Design and development of multitarget-directed N-Benzylpiperidine analogs as potential candidates for the treatment of Alzheimer's disease

Article abstract of DOI:10.1016/j.ejmech.2019.02.030

The multitarget-directed strategy offers an effective and promising paradigm to treat the complex neurodegenerative disorder, such as Alzheimer's disease (AD). Herein, a series of N-benzylpiperidine analogs (17–31 and 32–46) were designed and synthesized

Full text of DOI:10.1016/j.ejmech.2019.02.030

Accepted Manuscript
Design and development of multitarget-directed N-Benzylpiperidine analogs as
potential candidates for the treatment of Alzheimer's disease
Piyoosh Sharma, Avanish Tripathi, Prabhash Nath Tripathi, Santosh Kumar
Prajapati, Ankit Seth, Manish Kumar Tripathi, Pavan Srivastava, Vinod Tiwari, Sairam
Krishnamurthy, Sushant Kumar Shrivastava
PII:
S0223-5234(19)30144-8
DOI:
https://doi.org/10.1016/j.ejmech.2019.02.030
EJMECH 11122
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 22 January 2019
Revised Date: 8 February 2019
Accepted Date: 9 February 2019
Please cite this article as: P. Sharma, A. Tripathi, P.N. Tripathi, S.K. Prajapati, A. Seth, M.K. Tripathi,
P. Srivastava, V. Tiwari, S. Krishnamurthy, S.K. Shrivastava, Design and development of multitarget-
directed N-Benzylpiperidine analogs as potential candidates for the treatment of Alzheimer's disease,
European Journal of Medicinal Chemistry (2019), doi: https://doi.org/10.1016/j.ejmech.2019.02.030.
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
GRAPHICAL ABSTRACT

ACCEPTED MANUSCRIPT
Design and Development of Multitarget-Directed N-Benzylpiperidine
Analogs as Potential Candidates for the Treatment of Alzheimer’s
Disease
1
2
3
4
5
6
Piyoosh Sharma, Avanish Tripathi, Prabhash Nath Tripathi, Santosh Kumar Prajapati, Ankit
Seth, Manish Kumar Tripathi, Pavan Srivastava, Vinod Tiwari, Sairam Krishnamurthy,
Sushant Kumar Shrivastava*
7
8
Department of Pharmaceutical Engineering & Technology, Indian Institute of Technology
(Banaras Hindu University), Varanasi 221005, India
9
*Corresponding Author: skshrivastava.phe@itbhu.ac.in; Tel.: +91-945-2156-527
1

ACCEPTED MANUSCRIPT
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
ABSTRACT
The multitarget-directed strategy offers an effective and promising paradigm to treat the
complex neurodegenerative disorder, such as Alzheimer’s disease (AD). Herein, a series of
N-benzylpiperidine analogs (17–31 and 32–46) were designed and synthesized as multi-
functional inhibitors of acetylcholinesterase (AChE) and β-secretase-1 (BACE-1) with
moderate to excellent inhibitory activities. Amongst tested inhibitors, 25, 26, 40, and 41
presented the most significant and balanced inhibition against both the targets. Compounds
40 and 41 exhibited high brain permeability in PAMPA-BBB assay, significant displacement
of propidium iodide from peripheral anionic site (PAS) of AChE, and were devoid of
neurotoxicity towards SH-SY5Y neuroblastoma cell lines up to the maximum tested
concentration of 80 µM. Meanwhile, both these compounds inhibited self- and AChE-
induced Aβ aggregation in thioflavin T assay, which was also re-affirmed by morphological
characterization of Aβ aggregates using atomic force microscopy (AFM). Moreover, 40 and
41 ameliorated the scopolamine-induced cognitive impairment in elevated plus and Y-maze
experiments. Ex vivo and biochemical analysis established the brain AChE inhibitory
potential and antioxidant properties of these compounds. Further, improvement in Aβ1-42
-
induced cognitive impairment was also observed by compound 41 in the Morris water maze
experiment with significant oral absorption characteristics ascertained by the
pharmacokinetic studies.
29
30
KEYWORDS: Alzheimer’s disease; acetylcholinesterase (AChE); β-secretase-1 (BACE-1);
Aβ aggregation; multitarget-directed ligands; structure-based drug design.
2

ACCEPTED MANUSCRIPT
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
1. INTRODUCTION
Alzheimer’s disease (AD) is a multifactorial disease characterized by progressive
neurodegeneration, cognitive deficits, memory loss, and behavioral alterations [1]. It causes
substantial structural and functional damage to the brain [2]. AD is a big challenge for
healthcare and currently accounts for around 50 million cases worldwide. Recently published
WHO report has predicted to exalt this number to breach 152 million mark by 2050 [3]. AD
progresses symptomatically from mild to severe and its prevalence increases in aged
population [4]. There are currently no therapeutic regimens available that can prevent or
mitigate the disease progression. Several pathophysiological mechanisms were proposed in
last two decades to have significant role in AD such as amyloid beta (Aβ) aggregation and
deposition [5], tau protein hyperphosphorylation and formation of neurofibrillary tangles [6],
NMDA receptor activation [7], CREB signaling pathways [8], oxidative stress [9], biometal
dyshomeostasis [10], and decline in acetylcholine (ACh) levels [11].
Cholinesterase is a ubiquitous class of serine hydrolase enzyme comprising of two types:
acetylcholinesterase (AChE) and butyrylcholinesterase (BChE) that catalyze the hydrolysis of
ACh to choline and acetic acid [12]. Accordingly, cholinesterase inhibitors have been used to
enhance the pathologically reduced ACh levels by modulating its native hydrolytic
degradation and could be the vital therapeutic target in improving cognitive dysfunctions [13,
14]. AChE has also been recognized to play a vital role in increasing the deposition of Aβ
aggregates in AD [15]. Aβ is a small fragment formed by proteolytic cleavage of large
protein structure known as an amyloid precursor protein (APP). The activation of APP leads
to its fragmentation by an enzyme β-secretase-1 and also known as beta-site APP cleaving
enzyme-1 (BACE-1), followed by a series of reactions to form oligomers, fibrils, plaques,
and β-sheets [16]. Altogether, these reactions are responsible for the aggregation of Aβ and
disruption of cellular communication, inflammatory responses, progressive neuritic injury,
3

ACCEPTED MANUSCRIPT
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
neuronal deficits, and cognitive dysfunctions [17]. Accumulation of Aβ could also lead to
oxidative damage in the brain by entering into mitochondria and increasing generation of free
radicals [18].
Current treatment strategies encompass the use of FDA approved medications like AChE
inhibitors (AChEIs) (donepezil, rivastigmine, and galantamine) and NMDA-receptor
antagonist (memantine) [19, 20]. However, drug therapy for this ailment is still in its infancy
and fails to delay the progression of the disease. The new strategy of multi-targeted inhibitors
has been adopted recently, which involves targeting multiple enzymes simultaneously with a
single molecule [21-23]. This strategy was built upon the fact that AD is a multifactorial
disorder, linked with multiple targets. Thereby, design and development of multitarget-
directed ligands could prove beneficial for successfully mitigating the disease progression
rather than providing symptomatic relief only.
However, AChE inhibition remains to be the most promising target for ameliorating
cognitive dysfunctions exacerbated in AD. BACE-1 inhibition is another important target
with a significant role in cleavage of APP, generation of Aβ aggregates, and formation of
senile plaques. Further, as Aβ is linked with oxidative damage, molecules with BACE-1
inhibitory potential could also possess significant antioxidant effect [24]. The most potent
inhibitor of AChE, donepezil bears the N-benzylpiperidine nucleus as a core group, which is
bound to the catalytic active site (CAS) of AChE [25]. The literature also supports the
modification of other terminal part (indanone nucleus) leading to improved BACE-1
inhibitory potential [26]. Based on the above-mentioned facts, we propose that ligands
bearing N-benzylpiperidine nucleus could possess multi-functional enzyme inhibitory
potential against AChE and BACE-1.
On the basis of these conjectures and our quest to identify potential multi-target inhibitors for
the treatment of AD, we have designed and developed a series of N-benzylpiperidines as
4

ACCEPTED MANUSCRIPT
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
multi-target inhibitors of AChE and BACE-1 with a significant role in Aβ aggregation
inhibition and antioxidant activity. The structure-based drug design approach was used,
which utilizes e-pharmacophore models of protein structures (AChE and BACE-1) having the
presence of co-crystallized ligands (donepezil and F1M, respectively) bearing N-
benzylpiperidine nucleus. The generated e-pharmacophore models were used to screen the
library of compounds to identify potential hits, which were further filtered through several
computational frameworks, such as virtual screening (VS), docking-post processing (DPP),
and Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) analysis. The
identified potential hit was used to rationally design the library of compounds with improved
binding characteristics at peripheral anionic site (PAS) of AChE and aspartate dyad of
BACE-1. The designed compounds were synthesized, characterized, and biologically
evaluated. The designed analogs exhibited moderate to good inhibitory potential against
AChE and BACE-1 in micromolar to sub-micromolar ranges. Compounds 40 and 41
demonstrated significant displacement of propidium iodide from PAS-AChE, and SH-SY5Y
neuroblastoma cell lines were viable at the tested concentration range of 10–80 µM. These
compounds also exhibited self- and AChE-induced Aβ aggregation inhibition in thioflavin T
assay, which was also confirmed by morphological characterization of Aβ aggregates through
atomic force microscopy (AFM) analysis. Most promising compounds 40 and 41 further
exhibited alleviation of cognitive impairment in AD rat models.
100
2. RESULTS AND DISCUSSION
101
102
103
104
105
2.1 Computational studies and designing considerations
The structure-based drug design approach was used to generate e-pharmacophore models
using crystal structures of AChE (PDB code: 4EY7) [27] and BACE-1 (PDB code: 2ZJM)
[28], which were selected owing to the presence of N-benzylpiperidine nucleus in their co-
crystallized ligands. The generated e-pharmacophore models suggested the presence of two
5

ACCEPTED MANUSCRIPT
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
aromatic rings (R9 and R10), two hydrophobic residues (H7 and H8), and one hydrogen bond
acceptor (A3) as essential features for binding of inhibitors to AChE (Figure S1A and S2A,
Supporting Information), while two aromatic rings (R14 and R15), two hydrogen bond
donors (D7 and D8), and single positive ionic group (P13) were necessary for favored
binding within BACE-1 active pocket (Figure S1B and S2B, Supporting Information). On the
basis of these pharmacophoric features, ligands of Maybridge HitFinder database
(https://www.maybridge.com as accessed on June 2017) were screened using Phase module
of Schrödinger Maestro 2018.1. Initial screening of total 18930 ligands by pharmacophore
model of AChE resulted into 917 hits, which were screened by BACE-1 model to identify
206 hits.
The virtual screening (VS) workflow was employed, which comprised of three steps: HTVS,
standard precision (SP), and extra precision (XP) docking with the criteria of screening as
30% ligands in each step. The results of these protocols yielded 15 potential hits on the basis
of their promising glide and dock scores on both the target structures (AChE and BACE-1).
Pose filtration was performed on identified hits using DPP function. The X-ray
crystallographic structure of AChE revealed the presence of two major binding sites: PAS;
and CAS. The inhibitor of AChE should be extended from PAS (Tyr72, Asp74, Tyr124,
Trp286, and Tyr341) to CAS (Ser203, Glu334, and His447) [19]. BACE-1 is a membrane-
bound enzyme having the presence of catalytic aspartic dyad residues (Asp32 and Asp228) at
the interface of N- and C-terminus [29]. Therefore, ligands were screened by DPP on the
basis of their interactions with PAS and CAS residues of AChE and aspartate dyad of BACE-
1. Only 3 common hits were identified (Figure S3, Supporting Information) showing
significant interactions with the active site residues of both the targets (Table S1, Supporting
Information).
6

ACCEPTED MANUSCRIPT
130
131
132
133
134
135
136
137
138
139
140
141
142
MM-GBSA analysis of identified 3 potential hits (SEW06622, AW01119, and PD00698) was
performed to ascertain the binding free energy of docked ligands into the respective proteins.
The results of MM-GBSA analysis (Table S1, Supporting Information) showed SEW06622
as the top hit with the minimum binding free energy of -82.86 and -52.45 kcal/mol for AChE
and BACE-1, respectively.
AChE was demonstrated to be involved in the promotion of Aβ aggregation by binding
specifically to the PAS residues. AChEIs bound to the active site and PAS could not only
result in AChE inhibition but also prevent the Aβ production and deposition [30]. Therefore,
2,3-diOCH3 functionality present on phenyl group of SEW06622 was substituted with several
electron donating and withdrawing functional groups to improve the binding characteristics
with PAS and increase the inhibition of Aβ aggregation. Further, binding with aspartate dyad
residues of BACE-1 was cemented with the reduction of central methanimine functionality to
methanamine, while retaining the PAS-AChE binding characteristics (Figure 1).
7

ACCEPTED MANUSCRIPT
143
144
Figure 1. Designing strategy for a series of N-benzylpiperidine analogs.
145
146
147
148
149
150
151
152
153
154
Molecular docking and dynamics simulations studies were performed to affirm the binding
characteristics of designed ligands on AChE and BACE-1. Initially, docking parameters and
prepared grids were validated by re-docking the co-crystallized ligands into respective grids
of AChE and BACE-1. The superposition tool was used, which showed RMSD values within
2Å between co-crystallized and re-docked ligands to confirm the suitability of docking
protocols and generated grids [31]. The results of docking studies confirmed that designed
ligands (26 and 41) and hit compound SEW06622 accommodated within the active pocket of
AChE with substituted phenyl group occupied PAS, while benzylpiperidine nucleus was
extended deep into CAS and formed polar interactions with Ser203 and His447. SEW06622
showed charged and hydrophobic interactions with Asp74 and other PAS residues (Tyr124,
8

ACCEPTED MANUSCRIPT
155
156
157
158
159
160
161
162
163
164
165
166
Trp286, and Tyr341), respectively (Figure S4A, Supporting Information). While, 26 (Figure
S5A, Supporting Information) and 41 (Figure 2A) elicited similar interaction pattern at PAS,
except additional π-π stacking and π-cation interactions with Tyr341 residue. At anionic
subsite, benzyl portion of all the compounds exhibited interactions with Trp86 (π-π stacking
or π-cation) and Glu202 (charged) residues. Other active site residues of oxyanion hole and
acyl binding pocket were also observed similarly in docked poses of SEW06622, 26, and 41.
The binding mode of docked ligands on BACE-1 indicated significant interactions with both
the catalytic aspartate residues (Asp32 and Asp228) through either ionic salt bridge or H-
bonding. SEW06622 showed both the interactions with single —NH present in piperidine
ring (Figure S4B, Supporting Information), while designed series of compounds 26 (Figure
S5B, Supporting Information) and 41 (Figure 2B), showed these interactions with both —NH
groups present in piperidine ring and a side chain.
167
168
Figure 2. Binding pattern of 41 depicted in golden colored ligand binding surface around the
169
active pockets. [A] AChE (4EY7) and; [B] BACE-1 (2ZJM).
170
171
172
173
174
175
Further, to evaluate the stability of docked poses, 20 ns molecular dynamics simulation runs
were performed for docked complexes of SEW06622, 26, and 41 on both the targets AChE
and BACE-1. The results of molecular dynamics studies on AChE suggested that SEW06622
interacted significantly with PAS residues (Figure S6A–C, Supporting Information). The
methanimine (CH=N) functionality involved in H-bonding interaction with Tyr124 accounted
for 93% interaction in selected trajectories. Other PAS residues such as Asp74 and Trp286
9

ACCEPTED MANUSCRIPT
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
residues interacted through salt bridge formation and π-π stacking interactions, respectively.
While Tyr72 and Tyr341 residues interacted hydrophobically with SEW06622. The results of
dynamics simulations of designed ligands 26 (Figure S7A–C, Supporting Information) and 41
(Figure 3A–C) on AChE showed more prominent interactions with PAS residues compared
to SEW06622. The imine/amine functionalities of 26 and 41 interacted with Asp74 and
Tyr124 residues at PAS through H-bonding interaction. Both the ligands also interacted
through π-π stacking and hydrophobic interactions with Tyr341 and Trp286 residues at PAS,
while 41 additionally engaged in H-bonding interaction with Tyr341. Furthermore, these
ligands also interacted with His447 at CAS through hydrophobic interaction.
The outcome of molecular dynamics simulations on BACE-1 suggested that interactions of
SEW06622 with aspartate dyad residues were unstable. SEW06622 has formed minimal ionic
interaction with Asp32 and did not interact with Asp228 throughout the simulation run
(Figure S8A–C, Supporting Information). Interestingly, designed ligands 26 and 41 on
BACE-1 showed H-bonding with both catalytic aspartate dyad residues (Asp32 and Asp228).
Compound 26 exhibited H-bonding for 10% and 51% interactions in selected trajectories
with Asp32 and Asp228, respectively (Figure S9A–C, Supporting Information). While 41
having reduced functionality of –NH interacted more effectively with both dyad residues
(Asp32 and Asp228) having 100% and 99% interactions, respectively (Figure 4). Thereby,
observations of molecular dynamics simulations further strengthen our hypothesis that
designed series of ligands might improve the inhibitory potential against both the targets:
AChE and BACE-1.
10

ACCEPTED MANUSCRIPT
197
198
199
200
201
202
Figure 3. Molecular dynamics studies of 41-AChE (4EY7) docked complex. [A] Histogram
showing interaction fractions with active amino acid residues; [B] Graphical representation
showing percentage interactions with active site residues; [C] Timeline representation
showing interaction with all the amino acid residues at each time frame.
203
204
Figure 4. Molecular dynamics studies of 41-BACE-1 (2ZJM) docked complex. [A]
Histogram showing interaction fractions with active amino acid residues; [B] Graphical
representation showing percentage interactions with active site residues; [C] Timeline
representation showing interaction with all the amino acid residues at each time frame.
11
205
206
207

ACCEPTED MANUSCRIPT
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
2.2 Chemistry
The series of designed compounds 17–46 were synthesized following the Scheme 1. Briefly,
1-benzylpiperidin-4-amine (1) was treated with different aromatic aldehydes (2–16) under
acidic condition (glacial acetic acid) in absolute ethanol at room temperature (4–6 h) to yield
target compounds (17–31). FT-IR spectra of compounds 17–31 showed the appearance of a
characteristic stretching band for (N=CH) in the range of 1604–1557 cm-1. H NMR spectra
1
showed characteristic singlet peak for (N=CH) at δH = 8.66–8.27 ppm. The presence of
(N=CH) signal at δC = 161.83–154.12 ppm value confirmed the formation of compounds 17–
31. The resulted compounds (17–31) were treated with NaBH4 in dry methanol at 0–5 °C (2–
4 h) to yield their reduced counterparts (32–46). FT-IR spectra of 32–46 showed an
appearance of stretching band of —NH in the region of 3340–3204 cm-1. The disappearance
of sharp singlet peak of N=CH and appearance of two protons signal at δH = 3.99–3.3.82 ppm
1
13
for NH—CH2 in H NMR spectra confirmed the formation of compounds 32–46. Also, C
NMR spectra of compounds 32–46 showed the appearance of a signal in the range of δC =
51.82–47.12 ppm for N—CH2. The molecular weights of all synthesized compounds were
confirmed by mass spectrophotometry used with electrospray ionization (ESI) source. All
synthesized compounds were confirmed to have purity ≥95% (Table S2, Supporting
Information).
226
227
Scheme 1. Synthesis of target compounds. Reagents and conditions: (I) Glacial acetic acid,
228
absolute ethanol, room temperature, 4–6 h; (II) NaBH4, dry methanol, 0–5 °C, 2–4 h.
12

ACCEPTED MANUSCRIPT
229
230
2.3 Pharmacology
2.3.1 In vitro studies
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
2.3.1.1 Cholinesterase inhibition assay
The synthesized compounds 17-46 were assessed for their inhibitory potential against AChE
and BChE by Ellman method [32] using donepezil and rivastigmine as positive reference
standards. The results showed moderate to excellent inhibition of AChE in micromolar to
sub-micromolar concentration range except for the compounds 17, 19, and 23 (IC50 > 15 µM).
All the tested compounds elicited poor to moderate inhibitory potential against BChE. The
compound 41 of series showed 28.2 fold AChE selective inhibition than BChE comparable to
donepezil having 42.1 fold AChE selectivity. Amongst all the evaluated compounds,
compounds 32–46 has slightly pronounced AChE inhibitory activity than their respective
methanimine derivatives (17–31). The unsubstituted compounds 18 and 33 exhibited
moderate AChE inhibition at micromolar levels (18: IC50 = 2.1 µM; 33: IC50 = 1.7 µM). The
substitution of electron donating groups (—OH, —OCH3, and —CH3) weakened AChE
inhibitory potential in compounds 17, 19–24, 32, and 34–39. The substitution of electron
withdrawing groups drastically improved the inhibition against AChE. The 2,4-disubstitution
with weak electron withdrawing groups (—Cl and —F) (29: IC50 = 1.3 µM; and 31: IC50 =
1.3 µM; 44: IC50 = 0.62 µM; 46: IC50 = 0.71 µM) exhibited slightly higher inhibitory
potential than their mono-substituted (4-) counterparts (28: IC50 = 1.5 µM; and 30: IC50 = 2.9;
43: IC50 = 0.85 µM; 45: IC50 = 1.2 µM). The strong electron withdrawing substituted groups
(—NO2, —CF3, and —OCF3) showed significantly high AChE inhibitory potential in sub-
micromolar range (25: IC50 = 0.90 µM; 26: IC50 = 0.44 µM, 40: IC50 = 0.72 µM; 41: IC50 =
0.11 µM; and 42: IC50 = 0.59 µM) exception being compound 27, which elicited inhibition at
micromolar level (27: IC50 = 1.0 µM). The results of AChE inhibition assay also corroborated
with the outcome of docking and dynamics study with 41 being the most potent compound of
13

ACCEPTED MANUSCRIPT
254
255
the series by interacting significantly with PAS and CAS residues of AChE. The results are
tabulated in Table 1.
256
257
Table 1. Structures, cholinesterase (hAChE and hBChE), BACE-1 inhibitory potential, and
AChE selectivity index of tested compounds.
258
IC50 (µM) ± SEM
hAChE
Comd.
R
SIa
hAChE
>15 µM
hBChE
>30 µM
>30 µM
11±0.08
16±0.10
>30 µM
19±0.06
>30 µM
>30 µM
7.3±0.07
3.9±0.07
5.5±0.07
4.6±0.05
5.8±0.06
>30 µM
>30 µM
7.2±0.04
12±0.08
11±0.06
14±0.11
>30 µM
18±0.04
>30 µM
>30 µM
7.8±0.09
3.0±0.06
3.5±0.06
hBACE-1
6.3±0.13
3.4±0.10
14±0.09
2.7±0.07
12±0.06
10±0.11
3.3±0.07
19±0.09
0.73±0.03
0.28±0.03
1.8±0.06
1.7±0.04
3.9±0.10
1.5±0.05
3.5±0.05
5.3±0.09
2.9±0.10
13±0.08
2.0±0.08
9.9±0.08
5.9±0.09
2.6±0.09
17±0.04
0.43±0.04
0.22±0.02
1.4±0.04
2,3-diOCH3
H
-
-
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
2.1±0.06
>15 µM
3,4,5-triOCH3
2-OH
-
2.8±0.05
4.8±0.09
2.6±0.08
>15 µM
5.9
-
4-OH
2,4-diOH
3-CH3
7.4
-
4-CH3
4.4±0.03
0.90±0.03
0.44±0.05
1.0±0.06
1.5±0.06
1.3±0.06
2.9±0.06
1.3±0.03
2.2±0.08
1.7±0.07
2.2±0.06
2.4±0.06
4.0±0.07
2.3±0.07
4.1±0.10
3.7±0.08
0.72±0.04
0.11±0.02
0.59±0.05
-
4-NO2
8.1
8.9
5.3
3.2
4.3
-
4-CF3
4-OCF3
4-Cl
2,4-diCl
4-F
2,4-diF
3,4-diOCH3
H
-
3.3
6.9
4.8
6.0
-
3,4,5-triOCH3
2-OH
4-OH
2,4-diOH
3-CH3
8.0
-
4-CH3
-
4-NO2
10.8
28.2
5.9
4-CF3
4-OCF3
14

ACCEPTED MANUSCRIPT
4-Cl
2,4-diCl
4-F
0.85±0.04
0.62±0.05
1.2±0.08
4.0±0.06
3.5±0.09
3.9±0.06
3.1±0.10
1.4±0.06
1.1±0.04
0.66±0.06
3.0±0.09
0.55±0.03
2.9±0.03
0.24±0.03
-
4.7
5.7
43
44
45
46
3.3
2,4-diF
0.71±0.06
0.033±0.01
1.9±0.06
4.3
Donepezil
42.1
Rivastigmine
0.6
259
260
261
262
263
264
265
266
267
268
269
270
All the results are reported as IC50±SEM (n = 3). a SI (Selectivity Index) = IC50 of hBChE/IC50 of hAChE.
The enzyme kinetics study for compound 41 on hAChE was performed by Lineweaver-Burk
method to establish the mechanism of enzyme inhibition [33]. The results of the Lineweaver-
Burk plot (Figure 5) and Michaelis-Menten graph (Figure S30, Supporting Information)
showed decreased Vmax, while Km remained constant with an increase in the substrate
concentration. The plotted result suggested a non-competitive type of enzyme inhibition by
compound 41 on both free enzyme and the enzyme-substrate complex. The intersection
points on both axes of 1/Vmax and 1/[S] were found to be higher than zero, suggesting that the
inhibitor was preferentially bound to free enzyme rather than enzyme-substrate complex.
Further, the Dixon plot has been plotted (Figure S31, Supporting Information) using three
concentrations of inhibitor (0.03, 0.15, and 0.30 µM) and the outcome suggested that
compound 41 has Ki = 0.10 µM.
0.120
0.30 µM, R² = 0.991
0.080
0.040
0.15 µM, R² = 0.995
0.03 µM, R² = 0.985
Control, R² = 0.986
-10.0
10.0
20.0
1/[S] X 10-6 mM
30.0
-0.040
271
272
Figure 5. Lineweaver-Burk plot for the kinetic study of hAChE inhibition by compound 41.
273
274
275
2.3.1.2 BACE-1 inhibition assay
All the synthesized derivatives were evaluated for their BACE-1 inhibitory potential by
FRET-based BACE-1 fluorescence assay kit (Sigma Aldrich, Catalog No. CS0010). The
15

ACCEPTED MANUSCRIPT
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
unsubstituted derivatives 18 and 33 showed moderate inhibition against BACE-1 (18: IC50 =
3.4 µM; 33: IC50 = 2.9 µM). However, the substitution of electron donating groups (17, 19–
24, 32, and 34–39) significantly reduced the BACE-1 inhibitory potential. Compounds
substituted with 2-OH and 3-CH3 electron donating groups were the exceptions and showed
moderately higher inhibition potential (20: IC50 = 2.7 µM; 23: IC50 = 3.3 µM; 35: IC50 = 2.0
µM; 38: IC50 = 2.6 µM). These outcomes suggested that substitutions of electron donating
groups were favored at ortho than the para-position. Another electron donating substitution
with 3,4,5-triOCH3 (19: IC50 = 14 µM; 34: IC50 = 13 µM) elicited less pronounced BACE-1
inhibition than 2,3-diOCH3 substituted analogs (17: IC50 = 6.3 µM; 32: IC50 = 5.3 µM). The
results indicated the substitution of electron donating groups at 4-position significantly
declined the BACE-1 inhibitory potential of compounds. Conversely, substitution of electron
withdrawing groups (compounds 25–31 and 40–46) drastically improved the BACE-1
inhibition. The mono-substituted 4-Cl and 4-F showed higher inhibition (28: IC50 = 1.7 µM;
30: IC50 = 1.5 µM; 43: IC50 = 0.66 µM; and 45: IC50 = 0.55 µM) compared to their 2,4-
disubstituted counterparts (29: IC50 = 3.9 µM; 31: IC50 = 3.5 µM; 44: IC50 = 3.0 µM; and 46:
IC50 = 2.9 µM). The potent BACE-1 inhibitors 25, 26, 40, 41, 43, and 45 presented the IC50
values 0.73 µM, 0.28 µM, 0.43 µM, 0.22 µM, 0.66 µM, and 0.55 µM, respectively. The
higher BACE-1 inhibitory potential in these compounds might be attributed due to their
significant H-bonding interactions with the catalytic dyad residues Asp32 and Asp228
validated by the outcome of docking and dynamic simulations studies. Common compounds
25, 26, 40, and 41 with significant and balanced inhibitory potential against both the target
enzymes (AChE and BACE-1) were selected for further screening.
298
299
300
2.3.1.3 PAMPA-BBB assay
The blood-brain barrier (BBB) permeability is one of the essential prerequisites to target the
compounds for the treatment of AD. The BBB permeability of the potent inhibitors (25, 26,
16

ACCEPTED MANUSCRIPT
301
302
303
304
305
306
307
308
309
310
311
312
313
40, and 41) on AChE and BACE-1 were assessed by PAMPA-BBB assay [34]. The method
involved the use of lipid-infused parallel artificial membrane through which the test
compounds permeate from a donor compartment to the acceptor compartment [35]. The
permeability of nine commercial drugs as reported in our previously published report was
used as a benchmark to establish the correlation between experimental Pe(exp) and reference
permeability Pe(ref) [36]. The CNS permeability cut-off values indicated that compounds with
permeability (Pe) over 4.3926×10-6 cm.s-1 suggest excellent brain penetrability, whereas with
Pe less than 1.7766×10-6 cm.s-1 has poor brain permeation. The permeability value in between
them was considered as uncertain. The results of assay (Table 2) suggested that all the tested
compounds 25 (Pe = 4.9863×10-6 cm.s-1), 26 (Pe = 6.243×10-6 cm.s-1), 40 (Pe = 4.4069×10-6
cm.s-1), and 41 (Pe = 6.0902×10-6 cm.s-1) have appreciable and high brain permeability. The
higher brain permeability of compounds 26 and 41 could be attributed to the presence of
highly lipophilic trifluoromethyl (4-CF3) substituent at terminal phenyl group.
314
315
316
317
318
319
320
321
322
323
324
325
2.3.1.4 Propidium iodide displacement assay
Compounds 25, 26, 40, and 41 were the most potent inhibitors of AChE with high brain
permeability characteristics in PAMPA-BBB assay. Also, molecular docking and dynamic
simulations suggested significant interactions with PAS residues. Therefore, the affinity of
these ligands at 10 and 50 µM concentrations for the PAS-binding was evaluated by
propidium iodide displacement assay (Table 2). Propidium iodide is a known AChE ligand
binding specifically at its PAS and enhances the fluorescence intensity up to eight-fold [37].
The binding of inhibitors to PAS displaces propidium iodide and resulted into decreased
fluorescence intensity. Compounds 40 and 41 exhibited higher displacement of propidium
iodide (40: 10 µM = 23.6%, 50 µM = 33.8%; 41: 10 µM = 27.2%, 50 µM = 37.6%)
compared to donepezil (10 µM = 22.6%; 50 µM: 33.6%). Compounds 25 and 26 appeared to
have considerably less capability in displacing the propidium iodide from PAS-AChE (25: 10
17

ACCEPTED MANUSCRIPT
326
327
328
329
330
331
µM = 15.0%, 50 µM = 20.0%; 26: 10 µM = 12.4%, 50 µM = 19.3%). The results of the assay
were found to be in agreement with computational studies.
Considering the balance between inhibitory potential against both the targets (AChE and
BACE-1), along with the significantly high displacement of propidium iodide from PAS-
AChE, compound 40 and 41 were chosen for further assessments.
Table 2. The results of PAMPA-BBB and propidium iodide displacement assay.
Propidium iodide displacement from
PAMPA-BBB permeability
AChE PAS (% inhibition)b
Comd.
Permeability prediction
Pe(exp) (10-6 cm.s-1)
At 10 µM
At 50 µM
(CNS+a)
CNS+
CNS+
CNS+
CNS+
CNS+
4.9863 ± 0.2
6.2433 ± 0.3
4.4069 ± 0.1
6.0902 ± 0.3
6.8612 ± 0.2
15.0±1.5
12.4±1.3
23.6±2.5
27.2±2.9
22.6±1.7
20.0±1.2
19.3±1.4
33.8±3.1
37.6±3.0
33.6±1.7
25
26
40
41
donepezil
332
333
334
a‘CNS+’ (prediction of high BBB permeation); Pe (10-6 cm.s-1) > 4.3926. bAll the results are reported as mean ±
SEM (n=3); Propidium iodide displacement assay was performed on AChE to test the ability of compounds to
displace propidium with reference to the donepezil at 10 and 50 µM.
335
336
337
338
339
340
341
342
343
344
345
346
2.3.1.5. Self- and AChE-induced Aβ aggregation inhibition by thioflavin T assay
The results of propidium iodide displacement assay suggested significant binding of 40 and
41 at PAS- AChE. The AChE was also demonstrated to be involved in the promotion of Aβ
aggregation by specifically binding to PAS [30, 38]. Thus, AChEIs bound to the active site
and PAS could not only result in AChE inhibition but also prevent the Aβ production and
deposition. Therefore, thioflavin T based fluorometric assay was performed on compounds
40 and 41 to ascertain their PAS-AChE binding to be beneficial in preventing Aβ
aggregation. The test was conducted at three different ratios of Aβ: inhibitors (10:5, 10:10,
and 10:20 µM). The results were reported as % Aβ aggregation inhibition (Figure S32A and
S32B, Supporting Information) and normalized fluorescence intensity (NFI) (Figure 6A and
6B). Tested inhibitors showed higher anti-aggregatory property in self- (40: 18.5%–44.1%;
41: 26.1%–50.1%) and AChE-induced (40: 31.0%–69.3%; 41: 61.1%–89.0%) experiments
18

ACCEPTED MANUSCRIPT
347
348
349
compared to donepezil (Self-induced: 16.7%–41.6%; AChE-induced: 28.5%–62.8%). The
results confirmed the PAS-AChE binding ability of these inhibitors and are in concurrence
with the outcome of molecular docking and dynamics simulations.
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
2.3.1.6 AFM studies
AFM experiments were conducted to visualize high-resolution images of surface
topographical maps and calculate the nanometer size ranges of aggregates [39]. The changes
in morphological characterization of Aβ aggregates owing to the presence or absence of
inhibitor (41) were observed at different time intervals over 7 days of incubation period (0, 3,
5, and 7 days). Aβ aggregates, fibrils, and large oligomer formation were visually confirmed
through AFM images. Further, maximum peak height and roughness index of samples were
calculated to validate the inhibitory potential of compound 41. Aβ1-42 without inhibitor
resulted into formation of large oligomers and fibrils. The AFM histogram analysis of Aβ1-42
without inhibitor showed to increase the maximum peak height of aggregates from 23.72–
226.55 nm owing to its aggregation during 7 days of the incubation period (Figure 6C). Also,
the roughness index was observed to be very high in the range of 2.33–24.36 nm. The
addition of compound 41 halted the aggregation of Aβ1-42 with the maximum peak height of
aggregates was reduced drastically and observed in the range of 44.46–104.01 nm (Figure
6C). The roughness index was also decreased remarkably to 1.70–4.40 nm in the presence of
inhibitor. The results of the AFM analysis confirmed the significant inhibition of Aβ
aggregation by compound 41.
19

ACCEPTED MANUSCRIPT
[A]
[B]
1.00
0.75
0.50
0.25
0.00
1.00
0.75
0.50
0.25
0.00
41 + Aβ
(µM)
41 + Aβ
(µM)
DONE + Aβ
(µM)
DONE + Aβ
(µM)
40 + Aβ
(µM)
40 + Aβ
(µM)
368
369
370
Figure 6. Effect of tested compounds on Aβ aggregation. [A] Self-induced and; [B] AChE-
induced Aβ aggregation. Bars are showing the normalized fluorescence intensity (NFI) as
mean ± SEM of three separate experiments. [C] AFM images of Aβ1-42 aggregates (10 µM)
incubated with or without inhibitor (41) at different time intervals (0, 3, 5, and 7 days).
DONE = Donepezil. Images were visualized at 5 × 5 µM using the Nova Px image analysis
software (NT-MDT, Russia).
371
372
373
374
375
376
377
378
379
380
381
382
2.3.1.7 Neurotoxicity studies on SH-SY5Y cell lines
The neurotoxicity liability of 40, 41, and donepezil was evaluated against human
neuroblastoma SH-SY5Y cell lines following the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-
tetrazolium bromide (MTT) assay. The results suggested that compound 40 (Figure S33A,
Supporting Information) and 41 (Figure S33B, Supporting Information) exposed to
neuroblastoma cell lines SH-SY5Y exhibited non-significant reduction in cell viabilities up to
the maximum tested concentration of 80 µM. The percentage cell viabilities at tested
20

ACCEPTED MANUSCRIPT
383
384
385
386
concentrations were observed in the range of 76.3%–96.3% and 84.6%–99.2% for
compounds 40 and 41, respectively. The results asserted that SH-SY5Y neuroblastoma cell
lines were viable against both the tested compounds in the range of 10–80 µM
concentrations.
387
2.3.2 Behavioral studies
388
389
390
391
392
2.3.2.1 Acute toxicity studies
The acute toxicity was evaluated as per OECD guidelines 423 on healthy female Wistar rats.
Compounds 40 and 41 were well tolerated and showed no sign of abnormal reactions or
toxicity in the doses of 100–500 mg/kg, p.o. up to 14 days of testing. The results suggested
that tested compounds possessed significant margin of safety.
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
2.3.2.2 Scopolamine-induced amnesia model for testing cognition enhancement in rats
Behavioral studies to evaluate the cognition and memory for compounds 40 and 41 were
performed by the scopolamine-induced elevated plus maze and Y-maze models. Scopolamine
is a gold standard drug, and its administration is the most effective way for inducing
cognitive impairment due to the cholinergic deficit [40]. Test compounds were daily
administered daily to healthy male Wistar rats at the dose of 2.5, 5, and 10 mg/kg, p.o.
On the first three days of trials, elevated plus maze experiment was conducted to assess the
spatial working memory in rats after intraperitoneal administration of scopolamine [41, 42].
The treatment was administered 30 min prior to scopolamine injection and transfer latency
and time spent in open arms were calculated. The results were observed in accordance with
the in vitro studies and showed an anticipated reversal of scopolamine-induced dementia by
compounds 40 and 41. The control group has shown significantly lowered transfer latency
and time spent in open arms compared to all other groups. The scopolamine group
significantly prolonged the transfer latency (Figure 7A, p < 0.001) and time spent in open
arms (Figure 7B, p < 0.001) compared to control group indicating memory and learning
21

ACCEPTED MANUSCRIPT
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
impairment. Donepezil at the dose of 5 mg/kg significantly decreased the transfer latency
(Figure 7A, p < 0.001) and time spent in open arms (Figure 7B, p < 0.001) compared to the
scopolamine group. Compounds 40 and 41 at all the dose levels (2.5, 5, and 10 mg/kg)
showed significantly lowered transfer latency (Figure 7A, p < 0.001) and time in open arms
(Figure 7B, p < 0.001) compared to scopolamine group. Interestingly, 40 (10 mg/kg) and 41
(5 and 10 mg/kg) showed a statistically non-significant difference compared to donepezil.
On the seventh day of trial, the Y-maze test was performed to evaluate the effect of
compounds (40 and 41), and spontaneous alternations were calculated, which is a measure of
spatial working memory [43, 44]. The % spontaneous alternations was significantly reduced
in the scopolamine group of animals (Figure 7C, p < 0.001) compared to control group
indicative of induction of memory and learning impairment. Donepezil (5 mg/kg) showed
significantly increased spontaneous alternations (Figure 7C, p < 0.001) compared to
scopolamine group. Similar to the elevated plus maze experiment, 40 (10 mg/kg) and 41 (5
and 10 mg/kg) showed the statistically non-significant difference in % spontaneous
alternations compared to donepezil group. The total arm entries (Figure 7D) remained
unchanged in all groups suggested that scopolamine did not hamper the locomotor activity in
animals. The overall results of scopolamine-induced models advocated the potential of
compounds 40 and 41 in improving spatial and immediate memory.
22

ACCEPTED MANUSCRIPT
[B]
[A]
ns
***
ns
90
75
60
45
30
15
0
80
###
###
60
40
20
0
πππ
***
πππ
***
π
***
ππ
πππ
***
***
***
***
***
***
***
***
***
***
41 (mg/kg)
40 (mg/kg)
41 (mg/kg)
40 (mg/kg)
Scopolamine (0.5 mg/kg)
Scopolamine (0.5 mg/kg)
[C]
[D]
ns
15
10
5
80
***
***
***
ππ
60
40
20
0
***
ππ
***
***
πππ
***
###
0
41 (mg/kg)
41 (mg/kg)
40 (mg/kg)
40 (mg/kg)
Scopolamine (0.5 mg/kg)
Scopolamine (0.5 mg/kg)
426
427
Figure 7. Effect of compounds 40, 41, and donepezil on scopolamine-induced cognition and
memory improvement. [A] Transfer latency and; [B] Time spent in open arms during the
elevated plus maze test; [C] Spontaneous alternations (%) and; [D] Total number of arm
entries in the Y-maze experiment. All results are expressed as mean ± SEM (n = 6). ###p <
428
429
430
431
432
ππ
0.001 versus control; ***p < 0.001 versus scopolamine; ns p > 0.05, πππp < 0.001, p < 0.01,
πp < 0.05 versus DONE5 (donepezil 5 mg/kg).
433
434
435
436
437
438
439
2.3.2.3 Ex vivo and biochemical analysis
The effect of 40 and 41 on brain AChE level was evaluated by ex vivo study as per Ellman’s
method [32]. The results showed a significantly higher rate (Figure 8A, p < 0.001) of
substrate hydrolysis or elevated levels of AChE in the scopolamine- group of animals
compared to control group. The levels of AChE were remarkably attenuated (Figure 8A, p <
0.001) by treatment of 40 and 41 at the dose of 10 mg/kg compared to the scopolamine
group. The ex vivo brain AChE inhibitory potential of both the compounds at the tested
23

ACCEPTED MANUSCRIPT
440
441
442
443
444
445
446
447
448
449
450
451
452
dosage was found to be statistically non-significant in comparison to standard donepezil.
These results also reflected the ability of compounds to permeate the BBB.
The antioxidant potential of compounds was elucidated by estimation of oxidative
biochemical markers such as malonaldehyde (MDA, lipid peroxidation by-product) and
–
superoxide dismutase (SOD, involved in dismutation of O2 radicals). Biochemical analysis
showed that scopolamine group has significantly elevated the levels of MDA (Figure 8B, p <
0.001), while the levels of SOD (Figure 8C, p < 0.001) were considerably reduced compared
to control group. However, the treatment of 40 and 41 at 10 mg/kg dose attenuated the levels
of MDA (Figure 8B, p < 0.001), while significantly increasing the levels of SOD (Figure 8C,
p < 0.001) compared to scopolamine group. Also, the results suggested that donepezil has
non-significant antioxidant potential. Overall, the results of ex vivo and biochemical analysis
suggested 41 to possess significant brain AChE inhibitory potential along with antioxidant
property.
[A]
[B]
[C]
0.008
0.006
0.004
0.002
0.000
4
3
2
1
0
20
15
10
5
###
###
###
***
***
***
###
###
###
***
##
***
***
###
#
***
###
**
###
###
0
Control
DONE5
40
41
Control
DONE5
40
41
Control
DONE5
40
41
Scopolamine (0.5 mg/kg)
Scopolamine (0.5 mg/kg)
Scopolamine (0.5 mg/kg)
453
454
Figure 8. Results of ex vivo AChE and biochemical analysis. [A] AChE activity- rate of
substrate hydrolyzed; [B] TBARS assay- levels of MDA; [C] Superoxide dismutase assay-
levels of SOD units. DONE5- donepezil (5 mg/kg); Values are expressed as mean ± SEM (n
455
456
457
458
#
**
*
= 6). ###p < 0.001, ##p < 0.01, p < 0.5 versus control; ***p < 0.001, p < 0.01, p < 0.05
versus scopolamine.
459
460
461
462
2.3.2.3 Aβ1-42-induced ICV rat model: Morris water maze test
Intracerebroventricular (ICV) administration of Aβ1-42 is one of the most promising method to
develop learning and memory deficits along with the AD-like phenotypic condition [45]. Aβ
is the major constituent of senile plaques in the brains of AD patients [46], and its deposition
24

ACCEPTED MANUSCRIPT
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
also augmented with the increased activity of BACE-1 [47]. The effect of most potent
compound 41 was assessed in healthy Wistar rats by ICV injection of Aβ1-42 in the
hippocampal region of the brain. All the groups of animals have received the ICV Aβ1-42
injection except the sham group. After seven days of the recovery period, rats of all groups
were administered the respective treatments for nine consecutive days. During the last five
days of treatment, the Morris water maze experiment was performed to evaluate escape
latency time (ELT) and numbers of platform crossings over the period of 90 sec. The ELT
was significantly prolonged (Figure 9A, p < 0.001), while the number of platform crossings
(Figure 9B, p < 0.001) progressively declined in the model group of animals owing to Aβ1-42
administration compared to the sham group, which strongly suggested the impairment of
learning and memory. The treatment with donepezil (72.5–22.5 sec, p < 0.001) and 41 (74.5–
20.3 sec, p < 0.001) significantly decreased the ELT over the period of five days of trial
compared to model group (74.0–49.3 sec). Also, the number of platform crossings was
remarkably improved by donepezil (1.3–13.7, p < 0.001) and 41 (1.7–14.5, p < 0.001) group
with definite spatial preference in correct quadrant over five days of trials compared to model
group (1.0–5.7). Interestingly, treatment with compound 41 demonstrated remarkable
improvement in cognitive and memory functions at the tested dose.
[B]
[A]
16
90
75
12
Sham
60
Model
8
4
0
45
30
15
0
Donepezil
41
1
2
3
4
5
1
2
3
4
5
Days
Days
480
481
Figure 9. Effect of compound 41 in spatial memory improvement by ICV Aβ1-42-induced
482
483
484
Morris water maze experiment. [A] Escape latency time (ELT); [B] Number of platform
crossings in target quadrant during the last five days of trials. The values represented mean ±
SEM (n = 6).
25

ACCEPTED MANUSCRIPT
485
486
487
488
489
490
491
492
493
494
2.3.3 Pharmacokinetic studies
Preliminary pharmacokinetic investigations of compound 41 were envisaged on healthy male
Wistar rats at the dose of 10 mg/kg, p.o. The blood sampling was performed from retro-
orbital plexus at several time points (See Experimental Section). The parameters were
calculated by following the extravascular non-compartmental model using Kinetica software
(Thermo Fisher Scientific, version 5.0.11) and results are summarized in Table 3. The results
suggested that 41 has peak plasma concentration Cmax = 32.8 ± 1.7 ng/ml achieved at Tmax
=
5.3 ± 1.3 h. Compound 41 has shown elimination half-life t1/2 = 29.2 ± 2.2 h and mean
residence time MRT = 45.7 ± 2.0 h. Overall, results showed significant oral absorption of test
compound 41 after oral administration.
495
496
Table 3. The results of pharmacokinetic parameters after oral administration of 41 (10
mg/kg, p.o.).
Parameters
Cmax (ng/ml)
Tmax (h)
Effect of 41*
32.8 ± 1.7
5.3 ± 1.3
(AUC)0-24 (ng/ml*h)
633.3 ± 41.4
29.2 ± 2.2
45.7 ± 2.0
t1/2 (h)
MRT (h)
497
498
*All values are expressed as mean ± SEM (n = 3).
3. CONCLUSION
499
500
501
502
503
504
505
506
AD is a multifactorial complex disorder, and several pathophysiological mechanisms are
involved in its progression and development. Current therapeutic regimens only provide
symptomatic relief and not halt the disease progression. A single molecule- multiple target
strategy has been proven to be promising and explored for the development of effective drug
candidates to treat AD. We have identified the potential hit bearing N-benzylpiperidine
nucleus as multitarget-directed ligand against AChE and BACE-1 using e-pharmacophore
models with their co-crystallized ligands donepezil, and F1M, respectively. Several
computational tools such as phase-based screening, VS, DPP, and MM-GBSA were used to
26

ACCEPTED MANUSCRIPT
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
screen the top hit SEW06622. A series of N-benzylpiperidine analogs were rationally
designed by modifying the identified hit to improve its binding to AChE and BACE-1. The
modification of 2,3-diOCH3 functional group of SEW06622 with several electron
withdrawing and donating groups resulted into the increased binding of compounds to PAS-
AChE. Further, the modification by reduction of methanimine to methanamine leads to
significant improvement in binding of compounds to aspartate dyad of BACE-1. These
results of designed ligands were also corroborated by molecular docking and dynamics
simulation studies.
The designed series of ligands (17–31 and 32–46) were synthesized and characterized by
spectroscopic techniques (1H NMR, 13C NMR, and MS). The results of in vitro assays
indicated balanced inhibitory potential against AChE and BACE-1 in sub-micromolar ranges
by compounds 25 (AChE: IC50 = 0.90 µM; BACE-1: IC50 = 0.73 µM), 26 (AChE: IC50 = 0.44
µM; BACE-1: IC50 = 0.28 µM), 40 (AChE: IC50 = 0.72 µM; BACE-1: IC50 = 0.43 µM), and
41 (AChE: IC50 = 0.11 µM; BACE-1: IC50 = 0.22 µM). Compounds 40 and 41 showed
excellent displacement of propidium iodide from PAS-AChE, and inhibition of Aβ
aggregation in self- and AChE-induced experiments. The inhibition of Aβ was also
ascertained by morphological characterization of Aβ aggregates through AFM studies.
Compounds 40 and 41 were also found to be devoid of neurotoxicity towards SH-SY5Y
neuroblastoma cell lines even at maximum tested concentration of 80 µM. Compounds 40
and 41 ameliorated the scopolamine-induced cognitive dysfunctions in a dose-dependent
manner as evaluated by the elevated plus maze and Y-maze experiments. Ex vivo and
biochemical analysis of evaluated compounds indicated significant inhibition of brain AChE
and favorable alteration of oxidative stress biomarkers (MDA and SOD), respectively.
Moreover, 41 improved Aβ1-42-induced cognitive impairment in rats by Morris water maze
experiment and showed significant oral absorption ascertained by pharmacokinetic analysis.
27

ACCEPTED MANUSCRIPT
532
533
534
The results of in vivo experiments also substantiated the results of in vitro assays and
computational studies. Overall results indicate 41 to be the most promising multi-target
directed ligand and could be considered as a promising “lead” for the treatment of AD.
535
4. EXPERIMENTAL
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
4.1 Computational studies
All the computational studies were performed using Schrödinger Maestro 2018.1 on HP Z410
3.10 GHz workstation with Kernel Linux based operating system. Structure-based drug
designing approach was accomplished through the generation of e-pharmacophore models
using protein structures of AChE (4EY7) [27] and BACE-1 (2ZJM) [28]. Protein structures
were selected on the basis of their co-crystallized ligands bearing N-benzylpiperidine nucleus.
Briefly, both protein structures were pre-processed using Protein Preparation Wizard and
Prime modules to add missing hydrogens atoms, assign bond orders, and fill in the missing
side chains and loops. Protein structures were optimized using PROPKA pH 7.0 and
minimized using an OPLS3 force field. The prepared protein structures were used to generate
the respective e-pharmacophore models of AChE and BACE-1. Maybridge HitFinder
database was selected and screened by Phase module using generated e-pharmacophore
models and VS workflow.
The docking post-processing (DPP) function was performed to identify common ligands
interacting with active site residues of AChE (at PAS and CAS) and BACE-1 (Asp32 and
Asp228 dyad). Molecular mechanics/generalized born surface area (MM-GBSA) analysis of
screened ligands was performed to determine the binding free energy of docked ligands into
the respective proteins, and the potential top hit was identified. A series of analogs were
designed using top hit to improve their binding characteristics towards PAS-AChE and
aspartate dyad residues of BACE-1.
28