compound 1k
(296.3 mg, 2.0 mmol) and compound 2 (372.1 mg, 2.0 mmol) as reactants and obtained as a
1
white solid. (Yield 140.1 mg, 22%). H NMR (300 MHz, DMSO-d ,
δ
): 4.02 (s, 2H), 6.86 (dd,
= 8.4 Hz), 7.83 (d, 1H, = 5.1
= 5.1 Hz), 8.84 (s, 1H). C NMR (75 MHz, DMSO-d , ): 31.0, 111.9, 116.6,
6
1
H,
J
= 8.4,
J
= 1.8 Hz), 6.92 (s, 2H), 7.48 (s, 1H), 7.68 (d, 1H,
J
J
1
3
Hz), 8.63 (d, 1H,
J
δ
6
27

ACCEPTED MANUSCRIPT
+
1
22.9, 124.1, 125.9, 126.3, 128.7, 141.9, 142.3, 148.8, 152.1, 153.4, 164.7, 190.4. HRMS (ESI ):
+
calcd for [M+H] C H NO Br m/z 315.9973; found 315.9973.
1
5
11
2
4
.2.4.3. (2E)-2-[(3-bromo-4-pyridyl)methylene]-5-(2,2,2-trifluoroethoxy)indan-1-one
3l
(
)
The title compound 3l was prepared according to the general procedure A with compound 1l
(200.0 mg, 0.87 mmol) and compound 2 (161.2 mg, 0.87 mmol) as reactants and obtained as a
1
white solid. (Yield 340.1 mg, 98%). H NMR (300 MHz, CDCl ,
δ
): 3.91 (s, 2H), 4.43 (q, 2H,
J
3
=
7.8 Hz), 6.86 (dd, 1H,
J
= 8.4,
= 9.0 Hz), 8.57 (d, 1H,
m/z 398.0004; found 398.0003.
J
= 1.8 Hz), 6.92 (s, 2H), 7.48 (s, 1H), 7.46 (d, 1H,
J
= 5.1 Hz),
+
7
.74 (s, 1H), 7.90 (d, 1H,
J
J
= 4.8 Hz), 8.80 (s, 1H). HRMS (ESI ): calcd
+
for [M+H] C17
H12NO
2
BrF
3
4
.2.4.4. (2E)-2-[(3-bromo-4-pyridyl)methylene]-6-hydroxy-5-methoxy-indan-1-one (3m)
The title compound 3m was prepared according to the general procedure A with compound 1m
(3.22 g, 18.07 mmol) and compound 2 (3.36 g, 18.07 mmol) as reactants and obtained as a white
1
solid. (Yield 4.30 g, 68%). H NMR (300 MHz, , DMSO-d
6
,
δ
): 3.89 (s, 3H), 3.95 (s, 2H), 7.13
= 5.1 Hz), 8.83 (s, 1H), 9.77 (s, 1H).
): 30.7, 56.0, 108.2, 108.2, 122.9, 124.1, 125.8, 129.7, 142.0,
(
s, 2H), 7.46 (s, 1H), 7.81 (d, 1H,
J
= 5.1 Hz), 8.63 (d, 1H,
J
1
3
C NMR (75 MHz, , DMSO-d6,
δ
+
+
1
42.4, 144.1, 147.3, 148.8, 152.1, 155.4, 191.0. HRMS (ESI ): calcd for [M+H] C16
3
H13NO Br
m/z 346.0079; found 346.0077.
28