Novel carbazole-stilbene hybrids as multifunctional anti-Alzheimer agents

Article abstract of DOI:10.1016/j.bioorg.2020.103977

Molecules capable of engaging with multiple targets associated with pathological condition of Alzheimer's disease have proved to be potential anti-Alzheimer's agents. In our goal to develop multitarget-directed ligands for the treatment of Alzheimer's disease, a novel series of carbazole-based stilbene derivatives were designed by the fusion of carbazole ring with stilbene scaffold. The designed compounds were synthesized and evaluated for their anti-AD activities including cholinesterase inhibition, Aβ aggregation inhibition, antioxidant and metal chelation properties. Amongst them, (E)-1-(4-(2-(9-ethyl-9H-carbazol-3-yl)vinyl)phenyl)-3-(2-(pyrrolidin-1-yl)ethyl)thiourea (50) appeared to be the best candidate with good inhibitory activities against AChE (IC₅₀ value of 2.64 μM) and BuChE (IC₅₀ value of 1.29 μM), and significant inhibition of self-mediated Aβ_1–42 aggregation (51.29% at 25 μM concentration). The metal chelation study showed that compound (50) possessed specific copper ion chelating property. Additionally, compound (50) exhibited moderate antioxidant activity. To understand the binding mode of 50, molecular docking studies were performed, and the results indicated strong non-covalent interactions of 50 with the enzymes in the active sites of AChE, BuChE as well as of the Aβ_1-42 peptide. Additionally, it showed promising in silico ADMET properties. Putting together, these findings evidently showed compound (50) as a potential multitarget-directed ligand in the course of developing novel anti-AD drugs.

Full text of DOI:10.1016/j.bioorg.2020.103977

Journal Pre-proofs
Novel carbazole-stilbene hybrids as multifunctional anti-Alzheimer agents
Dushyant V. Patel, Nirav R. Patel, Ashish M. Kanhed, Divya M. Teli, Kishan
B. Patel, Prashant D. Joshi, Sagar P. Patel, Pallav M. Gandhi, Bharat N.
Chaudhary, Navnit K. Prajapati, Kirti V. Patel, Mange Ram Yadav
PII:
S0045-2068(20)31113-5
DOI:
https://doi.org/10.1016/j.bioorg.2020.103977
YBIOO 103977
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
6 May 2020
20 May 2020
24 May 2020
Please cite this article as: D.V. Patel, N.R. Patel, A.M. Kanhed, D.M. Teli, K.B. Patel, P.D. Joshi, S.P. Patel, P.M.
Gandhi, B.N. Chaudhary, N.K. Prajapati, K.V. Patel, M.R. Yadav, Novel carbazole-stilbene hybrids as
multifunctional anti-Alzheimer agents, Bioorganic Chemistry (2020), doi: https://doi.org/10.1016/j.bioorg.
2
020.103977
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 Elsevier Inc. All rights reserved.

Novel Carbazole-Stilbene Hybrids as Multifunctional anti-Alzheimer agents
1
1
2
3
1
Dushyant V. Patel , Nirav R. Patel , Ashish M. Kanhed , Divya M. Teli , Kishan B. Patel ,
1
1
1
1
Prashant D. Joshi , Sagar P. Patel , Pallav M. Gandhi , Bharat N. Chaudhary , Navnit K.
1
1
1,4*
Prajapati , Kirti V. Patel , Mange Ram Yadav
1Faculty of Pharmacy, Kalabhavan Campus, The Maharaja Sayajirao University of Baroda,
Vadodara-390001, Gujarat, India
2Shobhaben Pratapbhai Patel - School of Pharmacy & Technology Management, SVKM’s
NMIMS University, Vile Parle, Mumbai-400056, India
3Department of Pharmaceutical Chemistry, L. M. College of Pharmacy, Navrangpura,
Ahmedabad-380009, Gujarat, India
4Current Position: Director (R & D), Parul University, Limbda, Waghodia Road, Vadodara,
Gujarat, India
1

Abstract:
Molecules capable of engaging with multiple targets associated with pathological condition
of Alzheimer's disease have proved to be potential anti-Alzheimer's agents. In our goal to develop
multitarget-directed ligands for the treatment of Alzheimer’s disease, a novel series of carbazole-
based stilbene derivatives were designed by the fusion of carbazole ring with stilbene scaffold.
The designed compounds were synthesized and evaluated for their anti-AD activities including
cholinesterase inhibition, A aggregation inhibition, antioxidant and metal chelation properties.
Amongst
them,
(E)-1-(4-(2-(9-ethyl-9H-carbazol-3-yl)vinyl)phenyl)-3-(2-(pyrrolidin-1-
yl)ethyl)thiourea (50) appeared to be the best candidate with good inhibitory activities against
AChE (IC value of 2.64 μM) and BuChE (IC value of 1.29 μM), and significant inhibition of
5
0
50
self-mediated A1–42 aggregation (51.29 % at 25 μM concentration). The metal chelation study
showed that compound (50) possessed specific copper ion chelating property. Additionally,
compound (50) exhibited moderate antioxidant activity. To understand the binding mode of 50,
molecular docking studies were performed, and the results indicated strong non-covalent
interactions of 50 with the enzymes in the active sites of AChE, BuChE as well as of the A1-42
peptide. Additionally, it showed promising in silico ADMET properties. Putting together, these
findings evidently showed compound (50) as a potential multitarget-directed ligand in the course
of developing novel anti-AD drugs.
Keywords: Alzheimer’s disease, MTDL, Acetylcholinesterase, Butyrylcholinesterase, Aβ
aggregation, Metal chelation, Carbazole, Stilbene
2

Introduction
Alzheimer’s disease (AD) is an age-related devastating neurodegenerative disorder,
characterized by slow and inexorable memory loss that begins many years before the symptoms
emerge [1]. It is the most common type of senile dementia with prevalence of 50 million people
worldwide. By 2050, it is estimated that the number will rise to 150 million if no preventative
means are made available [2]. The two major hallmarks of AD are the accumulation of beta-
amyloid (Aβ) plaques outside the neurons and the twisted strands of tau protein inside the neurons
in the brain. Although the etiology of AD is very complex and not fully resolved, several factors
like deficit of acetylcholine (ACh) [3], abnormal Aβ deposition and accumulation [4], tau
hyperphosphorylation [5], oxidative stress [6], and dyshomeostasis of biometals [7] are considered
to play notable roles in the pathophysiology of AD. Currently, four drugs are approved for the
treatment of AD namely: donepezil, rivastigmine, galantamine and memantine [8]. They only give
symptomatic relief to the patient while none of the drugs available today stops the damage and
destruction of neurons that make the AD fatal.
The cholinergic hypothesis reveals that the cognitive deficit associated with AD is due to
cholinergic system dysfunction, mainly due to the declining levels of ACh in the brain [3, 9]. ACh
is rapidly hydrolyzed by acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE) in the
brain. Therefore, ChE inhibition is a useful strategy to increase ACh levels within the brain [10].
Besides its catalytic role, AChE also plays proaggregatory role by accelerating A peptide
aggregation and deposition into the fibrils. It is reported that AChE binds through peripheral
anionic site (PAS) to non-amyloidogenic form of A, acting as a pathological chaperone inducing
conformational changes in amyloidogenic form of A with subsequent A fibril formation [11].
In a healthy brain, AChE plays a major role in the hydrolysis of ACh. As the disease progresses,
3

AChE level decreases, but BuChE level increases up to 40 to 90 % in the hippocampus and
temporal cortex region of the brain [12]. BuChE plays several roles both in neural and non-neural
functioning. Clinical data suggested that high cortical levels of BuChE were associated with some
important AD hallmarks, such as the extracellular deposition of the Aβ and aggregation of
hyperphosphorylated tau protein [13]. This attests to the crucial role played by both the
cholinesterases and the imperative to develop multi-targeted directed ligands (MTDLs) which
could act on both of these ChEs.
Amyloid hypothesis states that Aβ plaques in the brain play a critical role in AD
pathogenesis [14, 15]. The plaques consist of aggregated oligomeric A of variable lengths which
are produced by sequential cleavage of the amyloid precursor protein (APP) by - and -secretase.
These aggregates initiate pathogenic cascade and eventually lead to neuronal loss and dementia
[16]. The A plaques generated from A1-42 are neurotoxic which continuously activate
inflammatory mediators, such as TNF- and IL-6. Furthermore, A1-42 itself can serve as an
oxygen-free radical donor that produces reactive oxygen species and directly affects the normal
physiological functions of neurocytes [17]. Hence, the inhibition of A1-42 aggregation could serve
as a rational approach for the treatment of AD.
Recent research indicates that oxidative stress is an event that precedes the appearance of
other hallmarks of AD. The subtle equilibrium between oxidants/antioxidants disturbed
by inequity between generation and scavenging of free radicals cause oxidative stress [18]. By
pathological oxidation-reduction steps, reactive oxygen species (ROS) and reactive nitrogen
species (RNS) can denature biomolecules like proteins, lipids and nucleic acids. This can induce
tissue damage through necrosis and apoptosis [19]. Thus, anti-oxidants endowed with additional
4

pharmacological properties are thought to offer a hope to combat this complex disease, in which
free radicals are important culprits but not the sole drivers.
Likewise, dyshomeostasis of metal ions such as iron, zinc, and copper, clearly occur in AD
brains. The elevated concentrations of metal ions hasten the formation of A aggregates and
neurofibrillary tangles, which activate neurotoxic pathways and promote inflammation, leading to
dysfunction and death of brain cells [20, 21]. Additionally, redox-active metal ions, Cu (I/II), and
Fe(II/III) associated with Aβ, were demonstrated to generate ROS under physiological conditions
through Fenton-like reactions [22]. The overproduction of ROS leads to oxidative stress and
eventually neuronal death in AD patients. Therefore, A aggregation and ROS production induced
by metal ions can be modulated by metal chelators, which highlights metal-ion chelation therapy
as a promising AD treatment. However non-selective metal chelators are likely to exhibit adverse
side effects which is liable to limit their long-term clinical use.
In continuation of our research for unraveling potential novel MTDLs to combat AD, a
series of novel carbazole-based stilbene derivatives as multifunctional anti-AD agents are
presented here. Herein, we report the designing, synthesis and anti-AD activities of these novel
compounds including cholinesterase inhibition, A aggregation inhibition, antioxidant and metal
chelation properties. Molecular modeling studies were also performed to know the binding mode
of these compounds with the target proteins. ADMET properties of the synthesized compounds
were also predicted using in silico methods.
Rationale of designing
The multifactorial nature of AD demands assaulting its key pathological hallmarks using
MTDLs. Although ChEIs provide only symptomatic and transient benefits to the patients, they
still remain the drugs of choice. Neurotoxic A plaques, metal ion dyshomeostasis and oxidative
5

stress play crucial roles in the pathogenesis of AD. However, targeting these factors all alone might
not be enough to combat such a highly complex pathological disease like AD. So, cholinesterase
inhibitors endowed with additional A aggregation inhibitory, and metal chelating and antioxidant
activities could prove to be the competent candidates to confront this multifaceted disease.
Carbazole is an important nitrogen-containing heterocycle present widely in many
phytochemicals with a range of biological activities [23]. It has been reported that the carbazole
derivatives possess cholinesterase inhibitory [24], Aβ aggregation inhibitory [25], ROS
scavenging, and neuron protective activities against oxidative damage [26]. Recently, various
hybrid molecules having the carbazole scaffold and some other biologically active moieties have
been developed as potential MTDLs for AD, like carbazole-tacrine hybrid (1) [27], carbazole-
ferulic acid hybrid (2) [24] and carbazole-thiazole hybrid (5) [28] (Figure 1). A plethora of
biological activities associated with carbazole makes it an interesting privileged scaffold in the
search for new anti-AD agents. Resveratrol (3,5,4'-trihydroxy-trans-stilbene) (6) is a naturally
occurring stilbene derivative possessing multiple anti-AD properties i.e. Aβ aggregation inhibitory
[29], neuroprotective [30] and antioxidant activities [31]. Several stilbene derivatives with
encouraging Aβ binding abilities have been developed as Aβ imaging probes and Aβ aggregation
inhibitors during the last two decades (Figure 1) [32]. In combination with other bioactive
moieties, these stilbene hybrids showed cholinesterase inhibitory, metal chelating, Aβ aggregation
inhibitory, and free radical scavenging activities [33–35]. A central lipophilic scaffold with a
terminal amine group is the salient feature of most of the reported stilbene derivatives which could
have an impact on their anti-AD properties.
6

Figure 1. Chemical structures of some previously reported (A) carbazole-based compounds (1-5)
and (B) stilbene-based compounds (6-10) as anti-AD agents.
A combination of pharmacophoric moieties having different activities offering some novel
hybrids have brought new hope for the treatment of AD. By molecular hybridization approach, a
fusion of one pharmacophore to other results into a highly integrated scaffold with lower molecular
weight. Here, a carbazole based stilbene scaffold was generated by the fusion of carbazole ring
with a stilbene scaffold. We have previously reported substituted triazinoindole derivatives as anti-
AD agents, in which pyrrolidine and piperidine side chains played an indispensable role in offering
7

cholinesterase inhibitory activity [36]. Herein, we have designed two series of carbazole based
stilbene derivatives in which the heterocyclic amines were linked with the designed scaffold using
linkers endowed with additional anti-AD property (Figure 2).
Figure 2. Molecular hybridization approach used to design carbazole-based stilbene derivatives.
8

Results and discussion
Chemistry
Based on the attachment of the heterocyclic amines through suitable linkers on either of the
two sites of the designed scaffold, these carbazole based stilbene derivatives have been divided
into two series i.e. Series-1, wherein the attachment of the linker was to the carbazole ring and
Series-2, in which the attachment was to the phenyl ring of the stilbene moiety (Figure 2).
Synthesis of the designed carbazole derivatives of Series 1 has been depicted in Scheme
1
and Scheme 2. The synthetic route for the required key intermediate (E)-9-ethyl-6-styryl-9H-
carbazol-3-amine (16) from the commercially available carbazole (11) is outlined in Scheme 1.
Ethylation of carbazole (11) by ethyl bromide in the presence of aqueous NaOH solution in DMSO
afforded N-ethylcarbazole (12) in excellent yield. Mono-formylation of N-ethylcarbazole (12) by
Vilsmeier-Haack formylation using DMF and phosphorus oxychloride generated 9-ethyl-3-
formylcarbazole (13). Treatment of 13 with concentrated nitric acid gave the nitro derivative (14),
which was reacted with benzyltriphenylphosphonium bromide under Wittig reaction condition to
yield a mixture of cis- and trans-stilbenes. This isomeric mixture was converted to a
single trans isomer (15) by performing the reaction with catalytic amounts of iodine in
toluene. Reduction of the nitro group in (E)-9-ethyl-3-nitro-6-styryl-9H-carbazole (15) by
stannous chloride offered the key amine intermediate (16).
9

O
H
(i)
(ii)
N
N
N
H
Me
Me
iii)
11
12
13
(
Ph
Ph
O
H
(v)
(iv)
H N
2
O N
2
O N
2
N
N
N
Me
Me
Me
16
15
14
Scheme 1. Synthesis of the key intermediate (E)-9-ethyl-6-styryl-9H-carbazol-3-amine (16).
Reagents and conditions: (i) EtBr, NaOH, DMSO, RT; (ii) POCl , DMF, CHCl ; (iii) conc. HNO ,
3
3
3
AcOH; (iv) (a) benzyltriphenylphosphonium bromide, LiOH, IPA, reflux, (b) I , toluene, reflux;
2
(
v) SnCl , THF, MeOH, reflux.
2
N-(9-Ethyl-3-styryl-9H-carbazole-6-yl)aminoalkylamide derivatives (20-25) have been
synthesized in two steps as shown in Scheme 2. Reaction of the amine intermediate (16) with the
respective acid chlorides in presence of K CO in acetone yielded the amide intermediates (17-
2
3
1
9), which were further linked with alicyclic amines to offer the designed N-(9-ethyl-3-styryl-9H-
carbazole-6-yl)aminoalkylamides (20-25).
Synthetic route for the designed (E)-1-(9-ethyl-6-styryl-9H-carbazol-3-yl)aminoalkylurea
derivatives (26-31) is shown in Scheme 2. The amine intermediate (16) was reacted with p-
nitrophenyl chloroformate in the presence of triethylamine, followed by reaction with alicyclic
amines and aminoalkylamines yielded the designed (E)-1-(9-ethyl-6-styryl-9H-carbazol-3-
yl)aminoalkylurea derivatives (26-31).
1
0

Ph
Ph
Ph
(i)
H
(ii)
H
n
n
R1
H N
2
N
N
X
N
R2
O
O
N
N
N
Me
Me
Me
16
17-19
X
20-25
1
2
1
2
(iii)
Compd.
n
Compd.
n
NR R
Compd.
n
NR R
1
18
7
2
3
4
Cl
Cl
Br
20
22
24
2
3
4
21
23
25
2
3
4
N
N
Ph
19
R1
N
H
1
2
1 2
A
N
Compd.
26
A
n
NR R
Compd.
A
n
NR R
R2
n
O
-
0
2
3
27
29
31
-
0
2
3
N
28
0
NH
NH
N
NH
NH
N
Me
3
26-31
Scheme 2. Synthetic route for the synthesis of compounds (20-31). Reagents and conditions: (i)
1
2
acid chloride, K CO , acetone; (ii) NHR R , THF, reflux; (iii) (a) p-nitrophenyl chloroformate,
2
3
TEA, DCM:THF (1:1), 0 °C to RT, (b) pyrrolidine/piperidine/aminoalkylamines, RT.
Synthesis of the designed carbazole derivatives of Series 2 has been outlined in Scheme
3
and Scheme 4. Synthetic route for the required key intermediate (E)-4-(2-(9-ethyl-9H-carbazol-
-yl)vinyl)aniline (34) from 9-ethyl-9H-carbazole-3-carbaldehyde (13) is shown in Scheme 3. The
3
reaction of 13 with 4-nitrophenylacetic acid (32) in the presence of piperidine under microwave
conditions afforded the trans nitro stilbene derivative (33) as the sole product, configuration of
which was confirmed by XRD. Reduction of the nitro group in compound (33) by stannous
chloride yielded the desired amine intermediate (34).
1
1

NO2
O
NO2
H
O
(i)
N
HO
Me
N
Me
13
32
33
(iii)
NH2
N
Me
34
Scheme 3. Synthesis of the key intermediate (E)-4-(2-(9-ethyl-9H-carbazol-3-yl)vinyl)aniline
(
34). Reagents and conditions: (i) Piperidine, MW; (ii) SnCl , THF, MeOH, reflux.
2
(E)-N-(4-(2-(9-Ethyl-9H-carbazol-3-yl)vinyl)phenyl)aminoalkylamide derivatives (38-
4
4
3) have been synthesized in a way similar to compounds (20-25) in two steps as shown in Scheme
. Reaction of the amine intermediate (34) with the respective acid chlorides offered the amide
intermediates (35-37), which were further reacted with alicyclic amines to afford the designed (E)-
N-(4-(2-(9-ethyl-9H-carbazol-3-yl)vinyl)phenyl)aminoalkylamides (38-43).
Synthesis of the designed urea derivatives (44-49) and thiourea derivatives (50-52) was
performed as depicted in Scheme 4. Reaction of the amine intermediate (34) with p-nitrophenyl
chloroformate in the presence of triethylamine, followed by reaction with aminoalkylamines
afforded the designed urea derivatives (44-49). Similarly, reaction of the amine intermediate (34)
with thiocarbonyldiimidazole followed by reaction with the respective aminoalkylamines afforded
the designed thiourea derivatives (50-52).
1
2

Scheme 4. Synthetic route for the synthesis of compounds (38-52). Reagents and conditions: (i) Acid
1
2
chloride, K CO , acetone; (ii) NHR R , THF, reflux; (iii) For 44-49 (a) p-nitrophenyl chloroformate, TEA,
2
3
DCM:THF (1:1), 0 °C to RT, (b) pyrrolidine/piperidine/aminoalkylamines, RT; For 50-52 (a)
thiocarbonyldiimidazole, DCM:THF (1:1), 0 °C to RT; (b) aminoalkylamines, RT.
Biological evaluation
In vitro cholinesterase inhibition studies
Anti-cholinesterase activity has been confirmed in the tested compounds using the
previously reported in vitro Ellman’s assay [36–38]. The obtained IC values of the compounds
5
0
for both AChE and BuChE enzymes and their selectivity indices (SI) are summarized in Table 1
and Table 2. All the compounds (20-31, 38-52) offered IC values in the range of 1.84-6.63 µM
5
0
for AChE and 1.02-5.01 µM for BuChE.
The inhibitory potential of the compounds from Series 1 having a side chain attached to
the carbazole (ring A) has been shown in Table 1. It was observed that changing the length of the
carbon chain affected the inhibitory activity. A comparative analysis of the inhibitory potential of
compounds (21, 23 and 25) having a piperidine ring, revealed that compound (25, n = 4) showed
comparatively better AChE inhibitory activity (IC value of 1.84 μM) while compound (21, n =
5
0
3
) and compound (23, n = 2) showed slightly poorer AChE inhibitory activities (IC values of
50
1
3

3
.54 and 4.59 μM, respectively) (Table 1). A similar activity pattern was also observed for the
compounds (20, 22 and 24) having a pyrrolidine ring. Among these compounds, 21 (n = 2) showed
better BuChE inhibition (IC value of 1.40 μM) in comparison to the other two derivatives. When
5
0
the pyrrolidinyl (compound 26) and piperidinyl (compound 27) moieties were attached directly (n
0, Table 1) to form urea derivatives, the inhibitory activities against both the enzymes decreased
=
notably. Due to direct attachment, these compounds lost their basic centers, which are seemingly
indispensable for cation- interaction with the enzymes. Compounds (26 and 27) showed
inhibitory activities (AChE; IC = 6.63 µM and 5.99 µM, respectively and BuChE; IC = 4.48
5
0
50
µM and 5.01 µM, respectively). There was no significant change observed in inhibitory activities
when the amide linkers (compounds 20-25) were substituted with urea linkers (compounds 28-31)
(Table 1).
Table 1. In vitro hAChE, EqBuChE and self-induced A aggregation inhibitory activities and
DPPH radical scavenging activity of compounds (20-31)
Ph
Ph
R1
N
H
A
N
R2
n
O
N
N
Me
Me
(CS-1)
20-31
RP of DPPHe
IC50 ± SEM (µM)
Aβ1-42 aggregation
Inhibition (%)
d
1
2
SIc
Compd
A
n
R R N
IC50 ± SEM (µM) or
% inhibition at 100 µM
hAChEa
EqBuChEb
at 25 µM
2
2
2
2
0
1
2
3
-
-
-
-
2
2
3
3
3.00 ± 0.52
4.69 ± 0.25
2.91 ± 0.14
3.54 ± 0.56
1.53 ± 0.31
0.51
0.29
0.52
0.72
42.72 ± 0.31
144.61 ± 2.32 (35.46 %)
135.42 ± 2.05 (39.31 %)
145.02 ± 3.61 (36.15 %)
122.41 ± 3.20 (41.63 %)
1.40 ± 0.22
1.51 ± 0.35
2.56 ± 0.29
52.08 ± 0.64
53.92 ± 0.28
49.88 ± 0.15
1
4

2
2
2
2
2
2
3
3
4
5
6
7
8
9
0
1
-
-
4
4
-
2.63 ± 0.31
1.84 ± 0.27
6.63 ± 0.54
5.99 ± 0.07
2.65 ± 0.32
3.79 ± 0.41
4.54 ± 0.52
3.57 ± 0.29
3.17 ± 0.43
2.51 ± 0.19
4.48 ± 0.15
5.01 ± 1.02
1.70 ± 0.18
1.99 ± 0.21
3.19 ± 0.35
1.02 ± 0.18
1.20
1.36
0.67
0.83
0.64
0.52
0.70
0.28
46.72 ± 0.33
48.09 ± 0.54
17.58 ± 0.42
21.55 ± 0.62
52.29 ± 0.34
50.08 ± 0.47
54.94 ± 0.76
54.35 ± 0.38
134.63 ± 1.29 (38.11 %)
138.18 ± 1.53 (36.97 %)
nd
-
-
-
nd
-NH
-NH
-NH
-NH
2
2
3
104.28 ± 3.87 (48.94 %)
110.42 ± 2.62 (47.54 %)
123.62 ± 3.53 (42.49 %)
118.58 ± 3.20 (45.62 %)
3
CS-1
> 100
> 100
-
27.53 ± 0.93
nd
> 500
> 500
> 500
Tacrine
0.056 ± 0.01
0.023 ± 0.01
0.008 ± 0.00
1.87 ± 0.08
0.14
81.3
Donepezil
nd
2
0.43 ± 0.72 µM
Curcumin
nd
nd
nd
nd
-
-
nd
(IC50)
Ascorbic acid
nd
13.91 ± 1.33 (98.25 %)
a
b
AChE from human erythrocytes, BuChE from equine serum, IC = 50 % inhibitory concentration (means ± SEM of three
5
0
c
d
experiments). Selectivity index = IC (BuChE)/IC50 (AChE). Aβ peptide/inhibitor 1:1 with 25 µM inhibitor
1-42
5
0
e
concentration. RP of DPPH (%) = reduction percentage of DPPH. CS-1 (carbazole-stilbene without amine side chain, see
supporting information).
Shifting the chain from the carbazole (ring A, Figure 2) to the phenyl ring (ring B) of
stilbene preserved the ChEs inhibitory potency (Table 2). A comparative analysis of the inhibitory
potential of compounds (38, 40 and 42) having a pyrrolidine ring, revealed that compound (40, n
=
2) showed the best profile of AChE and BuChE inhibitory activities (IC values of 2.36 μM and
50
1
.46 μM, respectively) while compound (38, n = 1) and compound (42, n = 3) showed slightly
lower AChE and BuChE inhibitory activities. A similar activity pattern was also observed for the
1
5

compounds (39, 41, and 43). Among these, compound (41, n = 2) showed the highest AChE and
BuChE inhibitory activities (IC values of 2.25 μM and 1.74 μM, respectively).
5
0
All the urea derivatives (46-49) showed good ChEs inhibition, whereas compounds
44 and 45) in which the heterocyclic amine was directly attached (n = 0) to form urea showed
(
moderate inhibitory activities (AChE; IC = 16.22 µM and 12.37 µM, respectively) and (BuChE;
5
0
IC = 11.65 µM and 8.58 µM, respectively). There was no significant change observed in AChE
5
0
inhibitory activities when the amide linkers (compounds 42 and 43) were substituted with urea
linkers (compounds 46 and 47) whereas the BuChE inhibitory activity was observed to be
increased by two fold. All the thiourea derivatives (50-52) showed the best ChEs inhibitory
activities profile amongst all of the derivatives. Among them, compound (50) was conferred with
the highest AChE and BuChE inhibitory activities (n = 2, IC value of 2.64 μM and 1.29 μM,
5
0
respectively).
Table 2. In vitro hAChE, EqBuChE and self-induced A aggregation inhibitory activities and
DPPH radical scavenging activity of compounds (38-52)
R1
H
N
A
N
R2
n
X
N
N
Me
Me
(38-52)
(CS-1)
RP of DPPHe
IC50 ± SEM (µM)
Aβ1-42
aggregation
Inhibition (%)
1
2
SIc
Compd
X
A
n
R R N
IC50 ± SEM (µM) or
hAChEa
EqBuChEb
d
%
inhibition at 100 µM
3
3
4
8
9
0
O
O
O
-
-
-
1
1
2
2.98 ± 0.78
2.49 ± 0.65
0.84
0.79
0.62
51.14 ± 0.31
132.42 ± 1.17 (42.34 %)
3.52 ± 0.87
2.36 ± 0.20
2.72 ± 0.98
1.46 ± 0.42
45.17 ± 0.64
54.27 ± 0.28
146.28 ± 3.21 (35.46 %)
189.10 ± 2.26 (30.15 %)
1
6

4
4
4
4
4
4
4
4
4
5
5
5
1
2
3
4
5
6
7
8
9
0
1
2
O
O
O
O
O
O
O
O
O
S
-
2
2.25 ± 0.31
4.77 ± 0.61
3.29 ± 1.09
16.22 ± 1.03
12.37 ± 0.87
4.71 ± 1.05
3.13 ± 0.71
3.04 ± 0.43
2.94 ± 0.98
2.64 ± 0.41
3.41 ± 0.25
3.19 ± 0.34
1.74 ± 0.19
4.76 ± 1.03
2.11 ± 0.69
11.65 ± 0.51
8.58 ± 1.02
2.32 ± 0.97
1.20 ± 0.57
1.92 ± 0.25
1.98 ± 0.84
1.29 ± 0.10
1.72 ± 0.25
1.32 ± 0.17
0.77
0.99
0.64
0.68
0.69
0.49
0.38
0.63
0.67
0.49
0.50
0.41
51.92 ± 0.15
53.08 ± 0.33
49.78 ± 0.54
42.84 ± 0.82
44.15 ± 0.67
38.90 ± 0.85
55.79 ± 0.78
53.86 ± 0.66
54.06 ± 0.49
51.29 ± 0.42
53.24 ± 0.44
53.29 ± 0.62
178.51 ± 3.66 (25.81 %)
126.43 ± 2.27 (41.34 %)
118.58 ± 1.12 (42.91 %)
nd
-
3
3
-
-
-
-
-
nd
-NH
-NH
-NH
-NH
-NH
-NH
-NH
2
2
3
3
2
2
3
161.45 ± 1.46 (13.67 %)
97.36 ± 1.46 (51.19 %)
106.01 ± 1.39 (45.95 %)
134.70 ± 1.17 (41.98 %)
86.13 ± 1.23 (72.36 %)
91.72 ± 3.43 (66.36 %)
90.33 ± 1.20 (70.36 %)
S
S
CS-1
Tacrine
> 100
> 100
-
27.53 ± 0.93
nd
> 500
> 500
> 500
0.056 ± 0.01
0.023 ± 0.01
0.008 ± 0.00
1.87 ± 0.08
0.14
81.3
Donepezil
Curcumin
nd
2
0.43 ± 0.72 µM
nd
nd
nd
nd
-
-
nd
(IC50)
Ascorbic acid
nd
13.91 ± 1.33 (98.25 %)
a
b
AChE from human erythrocytes, BuChE from equine serum, IC = 50 % inhibitory concentration (means ± SEM of three
5
0
c
d
e
experiments). Selectivity index = IC (BuChE)/IC (AChE). Aβ peptide/inhibitor 1:1 with 25 µM inhibitor concentration. RP
5
0
50
1-42
of DPPH (%) = reduction percentage of DPPH. CS-1 (carbazole-stilbene without amine side chain, see supporting information).
Self-mediated A1–42 aggregation inhibition study
1
7

A Peptides present in the extracellular amyloid plaques are produced by sequential
cleavage of APP by - and -secretases. A1-40 and A1-42 are the two main isoforms of A peptides
present in the plaques. A1-40 is the predominant product in the proteolytic cleavage, whereas A1-
is more fibrillogenic in nature [39, 40]. So, A1-42 was chosen for the A aggregation inhibition
4
2
study. The potential of the compounds to inhibit self-mediated A1-42 aggregation was assessed
using Thioflavin T (ThT) fluorescence assay. Curcumin was used as a positive control in this assay.
Percentage inhibitions of self-mediated A1-42 aggregation of all the tested compounds at 25 μM
concentrations are listed in Tables 1 and 2. All the tested compounds showed good A1-42
aggregation inhibition ranging from 38.9 to 55.79 %. Amongst them, compound (47) showed the
best A1-42 aggregation inhibition (55.79 %) at 25 μM concentration.
Antioxidant Activity
The antioxidant activity of the compounds was evaluated by their ability to reduce DPPH·
radical (purple) to DPPHH (yellow) and the corresponding radical-scavenging potential was
assessed by a decrease in the absorbance at 517 nm [36]. Ascorbic acid was employed as a positive
control in this assay. All the test compounds displayed moderate free radical scavenging activity
(Tables 1 and 2). Amongst the thiourea derivatives (compounds 50 - 52) showed moderate free
radical scavenging activity (IC50 values 86.13 μM, 91.72 μM and 90.33 μM, respectively)
compared to ascorbic acid (IC value of 13.9 μM) whereas tacrine and donepezil (IC values >
5
0
50
5
00 μM) were found to be practically devoid of significant free radical scavenging activity at this
concentration.
Metal Chelation study
1
8

2
+
2+
High levels, and at the same time deregulation of biometal ions, such as Cu , Zn , and
2
+
Fe , are closely involved in the pathogenesis of AD [41]. Thus, the potential of compounds to
form chelates with these biometals present in the brain of AD patients is like adding a feather in
the cap of ideal MTDLs to treat AD patients.
The ability of the test compound to chelate biometals was assessed using UV-
vis spectrophotometric assay [42]. The results demonstrated that when CuSO was added to
4
a solution of compound (50), its maximum absorbance at 343 nm decreased dramatically,
2
+
indicating the formation of ligand-Cu complex (Figure 3A). There were insignificant changes
in the positions and values of absorbance when FeSO , FeCl , ZnCl or AlCl were added into the
4
3
2
3
solution of the test compound, suggesting that the test compound (50) had poor chelating abilities
2
+
3+
2+
3+
for Fe , Fe , Zn , and Al . The test compound (50) was also assessed for its binding ability to
other biologically significant metal ions, such as Mg2+ and Ca wherein the compound (50)
exhibited very poor/no binding to these metal ions. 8-Hydroxyquinoline (HQ) was selected as a
positive control to validate the assay protocol. There were significant changes in the positions and
values of absorbance when metal ion solutions were added into the solution of HQ, confirming the
nonselective metal-chelating abilities of HQ (Figure 3B).
2+
1
9

Figure 3. Metal chelation study of compound (50) and HQ. UV-vis spectra of (A) compound (50)
and (B) HQ (25 μM) alone and in the presence of CuSO (25 μM), ZnCl (25 μM), FeSO (25
4
2
4
μM), FeCl (25 μM), AlCl (25 μM) MgSO (25 μM) and CaCl (25 μM) in methanol at room
3
3
4
2
temperature
The above results and comparison of Figure 3A and Figure 3B evinced that the target
2
+
compound (50) could selectively chelate Cu . This high specificity for a particular metal ion is of
prime importance in the design of a metal chelator to avoid chaotic binding to other critical
biometal ions, the depletion of which can lead to allied side effects.
Computational studies
Docking studies of compound (50) with target proteins
To understand the molecular interactions and binding mode of the most active compound
(50) with the ChEs, docking studies were carried out within the active sites of TcAChE (PDB code:
2
CKM) and hBuChE (PDB code: 4BDS) [43].
In the docking study of 50 with AChE, orientation of the compound along with the active
site was found to be similar to that of donepezil, extending from the active site amino acid residue
Trp84 to the peripheral site amino acid residue Tyr70. Binding to the dual sites having such an
interaction with these amino acids is crucial to display a strong affinity to the enzyme. In the
classical binding mode, the tricyclic scaffold without an amine side chain is generally positioned
near the CAS but in case of compound (50) an inverted binding mode is observed where instead
of the aromatic scaffold, the protonated tertiary amine is positioned in the CAS, deep in the binding
gorge. Pyrrolidinylethyl thiourea part of the compound (50) was observed interacting with the CAS
of the active site gorge, whereas the N-ethylcarbazole moiety of the scaffold was found to be
interacting with the receptor active site in PAS of the gorge (Figure 4). In PAS, the aromatic
2
0

carbazole moiety exhibited very strong - interactions with Tyr70 and Tyr334 (hAChE: Tyr72
and Tyr341). The phenyl ring of compound (50) was observed to be stabilized comfortably in the
active site of the enzyme by forming hydrophobic interactions with the aromatic amino acids
Tyr116, Phe330 and Phe331 (hAChE: Tyr119, Tyr337 and Phe338). Stability to the ligand-
receptor complex in CAS is mainly observed because of hydrogen bonding, cation- interaction
and salt bridge. The -NH of thiourea group interacted with Tyr130 (hAChE: Tyr133) by forming
a stable hydrogen bond. In addition to this, at physiological pH, the protonated nitrogen of
pyrrolidine moiety exhibited strong cation- interaction with Trp84 (hAChE: Trp86) along with a
hydrogen bonding and salt bridge interactions with Glu199 (hAChE: Glu202).
2
1

Figure 4. Docking model of compound (50) with TcAChE (PDB ID: 2CKM). (A) Binding mode of 50 in
the active site of TcAChE. The ligand is shown as green balls and sticks. AChE residues are shown as atom
type colour sticks. Hydrogen bonds formed between the ligand and the receptor are indicated by red lines.
(B) Ligand interaction diagram of 50 with TcAChE.
The binding mode of 50 with the BuChE enzyme indicated that it also occupied a large
catalytic cavity of BuChE (Figure 5). Carbazole ring was found to be stabilized in the hydrophobic
pocket of nonpolar amino acid residues Ala277, Val280, Pro285 and Leu286. The phenyl ring
stabilized the ligand-receptor complex by forming stable - interaction with Tyr332 residue.
Hydrogen bond between the -NH of thiourea and Tyr332 residue imparted stability to the ligand-
2
2

receptor complex. Further stability to this complex was also provided by the protonated nitrogen
of pyrrolidine by forming cation- interaction with Trp82 residue of the active site.
Figure 5. Docking model of compound (50) with hBuChE (PDB ID: 4BDS). (A) Binding mode of 50 in
the active site of hBuChE. Ligand is shown as green balls and sticks. hBuChE residues are shown as atom
type colour sticks. (B) Ligand interaction diagram of 50 with hBuChE.
To understand the binding interaction of compound (50) with Aβ1-42, a blind docking study
was performed using the X-ray crystal structure of human Aβ1-42 (PDB code: 1IYT) [44]. The
important regions involved in A aggregation include the N-terminal region, central hydrophobic
2
3

core (Leu17-Ala21), hinge regions (Arg5-Ser8 and Glu22-Asn27) and the hydrophobic region
Ile32-Ala42). The hydrophobic core around Leu17-Ala21 residues plays a crucial role in the β-
(
sheet formation. In this study, the most stable ligand-receptor complex offered promising
interactions (Figure 6). Compound (50) was observed to be aligned with the chain of Aβ1-42. The
carbazole ring showed stable interaction with Phe19 residue, whereas both the -NH of thiourea
group were observed to be interacting strongly with Asp7 residue by forming hydrogen bonding.
Further, the protonated nitrogen of pyrrolidine established stable salt bridge interactions with Asp7
and Glu11 residues. Along with this interaction, the ligand-receptor stability was further supported
by hydrogen bonding of the nitrogen of pyrrolidine with Asp7 residue of the active site. As
hydrogen bondings, salt bridges and hydrophobic interactions of Aβ1-42 monomers represent a
crucial factor governing their aggregation, binding of 50 with Aβ1-42 monomers by π-π interactions
and hydrogen bondings suggests that compound (50) could inhibit Aβ aggregation by blocking
intermolecular interactions among multiple Aβ monomers. It needs to be specified that this
docking is only a predictive tool, the real binding mode and interaction of compound (50) with
A1-42 has not been verified practically.
2
4

Figure 6. Docking model of compound (50) with Aβ1−42 (PDB code 1IYT): (A) Binding mode of 50 with
Aβ1−42. Ligand is shown as green balls and sticks. Aβ1−42 is shown as cartoon. Hydrogen bonds formed
between the ligand and the receptor are indicated by red lines. (B) Ligand interaction diagram of 50 with
hBuChE.
In silico physicochemical and pharmacokinetics parameters prediction
For considering any synthesized compound as a therapeutically important molecule, it
should not only be biologically active but should also be endowed with desirable physico-
chemical land pharmacokinetic properties. Approximately 40 % of the drug candidates failed in
the clinical trials due to the unacceptable ADME (absorption, distribution, metabolism, and
excretion) properties [45]. The in silico prediction of the ADME properties during the drug
development process is a useful strategy to identify these physicochemical and pharmacokinetic
liabilities and to identify and reject those compounds that are suspected to withstand the rigors of
2
5

the later stages of drug developments. Due to the significant progress made in the field of
computational sciences in recent times, in silico prediction of ADMET parameters becomes
relatively simple and reliable. The virtual physicochemical and pharmacokinetic parameters like
Lipinski’s parameters, NRB, PSA, QPPCaco, QPPMDCK, CNS, QPlogBB, QPlogKhsa were
predicted for compound (50) with QikProp module (Table 3) [46].
Table 3. Predicted ADMET parameters of compound (50) and donepezila
Parameter
MW
Limit
50
Donepezil
130-725
2-20
468.659
4.5
379.498
5.5
HBA
HBD
0-6
2
0
NRB
0-8
8
6
QPlogPo/w
PSA
-2 to 6.5
7 to 200
500-2000
0-2
7.002
38.55
1571.159
0
4.242
46.234
1248.451
0
Volume
ReFG
SASA
300 to 1000
878.316
681.675
Rule of Five
0
-1
1
0
(violation)
CNS
-
-
1
1284.443
0.219
1235.258
1.58
1
589.289
0.223
1070.771
0.516
-4.059
100
QPPMDCK
QPlogBB
QPPCaco
QPlogKhsa
QPlogS
-3 to 1.2
-
-1.5 to 1.5
-6.5 to 0.5
0-100
-8.12
100
%
HOA
#
star
0-5
4
0
2
6

aMW: molecular weight, HBA: hydrogen-bond acceptor atoms, HBD: hydrogen-bond donor
atoms, NRB: number of rotatable bonds, QPlogP : Predicted octanol/water partition coefficient,
o/w
PSA: polar surface area, #rtvFG: number of reactive functional groups; SASA: total solvent
accessible surface area, CNS: predicted central nervous system activity on a -2 (inactive) to +2
(active) scale, QPPMDCK: Predicted apparent MDCK cell permeability in nm/s, QPlogBB:
brain/blood partition coefficient, QPPCaco: Caco-2 cell permeability in nm/s, QPlogKhsa: binding
to human serum albumin, QPlogS: predicted aqueous solubility, % HOA: human oral absorption
on 0–100% scale, #star: number of parameters with values that fall outside the 95% range of
similar values for known drugs.
Lipinski’s rule-of-five states that most of the "drug-like" molecules should have molecular
weight ≤ 500, number of hydrogen bond acceptors ≤ 10, number of hydrogen bond donors ≤ 5 and
LogP ≤ 5 [47]. Poor absorption or permeation is more likely to be the case when molecules violate
more than one of these rules. Compound (50) satisfied all these parameters offering values in the
given acceptable ranges except for QPlogPo/w (value > 5). Compounds (50) violates only one limit
of the Lipinski’s rule-of-five, making it as a promising lead as a drug candidate. The number of
rotatable bond (NRB) and topological polar surface area (TPSA) are the two key parameters
introduced by Veber [48]. NRB is a simple topological parameter that indicates molecular
flexibility. It is an important descriptor for oral bioavailability of drugs. The compound could have
0
−8 rotatable bonds or less than 7 linear chains outside rings for good oral bioavailability. TPSA
is another important descriptor that is well correlated with passive transport through membranes
and therefore, it allows the prediction of bioavailability and penetration through blood-brain barrier
(BBB) and drug absorption, including intestinal absorption [49]. The mean value of a TPSA is
2
2
4
0.5 Å (range 4.63–108 Å ) for the marketed CNS drugs. Compound (50) possesses eight
2
rotatable bonds and a TPSA value of 38.55 Å [50]. QPCaco-2 value relates to the oral absorption
of a drug. It shows apparent permeability through gut-blood barrier. Values above 500 predict high
2
7

oral absorption which has been attained by compound (50). Good oral bioavailability of compound
50) is also supported by the predicted human oral absorption percent (% HOA) value. Brain/blood
(
partition coefficient (QPlogBB), CNS, n-octanol−water partition coefficient (QPlogPo/w), and
apparent MDCK cell permeability (QPPMDCK) all predict the ability of the compound to cross
the BBB. Compound (50) is predicted to be CNS active as it possesses a CNS value as 1 and
QPlogBB value as 0.219. QPPMDCK value is predicted apparent MDCK cell permeability in
nm/s. It is recognized as a good mimic for the BBB [51]. A QPPMDCK value higher than 25 is
viewed as good, and the compound (50) has shown considerably high value. The QPlogKhsa value
predicts the binding of the compound with human serum albumin. Compound (50) showed a
slightly higher value than the recommended QPlogKhsa values. #Star shows the number of
parameters with values that fall outside the 95% range of similar values for known drugs. A larger
number of #stars suggests that the compound is less druglike than the compound with few #stars.
The value of #star for compound (50) suggests its drug-likeness. Further, a compound having a
tertiary nitrogen-containing moiety, which is a common feature in many CNS active drugs,
normally exhibits a higher degree of brain permeation [50]. As discussed above, the
compound (50) is predicted to possess a good pharmacokinetic profile, which highlights its
biological significance.
Conclusion
By molecular hybridization approach, a combination of carbazole and stilbene rings resulted
in (E)-3-styryl-9H-carbazole scaffold. Modifications by attaching alicyclic amines with different
linkers to the (E)-3-styryl-9H-carbazole scaffold resulted in a novel series of anti-AD agents.
Among the series, compound (50) having pyrrolidine moiety and thiourea linker showed the most
promising inhibitory activities against AChE (IC value of 2.64 μM) and BuChE (IC value of
5
0
50
2
8

1
.29 μM). Compound (50) exhibited a significant self-mediated A1–42 aggregation inhibition
2
+
(
51.29 % at 25 μM concentration). It also displayed specific metal (Cu ) chelating ability and
moderate antioxidant activity. Molecular modeling studies indicated significant interactions
between this most potent compound (50) with PAS as well as CAS sites of both the enzymes as
well as with A1-42 peptide. Additionally, compound (50) exhibited favorable in silico ADMET
properties. All these results suggest that compound (50) could be a leading candidate with a high
potential for further development as a novel anti-AD drug.
EXPERIMENTAL SECTION
General: All of the commercial reagents and solvents required during the synthesis of the
compounds were procured from Sigma-Aldrich, Spectrochem, S. d. fine chemicals and Avra
chemicals, and were purified by general laboratory techniques whenever needed. Reaction
monitoring was carried out by thin-layer chromatography (TLC), using silica gel precoated plates
(60F254, Merck, 0.25 mm thickness) and visualizing in ultraviolet (UV) light (λ = 254 nm) or in
an iodine chamber. Compounds were purified by flash column chromatography with a Teledyne
ISCO CombiFlash Rf system using RediSep Rf columns. Yields reported here are unoptimized.
Melting points were determined in glass capillary tubes using a silicon oil-bath type melting point
apparatus (Veego) and the reported melting points are uncorrected. The IR spectra were recorded
on a Bruker ALPHA-T (Germany) FT-IR spectrophotometer for all the reported compounds and
1
13
are consistent with the assigned structures. H NMR and C NMR spectra were recorded on a
Bruker Advance-II 400 MHz spectrometer in CDCl or DMSO-d solvents. Chemical shifts (δ)
3
6
are expressed in parts per million (ppm) relative to the standard TMS, and the peak patterns are
indicated as s (singlet), d (doublet), t (triplet), m (multiplet), and br (broad signal). Mass spectra
were recorded using a Thermo Fisher mass spectrometer with an ESI ion source. LCMS analyses
2
9