N-alkylpiperidine carbamates as potential anti-Alzheimer's agents

Article abstract of DOI:10.1016/j.ejmech.2020.112282

Compounds capable of interacting with single or multiple targets involved in Alzheimer's disease (AD) pathogenesis are potential anti-Alzheimer's agents. In our aim to develop new anti-Alzheimer's agents, a series of 36 new N-alkylpiperidine carbamates was designed, synthesized and evaluated for the inhibition of cholinesterases [acetylcholinesterase (AChE) and butyrylcholinesterase (BChE)] and monoamine oxidases [monoamine oxidase A (MAO-A and monoamine oxidase B (MAO-B)]. Four compounds are very promising: multiple AChE (IC₅₀ = 7.31 μM), BChE (IC₅₀ = 0.56 μM) and MAO-B (IC₅₀ = 26.1 μM) inhibitor 10, dual AChE (IC₅₀ = 2.25 μM) and BChE (IC₅₀ = 0.81 μM) inhibitor 22, selective BChE (IC₅₀ = 0.06 μM) inhibitor 13, and selective MAO-B (IC₅₀ = 0.18 μM) inhibitor 16. Results of enzyme kinetics experiments showed that despite the carbamate group in the structure, compounds 10, 13, and 22 are reversible and non-time-dependent inhibitors of AChE and/or BChE. The resolved crystal structure of the complex of BChE with compound 13 confirmed the non-covalent mechanism of inhibition. Additionally, N-propargylpiperidine 16 is an irreversible and time-dependent inhibitor of MAO-B, while N-benzylpiperidine 10 is reversible. Additionally, compounds 10, 13, 16, and 22 should be able to cross the blood-brain barrier and are not cytotoxic to human neuronal-like SH-SY5Y and liver HepG2 cells. Finally, compounds 10 and 16 also prevent amyloid β_1–42 (Aβ_1–42)-induced neuronal cell death. The neuroprotective effects of compound 16 could be the result of its Aβ_1–42 anti-aggregation effects.

Full text of DOI:10.1016/j.ejmech.2020.112282

Journal Pre-proof
N-alkylpiperidine carbamates as potential anti-Alzheimer's agents
Urban Košak, Nika Strašek, Damijan Knez, Marko Jukič, Simon Žakelj, Abida
Zahirović, Anja Pišlar, Xavier Brazzolotto, Florian Nachon, Janko Kos, Stanislav
Gobec
PII:
S0223-5234(20)30251-8
DOI:
https://doi.org/10.1016/j.ejmech.2020.112282
EJMECH 112282
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 27 February 2020
Revised Date: 26 March 2020
Accepted Date: 26 March 2020
Please cite this article as: U. Košak, N. Strašek, D. Knez, M. Jukič, S. Žakelj, A. Zahirović, A.
Pišlar, X. Brazzolotto, F. Nachon, J. Kos, S. Gobec, N-alkylpiperidine carbamates as potential anti-
Alzheimer's agents, European Journal of Medicinal Chemistry (2020), doi: https://doi.org/10.1016/
j.ejmech.2020.112282.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. 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.
©
2020 Published by Elsevier Masson SAS.

N-alkylpiperidine carbamates as potential anti-
Alzheimer’s agents
‡
‡
‡
‡
‡
‡
Urban Košak, Nika Strašek, Damijan Knez, Marko Jukič, Simon Žakelj, Abida Zahirović,
‡
ø
ø
‡
‡,
Anja Pišlar, Xavier Brazzolotto, Florian Nachon, Janko Kos, Stanislav Gobec *
‡
University of Ljubljana, Faculty of Pharmacy, Aškerčeva 7, 1000 Ljubljana, Slovenia
ø
Institut de Recherche Biomédicale des Armées, 91223 Brétigny-sur-Orge, France
*
Corresponding Author: Stanislav Gobec
University of Ljubljana, Faculty of Pharmacy, Aškerčeva 7, 1000 Ljubljana, Slovenia
Tel: +386-1-4769500; Fax: +386-1-4258031; E-mail: stanislav.gobec@ffa.uni-lj.si
1

Abstract
Compounds capable of interacting with single or multiple targets involved in Alzheimer’s
disease (AD) pathogenesis are potential anti-Alzheimer’s agents. In our aim to develop new
anti-Alzheimer’s agents, a series of 36 new N-alkylpiperidine carbamates was designed,
synthesized and evaluated for the inhibition of cholinesterases [acetylcholinesterase (AChE)
and butyrylcholinesterase (BChE)] and monoamine oxidases [monoamine oxidase A (MAO-
A and monoamine oxidase B (MAO-B)]. Four compounds are very promising: multiple
AChE (IC = 7.31 µM), BChE (IC = 0.56 µM) and MAO-B (IC = 26.1 µM) inhibitor 10,
5
0
50
50
dual AChE (IC = 2.25 µM) and BChE (IC = 0.81µM) inhibitor 22, selective BChE (IC =
5
0
50
50
0
.06 µM) inhibitor 13, and selective MAO-B (IC50 = 0.18 µM) inhibitor 16. Results of
enzyme kinetics experiments showed that despite the carbamate group in the structure,
compounds 10, 13, and 22 are reversible and non-time-dependent inhibitors of AChE and/ or
BChE. The resolved crystal structure of the complex of BChE with compound 13 confirmed
the non-covalent mechanism of inhibition. Additionally, N-propargylpiperidine 16 is an
irreversible and time-dependent inhibitor of MAO-B, while N-benzylpiperidine 10 is
reversible. Additionally, compounds 10, 13, 16, and 22 should be able to cross the blood-brain
barrier and are not cytotoxic to human neuronal-like SH-SY5Y and liver HepG2 cells.
Finally, compounds 10 and 16 also prevent amyloid β1–42 (Aβ1–42)-induced neuronal cell
death. The neuroprotective effects of compound 16 could be the result of its Aβ1–42 anti-
aggregation effects.
Keywords
Alzheimer’s disease; Acetylcholinesterase; Butyrylcholinesterase; Monoamine oxidase;
Multi-target-directed ligands; N-alkylpiperidine carbamates.
2

1
. Introduction
Alzheimer’s disease (AD) is a neurodegenerative brain disease with amyloid β
1
2
3
peptide (Aβ) deposits (amyloid plaques), and products of oxidative stress contributing to the
pathogenesis of the disease. Amyloid plaques are mainly composed of a 42-amino-acid-long
4
Aβ isoform (Aβ1–42) , which forms cytotoxic structures, ranging from monomers to oligomers
2
,5
(
i.e., 2–6 monomers) and fibers . Several small molecules [e.g. elenbecestat (E2609),
umibecestat (CNP520)] and macromolecules (e.g. gantenerumab, CAD106, UB-311)
targeting the production and clearance of Aβ monomers, oligomers and fibres are currently
6
,7
evaluated in clinical trials for the treatment of AD .
One of the several proteins that contribute to oxidative stress in AD is monoamine
8
oxidase (MAO) , a flavin adenine dinucleotide (FAD)-containing enzyme located in the outer
9
mitochondrial membrane that catalyses the oxidative deamination of monoamines . MAO-A
8
and MAO-B are the two isoenzymes present in most mammalian tissues . The enzymatic
8
,9
activity of MAO-B is increased in the brains of patients with AD . This results in increased
levels of oxidative deamination reaction products such as hydrogen peroxide, a source of
8
–10
8
,
hydroxyl radicals, aldehydes and ammonia . These products contribute to oxidative stress
2
which enhances the neurodegeneration and synaptic dysfunction in AD . For example,
selective MAO-B inhibitor rasagiline (Figure 1) is currently evaluated in a phase II clinical
7
trial for the treatment of AD .
The neurodegeneration and synaptic dysfunction in AD severely affect the cholinergic
1
1
12
system and result in a decrease in the levels of the neurotransmitter acetylcholine (ACh) ,
1
3
which then produces memory loss and cognitive impairment , distinctive for patients with
AD. Acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) are the two
cholinesterases (ChEs) that terminate cholinergic neurotransmission by catalysing the
1
4
hydrolysis of ACh . Inhibition of ACh hydrolysis in the brain is used to increase levels of
3

1
5
ACh and consequently restore cognitive functions and alleviate symptoms of AD .
Accordingly, three out of the four approved drugs for alleviating symptoms of AD are ChE
16
17
18
inhibitors : selective AChE inhibitors donepezil and galantamine , and dual ChE inhibitor
1
9
rivastigmine (Figure 1).
Considering the intricate nature of AD pathogenesis, the multitarget-directed ligand
MTDL) approach has also been used to develop compounds capable of interacting with more
(
2
0–22
than one target involved in the pathogenesis of the disease
compounds ladostigil (inhibition of AChE, BChE, MAO-A, and MAO-B) and AD-35
. For example, multitarget
2
3
2
4
(
inhibition of AChE and metal-induced Aβ aggregation) (Figure 1) are currently evaluated
2
5,7
in a phase II clinical trial
.
Figure 1. Structures of the currently approved ChE inhibitors for the treatment of Alzheimer’s disease
donepezil, galantamine, and rivastigmine), and structures of some of the compounds currently
evaluated in clinical trials for the treatment of AD (rasagiline, ladostigil and AD-35).
(
4

Among all therapeutic areas, AD has one of the lowest numbers of compounds
progressing to regulatory review. No new drugs have been approved since 2003, and the
overall success rate during the 2002 to 2012 period was only 0.4%. This very high attrition
rate (99.6%) requires a constant supply of new compounds that can be evaluated for efficacy
2
6
in AD . In the current study, our hypothesis was that novel anti-AD compounds, capable of
interacting with one or more targets in AD pathogenesis, can be developed simply by
combining moieties of known anti-AD compounds. We report herein design, straightforward
two-step synthesis, and comprehensive biochemical evaluation of a series of N-
alkylpiperidine carbamates as potential anti-Alzheimer agents.
2
2
. Results and discussion
.1. Design
N-Alkylpiperidine carbamates were designed by fusing moieties of known ChE and
MAO-B inhibitors (Figure 2). The 1,3- and 1,4-disubstituted piperidine scaffold, the hallmark
2
7
28
of the series, was taken from selective BChE inhibitors with neuroprotective activity 1 , 2 ,
2
9
17
and 3 , and selective AChE inhibitor donepezil . Compound 1 is a hit compound discovered
using a hierarchical structure-based virtual screening protocol to identify novel BChE
27
inhibitors . Compounds 2 and 3 are lead compounds developed in a hit-to-lead optimization
2
8,29
campaigns
using compound 1 as the starting point. The N-benzyl group was taken from
2
8
17
sulfonamide 2 and donepezil , N-(2,3-dihydro-1H-inden-2-yl) group was appropriated from
2
7
29
amides 1 and 3 . Rasagiline is a selective MAO-B inhibitor with neuroprotective activity,
which inhibits MAO-B by forming a covalent bond between N-propargyl group and the N5
3
0
atom of FAD cofactor . The N-propargyl group of rasagiline was thus introduced to the
piperidine nitrogen to design compounds capable of inhibiting ChEs and MAO-B. Dual ChE
5

1
9
31
inhibitor rivastigmine , selective AChE inhibitor phenserine and selective BChE inhibitor
3
2
bisnorcymserine are pseudo-irreversible ChE inhibitors, which inhibit their target enzymes
by carbamoylation of the catalytic serine in the active site of ChEs. The active site is then
3
3
slowly regenerated by decarbamylation . The carbamate moieties of these three compounds
were thus introduced to the 3- or 4- position of piperidine ring directly (n = 0, Figure 2), or
via a methylene group (n = 1, Figure 2).
Figure 2. Design of N-alkylpiperidine carbamates.
2
.2. Synthesis
The synthesis of N-alkylpiperidine carbamates 4–39 is presented in Table 1. First, the
nitrogen of commercially available 4-hydroxypiperdine (40), 3-hydroxypiperidine (41), 4-
6

(hydroxymethyl)piperidine (42), and 3-(hydroxymethyl)piperidine (43) was alkylated with
propargyl bromide or benzyl bromide in the presence of Cs CO in acetone at room
2
3
3
4
temperature , to produce N-propargylpiperidine alcohols 44–47 and N-benzylpiperidine
alcohols 48–51 in moderate to high yields (63–88%). N-(2,3-dihydro-1H-inden-2-yl)
piperidine alcohols 52–55 were prepared in modest yields (21–33%) by alkylating the
nitrogen of starting piperidines 40–43 with 2-bromo-2,3-dihydro-1H-indene (56) in the
presence of K CO and a catalytic amount of KI in MeCN at reflux. Bromide 56 was prepared
2
3
from 2,3-dihydro-1H-inden-2-ol (57) and bromine in the presence of triphenylphosphine in
3
5
CH Cl in 83% yield.
2
2
The designed phenyl carbamates 4–15 and 4-isopropylphenyl carbamates 16–27 were
synthesized in high yields (85–96%) by treating N-alkyl alcohols 44–55 with phenyl
isocyanate or 4-isopropylphenyl isocyanate in the presence of a catalytic amount of N,N-
3
6
dimethylpyridin-4-amine (4-DMAP) in CH Cl at room temperature . Ethyl(methyl)
2
2
carbamates 28–39 were prepared in modest to moderate yields (28–54%) by treating N-
alkylpiperidine alcohols 44–55 with N-ethyl-N-methylcarbamoyl chloride in pyridine at
3
7
reflux .
Table 1. Synthesis of designed compounds. Reagents and conditions: (a) propargyl bromide (80 wt%
solution in PhMe) or benzyl bromide, Cs CO , acetone, rt, 24 h (82–86%); (b) 2-bromo-2,3-dihydro-
2
3
1
H-indene, K CO , cat. KI, MeCN, reflux (85 °C), 24 h (21–33%); (c) PhNCO or 4-iPrPhNCO, cat. 4-
2 3
DMAP, CH Cl , rt, 24 h (85–96%); (d) N-ethyl-N-methylcarbamoyl chloride, pyridine, reflux (120
2
2
°C), 24 h (28–54%).
7

a
b
1
2
a
b
1
2
Compds Pdp
R
n R
R
Compds Pdp
R
n R
R
4
5
1,4
1,3
0 H
0 H
30
31
1,4
1,3
1 Me Et
1 Me Et
6
7
8
9
1,4
1,3
1,4
1,3
1 H
1 H
0 H
0 H
32
33
34
35
1,4
1,3
1,4
1,3
0 Me Et
0 Me Et
1 Me Et
1 Me Et
1
0
1
2
3
4
5
1,4
1,3
1,4
1,3
1,4
1,3
1 H
1 H
0 H
0 H
1 H
1 H
36
37
38
39
40
41
1,4
1,3
1,4
1,3
1,4
1,3
0 Me Et
0 Me Et
1 Me Et
1 Me Et
1
1
1
1
1
/
/
0
0
/
/
/
/
8

1
1
1
1
2
2
2
2
6
7
8
9
0
1
2
3
1,4
1,3
1,4
1,3
1,4
1,3
1,4
1,3
0 H
0 H
1 H
1 H
0 H
0 H
1 H
1 H
42
43
44
45
46
47
48
49
1,4
1,3
1,4
1,3
1,4
1,3
1,4
1,3
/
/
1
1
0
0
1
1
0
0
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
2
2
4
5
1,4
1,3
0 H
50
51
1,4
1,3
1
1
/
/
/
/
0 H
2
6
1,4
1 H
52
1,4
0
/
/
9

2
2
2
7
8
9
1,3
1,4
1,3
1 H
53
54
55
1,3
1,4
1,3
0
1
1
/
/
/
/
/
/
0 Me
Et
Et
0 Me
a
Compds = compound number
b
Pdp = piperidine disubstitution pattern
2
.3. In vitro enzyme inhibition
The inhibitory potencies against human (h)BChE and hAChE of all synthesized N-
3
8
alkyl piperidines were determined using the method of Ellman , and the inhibitory potencies
against human (h)MAO-A and hMAO-B were determined using a previously described
3
9
fluorescence-based Amplex Red assay with minor modifications. Table 2 and Table 3 report
the structures and inhibitory potencies of 1,4-disubstituted N-alkylpiperidines carbamates and
1
,3-disbstituted N-alkylpiperidines carbamates, respectively. Compounds with IC50 values
higher than 100 µM were considered inactive (no inhibition). Since N-alkylpiperidine
alcohols 44–55 did not inhibit (IC > 100 µM) any of the four enzymes, they have not been
5
0
included in Table 1. Since none of the compounds inhibited hMAO-A (IC > 100 µM), these
5
0
values have not been included in Table 1. Rasagiline and rivastigmine were used as controls
for hMAO and hChE inhibition, respectively.
1
0

Table 2. Inhibitory potencies and structures of 1,4-disubstituted N-alkylpiperidine carbamates.
b
b
c
hAChE
hBChE
hMAO-B
a
1
2
Compds
R
n
R
R
IC ±SEM IC ±SEM IC ±SEM
50
50
50
(
µM)
(µM)
(µM)
4
0
H
H
>100
>100
19.06 ±3.67
>100
6
8
1
0
14.38 ±1.11 10.3 ±0.8
H
H
>100
6.55 ±0.79
>100
1
1
0
2
1
0
7.31 ±1.4
0.555 ±0.097 26.1 ±3.9
H
H
19.2 ±1.3
>100
18.1 ±1.5
>100
>100
1
4
1
1.433 ±0.306
0
.178
1
1
2
6
8
0
0
1
0
H
H
H
>100
>100
>100
>100
±
0.0093
0
.316
24.3 ±0.3
1.74 ±0.55
±
0.0293
>100
1
1

2
2
2
4
1
0
H
H
2.248 ±0.360 0.809 ±0.372
>100
>100
2
4
.308 ±1.597
.083 ±0.441
>100
>100
2
6
1
H
>100
2
3
8
0
0
1
Me
Me
Et
Et
>100
>100
11.4 ±2.6
>100
>100
>100
3
3
2
4
0
1
Me
Me
Et
Et
>100
>100
>100
>100
72.0 ±30.9
46.0 ±12.7
4
0.8 ±11.2
3
3
6
8
0
1
Me
Me
Et
Et
>100
>100
>100
4.46 ±1.46
>100
0.036
e
Rasagiline
/
/
/
/
/
/
/
>100
>100
±
0.004
Rivastigmine
/
19.8 ±2.6
1.439 ±0.228
>100
a
Compds = compound number
b
determined after 30 min preincubation of inhibitor with ChEs; data are expressed as mean ± SEM (n
=
3)
c
determined after 15 min preincubation of inhibitor with MAO-A/B; data are expressed as mean ±
SEM (n = 3).
d
c
=
IC (hMAO-A) = 29.5 ±4.4 µM .
50
1
2

Table 3. Inhibitory potencies and structures of 1,3-disubstituted N-alkylpiperidine carbamates.
b
b
c
hAChE
hBChE
hMAO-B
a
1
2
Compds
R
n
R
R
IC ±SEM IC ±SEM IC ±SEM
50
50
50
(
µM)
(µM)
(µM)
5
0
H
H
>100
>100
84.5 ±19.8
29.6 ±4.7
>100
>100
7
9
1
0
H
H
>100
0.46 ±0.049 33.3 ±7.5
1
1
1
3
1
0
67.1 ±13.7 0.520 ±0.078 39.5 ±8.2
0
.0645
H
H
>100
>100
>100
±
0.0115
0
.1795
1
5
1
>100
±
0.0233
1
1
2
2
7
9
1
3
0
1
0
1
H
H
H
H
>100
>100
>100
>100
71.83 ±18.7
2
.857
17.75 ±3.7
±
0.267
3.25 ±0.94 29.3 ±3.4
15.1 ±5.8 0.118 ±0.015
>100
1
3

7
.914 ±1.227
2
2
5
7
0
1
H
H
>100
>100
>100
1
8.7 ±3.0
0.442
0.1752
±
2
3
9
1
0
1
Me
Me
Et
Et
>100
>100
>100
>100
>100
>100
3
3
3
5
0
1
Me
Me
Et
Et
>100
21.0 ±6.8
>100
>100
96.6 ±21.3
>100
7.23 ±0.75
3
3
7
9
0
1
Me
Me
Et
Et
9.69 ±0.51
>100
>100
>100
9.39 ±2.20
>100
>100
0.036
d
Rasagiline
/
/
/
/
/
/
/
/
±
0.004
Rivastigmine
19.8 ±2.6 1.439 ±0.228
>100
a
Compds = compound number
b
determined after 30 min preincubation of inhibitor with ChEs; data are expressed as mean ± SEM (n
=
3)
c
determined after 15 min preincubation of inhibitor with MAO-A/B; data are expressed as mean ±
SEM (n = 3).
d
c
=
IC (hMAO-A) = 29.5 ±4.4 µM .
50
2
.4. Structure–activity relationships
1
4

2
.4.1. Structure–activity relationships of 1,4-disubstituted N-alkylpiperidine carbamates
Structure–activity relationships (SARs) of 1,4-disubstituted N-alkylpiperidine
carbamates is presented in Figure 3. N-Benzylpiperidine 22 (IC = 2.25 µM) and N-(2,3-
5
0
dihydro-1H-inden-2-yl)piperidine 24 (IC = 2.31 µM) are the most potent hAChE inhibitors
5
0
in this series. The iPr- group on the benzene ring of the carbamate group is needed for hAChE
inhibition, as removing it from compounds 22 and 24 reduces the inhibitory potencies (10,
IC = 7.31 µM; 12, IC = 19.2 µM). N-Propargylpiperidines and (ethyl)methylcarbamates,
5
0
50
with the exception of the very weak hAChE inhibitor 34 (IC = 72.0 µM), are inactive as
5
0
hAChE inhibitors (IC > 100 µM).
5
0
N-Benzylpiperidine 10 (IC = 0.56 µM) is the most potent hBChE inhibitor of the
5
0
series. Replacing the benzyl group with a 2,3-dihydro-1H-inden-2-yl group in 14 (IC = 1.43
5
0
µM) or a propargyl group in compound 6 (IC = 14.38 µM) reduces the inhibitory potencies
5
0
by up to 25-fold. In contrast to hAChE inhibition, the presence of the iPr- group on the
benzene ring of the carbamate group reduces hBChE inhibitory potency. Carbamate 20 (IC50
=
1.74 µM) is the exception to this rule, since it is almost 4-fold more potent than the benzene
counterpart 8 (IC50 = 6.55 µM). 1,4-Disubstituted N-alkylpiperidine carbamates with a
methylene linker between the piperidine ring and the carbamate moiety (n = 1, Figure 3) are
more potent hBChE inhibitors than compounds with the carbamate group bound directly to
the piperidine ring (n = 0, Figure 3). Again, there is an exception to this rule: compound 28
(
IC = 11.40 µM), which has an inactive “methylene” counterpart 30 (IC > 100 µM).
5
0
50
N-Propargylpiperidine 16 (IC = 0.18 µM) is the most potent hMAO-B inhibitor of
5
0
this series. Inserting a methylene group between the piperidine ring and the carbamate moiety
n = 1, Figure 3) does not affect the inhibition to a significant extent (18, IC = 0.32 µM).
(
5
0
Similar to hAChE inhibition, the iPr- group on the benzene ring of the carbamate group is
needed for hMAO-B inhibition. Removing this group from compound 18 reduces the
1
5

inhibitory potency by more than 30-fold (6; IC = 10.3 µM). N-Benzylpiperidine 10 (IC =
5
0
50
2
6.1 µM) is the only to inhibit hMAO-B inhibitor devoid of archetypal N-propargylpiperidine
moiety. Nonetheless, the inhibition of hMAO-B is comparable to the propargyl counterpart 6
IC = 10.3 µM). All N-(2,3-dihydro-1H-inden-2-yl)piperidines and (ethyl)methylcarbamates
(
5
0
are inactive as hMAO-B inhibitors (IC > 100 µM).
5
0
Figure 3. SARs of 1,4-disubstituted N-alkylpiperidine carbamates.
2
.4.2 The structure–activity relationships of 1,3-disubstituted N-alkylpiperidine
carbamates
Structure–activity relationships (SARs) of 1,3-disubstituted N-alkylpiperidine
carbamates is presented in Figure 4. Compound 25 (IC50 = 7.91 µM) is the most potent
1
6

hAChE inhibitor of the series. Inserting a methylene group between the piperidine ring and
the carbamate moiety (n = 1, Figure 4) slightly reduces the inhibitory potency (27, IC =
5
0
1
8.70 µM). Similar to 1,4-disubstituted piperidines, the iPr- group on the benzene ring of the
carbamate group is again needed for hAChE inhibition. Removing this group from compound
2
3 (IC = 15.10 µM) reduces the inhibitory potency (11, IC = 67.10 µM). As observed with
50 50
1
,4-disubstituted piperidines, N-propargylpiperidines and (ethyl)methylcarbamates, with the
exception of the very weak hAChE inhibitor 35 (IC = 96.6 µM), are inactive as hAChE
5
0
inhibitors (IC > 100 µM).
5
0
N-(2,3-Dihydro-1H-inden-2-yl)piperidine 13 (IC50 = 0.06 µM) is the most potent
hBChE inhibitor of this series. Inserting a methylene group between the piperidine ring and
the carbamate moiety (n = 1, Figure 4) increases the inhibitory potency of all 1,3-disubstituted
N-alkylpiperidine carbamates, except compound 15 (IC = 0.18 µM), which is less potent
5
0
than its “non-methylene” counterpart 13 (IC = 0.06 µM). As observed with 1,4-disubstituted
5
0
piperidines, the presence of the iPr- group on the benzene ring of the carbamate group reduces
hBChE inhibitory potency. Carbamate 23 (IC = 0.12 µM) is the exception to this rule, since
5
0
it is almost four-fold more potent than its benzene counterpart 11 (IC50 = 0.52 µM). N-
Propargylpiperidines and (ethyl)methylcarbamates are practically inactive as hBChE
inhibitors.
N-Propargylpiperidine 19 (IC = 2.86 µM) is the most potent hMAO-B inhibitor of
5
0
this series. Removing the methylene group between the piperidine ring and the carbamate
moiety (n = 0, Figure 4) reduces the inhibitory potency significantly (17, IC = 71.83 µM).
5
0
Replacing the propargyl group of compound 19 with a benzyl group (21, IC = 29.30 µM)
5
0
has a similar effect: the inhibitory potency is reduced by more than tenfold. Similarly as with
,4-disubstituted piperidines, all N-(2,3-dihydro-1H-inden-2-yl)piperidines and
ethyl)methylcarbamates are inactive as hMAO-B inhibitors (IC > 100 µM).
1
(
5
0
1
7

Figure 4. SARs of 1,3-disubstituted N-alkylpiperidine carbamates.
2
.5. Mechanism of hChE and hMAO-B inhibition
To determine the mechanism of inhibition, reversibility 100-fold dilution assay and
3
4
IC curve shift assays were performed with hAChE, hBChE and hMAO-B inhibitor 10, the
5
0
most potent selective hBChE inhibitor 13, the most potent selective hMAO-B inhibitor 16,
and dual hAChE and hBChE inhibitor 22.
In the 100-dilution assay, irreversible inhibitors maintain inhibition after dilution of
the preincubated mixture (Figure 5). In the IC50 curve shift assay, IC50 values of time-
dependent inhibitors decrease with prolongation of incubation time (Figures 6 and 7). Results
of these assays show that compounds 10, 13, and 22 are reversible (Figure 5A and 5B) and
1
8

non-time-dependent inhibitors (Figure 6) of hAChE and/or hBChE. The results also show that
compound 16 is an irreversible (Figure 5D) and time-dependent inhibitor (Figure 7) of
hMAO-B, while compound 10 is not (Figures 5D and 7). This was expected since compound
1
6 is an N-propargylamine and compound 10 is a N-benzylpiperidine derivative, not capable
of forming a covalent bond with hMAO-B.
Figure 5. Dilution assay. Structures of compounds 10, 13, 16, and 22 (A). Recovery of hAChE (B),
hBChE (C), and hMAO-B (D) activity after 100-fold dilution following 15 min incubation for hMAO-
B (30 min for ChEs) at 37 °C (at room temperature for ChEs) of the 100× enzyme concentration with
1
0-fold the IC of compounds 10, 13, 16, 22, rivastigmine (irreversible hChE inhibitor), donepezil
50
(
reversible hAChE inhibitor), rasagiline, paragyline,
L-deprenyl (irreversible hMAO-B inhibitors) and
isatin (reversible hMAO-B inhibitor). The control (DMSO) was carried out in the absence of inhibitor
1
9

and diluted in the same way. Data are expressed as percentages of control ± SEM of three independent
experiments (each performed in quadruplicate).
2
0

2
1

Figure 6. IC curve shift assay for hChE inhibition. Inhibiton of hAChE (A) and hBChE (B) by
5
0
rivastigmine (positive control for time-dependent hChE inhibition) and compounds 10, 13, and 22 at
various pre-incubation times (indicated in the legends). For inhibition of hBChE by compound 13 at
additional pre-incubation times (1, 5, 15, 30, and 60 min) see Supporting Figure S1. RA = residual
activity.
Figure 7. IC50 curve shift assay for hMAO-B inhibition. Inhibiton of hMAO-B by rasagiline
(
positive control for time-dependent hMAO-B inhibition) and compounds 10 and 16 at various pre-
incubation times (indicated in the legends). RA = residual activity.
2
.6. Co-crystal structure of hBChE in complex with compound 13
The crystal structure of compound 13 bound to hBChE was resolved at 2.5 Å
resolution (Supporting Table S1). The crystals were obtained by the hanging drop method,
then soaked with compound 13 at 1 mM final ligand concentration (1% DMSO). Close
2
2

examination of the binding mode explains why compound 13, in particular, does not inhibit
hBChE in a covalent manner. The phenyl substituent occupies the acyl-binding pocket of
hBChE, which is defined by residues Trp231, Leu286, Val288 and Phe398, and forms a T-
stacking interaction (i.e., π-stacking) with Trp231 (Figure 8A). It is important to compare the
binding poses of compounds 13 and 1 because compound 13 is a new lead compound derived
from hit compound 1. Importantly, despite the 2D similarity of compounds 1 and 13 (Figure
8
C), this does not necessarily lead to comparable binding poses. Thus, special attention
should always be placed to the structural elucidation or molecular modelling studies, which
can reveal the reasons for the deviations in the modes of inhibition. Contrary to the hit
27
inhibitor [IC (hBChE) = 21.3 nM] , the piperidine ring of carbamate 13 is flipped with the
5
0
indene moiety pointing towards Trp82 in the choline-binding pocket (Figure 8B). The
piperidine nitrogen occupies exactly the same position as in the parent inhibitor, and forms a
π-cation interaction with Tyr332, which is a pivotal interaction for the inhibition of hBChE.
Additionally, compound 13 is stabilized by H-bonds between the carbamate nitrogen and
oxygen with two water molecules. The latter are also H-bonded to the backbone carbonyl
groups of Pro285 and Ser287, and the piperidine nitrogen. The stabilization of the carbamate
core with several H-bonds places the inhibitor in the middle of the active site gorge with the
carbamate carbonyl group at the distance of 6.6 Å from Ser198, which precludes the covalent
binding of compound 13. This pose and inhibition data clearly demonstrate that introduction
of a carbamate group into a compound does not necessarily produce a pseudo-irreversible
ChE inhibitor.
2
3

Figure 8. Crystal structure of hBChE in complex with compound 13 (PDB 6SAM; resolution 2.5
Å). (A) Compound 13 (cyan) and the key residues of hBChE (green) involved in the binding are
shown as sticks. Water molecules are presented as orange spheres and hydrogen bonds are blue dotted
lines. (B) Alignment of the crystal structures of compound 13 (cyan) and the parent hit compound 1
(
magenta) in the active site of hBChE. (C) Structures of compounds 13 and 1.
2
.7. Inhibition of Aβ1–42 aggregation
The inhibitory activity of carbamates 10, 13, 16, and 22 against Aβ1–42 aggregation
4
0
was evaluated with the thioflavine-T (ThT) fluorometric assay . Compound 16 showed
modest Aβ1–42 anti-aggregation effects, with 25.7% inhibition of Aβ1–42 aggregation (Table 4).
Table 4. Inhibition of Aβ1-42 aggregation by N-alkylpiperidine carbamates 10, 13, 16, 22,
rivastigmine, rasagiline, resveratrol and donepezil.
a
Compound
Inhibition of Aβ1–42 aggregation
2
4

b
b
1
1
1
2
0
3
6
2
n.i.
n.i.
25.7 ± 0.8%*
b
n.i.
b
Rivastigmine tartrate n.i.
b
Rasagiline mesylate
Resveratrol
n.i.
86.6 ± 11.6%*
b
Donepezil
n.i.
a
Percent inhibition at 10 µM compound and 1.5 µM Aβ1–42. Percentage of inhibition means ± SD of
b
three independent experiments (each performed in quadruplicate); No inhibition (percent inhibition
lower than 20%). Statistical analysis: one-way ANOVA, followed by post-hoc Bonferroni t-test
(
SigmaPlot v 12.0), compared to control experiment (Aβ1–42 + DMSO); *p < 0.05.
2
.8. In vitro blood-brain barrier permeation
As AD is a brain disease, drugs used for its treatment must partition into the brain by
crossing the blood-brain barrier (BBB). Therefore, the brain permeation of compounds 10, 13,
6 and 22 was evaluated using the parallel artificial membrane permeation assay (PAMPA)–
1
BBB method, which is a high-throughput and low-cost assay used to exclude poorly
permeable compounds from further testing (e.g., in-vitro assays on cell lines and in-vivo
4
1
assays) . Table 5 shows the negative logarithms of the effective permeabilities (–logP ) of N-
e
alkylpiperidine carbamates 10, 13, 16, and 22 and five reference drugs with known central
nervous system penetration. For all of the N-alkylpiperidine carbamates, the –logP values are
e
lower than the limit for high permeability (5.6), which suggests that these compounds should
be able to cross BBB.
Table 5. In-vitro BBB permeation of carbamates and reference drugs.
PAMPA-BBB
Compounds –logP
(cm/s) Permeabilty
e
a
mean ± SD
prediction
2
5

1
1
1
2
0
3
6
2
5.31 ± 0.19
5.36 ± 0.04
5.01 ± 0.22
4.84 ± 0.21
5.11 ± 0.04
4.79 ± 0.03
5.17 ± 0.03
5.12 ± 0.05
high
high
high
high
high
high
high
high
low
Propranolol
Lidocaine
Verapamil
Quinidine
Theophylline 6.40 ± 0.05
a
–logPe < 5.6, high permeability; –logPe > 6.2, low permeability; intermediate –logPe values (5.6 –
6
.2) were labeled as intermediate BBB permeability.
2
.9. Cytotoxicity and neuroprotective effects of compound 10, 13, 16 and 22
To determine whether N-alkylpiperidine carbamates share the non-cytotoxic and
27
28
29
neuroprotective properties of the parent compounds 1 , 2 , and 3 , inhibitors 10, 13, 16 and
2
2 were characterized accordingly.
First, their cytotoxicity profiles were determined using the MTS (3-(4,5-
dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner
salt) assay. Results show that all four compounds (10, 13, 16 and 22) were non-cytotoxic at 20
µM on SH-SY5Y cells (Figure 9A) and at 50 µM on HepG2 cells (Figure 9B).
We then determined whether compounds 10, 13, 16 and 22 can protect neuronal-like
cells from toxic Aβ-species. Therefore, we examined neuronal death induced by 5 µM Aβ1–42
in the absence and presence of 20 µM compounds 10, 13, 16 and 22. In this assay, compounds
that decrease the percentage of 7-aminoactinomycin D (7-AAD) positive cells have a
neuroprotective effect against Aβ1–42-induced toxicity. Treatment of SH-SY5Y cells with 5
µM of Aβ1–42 caused significant toxicity, as evident from the significant increase of 7-AAD
positive cells (Figure 9C), whereas hMAO-B inhibitors 10 and 16 diminished Aβ1–42-induced
cell death (Figure 9C). The neuroprotective effect of compound 10 is independent of Aβ1–42
aggregation, as, compounds 10 at 10 µM did not inhibit Aβ1–42 aggregation in the ThT
2
6

fluorometric assay (Table 4). On the other hand, the neuroprotective effect of compound 16
could be the result of its Aβ1–42 anti-aggregation effects (Table 4).
Figure 9. Cytotoxicity and neuroprotective effects of compound 10, 13, 16 and 22. Cytotoxicity in
SH-SY5Y (A) and HepG2 cells (B). Cells were incubated in the presence of increasing concentrations
of compounds (1–100 µM) in reduced-serum medium. After 48 h, the cell viability was evaluated
using the MTS assay. The control group (DMSO) was considered as 100% cell viability. Data are
means ± SD of three independent experiments, each carried out in quadruplicate. (C) The effect on
Aβ1-42-induced toxicity in SH-SY5Y cells. Cells were treated with Aβ1-42 (5 µM) in the absence or
presence of compounds (20 µM) in reduced-serum medium. After 48 h treatment, neuroprotective
effect was evaluated by flow cytometry analysis of 7AAD staining. The graph shows the results of
2
7

quantitative analysis and indicates the percentage of dead cells, a fraction of 7AAD positive cells
pos
(
7AAD ). Results are means ± SD of at least two independent assays, each carried out in duplicate.
*
P < 0.05, **P < 0.1
It has been shown recently that Aβ1–42 aggregation initiates the cellular uptake of Aβ
4
2
oligomers and fibrils and subsequently the cytotoxicity . Therefore, the next step was to
evaluate the effects of neuroprotective compounds 10 and 16 on the distribution of Aβ1-42 in
SH-SY5Y cells. For this experiment, cells were stained for amyloid fibrils after the treatment
with monomeric Aβ1–42 peptide in the absence or presence of compounds 10 and 16. Staining
with the amyloid precursor protein (APP)/Aβ antibody revealed large extent of the Aβ
aggregates present in Aβ1–42 peptide-treated SH-SY5Y cells, whereas no such staining was
observed in non-treated (Control) cells as seen in Figure 10. Moreover, no noticeable
difference in the distribution of the amyloid fibrils in the Aβ1–42 peptide-treated SH-SY5Y
cells was observed in the presence of the compound 10, whereas less Aβ aggregates were
observed in the treated cells in the presence of compound 16 (Figure 10). This is in agreement
with the results of compound 10 and 16 obtained in the ThT fluorometric assay (Table 2).
2
8