Rational design, synthesis and biological screening of triazine-triazolopyrimidine hybrids as multitarget anti-Alzheimer agents

Article abstract of DOI:10.1016/j.ejmech.2017.04.064

In our endeavor towards the development of potent multitarget ligands for the treatment of Alzheimer's disease, a series of triazine-triazolopyrimidine hybrids were designed, synthesized and characterized by various spectral techniques. Docking and scoring techniques were used to design the inhibitors and to display their interaction with key residues of active site. Organic synthesis relied upon convergent synthetic routes were mono and di-substituted triazines were connected with triazolopyrimidine using piperazine as a linker. In total, seventeen compounds were synthesized in which the di-substituted triazine-triazolopyrimidine derivatives 9a-d showed better acetylcholinesterase (AChE) inhibitory activity than the corresponding tri-substituted triazine-triazolopyrimidine derivatives 10a-f. Out of the disubstituted triazine-triazolopyrimidine based compounds, 9a and 9b showed encouraging inhibitory activity on AChE with IC₅₀ values 0.065 and 0.092?μM, respectively. Interestingly, 9a and 9b also demonstrated good inhibition selectivity towards AChE over BuChE by ～28 folds. Furthermore, kinetic analysis and molecular modeling studies showed that 9a and 9b target both catalytic active site as well as peripheral anionic site of AChE. In addition, these derivatives effectively modulated Aβ self-aggregation as investigated through CD spectroscopy, ThT fluorescence assay and electron microscopy. Besides, these compounds exhibited potential antioxidants (2.15 and 2.91 trolox equivalent by ORAC assay) and metal chelating properties. In silico ADMET profiling highlighted that, these novel triazine derivatives have appropriate drug like properties and possess very low toxic effects in the primarily pharmacokinetic study. Overall, the multitarget profile exerted by these novel triazine molecules qualified them as potential anti-Alzheimer drug candidates in AD therapy.

Full text of DOI:10.1016/j.ejmech.2017.04.064

Accepted Manuscript
Rational design, synthesis and biological screening of triazine-triazolopyrimidine
hybrids as multitarget anti-Alzheimer agents
Ehtesham Jameel, Poonam Meena, Mudasir Maqbool, Jitendra Kumar, Waqar
Ahmed, Syed Mumtazuddin, Manisha Tiwari, Ph.D., Assistant Professor, Nasimul
Hoda, Ph.D., Professor, B. Jayaram
PII:
S0223-5234(17)30338-0
10.1016/j.ejmech.2017.04.064
EJMECH 9415
DOI:
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 26 January 2017
Revised Date: 2 April 2017
Accepted Date: 23 April 2017
Please cite this article as: E. Jameel, P. Meena, M. Maqbool, J. Kumar, W. Ahmed, S. Mumtazuddin,
M. Tiwari, N. Hoda, B. Jayaram, Rational design, synthesis and biological screening of triazine-
triazolopyrimidine hybrids as multitarget anti-Alzheimer agents, European Journal of Medicinal
Chemistry (2017), doi: 10.1016/j.ejmech.2017.04.064.
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
Rational design, synthesis and biological screening of
triazine-triazolopyrimidine hybrids as multitarget anti-Alzheimer agents
a
b
a
a
a
Ehtesham Jameel , Poonam Meena , Mudasir Maqbool , Jitendra Kumar , Waqar Ahmed , Syed
c
b
a
d,e,f
Mumtazuddin , Manisha Tiwari *, Nasimul Hoda * and B. Jayaram
a
Department of Chemistry, Jamia Millia Islamia (Central University), Jamia Nagar, New Delhi
1
10025, India
b
Dr. B. R. Ambedkar Centre for Biomedical Research, University of Delhi, New Delhi 110007,
India
c
Department of Chemistry, B. R. Ambedkar Bihar University, Muzaffarpur 842001, Bihar, India
d
Department of Chemistry, Indian Institute of Technology Delhi, Hauz Khas, New Delhi 110016,
India
e
Kusuma School of Biological Sciences, IIT Delhi, New Delhi 110016, India
f
Supercomputing Facility for Bioinformatics & Computational Biology, IIT Delhi, New Delhi
1
10016, India.
*
Corresponding authors.
#
$
Nasimul Hoda, Ph.D.
Professor
Manisha Tiwari, Ph.D.
Assistant Professor
Department of Chemistry
Jamia Millia Islamia,
Jamia Nagar, New Delhi-110025,
India
Dr. B. R. Ambedkar
Centre for Biomedical Research,
University of Delhi, New Delhi-110007,
India
Email: nhoda@jmi.ac.in
Email: mtiwari07@gmail.com

ACCEPTED MANUSCRIPT
Research highlights:
•
•
A series of triazine-triazolopyrimidine hybrids were designed and synthesised.
Docking studies revaeled that compounds 9a and 9b had appreciable binding free
energies and affinities with AChE.
•
•
The most promising compound (9a) had an inhibition selectivity of ~28 fold against
AChE over BuChE.
Aβ1–42 disaggregation, metal binding studies, CD spectroscopic, ADME property and
TOPKAT analysis were also performed.

ACCEPTED MANUSCRIPT
Graphical Abstract

ACCEPTED MANUSCRIPT
Abstract
In our endeavor towards the development of potent multitarget ligands for the treatment
of Alzheimer’s disease, a series of triazine-triazolopyrimidine hybrids were designed,
synthesized and characterized by various spectral techniques. Docking and scoring techniques
were used to design the inhibitors and to display their interaction with key residues of active site.
Organic synthesis relied upon convergent synthetic routes were mono and di-substituted triazines
were connected with triazolopyrimidine using piperazine as a linker. In total, seventeen
compounds were synthesized in which the di-substituted triazine-triazolopyrimidine derivatives
9
a-d showed better acetylcholinesterase (AChE) inhibitory activity than the corresponding tri-
substituted triazine-triazolopyrimidine derivatives 10a-f. Out of the disubstituted triazine-
triazolopyrimidine based compounds, 9a and 9b showed encouraging inhibitory activity on
AChE with IC50 values 0.065 and 0.092 µM, respectively. Interestingly, 9a and 9b also
demonstrated good inhibition selectivity towards AChE over BuChE by ~28 folds. Furthermore,
kinetic analysis and molecular modeling studies showed that 9a and 9b target both catalytic
active site as well as peripheral anionic site of AChE. In addition, these derivatives effectively
modulated Aβ self-aggregation as investigated through CD spectroscopy, ThT fluorescence
assay and electron microscopy. Besides, these compounds exhibited potential antioxidants (2.15
and 2.91 trolox equivalent by ORAC assay) and metal chelating properties. In silico ADMET
profiling highlighted that, these novel triazine derivatives have appropriate drug like properties
and possess very low toxic effects in the primarily pharmacokinetic study. Overall, the
multitarget profile exerted by these novel triazine molecules qualified them as potential anti-
Alzheimer drug candidates in AD therapy.

ACCEPTED MANUSCRIPT
Keywords: Alzheimer’s disease, Acetylcholinesterase, Butyrylcholinesterase, Triazine,
Triazolopyrimidine quinoline, Molecular docking, ADME analysis.

ACCEPTED MANUSCRIPT
1
. Introduction
Alzheimer’s disease (AD) is an age related irreversible neurodegenerative disorder
characterized by memory loss [1]. About 46.8 million people have been reported to suffer
globally in 2015 from AD which is expected to increase by three fold in 2050 [2]. No such
treatment has yet been developed that could prevent, cure or even slow down the progress of this
disorder. It is the third leading cause of death after cardiovascular disease and cancer globally.
The mechanism of this dementia has not been fully elucidated due to multifactorial progression
and complex etiology. Since, its exact etiology is unclear, current evidence suggests that the
abnormal accumulation of amyloid-β (Aβ) and tau protein in the form of neurofibrillary tangles
(
NFTs) are the main hallmark of AD [3]. Aβ is produced by the proteolytic break down of
amyloid precursor protein (APP), a transmembrane protein by the action of β- and ɣ-secretases
4]. NFTs are formed by the hyperphosphorylation of microtubule associated protein called tau
which is overseen by overexpression of GSK-3β [5]. Beside this, decrease of acetylcholine
ACh) an important neurotransmitter is also associated with the symptoms of dementia and
[
(
learning difficulties. A serine hydrolase enzyme called acetylcholinesterase (AChE), promotes
the breakdown of ACh. Drugs that target the clearing or preventing the formation of Aβ and
NFTs can prove to be a successful approach towards the treatment of AD. Donepezil,
rivastigmine, galantamine, and tacrine are some FDA approved acetylcholinesterase inhibitors
(AChEIs) used for the treatment of AD. These AChEIs lack the long-term efficacy and possess
side effects, like, gastrointestinal disturbance, hepatotoxicity, dizziness, liver damage, vomiting
and nausea [6]. The limited clinical efficacy of these approved drugs further propel medicinal
chemists to put their efforts towards the development of molecules that can simultaneously hit
multiple targets of AD etiology.

ACCEPTED MANUSCRIPT
1
,3,5-Triazine scaffold has a great history for the fabrication of antiviral [7], antifungal [8],
antimalarial [9], antibacterial [10], and anticancer agents [11][12]. Recently, a group of 2,4,6-
trisubstituted 1,3,5-triazine compounds was found to be effective against various cysteine
cathepsins with endopeptidase activity [13]. One more important pharmacological usage of
substituted 1,3,5-triazines has been explored that these compounds possess the antagonist
selectivity effect against adenosine receptors [14]. There are many more effects of triazine
analogues [15], so to say triazine is a fertile moiety and has been serving medicinal chemists for
a long time. The triazine and its derivatives could be synthesized via feasible convergent
synthetic route. Since, triazine possesses a roughly planar structure and was expected to favor the
Aβ disaggregation. Triazolopyrimidine scaffold was expected to engage with the important
residues at PAS of AChE, so could enhance the ability of the drug molecules to inhibit the
enzyme. Indeed, triazole scaffold has been efficiently utilized in the design of multifunctional
AChEIs [16]. Since the active site of AChE is roughly 20 Å deep narrow gorge encrusted
primarily by aromatic residues, so piperazine as a linker between the two main moieties was
expected to increase the length of the drug molecule and engage with in π-cation interaction with
the aromatic gorge. Hence, combined scaffolds of quinoline and triazolopyrimidine through
piperazine was preferred choice due to their various valuable properties. In our previous work we
attempted to evaluate these two moieties separately for the development of anti-AD drugs [17]
[18].

ACCEPTED MANUSCRIPT
Fig. 1. Design of multifactorial anti-AD agent
Herein, a series of seventeen compounds was designed and synthesized. In vitro inhibitory
activity towards cholinesterases and Aβ disaggregating ability of all the synthesized compounds
were assessed. Further, the mechanism of AChE inhibition was investigated by kinetic studies. In
order to obtain a better understanding of possible interactions with biological targets, molecular
docking studies and binding free energy calculations were also performed.
2
. Results and Discussion
2
.1.Chemistry
The synthesis of the target molecules were achieved by multistep reactions which are
outlined in Scheme 1 and 2. The triazine derivatives of the intermediate compounds i.e. mono
and disubstituted compounds 2a-e and 3a-g were synthesized by the reported literature[17]. The
cyanuric chloride was first reacted with 3-(trifluoromethyl)aniline, p-anisidine, p-fluoroaniline,
2 3
o-fluoroaniline and 3-chloro-4-fluoroaniline at -10 °C in THF and K CO to obtain
monosubstituted compounds. The obtained monosubstituted compounds were treated with
different amines to get 3a-g as disubstituted compounds. Synthetic route for the synthesis of 8,
i.e. 1-{5-methyl- [1,2,4]triazolo[1,5-a]pyrimidin-7-yl} piperazine attached to triazine derivatives

ACCEPTED MANUSCRIPT
of the mono and disubstituted compounds have been depicted in Scheme 2. Finally, the
monosubstituted triazine derivatives 2a-e were treated with 8 in the presence of K CO and 1,4-
2
3
dioxane at room temperature in N medium to obtain the target compounds di-substituted triazine
2
triazolopyrimidine piperazine as 9a-d. Similarly, the disubstituted triazine compounds 3a-g were
0
also treated with 8 in presence of K CO and 1,4-dioxane at 110 C in N medium to obtain the
2
3
2
tri-substituted triazine triazolopyrimidine piperazine 10a-e as desired compounds. In these
schemes we see that the reactions involved in the synthesis of the target compounds were
temperature dependent nucleophilic aromatic substitution reactions. The intermediates i.e.
0
monosubstituted triazines were synthesized at -5 to -10 C, disubstituted at room temperature
0
while the final or the targeted compounds were obtained at higher temperature, i.e., 110 C.
Finally, all the newly synthesized compounds were stable enough to be isolated by column
chromatography on silica gel and their structures were unambiguously verified by different
1
13
elemental analysis and spectroscopic data i.e. H NMR, C NMR, ESI MS and CHNS.

ACCEPTED MANUSCRIPT
2 3
Scheme: 1. Reagents and conditions: (i) aromatic amines, tetrahydrofuran (THF), -10 °C, K CO
(
ii) aliphatic/ aromatic amines, THF, room temperature, K CO , (iii) K CO , 1,4-dioxane, room temperature,
2 3 2 3
N ; (iv) & (v) 1,4-dioxane, K CO , 100 °C, N .
2
2
3
2
3 2 3
Scheme: 2. Reagents and conditions: (i) Acetic acid, 110 °C; (ii) POCl , 110 °C; (iii) piperazine, K CO , 1,4-
dioxane, 100 °C.

ACCEPTED MANUSCRIPT
Compound
R
1
R
2
Compound
R
1
3
(a)
2
2
(a)
(b)
3
(b)
2
2
2
(c)
(d)
(e)
3
3
(c)
(d)
3 (e)
9
9
(a)
(b)
3
3
(f)
(g)
9
9
(c)
1
0 (a)
(d)
10 (b)
1
0 (c)
1
0 (d)
1
0 (e)
10 (f)

ACCEPTED MANUSCRIPT
2
.2. Molecular modeling studies
Structure based drug design approach was used to obtain an influential drug candidate for
multi-faceted AD. Triazine, because of its roughly planar structure was selected as a preferred
scaffold. It was expected to interpolate between beta-amyloid sheets was expected to minimize
the beta-amyloid aggregation. Due to the aromaticity of triazine, it was also expected to engage
efficiently with the active site residues of AChE via π-π stacking interaction and alkyl-π-
interaction. Substitution of triazine with p-anisidine was a natural choice because anisidine is
known to have neuroprotective and radical scavenging effect [19]. Piperazine was incorporated
inorder to connect two electrophilic centres and its nitrogen was expected to be in protonated
state and hence may participate in π-cation interactions with the residues of aromatic gorge.
ParDock module of Sanjeevini (SCFBIO, IITD) [20][21] was selected for docking and scoring
purposes. The computational binding free energy of synthesized molecules and reference
compound donepezil with AChE and BuChE is provided in Table 1. The Binding free energy
(∆G in kcal/mol) of compounds 9a, 9b and donepezil for AChE was found to be -10.87, -10.53
and -7.85 respectively, which proposes a very high binding affinity for 9a and 9b between the
ligands and the enzyme but least with the donepezil. Binding free energy value i.e. -10.32, -9.93
and -6.07 kcal/mol of compounds 9a, 9b and donepezil for BuChE was also recorded. All the
synthesized molecules revealed more inhibition selectivity for AChE over BuChE.
A docking study was executed on reported crystal structure of AChE with PDB ID 1EVE.
Better interaction of compounds 9a and 9b for AChE was predicted and same results were
obtained via their docking studies. In Fig. 2(a) and 2(b) aromatic ring of Tyr334 was expected to
show π- π stacking with the triazine core moiety of both compound 9a and 9b, respectively.
Triazolopyrimidine bicyclic system of compound 9a was found to have the aryl-π interaction

ACCEPTED MANUSCRIPT
with the bicyclic system of Trp279 which is located at the peripheral anionic site of AChE. In
Fig. 2(b) Trp84 interact with Triazolopyrimidine of compound 9b via π-π stacking.
The results of docking study publicized that the dual binding activity for selected
compound may be due to π-π (aromatic) interactions with CAS and PAS of AChE.
Fig. 2. (a), (b) and (c) are the docked inhibitors of compound 9a, 9b and donepezil with AChE. Interactions
with key residues are shown in different colors. Tyr with hot pink, Trp in yellow, Phe in cyan, His in green,
Glu in blue and Ser in magenta color. π-π stacking are shown in red dashed lines.
Docking studies were further extended to the reported crystal structure of Human BuChE
(
PDB ID 4TPK) with the synthesized compounds. The docked structures of best resulting
compounds (9a and 9b) and donepezil as a reference compound are given in Fig. 3(a), 3(b) and
(c) respectively. 3-trifluoromrthylaniline of compound 9a and p-anisidine of compound 9b was
3
found to have the aryl-π interaction with the bicyclic system of Trp82 which is located at the
catalytic anionic site of AChE. Aromatic ring of Tyr332 was expected to interact with the
piperazine ring of compound 9a and π- π stacking of triazolopyrimidine bicyclic system with
compound 9b, respectively. However, the calculated free binding energies of the synthesized
molecules suggested their more selectivity for AChE over BuChE as shown in Table 1.

ACCEPTED MANUSCRIPT
Fig. 3. (a), (b) and (c) are the docked inhibitors of compound 9a, 9b and donepezil with BuChE. Interactions
with key residues are shown in different colors. Tyr with hot pink, Trp in yellow, Glu in blue and Ser in
magenta, Thr in purple, Asp in forest green, Asn in red, Ala in dark grey, and Leu in orange color. π-π stacking
are shown in red dashed lines.
Table 1. Calculated free energy values of the docked compounds by ParDock[20]
Compounds
Free energy values
Free energy values
(kcal/mol) BuChE
-10.32
-9.93
(
kcal/mol) AChE
-10.87
-10.53
-9.12
9
9
9
9
1
1
1
1
1
1
3
3
3
3
3
3
3
a
b
c
-8.85
d
-8.15
-7.31
0a
0b
0c
0d
0e
0f
a
-9.16
-8.14
-9.75
-8.34
-9.84
-8.67
-9.35
-7.98
-8.98
-7.73
-9.02
-7.81
-8.71
-7.41
b
-8.53
-7.18
c
-8.19
-6.91
d
-8.34
-7.39
e
-8.83
-7.58
f
-7.68
-6.32
g
-7.82
-6.75
Donepezil
-7.85
-6.07

ACCEPTED MANUSCRIPT
Bolded values depict the most active compounds 9a and 9b.
2
2
.3. Pharmacological evaluation
.3.1. In vitro AChE and BuChE inhibitory activity
The inhibitory activity of triazine derivatives 9a-d, 3a-g and 10a-f was evaluated in vitro
against AChE (from electrophorus electricus) and BuChE (from equine serum) using Ellman's
method [22]. Donepezil and tacrine were used as reference standards for comparative analysis.
The corresponding IC50 values of all the synthesized derivatives are listed in Table 2. Most of
the target compounds displayed significant inhibitory activities against both cholinesterases
(ChEs) with IC50 values in micromolar range. Structure activity relationship (SAR) studies
indicated that the ChE inhibition and selectivity were sensitive to substituents with different
electronic properties on the core triazine scaffold. Within the series, it was observed that the di-
substituted triazine derivatives 9a-d displayed better AChE inhibition than the corresponding tri-
substituted derivatives 10a-f. Interestingly, compound 9a with trifluoromethyl substitution at the
C-3 position of phenyl ring exhibited the most potent AChE inhibitory activity with IC value of
5
0
0
.065 µM, which was ~3.5-fold stronger than the reference compound tacrine. In addition, 9a
also demonstrated a high selectivity for AChE over BuChE (~28-fold). The presence of methoxy
substituent at C-4 position of phenylamine moiety in 9b also revealed notable selective inhibition
towards AChE (IC = 0.092
µ
M and S.I = ~ 16). In case of 9c, the introduction of fluoro group
at C-4 position also showed significant inhibition against AChE with IC values of 0.115 M,
M). The presence of fluoro
5
0
µ
5
0
while the BuChE inhibitory activity improved slightly (IC = 0. 93
µ
5
0
group at C-2 position of phenylamine moiety attached to C-2 of heteroaromatic triazine (9d) core
led to reduce AChE activity as compared to the lead compound 9a. In the tri-substituted-1,3,5-
triazine derivatives, a range of substituents were varied over the phenyl amine ring. In case of

ACCEPTED MANUSCRIPT
compounds 10a and 10b, trifluoromethyl substitutions at the C-3 position were kept constant and
variations were performed at other terminal end. The presence of cyclopropane ring in 10a (IC50
=
0.903
µ
M) was observed to be unfavorable for ChE inhibition in comparison to the
M). The combined introduction of Cl and F at C-3
and C-4 position of the phenyl ring (10c) gave more potent AChE inhibition (IC = 0.338 M)
M). However, absence of triazolo
pyrimidine heterocycle at N-4 terminal of piperazine ring (10f) lead to a significant loss of
inhibitory potency (IC50 = 1.25 M) which was nearly ~19-fold lower than most potent
methylbenzene moiety (10b, IC = 0.558
µ
5
0
µ
5
0
with respect to fluoro substituent at C-4 (10d, IC = 0.724
µ
5
0
µ
compound. These results point out towards the relevant role played by the triazolo pyrimidine
scaffold on the AChE inhibition which is in consistent with our earlier investigation [18]. The
series of di-substituted-1,3,5-triazine derivatives represented by nonsubstituted counterparts 3a-g
at C-4 of the triazine core exerted slightly lower AChE inhibition. In general, it seems that the
presence of trifluoromethyl substitution at C-5 of phenylamine ring is particularly favorable for
anti-AChE activity. The common trend of substitution observed at the other terminal end of
phenylamine moiety was of the order: 3e [3-Chloro-4-fluorol] > 3a [2,4-difluoro] > 3b [4-fluoro]
>
3d [4-methoxy] > 3c [4-methyl], respectively. In this series, compound 3e was found to
demonstrate a higher inhibitory potency towards AChE with an IC value of 0.432 M which
µ
50
was about ~1.5–fold potent than those of compound 3c. However, the replacement of substituted
trifluoromethyl group with 3-chloro-4-fluoro substituents in compounds 3f and 3g has led to
significant decrease in AChE inhibitory potencies. In general, most of the target compounds
were significantly active for both ChEs with greater selectivity towards AChE over BuChE.
Overall the SAR results clearly suggested the influence of electronic and steric properties of
substituents on the modulation of ChE inhibition.

ACCEPTED MANUSCRIPT
Table 2. In vitro inhibition of AChE and BuChE, selectivity index and oxygen radical absorbance capacity
(ORAC, Trolox equivalents) of the target compounds.
a
b
c
d
Compounds
AChE
BuChE
IC50 (µM)
Selectivity index
BuChE/AChE
Trolox equiv
IC50
(
µ
M)
9
a
0.065 ± 0.002
0.092 ± 0.001
0.115 ± 0.11
0.286 ± 0.52
0.903 ± 0.30
0.558 ± 0.69
0.388 ± 0.19
0.724 ± 0.26
0.859 ± 0.07
1.25 ± 0.33
1.88 ± 1.27
1.52 ± 0.08
0.93 ± 1.03
2.23 ± 0.24
1.90 ± 0.29
1.75 ± 0.75
2.614 ± 0.13
2.49 ± 0.30
3.782 ± 0.22
3.65 ± 0.93
4.33 ± 0.67
2.45 ± 0.91
2.09 ± 0.09
2.961 ± 0.15
3.983 ± 0.52
2.77 ± 0.83
1.639 ± 0.38
2.72 ± 0.09
0.056 ± 0.14
28.92
16.52
8.08
7.79
2.10
3.13
6.73
3.43
4.40
2.92
8.72
4.65
3.11
5.01
9.21
3.02
2.32
57.87
0.25
2.15 ± 0.64
2.91 ± 0.58
1.68 ± 0.52
1.14 ± 0.62
0.98 ± 0.32
1.04 ± 0.38
1.37 ± 0.52
0.89 ± 0.16
0.97 ± 0.35
0.80 ± 0.33
1.16 ± 0.69
1.53 ± 0.82
1.20 ± 0.86
1.83 ± 0.40
1.22 ± 0.39
1.62 ± 0.51
1.78 ± 0.83
9
b
9
c
9
d
10a
10b
10c
10d
10e
1
0f
3
a
0.496 ± 0.09
0.526 ± 0.27
0.672 ± 0.35
0.591 ± 0.092
0.432 ± 0.44
0.917 ± 0.19
0.705 ± 0.08
0.047 ± 0.15
0.226 ± 0.07
3
b
3
c
3
d
3
e
3
f
3
g
e
Donepezil
Tacrine
n.d
e
n.d
a
AChE from electric eel; IC , inhibitor concentration (mean ± SD of three independent experiments) resulting
50
in 50% inhibition of AChE.
b
BuChE from equine serum; IC , inhibitor concentration (mean ± SD of three independent experiments)
50
resulting in 50% inhibition of BuChE.
c
SI selectivity index for AChE: IC50 (EqBuChE)/IC50 (EeAChE).
Data are expressed as µmol of trolox equivalent/µmol of tested compound.
n.d, not determined.
d
e

ACCEPTED MANUSCRIPT
2
.3.2. Kinetic study of AChE inhibition
To gain further insight into the mode of inhibition of this family of compounds on AChE,
kinetic measurements were carried out with the most active inhibitors (9a and 9b) of the series,
using EeAChE. The analysis was carried out by recording substrate concentration-enzyme
velocity curves in the presence of different concentrations of test inhibitors. Graphical analysis of
the reciprocal Lineweaver-Burk plots showed both increased slopes (decreased Vmax) and
intercepts (higher Km) at increasing concentration of the inhibitors (Fig. 4). This pattern
indicated that compounds 9a and 9b were both mixed-type inhibitors that could simultaneously
bind to CAS and PAS of the AChE enzyme. These kinetic measurements are in good agreement
with the results of molecular modeling studies which supports the dual site binding of these
compounds. Replots of the slope versus concentration of compounds 9a and 9b gave an estimate
of the inhibition constant, Ki, of 69 and 102 nM, respectively.
Fig. 4. Kinetic study on the mechanism of EeAChE inhibition by compounds 9a (A) and 9b (B). Overlaid
Lineweaver-Burk reciprocal plots of EeAChE initial velocity at increasing substrate concentration (0.05-0.50
mM) in the absence of inhibitors and in the presence of 9a and 9b are shown.
2
.3.3. In vitro antioxidant activity assays
Generation of reactive radical species have been identified as the major contributing
factor in AD pathogenesis [23]. Therefore, its reduction is another crucial aspect in designing

ACCEPTED MANUSCRIPT
multifunctional agents for AD treatment [24][25]. To evaluate the antioxidant activity of new
triazine derivatives, the oxygen radical absorbance capacity using fluorescein (ORAC-FL) was
employed by following a well-established protocol [26]. Trolox, a water-soluble vitamin E
analogue, was used as a standard, and the results were expressed as trolox equivalents (
µmol of
trolox/ mol of tested compound), in a relative scale where ORAC (trolox) = 1. As shown in
µ
Table 2, most of the target compounds demonstrated better antioxidant activity with radical
absorbance capacities ranging from 2.91 to 1.04-fold of the trolox value. In general, best results
were obtained with derivatives bearing a 4-methoxy substituent at the phenyl ring which seems
to be favourable to the compound’s overall radical scavenging ability. Especially, compound 9b
exhibited the most potent antioxidant activity within the series with ORAC-FL values of 2.91
trolox equivalents followed by compounds 9a, 9c, 3d and 3g with 2.15, 1.68, 1.83 and 1.78
trolox equivalents, respectively. However, compounds of tri-substituted triazine series mainly
1
0d, 10e and 10f showed relatively lower antioxidant activity.
2
.3.4. Inhibition of self-mediated A
The abnormal production and accumulation of misfolded A
major triggering factor in the AD pathogenesis. The A 1 and Aβ1-42 are the major components
β1-42 aggregation
β
peptides represents the
β
-40
of senile plaques found in AD brain. However, A 1
β
is markedly more fibrillogenic compared
-42
to relatively soluble A 1
β
form [27]. Therefore, prevention of Aβ1-42 aggregation is another
-40
potential approach for the treatment of AD. With this aim, di and tri-substituted triazine
derivatives were selected to assess their ability to inhibit A 1 peptide aggregation by
β
-42
employing thioflavin-T fluorescence assay using curcumin (a known anti-amyloidogenic agent)
as the standard. As shown in Table 3, most of the tested compounds exhibited significantly
higher Aβ aggregation inhibitory activity (1-1.4 folds) at 25 µM compared to the reference

ACCEPTED MANUSCRIPT
compound curcumin. Compounds 9a-d of di-substituted triazine series were found to be more
effective in blocking the A 1
β
self-aggregation with inhibition ratio of 75.3% (IC50 = 10.43
-42
µ
M), 68.5% (IC50 = 11.55 µM), 64.5% (IC50 = 15.89 µM) and 65.8% (IC50 = 13.4 µM),
respectively. The trend observed in these set of compounds was found to be similar as observed
with cholinesterase inhibition assay. In the screening results of tri-substituted series 10a-f, the
most effective compound was found out to be 10c, followed by 10f, 10d and 10e, their inhibitory
potency are 61.7% (IC50 = 16.29
µM), 58.8% (IC50 = 23.54), 54.7% (IC50 = 18.31 µM) and
4
9.3% (IC50 = 25 M), respectively. On the other hand, compounds 3a-g gave slightly better
µ
inhibitory activity with percentages of inhibition ranging from 65.2% to 46.2% than the
compounds of tri-substituted series. Overall, our results revealed that the presence of core
triazine moiety is playing a crucial role in binding with beta-amyloid sheets which further
prevents Aβ1-42 mediated aggregation process.
Table 3. Inhibition of self-induced Aβ1-42 aggregation by target compounds 9a-d, 10a-f, 3a-g and reference
compound curcumin.
Compounds
A
β1-42
A
β1-42
Compounds
A
β1-42
A
β1-42
aggregation
inhibition (%)
75.32 ± 0.34
aggregation
aggregation
inhibition (%)
58.80 ± 0.92
aggregation
IC (µM)
50
a
b
a
b
IC (
µM)
5
0
9
a
10.43 ± 0.18
11.55 ± 0.49
15.89 ± 0.88
13.40 ± 1.06
10f
23.54 ± 0.99
17.85 ± 0.87
18.55 ± 1.06
20.72 ± 1.18
17.60 ± 0.59
9
b
68.52 ± .039
64.53 ± 0.72
65.80 ± 1.03
44.33 ± 0.69
22.45 ± 0.52
61.70 ± 0.75
54.68 ± 0.98
49.33 ± 0.92
3a
65.23 ± 0.44
63.01 ± 0.81
64.77 ± 0.65
67.20 ± 0.19
46.16 ± 1.22
62.52 ± 1.04
60.74 ± 0.78
54.32 ± 0.95
9
c
3b
9
d
3c
c
1
0a
0b
0c
0d
0e
n.t
3d
c
c
1
n.t
3e
3f
n.t
1
16.29 ± 1.35
18.31 ± 0.59
25.0 ± 0.81
18.05 ± 1.36
23.80 ± 0.73
22.51 ± 0.89
1
3g
1
Curcumin

ACCEPTED MANUSCRIPT
a
Inhibition of self-induced Aβ1-42 aggregation (25 µM) by tested inhibitors at 25 µM by thioflavin-T based
fluorescence method (means ± SD of three experiments).
b
The concentration (µM) required for 50% inhibition was determined from dose-response curves. Data are
expressed as means ± SD of three independent experiments.
c
n.t. means not tested
2
.3.5. Effect on Aβ β-sheet formation by compounds 9a and 9b
fibril formation has been correlated with secondary structure transitioning from
disordered random structure to ordered -sheet-rich conformation [28][29]. On this premise,
finding small molecules that ubiquitously prevent the conformational transition from initial
random coil or a-helix into -sheet seems to be intriguing for AD therapeutic development. In
order to elucidate the structural details of the interactions between A 1-42 and the triazine
derivatives (9a and 9b), CD spectroscopy was utilized as a means to obtain information about the
mechanism of action of these inhibitors in preventing conformational changes in A 1 peptide.
Aβ
β
β
β
β
-42
As shown in Fig. 6A, the CD spectrum of A 1
β
monomer did not exhibit any spectral feature of
-42
α
-helix and
spectra recorded after the incubation period of 12 h indicated fall of unstructured conformation
content and a small rise in -helix (Fig. 6B). The spectra recorded in the presence of 9a and 9b
co-incubated with peptide for 12 h assumes shapes closer to those characteristic for the -helix
-sheet and random structures (Fig. 6B). The representative CD
β-sheet, but showed characteristic features of dominantly unordered structure. The
α
α
conformation with very low
spectra of aggregated A 1
β
β
after 24 h exhibited a characteristic pattern of β-sheet conformation
-42
with absorption minimum around 217 nm (Fig. 6C), which indicate that disordered Aβ1-42
monomers aggregated into
β-sheet rich fibrillar aggregates after the incubation. However, the
treatment of 9a and 9b to A 1
β
peptide permitted different structural transitions compared with
-42
those observed when A 1
β
was incubated alone (Fig. 6C). Intriguingly, CD spectrum clearly
-42
assessed that these compounds caused a decrease in the CD ellipticity measured at 217 nm,

ACCEPTED MANUSCRIPT
indicating their preventing effect on A
β
from adopting a
β
-sheet rich conformation. These results
revealed that test compounds could reduce the
further aggregation into mature fibrils.
β-sheet structure formation thereby hampering
Fig.5. CD spectroscopy of Aβ1−42 alone or with compounds 9a and 9b incubated for 0 h (A), 12 h (B) and 24 h
C). Aβ1-42 (20 µM) was incubated either alone in 20 mM sodium phosphate buffer (pH 7.4) at 37 °C, or in the
presence of 20 µM 9a and 9b.
(
2
.3.6. Inhibition of A
β1–42 fibril formation monitored by transmission electron microscopy
TEM
Next, TEM study was employed in order to monitor and probed the degree of A
β
aggregation after compound (9a and 9b) treatment. At 0 h time point, there was no aggregation

ACCEPTED MANUSCRIPT
in the samples containing A
β1–42 with (Fig. 6A) or without compounds (data not shown).
However, after 24 h of incubation at 37 °C, the sample of A 1
β
alone had mostly aggregated
–42
into dense, mature and bulky amyloid fibrils (Fig. 6B). In contrast, only fewer and shorter
aggregates were visible in the electron micrographs of A 1 samples after incubation with
β
–42
compounds 9a and 9b (Fig. 6D and E) under the same experimental conditions as compared to
standard curcumin (Fig. 6C). The TEM results were in agreement with the results of ThT
binding studies, which further proved that these compounds could effectively inhibit and slow
down the rate of Aβ1–42 fibrils formation in vitro.
Fig. 6. TEM image analysis of Aβ1–42 aggregation in the presence of 9a and 9b. (A) Aβ1–42 (25 µM), 0 h. (B)
Aβ1–42 alone (25 µM) was incubated at 37 °C for 24 h. (C) Aβ1–42 (25 µM) and curcumin (25 µM) were
incubated at 37 °C for 24 h. (D) Aβ1–42 (25 µM) and 9a (25 µM) were incubated at 37 °C for 24 h. (E) Aβ1–42
(
25 µM) and 9b (25 µM) were incubated at 37 °C for 24 h.
2
.3.7. Inhibition of AChE-induced A 1
β
aggregation
-42
The discovery that AChE plays additional roles, besides its "classical" function in
terminating synaptic transmission, has gained further attention in its role as target for AD
therapy. In particular, PAS of AChE has been identified as a site promoting non-cholinergic
functions by accelerating the assembly of Aβ fibrils that contribute towards the neurotoxic
cascade of AD [30][31][32]. On this premise, novel classes of AChE inhibitors targeting PAS

ACCEPTED MANUSCRIPT
have emerged as promising disease-modifying AD drug candidates. Thus, the triazine derivatives
(9a and 9b) showing best AChE inhibitory potency were further selected to assess their
inhibitory activity on the AChE-induced A 1 peptide aggregation by employing the same ThT-
β
-42
based fluorometric assay. In particular, test compounds were screened at a single concentration
and compared with reference compound donepezil. After 24 h of incubation, there observed a
remarkable increase in the intensity of ThT fluorescence signal from co-incubated samples of
AChE-A
fluorescence yield was significantly reduced in the presence of compounds 9a and 9b (Fig. 7) by
45% and ~58%, respectively, which was higher than that of donepezil (22.8%). The results of
ThT assay further confirmed the dual mode of inhibition and suggested that targeted compounds
a and 9b may have a potential disease-modifying role in AD therapy.
β which clearly indicated acceleration of Aβ fibrillogenesis by AChE. This increase
~
9

ACCEPTED MANUSCRIPT
Fig. 7. ThT emission fluorescence spectra (range 450-600 nm) in the absence of AChE (a) Aβ1-42 alone, and in
the presence of (b) Aβ1-42+AChE, (c) 9a+Aβ1-42+AChE, (d) 9b+Aβ1-42+AChE, (e) Overlay of ThT emission
fluorescence spectra, (f) The inhibitory activity of test compounds to AChE-induced Aβ1-42 aggregation in ThT
binding assay. The concentration of Aβ1-42 was 25 µM, and the concentration ratio of Aβ1-42, AChE, compound
was 100:1:100.
2
.3.8. Metal-chelating properties of compound 9a and 9b
The ability of 9a and 9b to chelate biometals such as Cu , Zn and Fe was studied by
UV-vis spectroscopy [33]. Upon the addition of CuSO , a red shift in the maximum absorption
2
+
2+
2+
4
2+
from 208 nm to 214 nm and 228 nm to 238 nm occurred, indicating the formation of a 9a-Cu
complex (Fig. 8a). The maximum absorption at 228 nm exhibited a significant shift when FeSO4
2
+
2+
and ZnCl was added, suggesting that 9a binds to Fe and Zn . Interestingly, when the same
2
experiment was performed with compound 9b, the maximum absorption at 228 nm was shifted
2
+
2+
to 234 or 236 nm, respectively, suggesting complexes formation of 9b-Fe and 9b-Zn ,
respectively (Fig. 8b).
2
+
The stoichiometry of the 9a-Cu complex was investigated using the molar ratio method,
by preparing solutions of compound 9a with increasing amounts of CuSO . The UV spectra were
4
used to obtain the absorbance of the complex of 9a and CuSO at different concentrations at 238
4
nm. As indicated in Fig. 8c, the absorbance linearly increased at first. When the mole fraction of
2
+
Cu to 9a was more than 0.9, the absorbance tended to be stable. Therefore, two straight lines
were drawn with the intersection point at a mole fraction of 0.9, which implied a 1:1
2
+
2+
stoichiometry for the 9a-Cu complex. The stoichiometric ratio of the 9b-Cu complex was
determined at 236 nm (Fig. 8d). The mole fraction was found to be 0.96, suggesting a 1:1
stoichiometry for the complex. These observations reflect the ability of test compounds to

ACCEPTED MANUSCRIPT
chelate biometals, and therefore could serve as potential metal chelators for AD therapy.
Fig. 8. (a) UV spectrum of compound 9a (50 µM) alone and in the presence of CuSO
4 4
(50 µM), FeSO (50
µM) and ZnCl (50 µM). (b) UV spectrum of compound 9b (50 µM) alone and in the presence of CuSO (50
2
4
µM), FeSO (50 µM) and ZnCl (50 µM). (c) Determination of the stoichiometry of complex 9a-Cu(II) by
4
2
Job’s method. (d) Determination of the stoichiometry of complex 9b-Cu(II) by Job’s method.
2
+
2
.3.9. Effect of compounds 9a and 9b on Cu -induced Aβ1–42 aggregation
To investigate the ability of the triazine derivatives to inhibit metal-induced Aβ
aggregation, we studied the effect of most potent compounds 9a and 9b on metal-induced Aβ1–42
aggregation by ThT fluorescence assay. Curcumin was used as reference compound. As shown
2
+
in Fig. 9, the fluorescence of Aβ treated with Cu is 156.4 % that of Aβ alone (100 %), which
2
+
demonstrates that Cu accelerates Aβ aggregation. On the contrary, the fluorescence of Aβ
2
+
treated with Cu and the tested compounds decreased dramatically (9a, 69.7 % inhibition of
2
+
Cu -induced Aβ aggregation; 9b, 66.6 % inhibition and curcumin, 60.28 % inhibition) at a
2
+
concentration ratio of 1:1:2 [Aβ (25 µM):Cu (25 µM):Inhibitor (50 µM)]. These results

ACCEPTED MANUSCRIPT
2
+
suggested that our compounds could inhibit Cu -induced Aβ1–42 aggregation noticeably through
2
+
chelating with Cu ions.
2+
Fig. 9. ThT binding assay for Cu -induced Aβ aggregation, and test compound induced Aβ disaggregation
Values are reported as the mean ± SD of three independent experiments. Statistically significant differences
2+
from Aβ1−42 + Cu were analyzed by ANOVA: ***p < 0.001.
2
+
2
.3.10. Inhibition of compounds 9a and 9b on Cu -induced Aβ
aggregation monitored by
–42
1
TEM
2
+
To further confirm the ability of compounds 9a and 9b to inhibit Cu -mediated Aβ
1
–42
2
+
aggregation, TEM analysis was performed. As depicted in Fig. 10C, Cu (25 µM) could
accelerate the aggregation of Aβ1–42 (25 µM) which was evident by the presence of denser
fibrillar aggregates compared with Aβ1–42 alone samples (Fig. 10B) under the same experimental
conditions. However, there observed a alteration in the metal-triggered Aβ aggregation upon 9a
and 9b (50 µM each) addition as indicated by the presence of only fewer Aβ aggregates which
clearly suggested that these compounds provided considerable disaggregation effect (Fig. 10E
and 10F). The observations of TEM are consistent with ThT binding assay results which

ACCEPTED MANUSCRIPT
suggested that metal chelation by these compounds may be one of the driving forces in altering
the structural organization of Aβ aggregates.
2
+
Fig. 10. TEM images for Cu -induced Aβ aggregation and test compound induced Aβ disaggregation. ([Aβ
1
–
2+
42]
B) Aβ1–42 alone; (C) Aβ1–42 + Cu , (D) Aβ1–42 + Cu + 9a; (E) Aβ1–42 + Cu + 9b; (F) Aβ1–42 + Cu
curcumin).
= 25 µM, [Cu ] = 25 µM, [9a] = 50 µM, [9b] = 50 µM, [curcumin] = 50 µM, 37 °C, 24 h; (A) Aβ , 0 h;
1–42
2+
2+
2+
2+
(
+
2
.3.11. Bioavailability and ADME parameters screening for drug likeness
The assessment of pharmacokinetic properties is one of the critical stages in the drug
development process. In silico ADMET analysis can be used to check the lead candidates for
their plausible drug likeness and bioavailability. The computational tool QikProp, v. 3.5
available in Schrödinger software has been used to study the pharmacokinetic profile of the
targeted compounds [34]. A few significant indicators of their pharmacokinetic profiles were
considered with special emphasis on the requirements of a CNS active drug (Table S1,
Supplementary material). Our results indicated that most of the compounds showed drug like
characteristics based on Lipinski’s rule of five (mol_MW < 500, QPlogPo/w < 5, donorHB ≤ 5,
accptHB ≤ 10). Based on the predicted values for QPlogBB and CNS activity, all compounds

ACCEPTED MANUSCRIPT
might be able to penetrate into the CNS. Generally, drugs aimed at the CNS tends to have lower
2
polar surface area (PSA) usually between 60 and 70 Å [35]. The calculated theoretical PSA
values for all compounds may support their ability to penetrate the blood-brain barrier. The
distribution of the compound in the human body depends on factors such as blood-brain barrier
(QPLog BB), permeability such as apparent Caco-2 permeability (QPPCaco), apparent MDCK
cell permeability (QPPMDCK), volume of distribution and plasma protein binding (QPLogKhsa
for Serum protein binding). All of these compounds exhibited high permeability for both in vitro
Caco-2 cells and in vitro MDCK cells. Moreover, all compounds displayed good oral absorption.
The predicted QPlogKhsa values of all compounds were found to be within acceptable range
which indicates their strong binding with plasma protein. The predicted ADME properties
revealed that these compounds possess appropriate pharmacokinetic profiles required topenetrate
BBB and therefore, could be considered as good candidate for drug development.
2
.3.12. Toxicity prediction
Toxicity predictor (TOPKAT module of Accelrys Discovery Studio 4.0) computes a
probable value of toxicity for a submitted chemical structure using quantitative structure-activity
relationship (QSTR) equation and locates fragments within the compound that indicate a
potential threat to toxicity risk. Parameters for toxicity risk such as carcinogenicity,
mutagenicity, skin irritancy, ocular irritancy, aerobic biodegradability and developmental
toxicity potential etc., were observed for synthesized compounds 9a-b, 10a-f and 3a-g. The
results of toxicity risk for triazine derivatives showed moderate to good drug score. In general,
toxicity screening results of TOPKAT for USFDA and NTP predicted rodent carcinogenicity,
Ames mutagenicity, developmental toxicity potential, aerobic biodegradability, ocular irritancy
and skin irritancy showed positive response for most of the triazine derivatives (Table S2,