Fluoro-benzimidazole derivatives to cure Alzheimer's disease: In-silico studies, synthesis, structure-activity relationship and in vivo evaluation for β secretase enzyme inhibition

Full text of DOI:10.1016/j.bioorg.2019.102936

Bioorganic Chemistry 88 (2019) 102936
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Fluoro-benzimidazole derivatives to cure Alzheimer’s disease: In-silico
studies, synthesis, structure-activity relationship and in vivo evaluation for β
secretase enzyme inhibition
a,b
a,c,⁎
b
a
Sayyad Ali , Muhammad Hassham Hassan Bin Asad
, Soham Maity , Wahid Zada ,
c
a
b
a,⁎
Albert A. Rizvanov , Jamshed Iqbal , Borhan Babak , Izhar Hussain
a
b
c
Department of Pharmacy, COMSATS University Islamabad, Abbottabad Campus, 22060, Pakistan
Department of Chemistry, Michigan State University, East Lansing, MI 48824, USA
Department of Genetics, Institute of Fundamental Medicine and Biology, Kazan Federal University, 420021, Russia
1
. Introduction
drugs and to approach a conclusive diagnosis at its onset. Though
several BACE1 inhibitors were evaluated, however, majority proved
ineffective in clinical trials [7]. Peptidic based inhibitors (tetra-dec-
apeptide Lys-Thr-Glu-Glu-Ile-Ser-Glu-Val-Leu-Statine-Val-Ala-Glu-Phe-
OH, called STA-200) were proved effective for BACE1 inhibition
(IC50 = 30.0 nM), however, they were unable to penetrate BBB [8].
Hydroxy ethylene (HE) derivatives of homo-statine (HE bio-isostere)
were found previously to produce and mimic BACE1 inhibitory effects
(IC50 = 1.6 nM) [9]. Similarly, compound KMI-008 (IC50 = 413 nM)
was documented in literature to decrease 38% sAPPβ load from COS-7
cell-lines, although it was found to overexpress APP moieties [10].
Based on isophthalamide skeleton attached to the small amide isosteres
were remained an effective BACE1 inhibitor (IC50 = 27.2 nM) [11].
Moreover, family of pyridinyl aminohydantoins-4 derivative was re-
ported to produce an improved pose in S3 domain of BACE1
(IC50 = 20 nM) enzyme [12]. Godemann et al. (2009) enlisted below
mentioned BACE1 inhibitors, however, proved inadequate to cope AD
[13].
Amyloid-β (Aβ) accumulation is linked with the senile neurofi-
brillary tangles and plaques formation in brain stem lead to
Alzheimer’s disease (AD). It is composed of 38 to 43 amino acids and
produced by the catalytic detachment of Amyloid Precursor Protein
(
APP) [1]. APP has an amyloid beta encoding domain usually cleaved
by a sequence of proteases enzyme called secretases (Fig. 1) [2]. The
beta site amyloid precursor protein cleaving enzyme 1 (BACE1 or β-
secretase) liberates a soluble N-terminus sAPPβ, while γ- secretase
producing APP intracellular domain (AICD) and an insoluble Aβ
plaques [3]. Aβ-42 is a principal toxic kind of Aβ involved in plaques
formation due to the eminent hydrophobicity, accumulation and fi-
brillization potency that results a delay in clearance and inflamma-
tion of the brain [4].
As documented earlier, AD is a progressive neurodegenerative ir-
reparable illness cause of dementia in aged people with high mortality
rate [5]. It is marked with a continuous decline in memory, thinking,
language, behaviors, judgment, visuospatial, impaired cognitive cap-
ability and whole lifestyle. AD is devastating not only for the individual
but also for the caretaking family and community [6]. Presently, no
absolute treatment is available for AD, however, intended therapeutics
used for symptomatical treatment have short-term benefits [1]. The
development of medications for AD has been restricted due to the
deficit in skills to monitor the advancement in disease, effectiveness of
Eli Lilly and Co. introduced LY2886721 after phase-II clinical trial,
nevertheless revealed hepatotoxic in nature [14]. Recent failure rate in
finding and screening of BACE1 inhibitors has encouraged to continue
our efforts in the quest of small molecules (fragments) having ther-
apeutic efficacy for AD. BACE1 inhibitors are meant to arrest Aβ pla-
ques accumulation and to prevent further proliferation of AD. Current
advances in computational approaches have been extensively used to
⁎
Corresponding authors at: Department of Genetics, Institute of Fundamental Medicine and Biology, Kazan Federal University, 420021 Kazan, Russia and
Department of Pharmacy, COMSATS University Islamabad, Abbottabad Campus, 22060, Pakistan.
E-mail addresses: hasshamasad@cuiatd.edu.pk (M.H.H.B. Asad), Izharhussain@cuiatd.edu.pk (I. Hussain).
https://doi.org/10.1016/j.bioorg.2019.102936
Received 23 November 2018; Received in revised form 7 April 2019; Accepted 15 April 2019
Available online 24 April 2019
0045-2068/ © 2019 Elsevier Inc. All rights reserved.

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
Fig. 1. The APP is initially cleaved by the enzyme β-secretase (BACE1) at the N-terminus of the Aβ domain. This cleavage produces the soluble sAPPβ and a C-
terminal portion, which experiences a second cleavage by another protease called γ-secretase, which cleaves the transmembrane domain of APP. The final production
of Aβ is the main player of the pathogenesis of AD. In the above process β-secretase (BACE1) is rate-limiting enzyme for the initiation of the AD pathogenesis.
discover and optimize novel entities with an affinity for a targeted
enzyme. In silico paradigm represents a rich array of possibilities in
expediting new targets identification for newly discovered compounds
with proposed activity against those target [7]. Virtual screening of
enzyme-ligand interaction (docking), synthesis of fluoro-benzimidazole
analogs (SAR), βsecretase activity (in-vitro assay) and in-vivo (animal
model) studies was the outlined in the present manuscript.
2.2. Docking study
Results from molecular docking deduced reliable bonding of ben-
zimidazoles derivatives with the aspartic acid portions (active domain)
of BACE1 as shown in Table 5. The most potent compound 7C was
found to pose the best bonding with two key aspartic acids of an active
pocket of the BACE1 enzyme. Complete detail about structure of BACE1
and flap region along with optimum polar interaction with the most
potent compound 7c is shown in Fig. 2.
2
. Results and discussion
2.3. FRET activity
2.1. Synthesis
Benzimidazoles derivatives were found to show different BACE1
The condensation reaction of o-phenylenediamine with the re-
inhibitory effects (in vitro). Compound 7c was found the most potent
spective substituted aromatic aldehyde resulted in the synthesis of
seven compounds (∼90% yield) of series of aryl benzimidazoles. The
synthesized compounds were given chemical names and assigned
codes. These compounds were 2-(4-chlorophenyl)-1H-benzo[d]imida-
zole (6a), 2-(4-fluorophenyl)-1H-benzo[d]imidazole (7a), 2-(2,3,5-tri-
(
IC50 = 510 nM) and inhibited 98% BACE1 activity at 500 pmol. 6a was
noted the least potent among all having IC50 = 312.3 µM with 39%
inhibitory effect of the same concentration. Rest of all compounds 9a
(
IC50 = 1.3 μM, 72% inhibition), 10a (IC50 = 5.6 µM, 65% inhibition),
8
a (IC50 = 15.3 µM, 57% inhibition), 8b (IC50 = 19.5 µM, 52% inhibi-
tion), 7b (IC50 = 93.3 µM, 55% inhibition), 7a (IC50 = 112.3 µM, 80%
inhibition), 9b (IC50 = 117.3 µM, 58% inhibition), 8c
fluorophenyl)-1H-benzo[d]imidazole
fluorophenyl)-1H-benzo[d]imidazole
(7b),
(7c),
6-fluoro-2-(2,4,5-tri-
2-(3-(trifluoromethyl)
phenyl)-1H-benzo[d]imidazole (8a), 6-fluoro-2-(3-(trifluoromethyl)
phenyl)-1H-benzo[d]imidazole (8b), 6-Fluoro-2-(4-(trifluoromethyl)
phenyl)-1H-benzo[d]imidazole (8c). Among all 7c was found the most
potent, however, ClogP, Lipinski, BBB permeation and IC50 was calcu-
lated for each compound, and summarized in Table 1. Synthesis of
benzimidazoles with more substituted fluorine molecules were 2-(tri-
(
IC50 = 123.7 µM, 60% inhibition) were found in between in efficiency.
A complete list of compounds is mentioned in Tables 1–3 along with
their chemical names, structures and IC50 values.
2.4. Morris water maze assay
fluoromethyl)-1H-benzo[d]imidazole
(9a)
and
6-fluoro-2-(tri-
fluoromethyl)-1H-benzo[d]imidazole (9b). Complete detail about their
ClogP, Lipinski, BBB permeation and IC50 is highlighted in Table 2.
Table 3 described above mentioned parameters about 6-fluoro-1-(4-
In vivo studies designed to map out the efficacy of newly synthesized
compounds via behavioral tests. Morris water maze assay showed that
aluminum chloride inducted neurotoxicity to the mice exhibited raised
escape latency (35.70 ± 3.95) in comparison to the control mice
(15.60 ± 1.01) on the fifth day. The treated mice showed improved
learning curves and animals showed better spatial learning and memory
fluorobenzyl)-2-(2,4,5-trifluorophenyl)-1H-benzo[d]imidazole
com-
pound (10a). Moreover, Table 4 describes detailed NMR information
about the structure elucidation of eight benzimidazoles analogs, how-
1
13
19
ever, all NMR ( H NMR, C NMR and F NMR) spectra are available as
supplementary material (Figs. 1S−21S).
(20.99 ± 3.56; p < 0.05) when compared to the AlCl treated group.
3
The compound 7c treated mice took more time during exploring the
2

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
Table 1
Represents the synthesis of benzimidazoles derivatives from NaHSO
3
and o-phenylenediamine/halo-o-phenylenediamine with different aromatic aldehydes in
presence of DMF at 80 °C resulted in impressive yield of compounds with their IC50 values.
Code
Diamine
Fluro-aryl-aldehyde
Product (% yield)
ClogP, (Lipinski), BBB permeation, (other parameters)
IC50 BACE1
313.3 µM
6
a
3.75, 0 violation, Yes
(
95)
(Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 1)
7
a
b
3.25, 0 violation, Yes
112.3 µM
93.3 µM
(
(
98)
(
Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 2)
7
3.89, 0 violation, Yes
Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 4)
(
(
95)
7
c
a
4.11, 0 violation, Yes
Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 5)
510 nM
15.5 µM
(
95)
8
3.96, 0 violation, Yes
Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 4)
(
(
92)
8
b
4.23, 0 violation, Yes
Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 5)
19.5 µM
(
(
90)
8
C
2.20, 1 Violation, No,
Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 5)
123.7 µM
(
89)
(
target quadrant (31.10 ± 1.23) than the AlCl
3
models. The Alzheimer’s
Table 3
Represents
models crossed the said quadrant (5.23 ± 1.13) and the control has
crossed quite often the quadrant as compare to (12.43 ± 1.87) and the
treated animal with our synthesized compound gave results in between
the two described groups (9.32 ± 1.24). Overall results are expressed
in Fig. 3.
the
synthesis
of
6-fluoro-1-(4-fluorobenzyl)-2-(2,4,5-tri-
fluorophenyl)-1H-benzo[d]imidazole with benzyl chloride along with IC50
value.
Code Benzimidazole
Product
% Yield ClogP, (Lipinski),
IC50
derivative
BBB permeation
BACE1
1
0a
88
3.90, 1 violation,
No,
5.6 µM
2
.5. Novel object recognition assay
(
Lowly absorbed
by GIT;
Novel object recognition assay demonstrated no significant differ-
HB_Donor: 0;
HB_acceptor: 6)
ence in time spent by the mice in exploring of two identical objects in
the training session. In the testing phase, the negative control mice
explored the novel object for a longer time period with recognition test
(
RT) (67%) while positive control (AD model) and AD-mice treated
previously to develop computationally or synthesized BACE1 inhibitors,
however, a dire need has been felt nowadays [17]. Active domain of
BACE1 is huge, open, lipophilic and possess flexible conformation
usually holds a flexible β-hairpin loop (67–77 residues flap-region) to
position the substrate for catalytic activity of an enzyme optimally
[18,19]. BACE1 inhibitors have rigid hydrogen bonding framework
which occupied S1 to S3 binding domain and found to involve in es-
sential H-bonding along with hydrophobic-interactions. Perhaps 7c
showed inhibitory potential by following this principle (Fig. 2) which
accurately docked within the specified flexible binding domain. Back-
bone movements of the potential ligand(s) was due to the energy
minimization to reduce the intrinsic property of rigidity of both the
with the synthesized compounds showed a considerably lower RT value
at 45.8% and 57% respectively (Fig. 4). The average exploration time
and the recognition index was indicated significant at p < 0.0002,
however, 7c was found the best to enhance the three-dimensional
spatial learning/memory of AlCl induced model at p < 0.05.
3
Proteases BACE1 (β-secretase) is attributed for amyloid beta peptide
(
Aβ42) formation led to senile plaques accumulation common in
Alzheimer’s disease [15]. BACE1 is more prone to produce Aβ42 at all
level of the nervous system [16]. Being a substrate for BACE1 APP led
to rapid proteolytic action end up with the progression of AD. These
circumstances reinforced the amyloid cascade hypothesis and propose
BACE1 as a potential therapeutic target. Several attempts were made
Table 2
Represents the synthesis of benzimidazoles with substituted fluorine molecules with their IC 50 values.
Code
Diamine
Product
% Yield
ClogP, (Lipinski),
IC50
BBB permeation
BACE1
9
a
b
95
2.98, 0 violation, Yes,
1.3 µM
(
Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 4)
9
88
2.30, 0 violation, Yes,
Highly absorbed by GIT; HB_Donor: 1; HB_acceptor: 5)
117.3 µM
(
3

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
Table 4
Detailed NMR spectroscopic description to configure structures of newly synthesized compounds.
Sr. No
Code of
Compound name
Spectroscopic detail
compound
1
1
6a
7a
7b
7c
8a
2-(4-chlorophenyl)-1H-benzo[d]imidazole
Yield: 95%, Column chromatography [Ethylacetate:hexane (1:9), Rf = 0.8]; H NMR (500 MHz,
DMSO‑d
J = 7.8 Hz), 6.6 (2H, d, J = 7.8 Hz); C NMR (125 MHz, DMSO-d
6
): δ 11.20 (1H, s), 7.7 (2H, dd, J = 7 Hz), 7.15 (2H, d, J = 7 Hz), 6.9 (2H, dd,
13
6
): δ 150.1, 143.4 (2C), 131.6
+
(
2C), 133.6 (2C), 130.6 (2C), 123.2 (2C), 123.2 (2C); HR-EI MS: m/z 228.04539; [(M + 1)
9 2
Calcd for C13H ClN 228.04543]
1
2
3
4
5
2-(2,3,5-trifluorophenyl)-1H-benzo[d]imidazole
2-(2,3,5-trifluorophenyl)-1H-benzo[d]imidazole
Yield: 98%, Column chromatography [Ethylacetate:hexane (1:9), Rf = 0.9]; H NMR (500 MHz,
1
3
6
DMSO-d ): δ 10.5 (1H, s), 7.5 (4H, q), 7.1 (4H, q); C NMR (125 MHz, DMSO): δ 167.7 (d
19
J
C-F = 58.2 Hz), 156.0, 142.6 (2C), 134.4, 128.0 (2C), 121.6 (2C), 116.6 (4C); F NMR
+
(
470 MHz, DMSO-d
6
): δ −60.1; HR-EI MS: m/z 212.07492; [(M + 1) Calcd for C13
H
9
FN
2
2
12.07498]
1
Yield: 98%, Column chromatography [Ethylacetate:hexane (0.5:9.5), Rf = 0.9]; H NMR
(
6
500 MHz, DMSO-d ): δ 12.4 (1H, s), 8.2 (1H, q), 7.8 (1H, q), 7.7 (1H, d, J = 7.5 Hz), 7.6 (1H,d,
13
J = 8 Hz), 7.2 (2H, q); C NMR (125 MHz, DMSO): δ 169.9, 169.2, 161.2, 154.6 (2C), 144.3
+
(
2C), 133.5, 124.4 (2C), 109.8 (3C); HR-EI MS: m/z 266.04667; [(M + 1) Calcd for C13
H
7
F
3
N
2
2
66.04671]
Yield: 95%, Column chromatography [Ethylacetate:hexane (1:9), Rf = 0.85]; H NMR (500 MHz,
DMSO-d ): δ 12.5 (1H, NH, s), 7.9 (1H, d, J = 8.65 Hz), 7.6 (1H, d, J = 7 Hz), 7.4 (1H, s), 7.2 (1H,
d, J = 4.5 Hz), 6.9 (1H, t, J = 9 Hz); C NMR (175 MHz, DMSO-d ): δ 140.5(1C), 128.6 (1C),
27.6 (1C), 124.5(1C); F NMR (500 MHz, DMSO-d ) δ −115.0 (2C), −131.0, −141.7; HR-EI
MS: m/z 266.04665; [(M + 1) + Calcd for C13 266.04671]
Yield: 92%, Column chromatography [Ethylacetate:hexane (1:9), Rf = 0.8]; H NMR (500 MHz,
DMSO-d ): δ 12.4 (1H, NH, s), 8.5 (2H, t, J = 7.5 Hz), 8.0 (1H, s), 7.7 (2H, s), 7.3 (2H, t,
J = 3 Hz), 5.8 (1H, s); C NMR (175 MHz, DMSO-d ): δ 162.3, 149.6, 142.5 (2C), 133.7, 131.1,
1
5-fluoro-2-(2,3,5-trifluorophenyl)-1H-benzo[d]
imidazole
6
1
3
6
1
9
1
6
6 4 2
H F N
1
2-(3-(trifluoromethyl) phenyl)-1H-benzo[d]
imidazole
6
1
3
6
1
9
1
30.2, 129.9, 128.5 (d, J = 272.8 Hz), 126.2 (2C, d, J = 3.2 Hz), 122.8, 122.5 117.4, 113.4;
F
NMR (470 MHz, DMSO-d
62.07178]
6
9 3 2
) δ −61.3; HR-EI MS: m/z 262.07172; [(M + 1) + Calcd for C14H F N
2
1
6
7
8
9
8b
8c
5-fluoro-2-(3-(trifluoromethyl) phenyl)-1H-
Yield: 90%, Column chromatography [Ethylacetate:hexane (1:9), Rf = 0.92]; H NMR (500 MHz,
benzo[d]imidazole
DMSO-d
.1 (1H, s), 6.9 (1H, t, J = 7.5 Hz), 5.2 (1H, s); F NMR (470 MHz, DMSO-d
HR-EI MS: m/z 280.06228; [(M + 1) + Calcd for C14 280.06236]
Yield: 88%, Column chromatography [Ethylacetate:hexane (1:9), Rf = 0.9]; H NMR (500 MHz,
DMSO-d ): δ 12.5 (1H, NH, s), 8.2 (1H, s), 8.2 (1H, d, J = 8 Hz), 8.0 (1H, s), 7.5 (2H, m), 7.2 (1H,
t, J = 8 Hz), 6.9 (1H, m); F NMR (470 MHz,DMSO-d ) δ −62.9, −119.9; HR-EI MS: m/z
80.06228; [(M + 1) + Calcd for C14 280.06236]
Yield: 95%, Column chromatography [Ethylacetate:hexane (2:8), Rf = 0.8]; H NMR (500 MHz,
6
): δ 11.2 (1H, NH, s), 8.2 (1H, t, J = 8 Hz), 8.0 (1H, s), 7.5 (1H, s), 7.4 (1H, t, J = 10 Hz),
1
9
7
6
) δ −61.3, −120.9;
8 4 2
H F N
1
6-Fluoro-2-(4-(trifluoromethyl)phenyl)-1H-
benzo[d]imidazole
6
1
9
6
2
8 4 2
H F N
1
9a
2-(trifluoromethyl)-1H-benzo[d]imidazole
1
9
DMSO-d
6
) δ 13.2 (1H, NH, s), 7.71 (2H, t, J = 2.5 Hz), 7.3 (2H, d, J = 7.5 Hz); F NMR
(
470 MHz, DMSO-d
6
7 5 3 2
): δ −119.2; HR-EI MS: m/z 204.03099; [(M + 1) + Calcd for C H F N
2
04.03106]
1
9b
10a
5-fluoro-2-(trifluoromethyl)-1H-benzo[d]
Yield: 88%, Column chromatography [Ethylacetate:hexane (2:8), Rf = 0.85]; H NMR (500 MHz,
1
9
imidazole
DMSO-d
NMR (470 MHz,DMSO-d
204.03106]
6
): δ 11.3 (1H, NH, s), 7.7 (1H, q), 7.5 (1H, d, J = 8.5 Hz), 7.2 (1H, td. J = 2 Hz);
F
6
) δ −62.9, −75.1; HR-EI MS: m/z 204.03099; [(M + 1) + Calcd for
7 4 4 2
C H F N
1
1
0
6-Fluoro-1-(4-fluorobenzyl)-2-(2,3,5-
Yield: 88%, Column chromatography [Ethylacetate:hexane (1:9), Rf = 0.8]; H NMR (500 MHz,
trifluorophenyl)-1H-benzo[d]imidazole
DMSO-d
J = 8 Hz), 7.2 (5H, m); F -NMR (470 MHz,DMSO-d
m/z 374.08417; [(M + 1) + Calcd for C20 374.08424]
6
): δ 12.6 (1H, NH, s), 8.2 (1H, m), 7.8 (1H, m), 7.7 (1H, d. J = 8 Hz), 7.6 (1H, d.
1
9
6
) δ −116.5, −132.1, −142.8; HR-EI MS:
11 5 2
H F N
Note: Abbreviations used to interpret NMR spectra were: s, singlet; dd, doublet of doublets; m, multiplet; d, doublet; t, triplet; q, quartet; quint, br.t, broad triplet;
quint, quintet; sep, septet; dist., distorted; sex, sextet.
moieties [20]. Optimum polar interaction with both of the catalytic
aspartates (Asp32 & Asp 228) aligned the benzimidazole scaffold to-
ward S2 therefore bound via van der Waals interactions with Val 69,
Ile126 and Tyr 198 amino acids. Overall geometry, posing of scaffold
and optimum polar interaction within the active domain of 7c con-
ferred it as a lead to explore as a targeted BACE1 inhibitor [20]. Mo-
lecular docking described in detail about compound ‘7c’ from lowest
possible energy level to the higher state in bound form. Optimum
binding energy with aspartate at the catalytic domain was observed at
the lowest energy level. Lower optimum binding led to better bioac-
tivity along with strong polar interaction. No substantial deviation was
observed from first one and lower possible energy binding pose when
compared with root means square deviation lower bonded (rmsd l.b)
value equal to zero. Compound 7c had optimum binding at others poses
above rank 01 and were proved parallel in-silico and in-vitro studies
showed significant inhibitory activity. Fluorine for molecular me-
chanics calculations and in docking was adopted based on fluorine
chemistry which has tremendous role in creation of effective BACE1
inhibitors besides US patent [US15751765] [22]. Attachment of
fluorine at the aldehyde ring successfully enhanced the activity along
with marked bioavailability at the target site (7a & 7b). Introduction of
an extra fluorine atom on diamine nucleus conferred more potent en-
zyme inhibition [17,23]. A marked decline in activity was observed
with -CF3 moiety when shifted from Cβ to Cγ along with poor blood
brain barrier penetration value. Attached of trifluoro- moiety to the
benzimidazole ring led to optimum activity for BACE1 inhibition. In-
serting fluorine atoms on the benzene or carbon atom as in case of 7c
gave us pretty remarkable results (bioavailability enhanced at the
target site depicted by the virtual screening) [23,24]. The most active
compound 7c was reacted with 4-fluorobenzyl chloride while resulted
product was unable to show optimum attachment with aspartate,
however, in vitro results were more promising. The molecular docking
results showed that the ligand that already contained the two aryl
groups had a versatile level of inhibitory activity with different
(
Tables S1–S2) [21].
Present article enlisted ten synthesized compounds treated against
BACE1 enzyme. Structure activity relationship based on the data-set
was developed. Substitution of hydrogen atom with chlorine at Cγ
4

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
Table 5
Bonding of benzimidazoles derivatives with the aspartic acid (active domain) of BACE1, depicted as in red sticks, generated from AutoDock Vina and MOE molecular
docking software.
Codes
Ligand shown in the active(red) pocket of the
Ligand surrounded by amino acids including the active
The active domain is more highlighted as b elow with
entire enzyme.
members(aspartic acid 32/and 228)
bonds
7
7
a
c
8
a
8
8
b
c
9
9
a
b
1
0a
numbers/and positions with the attached fluorine moieties [24,25].
Morris water maze test remained as a typical technique for studying
the spatial memory/learning in rodents and is generally applied to
examine the deficits in behaviors in AD animal models. Despite this,
launching the water maze experiment in precise infective agents’ free
environment is a complex technique, the screening circumstance is
rather stress-inducing to the rodents. A less complicated and more fa-
vorable behavioral test might be more help-full to assess a huge
quantity of possibly useful compounds in AD rodents. The novel object
recognition test is founded on the learned instinct of mice to find out/
explore novelty/ unfamiliarity and is a complete working memory test
(Fig. 5) [26]. Both human and animals have hippocampus involved in
the formation of recognition memory. Novel object recognition test
does not need spatial exploration/learning and demand for a positive or
negative stimulus. Such stimuli might be the cause of stress in mice and
this stress has been concluded as a chief factor negatively affecting the
5

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
Fig. 2. Describes in detail the structure of BACE1 and flap region (a) along with optimum polar interaction with potential compound 7c (b). Enhanced view of polar
interaction with the most potent compound 7c is highlighted (c).
Fig. 3. The results of compound 7c on the memory and learning on mice models exploiting the Morris water maze screening assay. (a) Escape latency time to reach
the platform on various days of the AlCl induced neurotoxicity group, the control, and compound 7c treated group, (b) represent the fifth-day trials comparison
3
studies among the three groups. (c) Represent the percent total time the rodent spends in the explorations of the target quadrant and similarly (d) shows how many
times the mice crossed target quadrant.
learning and memory [27]. Keeping in view all the physiological re-
levancy in mind this test has been carried out for compound 7C po-
tentially beneficial for in-vitro evaluations [28].
3. Conclusions
The present study described the synthesis of eight fluoro-benzimidazole
derivatives relayed on virtual screening as a BACE1 inhibitor, in particular,
6

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
Fig. 4. Novel object recognition assay: Synthesized compound 7c enhanced the three-dimensional spatial learning/memory of AlCl induced model.
3
Fig. 5. A diagrammatic illustration of the novel object recognition test. A pre-habituation and habituation phases followed by a pair of objects located in the empty
boxes (10 min training session). Mice were put in the experimental field having familiar and unfamiliar (novel) objects after 60 min. Time was noted for exploring the
objects and for each object by the mice (video recording).
the most potent and effective 6-fluoro-2-(2,4,5-trifluorophenyl)-1H-benzo
d]imidazole (7c). Attachment of fluorine on the carbon (benzene/alde-
using MGL tools v1.5.6 and AutoDock v4.2. ChemDraw ultra 16.0 was
used to draw the structures, subsequently converted to the 3D via chem-
3D pro 16.0. Moreover, energies of the ligand structures were mini-
mized with MM2. The β-secretase (PDB ID: 1FKN) was downloaded
from RCSB protein data bank in less than 2.0 Å resolution. Addition of
water, charges, hydrogen atom along with the removal of co-crystal-
lized ligand, protein structures were prepared for molecular docking.
AutoDock vina exploits the Gasteiger partial charges estimation tech-
nique for conformers of protein calculations to ligand was adopted.
Similarly, the ligands were added with Gasteiger charges and all the
rotatable bonds were determined based on the ligand molecules [18].
Active site was specified via grids around the co-crystallized ligands
before the removal of ligands. For the sake of docking, nine poses of the
compound were generated via Lamarkian Genetic Algorithm. The
possible binding sites and docked poses were selected visualizing and
binding free energies. PyMol tool v2.1 and similarly visualizer dis-
covery studio v17.2 was used to acquire the possible binding pose
[
hyde/benzimidazole ring) led to the improved potency of the moieties and
also sometimes resulted in enhanced bioavailability at the target site.
Optimized in-silico results have been validated by the in-vitro and in-vivo
results which showed the arrest of the Aβ plaques with varying in-vitro IC50
results, and an orally bioactive brain-penetrating inhibitor that effectively
reduced the AD symptoms in mice models. Additional changes to this series
intended to improve the pharmacological activities and reduce the off-
target adverse reactions would be the focus of the future research.
4
. Material and methods
4.1. Virtual screening with molecular docking
Benzimidazole fragment synthesizes after thoroughly in silico au-
thentication. For all β-secretase inhibitors, molecular docking was done
7

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
figure [29]. The ligand-enzyme interaction was redocked and authen-
ticate via Molecular Operating Environment (MOE) drug discovery
software by Chemical Computing group and similarly the pharmaco-
kinetic properties of the ligand i.e. ClogP, (Lipinski), BBB permeation
were calculated through MOE [30].
ppm (δ) units and the coupling constants (J) were recorded in MHz.
4.4. Fluorescence resonance energy transfer (FRET) assay
BACE1 blocking activity for newly synthesized benzo[d]imidazole
derivatives was performed via Sigma –Aldrich FRET-based screening
assay activity kit (product # CS0010).
4.2. Synthesis of benzimidazoles derivatives
Briefly, assays were performed in a total 100 μl volume composed of
BACE1 substrate (20 μl, 50 μM, Catalog # A1472), fluorescent assay
buffer (78 μl/78-X μl, pH 4.5, catalog # F8303), test samples (X μl,
Based on the in-silico screening, a series of eight different benzimi-
dazoles were synthesized. The reactants and all the reagents were
purchased from Sigma Aldrich and Alfa Aesar. Several approaches are
available for the chemical synthesis of derivatives of benzimidazole
1
9
00 µM) and BACE1 enzyme (2 µl, ∼0.3 unit/μl, Catalog # B9059) in
6 well plate reaction mixture incubated at 37 °C for 1:15 min at opti-
mized assay condition (usually 5%-20% fluorescent product produced
from substrate within 1–2 h). Baseline fluorescence reading (time zero
reading) was noted immediately after the addition of BACE1 enzyme
[
31]. For the synthesis of 2-substituted benzimidazoles two approaches
were used;
(
a). Synthesis of 2-aryl benzimidazoles was done by the condensa-
(
fluorometer with excitation at 320 nm) while emission signal was read
tion reaction of o-phenylenediamine with the respective substituted
at 405 nm room temperature. Both BACE1 substrate and enzyme were
prepared in the buffer, while tested samples were initially dissolved in
DMSO (5%) subsequently serially diluted with provided buffer in the
desired volume (2 µl, 3 µl & 5 μl) and strength (200 pmol, 300 pmol &
aromatic aldehyde in the presence of NaHSO . An equimolar con-
3
centration of 0.5 mmol of ortho-phenylenediamine and the respective
aryl-aldehyde 0.5 mmol were systematically mixed in 2 ml of DMF
along with (0.15) mmol of sodium hydrogen sulphite. The reaction
mixture was stirred at 80 °C until completed and checked periodically
with TLCs. By adding 20 ml of distilled water (with constant stir) the
mixture was cooled at room temperature. The product was formed and
separated as free-floating solid. It was accumulated by suction filtra-
tion, washed with distilled water and finally dried. The gummy solid
material was extracted with ethyl acetate and washed with water and
brine solution. It was dried on sodium sulphate and the final residue
was extracted with a column on silica gel by using hexane: ethyl acetate
5
00 pmol) respectively. A standard curve was generated between
fluorescent unit (FU) against standard (100 µM, 1–5 μl) solution con-
centration (100–500 pmol) to find out 50% BACE1 cleavage. FU of the
blank was subtracted from all signal readings of the reaction mixtures.
Sample blank (buffer & substrate) was treated as a negative control,
however, positive control was a sample (buffer & substrate) with en-
zyme mixture (www.sigmaaldrich.com). IC50 of each of the tested
compounds were compared and calculated using GraphPad Prizm v5
software.
(
2–6:1–3) as eluent [32]. The overall reaction for the synthesis of 2-
Aryl-substituted benzimidazoles is given below.
4.5. In vivo study
4
.5.1. Experimental animals
All the experiments were conducted using male BALB/c mice ar-
ranged from NIH (Islamabad, Pakistan). Animals housing was in stan-
dard Type III cages of Makrolon TM in a group of 3 or 4 mice with
bedding of sawdust. Water and food were supplied ad libitum, water
was thoroughly been checked and freshly supplied on daily basis to
prevent infectious agents’ growth, similarly, sawdust bed was periodi-
cally checked and replaced to prevent all the possible infective organ-
isms results from mice’ feces. All animals were divided into six groups,
each group (n = 9) were between 3 and 5 months of age.
(
b). Ortho-phenylenediamine was mixed with trifluoroacetic acid
(
TFA) at an equimolar concentration of 0.5 M, at a reaction temperature
of 70 °C for 2 h. The excess TFA was evaporated to get the desired
product in a sufficient quantity/yield. The product was made purified
by a silica gel chromatography column to obtain total free analytical
yield (overall reaction is given below) [33].
4
.5.2. Animal preparation for behavioral tests
Animals were given drinking water with aluminum chloride
(
c) 5-Fluoro-2-(2,3,5-trifluorophenyl)-1H-benzo[d]imidazole (1 M)
(
17 mg/kg) for a period of 32 days as a positive control while the ne-
was added to round bottom flask having acetone (30 ml) along with
catalytic amount of triethylamine (10–15 drops) and stir for half an
hour under reflux condition followed by addition of 4-fluorobenzyl
chloride equimolar compared to the imidazole in the reaction mixture.
After 8 h, the final product was confirmed through TLC and were pur-
ified through column chromatography through silica gel.
gative control group was provided with normal tap water. Treated
animals were given synthesized compound (50 mg/Kg) with AlCl
Morris Water Maze test trials were carried out from day 27th to day
1st per oral treatment, whereas the remaining behavioral tests were
3
.
3
4
.3. Spectral analysis
executed on 32nd and 33rd day. Behavioral tests were carried out be-
tween 09:00 am to 5:00 pm. Mice were relocated to the testing room
The 1H NMR, C NMR, and F NMR spectra were recorded by
13
19
2
0 min before the initiation of the first trial to acquaint it with the
Agilent (DDR2 500 MHz NMR spectrometers) equipped with 7600AS 96
sample autosamplers running VnmrJ 3.2A. The instruments have TMS
as internal standards. The values of the chemical shift were described in
environment of the testing room. The temperature of the testing room
was well retained at the 25 ± 2 °C. The behavioral tests were carried in
8

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
the absence of an investigator and the entire tests were videotaped by a
cam and the results were analyzed for final outcomes [33,34].
References
[
1] W.W. Barker, C.A. Luis, A. Kashuba, M. Luis, D.G. Harwood, D. Loewenstein,
Relative frequencies of Alzheimer disease, lewy body, vascular and frontotemporal
dementia, and hippocampal sclerosis in the State of Florida Brain Bank, Alzheimer,
Dis. Assoc. Disord. 16 (2002) 203–212.
4
.5.3. Morris water maze test
It is used find out the memory and learning of the animal’s
[
[
2] A.K. Ghosh, E.R. Mannhold, H. Kubinyi, G. Folkers, G.F. Chiba, Aspartic acid pro-
teases as therapeutic targets, first ed., Wiley-VCH, 2010.
knowledge about the surrounding and its spatial remembrance. The
apparatus used for this procedure has a round pool with 120-cm dia-
meter and a depth of 60 cm. This pool is divided in 4 hypothetical
quadrants South, West, East, and North. The trial began on the 27th day
and 5 trials were executed each day. The mouse was given only 60 sec
to find out the hidden platform in each test and an interval of 10 min
between the two successive trials was given. The usual time essential by
a mouse to arrive at the platform was recorded and an average of 5
trials was presumed as the escape latency of the animal for that day. On
the 32nd day of intervention probe test was executed when there was
no platform, while the release location/point of the animal was main-
tained the same as North-East (NE). The spatial learning/memory of the
mouse was analyzed by computing the time-spent by the mouse in the
same quadrant in which the platform was formerly located. The number
of crossings across the earlier platform position was also found out
3] R. Vassar, B.D. Bennett, S. Babu-Khan, S. Kahn, E.A. Mendiaz, P. Denis, Beta-se-
cretase cleavage of Alzheimer’s amyloid precursor protein by the transmembrane
aspartic protease BACE, Science 286 (1999) 735–741.
[
4] D.S. Wang, D.W. Dickson, J.S. Malter, β-Amyloid degradation and Alzheimer’s
disease, J. Biomed. Biotechnol. (2006) 1–12, https://doi.org/10.1155/JBB/2006/
58406.
[5] C.P. Ferri, M. Prince, M.C. Brayne, C.H. Brodaty, L. Fratiglioni, M. Ganguli, Global
prevalence of dementia: a Delphi consensus study, Lancet 2005 (366) (2005)
2
112–2117.
[
[
[
6] R.J. O’Brien, P.C. Wong, Amyloid precursor protein processing and Alzheimer’s
disease, Annu. Rev. Neurosci. 34 (2011) 185–204.
7] C.A. Taft, C.H.T. de Paula Da, Silva, New Developments in Medicinal Chemistry,
Bentham Science Publishers, Amazon, France, 2014.
8] S. Sinha, J.P. Anderson, R. Barbour, G.S. Basi, R. Caccavello, D. Davis, M. Doan,
H.F. Dovey, N. Frigon, J. Hong, K. Jacobson-Croak, N. Jewett, P. Keim, J. Knops,
I. Lieberburg, M. Power, H. Tan, G. Tatsuno, J. Tung, D. Schenk, P. Seubert,
S.M. Suomensaari, S. Wang, D. Walker, J. Zhao, L. McConlogue, V. John,
Purification and cloning of amyloid precursor protein beta-secretase from human
brain, Nature 402 (1999) 537–540.
[
35,36].
[
9] L. Hong, G. Koelsch, X. Lin, S. Wu, S. Terzyan, A.K. Ghosh, X.C. Zhang, J. Tang,
Structure of protease domain of memapsin 2 (beta-secretase) complexed with in-
hibitor, Science 290 (2000) 150–153.
4
.5.4. Novel object recognition test
[
[
10] D. Shuto, S. Kasai, T. Kimura, P. Liu, K. Hidaka, T. Hamada, S. Shibakawa,
Y. Hayashi, C. Hattori, B. Szabo, S. Ishiura, Y. Kiso, KMI-008, a novel beta-secretase
inhibitor containing a hydroxymethylcarbonyl isostere as a transition-state mimic:
design and synthesis of substrate-based octapeptides, Bioorg. Med. Chem. Lett. 13
The novel object recognition test used as a modified form of the
methodology described in the literature [37]. All the testing procedures
were carried out at in a box of (25 cm × 25 cm × 25 cm), daytime at
(
2003) 4273–4276.
2
5 ± 2 °C in the same room. The methodology includes four phases:
11] S.J. Stachel, C.A. Coburn, S. Sankaranarayanan, E.A. Price, G. Wu, M. Crouthamel,
B.L. Pietrak, Q. Huang, J. Lineberger, A. Espeseth, L. Jin, J. Ellis, M.K. Holloway,
S. Munshi, T. Allison, D. Hazuda, A.J. Simon, S.L. Graham, J.P. Vacca, Macrocyclic
inhibitors of beta-secretase: functional activity in an animal model, J. Med. Chem.
(
1) pre-habituation; (2) habituation; (3) training and (4) testing (Fig. 5).
Mice were carried into the testing-room 30–35 min prior to the initia-
tion of the experiment on the first day, to get acquainted with the
surroundings. They were allowed for the free exploration of the box
without having any objects for five min [37]. Mice were familiarized
within the empty box for twenty min on the second and third days. On
the fourth day, every mouse was exposed to a trial of training followed
by a testing trial. In the training trial, 2 objects were located oppositely
to each other inside the box at a similar space from the adjacent corner
4
9 (2006) 6147–6150.
[
[
12] P. Zhou, Y. Li, Y. Fan, Z. Wang, R. Chopra, A. Olland, Y. Hu, R.L. Magolda,
M. Pangalos, P.H. Reinhart, M.J. Turner, J. Bard, M.S. Malamas, A.J. Robichaud,
Pyridinyl aminohydantoins as small molecule BACE1 inhibitors, Bioorg. Med.
Chem. Lett. 20 (2010) 2326–2329.
13] R. Godemann, J. Madden, J. Krämer, M. Smith, U. Fritz, T. Hesterkamp, T.J. Barker,
S. Höppner, D. Hallett, A. Cesura, A. Ebneth, J. Kemp, Fragment-based discovery of
BACE1 inhibitors using functional assays, Biochemistry 48 (2009) 10743–10751.
14] D.k. Lahiri, B. Maloney, J.M. Long, N.H. Greig, Lessons from a BACE1 inhibitor trial:
Off-site but not off base, Alzheimers, Dement 10 (2014) S411–S419.
15] M. Citron, Secretase as a target for the treatment of Alzheimer’s disease, J. Neurosci.
Res. 70 (2002) 373–379.
[
[
[
[
38]. The mice could discover the objects for 10 min and then brought
back to home cages. After an hour the mice were located again to the
same experimental box, whereas one of the two familiarized objects
was substituted by a new/novel object, for the start of a 10 min testing
phase [39]. The behavior of the mouse was recorded with a video-came.
Recognition index can be calculated by the formula given below;
16] M.E. Kennedy, W. Wang, L. Song, J. Lee, L. Zhang, G. Wong, Measuring human beta
secretase (BACE1) activity using homogeneous time-resolved fluorescence, Anal.
Biochem. 319 (2003) 49–55.
[17] D. Oehlrich, H. Prokopcova, H.J.M. Gijsen, The evolution of amidine-based brain
penetrant BACE1 inhibitors, Bioorganic. Med. Chem. Lett. 24 (2014) 2033–2045.
[18] J. Gasteiger, M. Marsili, Iterative partial equalization of orbital electronegativity—a
rapid access to atomic charges, Tetrahedron 36 (1980) 3219–3228.
Recognition Test = [Time spent with novel object/Time spent with object
1
+novel object] × 100
[19] H. Cheng, K.S. Vetrivel, P. Gong, X. Meckler, A. Parent, G. Thinakaran, Mechanisms
of disease: new therapeutic strategies for Alzheimer's disease targeting APP pro-
cessing in lipid rafts, Nat. Clin. Pract. Neuro. 3 (2007) 374–382.
[
[
20] W. Sherman, T. Day, M.P. Jacobson, R.A. Friesner, R. Farid, Novel procedure for
modeling ligand/receptor induced fit effects, J. Med. Chem. 49 (2006) 534–555.
21] O. Trott, A.J. Olson, AutoDock Vina: improving the speed and accuracy of docking
with a new scoring function, efficient optimization, and multithreading, J. Comput.
Chem. 31 (2010) 455–461.
4
.6. Statistics
All numerical values were given as mean ± SEM. Column statistic
was used to compare the results and level of significance was set at
p < 0.05.
[
22] J. Karsten, L. Tagmose, M. Marigo, 2-amino-3-fluoro-3-(fluoromethyl)-6-methyl-6-
phenyl-3, 4, 5, 6-tetrahydropyridines as BACE1 inhibitors, U.S. Patent Application
No. 10/011,596.
[
[
[
23] F. Leroux, P. Jeschke, M. Schlosser, α-fluorinated ethers, thioethers and amines:
anomerically biased species, Chem. Rev. 105 (2005) 827–856.
Acknowledgements
24] W.K. Hagmann, The many roles for fluorine in medicinal chemistry, J. Med. Chem.
5
1 (2008) 4359–4369.
This research work is partially supported by the COMSATS
University Islamabad, Abbottabad campus under the funding project #
25] C.A. Lipinski, Lead and drug-like compounds: the rule-of-five revolution, Drug.
Discov. Today. Technol. 1 (2004) 337–341.
1
6-60/CRGP/CIIT/ABT/14/635 and partly by the International
[26] C. Exley, Molecular mechanism of aluminium-induced Alzheimer’s disease? J.
Inorg. Biochem. 76 (1999) 133–140.
Research Support Initiative Program (IRSIP) higher education com-
mission Islamabad, Pakistan for abroad visit to the MSU, USA.
[
[
27] N. Jones, S.M. King, Influence of circadian phase and test illumination on pre-
clinical models of anxiety, Physiol. Behav. 72 (2001) 99–106.
28] N.P. Van Goethem, K. Rutten, F.J. van der Staay, L.A.W. Jans, S. Akkerman,
H.W.M. Steinbusch, Object recognition testing: rodent species, strains, housing
conditions, and estrous cycle, Behav. Brain. Res. 232 (2012) 323–334.
29] D. Santos-Martins, S. Forli, M.J. Ramos, A.J. Olson, AutoDock4Zn: An improved
AutoDock force field for small-molecule docking to zinc metalloproteins, J. Chem.
Inf. Model. 54 (2014) 2371–2379.
Appendix A. Supplementary material
[
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bioorg.2019.102936.
9

S. Ali, et al.
Bioorganic Chemistry 88 (2019) 102936
[
[
30] S. Vilar, G. Cozza, S. Moro, Medicinal chemistry and the molecular operating en-
vironment (MOE): application of QSAR and molecular docking to drug discovery,
Curr. Top. Med. Chem. 8 (2008) 1555–1572.
201–204.
[35] K. Bromley-Brits, Y. Deng, W. Song, Morris water maze test for learning and
memory deficits in Alzheimer’s disease model mice, J. Vis. Exp. 53 (2011) 2920.
[36] A. Mahboob, S.M. Farhat, G. Iqbal, M.M. Babar, S.S. Zaidi, N.S.M. Nabavi, Alpha-
lipoic acid-mediated activation of muscarinic receptors improves hippocampus- and
amygdala-dependent memory, Brain. Res. Bull. 122 (2016) 19–28.
31] B. Chu, F. Liu, L. Li, C. Ding, K. Chen, Q. Sun, A benzimidazole derivative exhibiting
antitumor activity blocks EGFR and HER2 activity and upregulates DR5 in breast
cancer cells, Cell. Death. Dis. 6 (2015) e1686.
[
[
32] C.X. Zhang, G.J. Zheng, F.Q. Bi, Y.L. Li, A simple and efficient synthesis of the
valsartan, Chinese. Chem. Lett. 19 (2008) 759–761.
[37] S. Akkerman, A. Blokland, O. Reneerkens, G.N.P. Van, E. Bollen, H.J.M. Gijselaers,
Object recognition testing methodological considerations on exploration and dis-
crimination measures, Behav. Brain. Res. 232 (2012) 335–347.
33] P. Bamborough, J.A. Christopher, G.J. Cutler, M.C. Dickson, G.W. Mellor,
J.V. Morey, 5-(1H-Benzimidazol-1-yl)-3-alkoxy-2-thiophenecarbonitriles as potent,
selective, inhibitors of IKK-ε kinase, Bioorganic. Med. Chem. Lett. 16 (2006)
[38] R.N. Hughes, Neotic preferences in laboratory rodents : Issues, assessment and
substrates, Neurosci. Biobehav. Rev. 31 (2007) 441–464.
6
236–6240.
[39] R. Zhang, G. Xue, S. Wang, L. Zhang, C. Shi, X. Xie, Novel object recognition as a
facile behavior test for evaluating drug effects in AβPP/PS1 Alzheimer’s disease
mouse model, J. Alzheimer’s. Dis. 31 (2012) 801–812.
[
34] O. René, A. Souverneva, S.R. Magnuson, B.P. Fauber, Efficient syntheses of 2-
fluoroalkylbenzimidazoles and –benzothiazoles, Tetrahedron. Lett. 54 (2013)
10