Pharmacological investigation of quinoxaline-bisthiazoles as multitarget-directed ligands for the treatment of Alzheimer's disease

Article abstract of DOI:10.1016/j.bioorg.2019.102992

Alzheimer's disease (AD) is the most prevalent disease of old age leading to dementia. Complex AD pathogenesis involves multiple factors viz. amyloid plaque formation, neurofibrillary tangles and inflammation. Herein we report of a new series of quinoxaline-bisthiazoles as multitarget-directed ligands (MTDLs) targeting BACE-1 and inflammation concurrently. Virtual screening of a library of novel quinoxaline-bisthiazoles was performed by docking studies. The most active molecules from the docking library were taken up for synthesis and characterized by spectral data. Compounds 8a-8n showed BACE-1 inhibition in micro molar range. One of the compounds, 8n showed BACE-1 inhibition at IC₅₀ of 3 ± 0.07 μM. Rat paw edema inhibition in acute and chronic models of inflammation were obtained at 69 ± 0.45% and 55 ± 0.7%, respectively. Compound 8n also showed noteworthy results in AlCl₃ induced AD model. The treated rats exhibited excellent antiamnesic, antiamyloid, antioxidant, and neuroprotective properties. Behavioural parameters suggested improved cognitive functions which further validates the testimony of present study. Moreover, compound 8n was found to have inherent gastrointestinal safety. This new string of quinoxaline-bisthiazoles were identified as effective lead for the generation of potent MTDLs and compound 8n was found to showcase qualities to tackle AD pathogenesis.

Full text of DOI:10.1016/j.bioorg.2019.102992

Accepted Manuscript
Pharmacological investigation of quinoxaline-bisthiazoles as multitarget-direc-
ted ligands for the treatment of Alzheimer’s disease
Sneha R. Sagar, Devendra Pratap Singh, Rajesh D. Das, Nirupa B. Panchal,
Vasudevan Sudarsanam, Manish Nivsarkar, Kamala K. Vasu
PII:
S0045-2068(18)31298-7
https://doi.org/10.1016/j.bioorg.2019.102992
102992
DOI:
Article Number:
Reference:
YBIOO 102992
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
10 November 2018
7 April 2019
17 May 2019
Please cite this article as: S.R. Sagar, D. Pratap Singh, R.D. Das, N.B. Panchal, V. Sudarsanam, M. Nivsarkar, K.K.
Vasu, Pharmacological investigation of quinoxaline-bisthiazoles as multitarget-directed ligands for the treatment
of Alzheimer’s disease, Bioorganic Chemistry (2019), doi: https://doi.org/10.1016/j.bioorg.2019.102992
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.

Pharmacological investigation of quinoxaline-bisthiazoles as multitarget-directed ligands
for the treatment of Alzheimer’s disease
Sneha R. Sagara,c, Devendra Pratap Singhb,#, Rajesh D. Dasa, Nirupa B. Panchala,c, Vasudevan
Sudarsanama, Manish Nivsarkarb, Kamala K. Vasua,*
aDepartment of Medicinal Chemistry, B. V. Patel Pharmaceutical Education and Research
Development (PERD) Centre, S. G. Highway, Thaltej, Ahmedabad - 380054, Gujarat, India.
bDepartment of Pharmacology and Toxicology, B. V. Patel Pharmaceutical Education and
Research Development (PERD) Centre, S. G. Highway, Thaltej, Ahmedabad - 380054, Gujarat,
India.
cInstitute of Pharmacy, NIRMA University, S. G. Highway, Chandlodia, Gota, Ahmedabad -
382481, Gujarat, India.
*Corresponding author. Tel.: + 91 79 27439375; Fax: + 91 79 27450449
E-mail: kamkva@gmail.com, kamalav@perdcentre.com
‡Communication Ref. No.: PERD96022018
#Present address: Manager R&D, (FDD-NDDS), Sun Pharmaceutical Industries Limited,
Tandalija, Vadodara, Gujarat.
Abstract
Alzheimer’s disease (AD) is the most prevalent disease of old age leading to dementia. Complex
AD pathogenesis involves multiple factors viz. amyloid plaque formation, neurofibrillary tangles
and inflammation. Herein we report of a new series of quinoxaline-bisthiazoles as multitarget-
directed ligands (MTDLs) targeting BACE-1 and inflammation concurrently. Virtual screening
of a library of novel quinoxaline-bisthiazoles was performed by docking studies. The most active
molecules from the docking library were taken up for synthesis and characterized by spectral
data. Compounds 8a-8n showed BACE-1 inhibition in micro molar range. One of the
compounds, 8n showed BACE-1 inhibition at IC50 of 3 ± 0.07 µM. Rat paw edema inhibition in
acute and chronic models of inflammation were obtained at 69 ± 0.45% and 55 ± 0.7%,
respectively. Compound 8n also showed noteworthy results in AlCl3 induced AD model. The
treated rats exhibited excellent antiamnesic, antiamyloid, antioxidant, and neuroprotective
properties. Behavioural parameters suggested improved cognitive functions which further
validates the testimony of present study. Moreover, compound 8n was found to have inherent
gastrointestinal safety. This new string of quinoxaline-bisthiazoles were identified as effective

lead for the generation of potent MTDLs and compound 8n was found to showcase qualities to
tackle AD pathogenesis.
Keywords
Quinoxalines, thiazoles, BACE-1, Inflammation, Multitarget-directed ligands

1. Introduction
Alzheimer’s disease (AD) is chronic, irreversible and the most prevalent
neurodegenerative disorder of the brain concomitant to old age and currently accounts over 46.8
million cases of dementia worldwide.[1-3] Primarily characterized by exacerbated cognitive
dysfunction and irreversible memory impairment. Clinically, AD can be explicated as mild to
severe cognitive decline accompanied by gradual dysfunction both in memory and areas of
higher intellectual functions such as linguistic skills, judgments in complicated situations and
basic routine works. Major victims of AD are people of the age groups of 65 years and older;
however the onset may occur earlier during their 40s or 50s which remains un-noticed. In 2015,
there were as many as 46.8 million cases of dementia globally and projections are likely that this
figure may double every 20 years, rising to 74.7 million by 2030 and 131.5 million by 2050 if an
efficient systematic treatment methodology is not developed.[4]
Although no substantial evidences to compel the exact pathophysiology of AD are yet
identified, certain epidemiological and preliminary researches largely attribute the underlying
mechanism to two major hallmarks viz. (i) β-amyloid (Aβ) plaques and (ii) neurofibrillary
tangles (NFTs).[5] Additionally, other factors affecting the pathophysiology of this malady in
some way or the other are inflammation, excitotoxicity, neurotransmitter imbalance, oxidative
stress, elevated cholesterol level, dysregulation of biometal ions (especially Cu+2, Zn+2 and Fe+2),
etc.[6,7] The treatment of AD has endured a challenge for the pharmaceutical community as yet.
The pathology of AD is intricately cumbersome owing to its association with abnormality and
dysfunction of multisystem. Thus, the prevention and real cure have been elusive till date and its
molecular etiology still remains enigmatic. Current pharmacotherapy being palliative in nature
prompts timely symptomatic relief and ameliorates patient’s condition but do not produce
profound disease modifying effects.
Accordingly, the search for viable disease modifying agents could be addressed by shifting the
drug discovery paradigm from single to multitarget-directed ligands which are more requisite in
addressing the labyrinthine nature of the disease. The MTDL strategy has emerged as a powerful
drug design exemplar after the multitarget approach started on the discovery of multiple
activities of natural products such as curcumin, resveratrol and some flavonoids.[8] Ever since the
benefits of MTDLs therapy over single target drug design were noticed, a lot of MTDLs were

studied to combat the disease modifying complications associated with AD.[9-19] In this
connection, we arrived at a cognizance that there is an unmet need to develop antiamyloid
therapy as well which can hit distinct molecular targets of the neurodegenerative cascade.[20-22] In
this study, we have designed MTDLs acting upon amyloid pathway enzyme viz. β-site APP
cleaving enzyme (BACE)-1 and inflammation. Our study aims at combining two scaffolds into
one single molecule to generate potential lead for complex AD intervention.
A retrospection of the concept of amyloid hypothesis revealed that the amyloid precursor
protein (APP) gets cleaved by two different pathways: “non-amyloidogenic” and
“amyloidogenic”.[23] Non amyloidogenic pathways were not detected to be involved in Aβ
generation whereas amyloidogenic pathway products were found to be primarily responsible for
the formation of β-amyloid plaques. The APP first gets cleaved by β-secretase enzyme and
subsequently by γ-secretase provoking the release of β-amyloid into the extracellular cerebral
site. Consequently, the cleaved products of APP form amyloid-β (Aβ) peptides: Aβ40 and Aβ42.
Of these, Aβ42 is identified as immensely neurotoxic and participates in plaque formation in the
pathogenesis of AD.[5,24] Besides the amyloid plaque formation, an activation of inflammatory
processes also affects the progression of AD. An inference made from the available literature is
that pro-inflammatory substances like reactive oxygen species (ROS), interleukin (IL) 1 & 6,
tumor necrosis factor (TNF) and cyclooxygenases (COXs) play a major role in degeneration and
death of the neuronal cells.[25] In short, we can conclude that BACE-1 is fundamentally
responsible for the generation of neurotoxic β-amyloid (Aβ) peptide which is broadly considered
to have a crucial early role in etiology of AD. As a result, BACE-1 has originated as a key drug
target to taper the extension of toxic Aβ in the AD brain, and the development of BACE-1 and
inflammation inhibitors as therapeutic agents is being eagerly executed. The morbid
pathophysiology of BACE-1 and neuronal inflammation in AD is pictured in Figure 1.
Insert Figure 1. Here
Figure 1. Pathogenesis of Alzheimer’s disease[26]
1.1.
Design of molecules
The therapeutic profiles of various MTDLs explored so far, seem to confer with the
ability of quinoxalines arising as a useful ring scaffold in AD therapy. For example, 1-(2,3-
di(furan-2-yl)quinoxalin-6-yl)-3-(3-hydroxypropyl)urea (1) showed inhibition of tau
fibrillization with IC50 of 2.4 µM.[27] Hexyl 2-cyano-2-(3-(4-methylpiperazin-1-yl)quinoxalin-2-

yl)acetate (2) was also evaluated as inhibitor of tau, α-synuclein and amyloid β1-42. IC50 of the
compound (2) against tau protein and amyloid β1-42 was obtained at 2.8 µM and 0.76 µM,
respectively.[28] N-(3-aminoquinoxalin-2-yl)-4-(piperidin-1-ylmethyl)benzamide (3) showed 47%
inhibition of BACE-1 at 20 µg/mL concentration.[29] These scaffolds warrant the therapeutic
ability of quinoxalines against AD which is further demonstrated in Figure 2. Additionally, our
lab has a profound interest in exploring various heterocycles like thiazoles, thiophenes, and
quinazolinones in varied therapeutic regions.[30-32] As a result, potent anti-inflammatory
properties of thiazoles have been proven in in vivo rat model. Ethyl (5-((hydroxyimino)(4-
(methylsulfonyl)phenyl)methyl)-4-methylthiazol-2-yl)carbamate (4) showed substantial anti-
inflammatory activity with 69% inhibition of rat paw edema in acute inflammation.[30] N,N'-(5,5'-
oxalylbis(4-methylthiazole-5,2-diyl))dibenzamide (5) showed 73% inhibition of edema in acute
inflammation model.[30] The collaborative assembly of two ring scaffolds into a single molecule
to act as BACE-1 and inflammation inhibitor is portrayed in Figure 2.
Insert Figure 2. Here
Figure 2. Design strategy for a novel series of MTDLs in AD
2. Results & Discussion
The present study encapsulates development of such chemical entities featuring abilities to hit
multiple targets associated with AD. The string of prototype MTDLs designed in this study
showed remarkable inhibition of BACE-1 enzymatic pathway and also exposed potential anti-
inflammatory tendencies. To delineate this fact, in silico docking studies, in vitro, in vivo and GI
safety studies have been carried out comprehensively and the interpretation of the results have
been discussed in following sections.
2.1. Molecular Docking
The rationale of the present work was initiated by molecular docking studies.[33] A library of
molecules was scrutinized by in silico docking studies employing BACE-1, BACE-2, Cathepsin
D, COX-1 and COX-2 enzymes to predict the binding affinity. To achieve selectivity towards
BACE-1 over BACE-2 and Cathepsin D, binding interaction with the catalytic pocket of BACE-
1 was mandatory. The catalytic pocket of BACE-1 has an aspartic dyad at the centre which
forms hydrogen bonding network in the active site. The molecular docking study bespoke that
some of the molecules displayed binding interactions with the aspartic dyad of BACE-1. Lower
binding affinity towards BACE-2 and Cathepsin-D was beneficial to get the selective BACE-1

inhibition as non-selective inhibition may lead to the cell death therefore only molecules
selectively inhibiting BACE-1 over BACE-2 and Cathepsin-D were further assessed.
Additionally, docking studies on inflammatory pathway enzymes COX-1 and COX-2 were
carried out. Higher binding affinity and docking score for COX-1 and COX-2 was advantageous.
From the library of designed compounds, 14 molecules 8a-n were selected for further studies.
The results of in silico screening of designed MTDLs for BACE-1 inhibitory and anti-
inflammatory pathways are depicted in Table 1. From the virtual library, compounds 8a-n
showed higher binding affinity with BACE-1, COX-1 and COX-2 enzymes. Compounds with
lesser docking score values are incorporated in Supplementary Material.
Insert Table 1 here
The binding interactions of compound 8n against BACE-1 and COX-2 enzymes are pictured in
Figure 3. Figure 3A illustrates an overlay of docked compound 8n and BACE-1 inhibitor
AZD3839 (reference molecule) with BACE-1 enzyme’s catalytic active site. AZD3839 showed
H-bonding interactions with catalytic aspartic dyad Asp32 and Asp228. In addition, it also
interacted with Phe108 and Trp76. π-π stacking interaction was observed with amino acid
residue Tyr71. Similarly, compound 8n also showed binding interactions with catalytic Asp32,
Asp228, Trp76 and Tyr71. Docking score value obtained for AZD3839 and compound 8n was
found to be -8.8 and -7.8, respectively. Figure 3B illustrates an overlay of docked compound 8n
and a COX-2 inhibitor celecoxib with COX-2 enzyme’s catalytic region. The H-bonding
interactions with amino acids Leu338, Ser339 and Phe504 and π-cation interaction with Arg106
were observed for celecoxib whereas in case of compound 8n interactions were observed with
Phe504, Leu338 and Arg106.
Insert Figure 3A and 3B here
Figure 3. Docking study results
3A) Overlay of compound 8n (green sticks) with AZD3839 (orange sticks): AZD3839 showed H-bonding
interactions (Yellow dotted line) with Asp32, Asp228, Trp76 and Phe108 and π-π stacking interaction (cyan dotted
line) with Tyr71. Compound 8n interacted with Asp32, Asp228, Trp76 and Tyr71.
3B) Overlay of compound 8n (green sticks) with Celecoxib (pink sticks): Celecoxib showed H-bonding
interactions (Yellow dotted line) with Leu338, Ser339 and Phe504 and π-cation interaction (green dotted line) with
Arg106. Compound 8n also interacted with Phe504, Leu338 and Arg106.
2.2. Physicochemical properties prediction

Drug likeliness prediction of drugs targeting BBB is an important criteria to look into. Qikprop
module of Schrödinger was used for the prediction of physicochemical properties (Table 2).[34]
Molecules obeyed the Lipinski’s rule of five and Jorgensen’s rule of three. They showed drug
likeliness properties with higher BBB permeation and lower toxicity according to HERG
inhibition which is evident from Table 2. Physicochemical parameters like, QPlogPo/w,
QPlogBB, %HOA, QPPCaco and QPlogHERG are enlisted in Table 2.
Insert Table 2 here
2.3. Chemistry
The active compounds identified through docking studies were synthesized as per scheme
mentioned in Figure 4. The substituted 2,3-bis(bromomethyl)quinoxalines 2a-c were obtained by
reaction of substituted benzene-1,2-diamines 1a-c with 1,4-dibromobutane-2,3-dione in ethanol
at 0°C for 2 hours. To afford intermediates N'-carbamothioylformimidamides 4a-d,
carboxamidines 3a-c were reacted with various isothiocyanates in tetrahydrofuran at room
temperature for about 3 to 4 hours.[35,36] Synthesis of substituted N'-carbamothioyl-N,N-
dimethylformimidamides 6a-e was carried out via reaction of substituted thiourea 5a-e with 1,1-
dimethoxy-N,N-dimethylmethanamine in methanol for 2 hours at ambient temperature. The
treatment of substituted 2,3-bis(bromomethyl)quinoxalines 2a-c with substituted N'-
carbamothioylformimidamides 4b-d afforded tritylated intermediates 7a-f. Upon deprotection
with trifluoro acetic acid final compounds 8a-f were obtained. Also, intermediates 2a-c were
reacted with 4a and 6a-e to produce final compounds 8g-n.
On the basis of 1H NMR spectra of intermediates 2a and 4a, the formation of final compound 8g
can be explained. Intermediate 2a, 2,3-bis(bromomethyl)quinoxaline has molecular formula of
C10H8Br2N2. Its 1H NMR spectra shows a singlet at 4.338 δ (a) which corresponds to four protons
of two –CH2 groups. Further, a triplet of two protons is obtained at 7.331 δ (b) and a doublet of
two protons is obtained at chemical shift value of 7.821 δ (c) (See Supplementary Material). An
another intermediate, N,N-diethyl-N'-(phenylcarbamothioyl)benzimidamide 4a has molecular
1
formula of C18H21N3S. The H NMR spectra of this compound shows a triplet of two –CH3
groups (6H) at chemical shift of 1.089 δ (a’). A quartet peak is obtained at 3.187 δ corresponds
to four protons of two –CH2 groups (b’). In the aromatic region, peaks corresponding to two

phenylic protons are obtained as multiplet at 7.142-7.472 δ (c’ to g’). A singlet for proton
corresponding to –NH group is obtained at 10.290 δ (h’) (See Supplementary Material).
The final compound 5,5'-(quinoxaline-2,3-diyl)bis(N,4-diphenylthiazol-2-amine) 8g has
molecular formula of C38H26N6S2. Compound 8g is formed by reaction of intermediate 2a (1 eq)
with intermediate 4a (2 eq). In compound 8g, two –CH2 group’s peak of intermediate 2a (a) is
disappeared as it is now cyclized to form thiazole scaffold. Also, peaks of –CH2-CH3 group
disappeared in final compound’s spectra (a’ and b’) which were seen in case of intermediate 4a
(–N(CH3)2 group acts as readily leaving group upon cyclization). Importantly, there are four
phenyl groups in compound 8g and the corresponding aromatic protons are obtained in the range
of 6.97-7.83 δ (20H, a’’ to g’’). Additionally, peaks of protons present in the quinoxaline ring
scaffold remain unchanged as in case of 2a (4H, f’’ and g’’). In 8g, there is a singlet peak
corresponding to two –NH group protons (h’’) (See Supplementary Material).
Insert Figure 4 Here
Figure 4. Synthetic scheme of quinoxaline-bisthiazoles
2.4. Biological activity
2.4.1. In vitro BACE-1 enzyme inhibition assay
The therapeutic profile of MTDLs as BACE-1 inhibitor was practically assessed by FRET based
in vitro enzymatic assay. All the synthesized compounds (8a-8n) were evaluated for their
BACE-1 enzyme inhibitory activity and the results were compared with the clinical candidate’s
(AZD3839) reported value. IC50 values in terms of µM ± SEM (n=3) are enlisted in Table 3.
Results suggested that all compounds inhibited BACE-1 enzyme in micro molar range.
Compounds 8b, 8l and 8n were found have maximal potency with IC50 equivalent to 4, 8 and 3
µM, respectively. This consequence of high potency can be attributed to their high binding
affinity with BACE-1 enzyme which was previously evaluated from docking study. Compound
8k and 8m showed minimal potency with IC50 values above 100 µM. Compounds 8a, 8c, 8d, 8e,
8f, 8g, 8h, 8i, and 8j showed moderate inhibition of BACE-1 enzyme. In order to inhibit BACE-
1 enzyme, one of the structural features required is ‘–NH2’ functionality as it forms tight H-bond
with catalytic aspartic acids Asp32 and Asp228 of BACE-1 enzyme. The substitution at R2
position gives less active compounds (see Table 3). Quinoxaline ring scaffold is responsible for
π-π stacking interaction with Tyr71. Ring ‘N’ of thiazole ring is also significant for H-bonding

interactions with Trp76. In addition, substitution at R3 position with ‘-N(CH3)2’ group did not
give potent BACE-1 inhibition as in case of compounds 8c, 8d and 8e. BACE-1 inhibition was
hardly affected by the substitution at R1 position.
Insert Table 3 here
2.4.2. In vivo analysis
2.4.2.1. Acute anti-inflammatory activity: Carrageenan induced rat paw edema study
In addition, the anti-inflammatory potential of these MTDLs was tested by studying rat model of
acute and chronic inflammation. All 14 molecules were examined via carrageenan induced acute
inflammation model. A subcutaneous injection of carrageenan induces inflammation by
metabolism of arachidonic acid. This causes plasma and leukocyte extravasations, increased
tissue water and plasma protein exudation along with neutrophil extravasations. The above
inflammatory event was diminished by our string of MTDLs at 50 mg·kg-1 dose. Compounds 8c,
8i, 8l and 8n showed anti-inflammatory activity with % edema inhibitions of 60%, 55%, 57%
and 69%, respectively. On the contrary, standard drug diclofenac given at a dose of 7.5 mg·kg-1
showed edema inhibition of 72% (See Table 3).
2.4.2.2. Chronic anti-inflammatory activity: Formalin induced rat paw edema study
Further, the most active molecules obtained from in vitro BACE-1 assay and in vivo acute
inflammatory studies i.e. compounds 8c, 8g, 8l, and 8n were taken forward to chronic model of
inflammation. Formalin induces early neurogenic response followed by tissue mediated
responses by releasing histamine, 5-HT, kinin and prostaglandins. A formalin induced chronic
rat paw edema model was developed and molecules were tested consecutively for 5 days. The
results indicated the % edema inhibition of 28%, 49%, 31% and 55%, respectively at a dose of
50 mg kg-1. In this instance, the standard drug celecoxib (40 mg·kg-1) given to rats showed 68%
edema inhibition when tested under same conditions and parameters. The results are shown in
Table 4.
One of the structural features required for the anti-inflammatory activity is indeed the presence
of thiazole ring moiety as ring “N” forms H-bond with both COX-1 and COX-2 enzymes.
Moreover, anti-inflammatory activity is favored by –NH2 functional group as seen in compounds
8c, 8l and 8n. Substitution at R1 position was not majorly affected by different groups. At R3
position, –CH3 and -C6H5 groups did not favor the anti-inflammatory activity so, no substitution
is required there.

Insert Table 4 here
The most optimal hybrid molecule 8n resulting from BACE-1 inhibition assay, acute and chronic
inflammation model was further taken up to analyze its anti-Alzheimer’s activity by employing
AlCl3 induced AD model. Compound 8n was screened by behavioural parameters, antioxidant
property and histopathological studies.
2.4.2.3. Behavioural parameters
2.4.2.3.1. Y-maze test
Simultaneous alteration performance (SAP) was determined using Y-maze apparatus and %
alteration was calculated.[37] During acquisition session, rats in all the groups showed preference
for novel environment and higher % alteration as depicted in Figure 5A. There was no significant
difference of % alteration during this period. In contrast, on 7th, 14th, 21st and 28th day DC group
animals showed significant decrease in the % alteration parameter compared to the NC group
(*p<0.05). Treatment with compound 8n reduced the amnesic effect aroused from AlCl3
injections and significantly improved % alteration (#p<0.05).
2.4.2.3.2. Conditioned avoidance response test
Using pole climbing apparatus, conditioned avoidance response (CAR) was evaluated in all test
groups. During the acquisition time, rats learned to escape from the foot shock and showed low
CAR (sec) values. None of the groups showed any significant differences of CAR during
acquisition session, whereas on day 7th, 14th, 21st and 28th day of the test session, CAR
substantially increased in DC group rats when compared to NC rats (*p<0.05). Treatment with
compound 8n improved the learning memory of rats and showed significant decrease in the
CAR, contrary to which is observed in DC group rats (#p<0.05) as shown in Figure 5B.
2.4.2.3.3. Elevated plus maze test
Spatial working memory of rats was investigated using EPM test.[38] Time (sec) consumed by
each rat to travel from an open arm to a closed arm (transfer latency) was recorded. During
training period, none of the rats showed any significant difference in their transfer latency after
having memorized the path. Further, AlCl3 treatment in DC group rats significantly increased the
transfer latencies (*p<0.05) on 7th, 14th, 21st and 28th day and the rats took more time to reach a
closed arm from an open arm than they usually did in training sessions. Improvement in spatial

working memory was noted in compound 8n treated rats and they showed significant decrease in
transfer latency (#p<0.05) compared to DC group rats, as depicted in Figure 5C.
Insert Figure 5A, 5B and 5C here
Figure 5A, 5B and 5C. Behavioral parameters estimation
A) Results of Y maze test expressed as % alteration B) results of conditioned avoidance response (CAR) using pole
climbing apparatus C) results of transfer latency using elevated plus maze test. Data expressed as mean ± SEM.
Asterisk denotes statistical significance (*p<0.05) when compared to NC group and (#p<0.05) when compared to
DC group, using two-way ANOVA followed by Tukey’s multiple comparison test.
2.4.2.4. Haematological parameters estimation
After 28 days of AlCl3 dosing, on the 29th day haematological parameters were observed.[39]
Haemoglobin and haematocrit levels were estimated using automated haematology analyser.
From the results of the mentioned study, it was observed that there was a significant reduction in
haemoglobin and haematocrit level in PC rats (*p<0.05) as given in Figure 5D & 5E. The cause
of decrease in the haemoglobin and haematocrit can be attributed to the gastric ulcers present in
rat’s stomach in meloxicam treated group. Treatment with compound 8n did not show decrease
in haemoglobin and haematocrit level. These results were further validated by gastrointestinal
lesion index study.
Insert Figure 5D and 5E here
Figure 5D and 5E. Estimation of haematological parameters
D) Estimation of haemoglobin (HGB) level E) Results of haematocrit (HCT) level. Data expressed as mean ± SEM
(*p<0.05) by using one-way ANOVA followed by Tukey’s multiple comparison test.
2.4.2.5. Estimation of serum biochemical parameters
Serum biochemical parameters were estimated by albumin and total protein level determination.
No significant difference was observed in albumin level in any of the groups (Figure 5F).
Nevertheless, significant reduction in the total protein content (*p<0.05) were observed in PC
rats as evident from Figure 5G. DC group and compound 8n treated rats did not show any
reduction in total protein level.
Insert Figure 5F and 5G here
Figure 5F and 5G. Estimation of serum biochemical parameters
F) Estimation of serum albumin level G) Estimation of total protein level. Data expressed as mean ± SEM (*p<0.05)
by using one-way ANOVA followed by Tukey’s multiple comparison test.
2.4.2.6. Antioxidant activity

The in vivo antioxidant activity was checked by lipid peroxidase assay (LPO) and superoxide
dismutase assay (SOD). Rat brain homogenates were subjected to MDA level determination
using UV/Visible spectrophotometer (Shimadzu UV-1800). MDA is an oxidative marker which
indicates oxidative damage of tissues. Aluminium causes oxidative damage in conjunction with
copper and iron and significantly increases the level of LPO in rat brain tissues. The mechanism
underlying the increase in LPO production is the stabilization of iron by aluminium in its
reduced state (Fe2+). AlCl3 induction increased the oxidative stress and stimulated MDA level
increment in DC animals. In contrast, treatment with 8n significantly reduced MDA level
compared to the DC rats (#p<0.05) as shown in Figure 5H. In addition to this, AlCl3 also causes
oxidative stress induced tissue damage by significant decrease in SOD activity. SOD value was
significantly increased in compound 8n treated rats in comparison to DC group rats (#p<0.05) as
depicted in Figure 5I. Importantly, the overall results of these studies indicates that compound 8n
showed promising free radical scavenging effects in the rat brain.
Insert Figure 5H and 5I here
Figure 5H and 5I. Anti-oxidative effect of compound 8n
H) Results of lipid peroxidation (LPO) assay expressed as MDA level I) Estimation of super oxide dismutase (SOD)
activity in rat brain homogenate. Data presented as mean ± SEM (n=6 animals/group) *p<0.05 compared to NC
group and #p<0.05 when compared to DC group using one-way ANOVA followed by Tukey’s multiple comparison
test.
2.4.2.7. Histopathological studies of rat brain
Histopathological evaluation of brain samples was carried out using Haematoxylin-Eosin (HE)
stain and Congo red stain. HE stained brain sections were observed under 100 X magnified
microscope and hippocampal region was closely examined (Figure 6A). NC rats were seen to
have intact healthy neurons in their hippocampus whereas, the DC group showed signs of
degeneration (indicated by an arrow). The PC group rats did not showed any degeneration but
slight thin hippocampal region was observed. Interestingly, compound 8n treated rats had an
intact and healthy hippocampus in their brain indicating that 8n protected neurons from
degeneration. Thus, results of this study signify the neuro-protective benefits of compound 8n.
Further, antiamyloid potential of compound 8n was estimated by using Congo red stain (Figure
6B). In the NC group brain cortex, no amyloid plaques were observed but the DC group had
amyloid plaques in their cortex region. No plaques were observed in PC group and 8n treated
rats as depicted in Figure 6B.

Insert Figure 6A here
Figure 6A. Typical microscopic HE stained sections (100X magnification) of brain
hippocampus
NC group rats with healthy and intact hippocampus, DC group rats had ruptured hippocampus and degenerated
neuronal cells, PC rats showed intact hippocampal region and Compound 8n treated rats with normal and healthy
neuronal cells in hippocampal region
Insert Figure 6B here
Figure 6B. Congo red stained sections (100X magnification) of brain cortex
NC group rat’s brain cortex without any amyloid plaques, Amyloid plaques were present in the cortex of DC group
rats (stained red) indicated by an arrow, PC group rats without plaques and Compound 8n treated brain cortex did
not show amyloid plaques
2.4.2.8. Gastrointestinal (GI) safety study
Nonsteroidal anti-inflammatory drugs (NSAIDs) have an inherent risk of causing gastro-
enreropathic damage to the GI track. On the contrary, test compound 8n interestingly didn’t
show any signs of GI damage. To check this aspect, macroscopic and microscopic evaluation of
gastrointestinal tract were carried out on 29th day of AlCl3 model. The internal portions of
stomach and intestines were examined and lesion indices were calculated for all rats. Control
group animals did not show any gastric damage in their stomach or intestines, whereas, PC group
rats had lesions in their stomach (lesion index = 12.19 ± 0.8 mm) and intestines (lesion index =
29.15 ± 1.6 mm). Compound 8n treated rats did not show any lesions in their intestine, however
a small damage was observed in the stomach region (lesion index = 3.07 ± 0.35 mm) as shown in
Figure 6C. Although compound 8n showed promising anti-inflammatory activity, gastric damage
was minimal as compared to the standard drug meloxicam. The gastric damage in stomach and
intestinal sections were further checked up by microscopic evaluation using HE stain. NC and
DC group rats did not show any erosions in their stomachs (see Figure 6D) and intestines (see
Figure 6E). PC group rats were noticed to have severe gastric injuries with focal erosions in
stomach and epithelium erosion and stratification in intestines. Importantly, compound 8n
treated rats showed minor focal erosion in their stomach but no damage was observed in
intestines.
Insert Figure 6C here
Figure 6C. Macroscopic evaluation of gastric damage by lesion index

Data given as mean ± SEM (*p<0.05) by using two-way ANOVA followed by Tukey’s multiple
comparison test.
Insert Figure 6D here
Figure 6D. Typical microscopic HE stained sections (under 100X magnification) of rat stomach
No stomach erosions present in NC rats, DC group rats without any erosions, PC rats stomach had large focal
erosions (indicated by an arrow), compound 8n treated rats had little erosions (indicated by an arrow)
Insert Figure 6E here
Figure 6E. Typical microscopic HE stained sections (under 100X magnification) of rat intestine
NC group rats did not show gastric damage, DC group rats without any focal erosions, PC group rats showed focal
erosions of the superficial epithelium and epithelial stratification indicated by an arrow, No erosions or epithelium
stratification was observed in compound 8n treated animals
3. Conclusion
The amyloidogenic cascade theory of Alzheimer’s disease suggest that secretases, the enzyme
acting upon β-APP facilitates the emanation of Aβ peptides and apparently emerges as one of the
key contributors in AD aetiology. The amyloid plaques and damaged neurons induce the release
of TNF-α, IL-1β, COXs and many other inflammatory cytokines leading to neuronal cell
degeneration and cell death. Therefore, BACE-1 promotion and neuronal cell inflammation are
critical processes related to the development of AD. Hence, it can be logically concluded that
BACE-1 inhibitory and anti-inflammatory leads can function as a therapeutic paragon in AD
intervention. The ongoing research circulates around development of disease-modifying
therapies targeting BACE-1 and inflammation for the treatment of AD. Quinoxaline and thiazole
scaffolds were conjugated owing to their BACE-1 inhibitory and anti-inflammatory responses to
develop novel MTDLs. In silico docking studies of library of quinoxaline-bisthiazoles suggested
their multitargeted profile. Compounds were found to interact with catalytic pocket of BACE-1,
COX-1 and COX-2 whereas, least binding score was obtained for BACE-2 and Cathepsin D
enzymes. Compounds were taken further for in vitro BACE-1 inhibition assay and in vivo acute
and chronic inflammation studies. Taken together, the outcomes from in silico, in vitro and in
vivo studies, compound 8n emerged as the most potent MTDL. In AlCl3 induced AD model,
compound 8n showed promising antiamnesic, antioxidant, antiamyloid and neuroprotective
attributes. In addition, despite having potent anti-inflammatory benefits, this molecule hardly
shows any GI toxicity unlike other NSAIDS. The functional consequences of targeting BACE-1
inhibitory and anti-inflammatory pathways turned out to be of pragmatic relevance to attempt

restoration of cognitive dysfunction with disease modifying properties related to AD pathology.
In future, compound 8n may emerge as promising MTDL in clinical studies.
4. Materials and Methods
4.1.
Molecular Docking Studies
Prior to the in silico evaluation, three-dimensional X-ray crystal structures of proteins required
for analysis were downloaded from the protein-data-bank. To evaluate the binding modes of
compounds, docking studies were carried out using Maestro 11.1.012 program of Schrödinger,
LLC, New York, NY, 2017, USA. Structures of various quinoxaline-bisthiazoles with different
R1, R2 and R3 substitutions were fed into the software (See Table 1 and Supplementary material).
A meticulous virtual screening of these fabricated molecules was performed to rationalize their
feasible inhibitory tendencies against BACE-1 and inflammatory pathway enzymes. Out of
these, the molecules with best docking score values amongst BACE-1, COX-1 and COX-2 were
selected for further assessment (See Table 1). Amongst these, the molecules displaying lesser
docking score for BACE-2 and Cathepsin D were chosen so as to procure selectivity of BACE-1
over BACE-2 and Cathepsin D.
Ligand preparation: All molecules were drawn and cleaned up in the Maestro 2D sketcher
software. The Ligprep module was employed for energy minimization by OPLS3 force field and
possible states were obtained using Epik (pH 7.0).
Protein preparation: To prepare BACE-1 protein, a crystal structure containing coherent
structure of BACE-1 bound ligand AZD3839 (clinical candidate) was retrieved from protein-
data-bank (PDB ID: 4B05).[40] The protein structure was constructed using protein preparation
wizard of Schrödinger 11.1.012 software. Bond orders were assigned and hydrogens were added
to the structures. Subsequently, zero-order bonds to metals and to disulfide bonds were added.
Missing side chains were affixed using prime and water molecules within 5 Å of het groups were
removed. All ions were deleted and protein structures were optimized using OPLS3 force field
and minimized for receptor grid generation which was obtained at the centroid of the ligand
(AZD3839) of BACE-1 receptor. Likewise, protein structure for BACE-2 was constructed using
the crystal structure of human BACE-2 receptor (PDBID: 2EWY) bound to hydroxyl-
ethylenamine transition-state inhibitor.[45] The crystal structure of Cathepsin D (PDB ID: 1LYW)
was used for docking process.[46] Binding sites were generated by using SiteMap module[47] in
Schrödinger. Amongst the top ranked potential binding sites identified by using SiteScore, the

topmost binding site was selected for receptor grid generation. Grid was generated at the binding
site pocket of the Cathepsin D enzyme and then used for the docking process.
Cyclooxygenases protein preparation: Crystal structure of COX-1 in complexation with
Celecoxib (PDB ID: 3KK6) and COX-2 with co-crystallized Celecoxib ligand (PDB ID: 3LN1)
were used for the docking process.[48,49] All the steps for the protein preparation were followed as
described in BACE-1 protein preparation protocol using chain A of the enzyme.
Docking study: Ligand docking was brought out using Glide module of Schrödinger suite and
extra precision (XP) mode was applied for the docking of all the molecules with proteins.
Docking score and interactions of ligands with protein were considered for activity prognosis of
the molecules. Re-docking of co-crystallized ligand with the proteins was brought to operation
for precise validation of docking process.[33]
4.2.
Physicochemical properties prediction
Physicochemical properties of all compounds were virtually perceived by using QikProp
application of Schrödinger software. Drug likeliness properties such as octanol/water partition
coefficient (QPlogPo/w), brain/blood partition coefficient (QPlogBB), % human oral absorption
(HOA), Caco-2 cell permeability (QPPCaco) and HERG inhibition (QPlogHERG) were taken
into evaluation to foretell brain permeability and toxicity of the molecules.[34]
4.3.
Chemistry
4.3.1 General methods
Unless stated otherwise, all starting material, reagents and solvents were purchased from
commercial suppliers Spectrochem, Sigma-Aldrich, Fisher Scientific, HiMedia and used without
any further purification. To check the progress of reaction, thin layer chromatography (TLC) was
employed using ready-made silica gel plates (pre-coated, Merck) and visualized under ultraviolet
lamp. Melting points (°C) were obtained using scientific melting point apparatus (Veego; Model
VMP-DS) and are uncorrected. Column chromatography was carried out using 100-200 mesh
silica gel. To check the purity of compounds HPLC was carried out using SHIMADZU-LC-
2010. Shimadzu’s IR Affinity 1 was used to record FTIR spectra. Mass spectra were obtained by
using Perkin Elmer Mass Spectrometer and values are quoted in m/z. HRMS analysis was done
1
13
by Q-TOF micromass (ESI-MS). H NMR and C NMR spectra were recorded using Bruker
Avance II 400 MHz NMR spectrophotometer. Chemical shift values are represented in parts per
million (ppm) in δ-scale. Coupling constant (J) values were observed in Hertz (Hz). Fine

splitting of the signals are described as s (singlet), d (doublet), t (triplet), q (quartet), m
(multiplet) and dd (doublet of doublets).
4.3.2 Synthesis
4.3.2.1. Synthesis and characterization of substituted 2,3-bis(bromomethyl)quinoxalines
(2a-c)
2,3-bis(bromomethyl)quinoxaline (2a)
To the solution of benzene-1,2-diamine (1.0 eq) 1a in ethanol, 1,4-dibromobutane-2,3-dione (1.0
eq) was added slowly. Reaction temperature was maintained at 0 °C. Reaction mixture was
allowed to stir for 2 hr. After completion of reaction, ethanol was evaporated in vacuo to get
dried compound 2a. Pale green solid; Yield: 75%; Melting point: 151-153 °C; LC-ESI-MS
(m/z): 313.9 [M]+, 315.9 [M+2]+, 317.9 [M+4]+
2,3-bis(bromomethyl)-6-chloroquinoxaline (2b)
Intermediate 2b was prepared by the procedure described in 2a using 4-chlorobenzene-1,2-
diamine 1b. Yield: 71%; Light pink solid; Melting point: 174-176 °C; LC-ESI-MS (m/z): 347.9
[M]+, 349.9 [M+2]+, 351.9 [M+4]+, 353.9 [M+6]+
2,3-bis(bromomethyl)-6,7-dimethylquinoxaline (2c)
Intermediate 2c was prepared by the procedure described in 2a using 4,5-dimethylbenzene-1,2-
diamine 1c. Yield: 74%; Light pink solid; Melting point: 181-183 °C; LC-ESI-MS (m/z): 341.9
[M]+, 343.9 [M+2]+, 345.9 [M+4]+
4.3.2.2. Synthesis and characterization of substituted N'-carbamothioylformimidamide (4a-
d)
N,N-diethyl-N'-(phenylcarbamothioyl)benzimidamide (4a)
To the solution of N,N-diethylbenzimidamide 3a (1.0 eq) in THF, addition of
isothiocyanatobenzene (1.0 eq) was done. Reaction was allowed to proceed at room temperature
for 3-4 hr. After completion of reaction, THF was evaporated in vacuo to get pale yellow solid of
4a. 4a was used without further purification. Yield: 81%; Pale yellow solid; Melting point: 195-
197 °C; LC-ESI-MS (m/z): 312.1 [M+H]+
N,N-diethyl-N'-(tritylcarbamothioyl)acetimidamide (4b)
Intermediate 4b was prepared by the procedure described in 4a using
(isothiocyanatomethanetriyl)tribenzene and N,N-diethylacetimidamide 3b. Yield: 70%; Colorless
solid; Melting point: 188-190 °C; LC-ESI-MS (m/z): 416.2 [M+H]+

1-(bis(dimethylamino)methylene)-3-tritylthiourea (4c)
Intermediate 4c was prepared by the procedure described in 4a using
(isothiocyanatomethanetriyl)tribenzene and 1,1,3,3-tetramethylguanidine 3c. Yield: 86%; Pale
yellow solid; Melting point: 182-184 °C; LC-ESI-MS (m/z): 417.2 [M+H]+
N,N-diethyl-N'-(tritylcarbamothioyl)benzimidamide (4d)
Intermediate 4d was prepared by the procedure described in 4a using
(isothiocyanatomethanetriyl)tribenzene and N,N-diethylbenzimidamide 3a. Yield: 82%; Melting
point: 195-197 °C; LC-ESI-MS (m/z): 478.2 [M+H]+
4.3.2.3 Synthesis and characterization of substituted N'-carbamothioyl-N,N-
dimethylformimidamide (6a-e)
N'-carbamothioyl-N,N-dimethylformimidamide (6a)
Thiourea 5a (1.0 eq) in methanol was reacted with 1,1-dimethoxy-N,N-dimethylmethanamine
(2.0 eq) at room temperature for 3 hr to get N'-carbamothioyl-N,N-dimethylformimidamide 6a.
After completion of reaction, methanol was evaporated in vacuo to get compound 6a. 6a was
used without further purification. Yield: 85%; Pale yellow solid; Melting point: 202-204 °C; LC-
ESI-MS (m/z): 132.1 [M+H]+
N,N-dimethyl-N'-(phenylcarbamothioyl)formimidamide (6b)
Intermediate 6b was prepared by the procedure described in 6a using 1-phenylthiourea 5b.
Yield: 79%; Melting point: 195-197 °C; LC-ESI-MS (m/z): 208.2 [M+H]+
N'-((4-fluorophenyl)carbamothioyl)-N,N-dimethylformimidamide (6c)
Intermediate 6c was prepared by the procedure described in 6a using 1-(4-fluorophenyl)thiourea
5c. Yield: 71%; Melting point: 208-210 °C; LC-ESI-MS (m/z): 226.2 [M+H]+
N'-((4-chlorophenyl)carbamothioyl)-N,N-dimethylformimidamide (6d)
Intermediate 6d was prepared by the procedure described in 6a using 1-(4-chlorophenyl)thiourea
5d. Yield: 73%; Melting point: 211-213 °C; LC-ESI-MS (m/z): 241.5 [M]+, 243.5 [M+2]+
N,N-dimethyl-N'-(p-tolylcarbamothioyl)formimidamide (6e)
Intermediate 6e was prepared by the procedure described in 6a using 1-(p-tolyl)thiourea 5e.
Yield: 78%; Melting point: 204-206 °C; LC-ESI-MS (m/z): 221.1 [M+H]+
4.3.2.4. Synthesis and characterization of final compounds: Substituted 5,5'-(quinoxaline-
2,3-diyl)bis(thiazol-2-amine) (8a-f)
5,5'-(quinoxaline-2,3-diyl)bis(4-methylthiazol-2-amine) (8a)

To a solution of 2,3-bis(bromomethyl)quinoxaline (1.0 eq) 2a in acetonitrile, K2CO3 (3.0 eq) was
added. Reaction mixture was allowed to stir and addition of N,N-diethyl-N'-
(tritylcarbamothioyl)acetimidamide 4b (2.0 eq) was done. Reaction was allowed to proceed at
room temperature for 4 hr. After completion of reaction, reaction mixture was poured into ice
cold water to get the precipitates of tritylated 7a. Deprotection of tritylated 7a was done using
trifluoro acetic acid (50.0 eq) in DCM for 3 hr. After completion of reaction, reaction mixture
was poured into ice cold water. Solution was neutralized with sodium bicarbonate to get
deprotected final compound 8a. 8a was purified by column chromatography (Chloroform:
Methanol = 9.5:0.5). Yellow solid; Yield: 76%; Purity: 99.8%; Melting point: 150-152 °C;
Molecular formula: C16H14N6S2; LC-ESI-MS (m/z): 355.1 [M+H]+; HRMS (TOF) m/z calcd for
C16H14N6S2 [M+H]+ 355.0721, found: 355.0824; IR (KBr, cm-1): 3469.994, 3329.14, 3062.96,
1
1681.93, 1490.97, 1157.24; H NMR (400MHz, DMSO-d6) δ: 2.452 (s, 6H), 7.601 (s, D2O
13
exchangeable, 4H), 7.684 (t, J = 3.2 Hz, 2H), 7.865 (d, J = 6.4 Hz, 2H); C NMR (400MHz,
DMSO-d6) δ: 15.07 (2C), 127.30 (2C), 127.43 (2C), 127.67 (2C), 141.72 (2C), 144.79 (2C),
147.70 (2C), 171.28 (2C)
5,5'-(quinoxaline-2,3-diyl)bis(4-phenylthiazol-2-amine) (8b)
Final compound 8b was synthesized as per the procedure described in 8a using starting materials
2,3-bis(bromomethyl)quinoxaline 2a and N,N-diethyl-N'-(tritylcarbamothioyl)benzimidamide
4d. 8b was purified by column chromatography (Ethyl acetate: Hexane = 7:3). Yellow solid;
Yield: 58%; Purity: 99.1%; Melting point: 190-192 ºC; Molecular formula: C26H18N6S2; LC-ESI-
1
MS (m/z): 479.1 [M+H]+; IR (KBr, cm-1): 3475.73, 3359.18, 3061.03, 3022.45, 1489.05; H
NMR (400MHz, DMSO-d6) δ: 7.151 (s, D2O exchangeable, 4H), 7.477 (t, J = 7.6 Hz, 2H), 7.560
(t, J = 6.8 Hz, 4H), 7.633-7.698 (m, 8H); 13C NMR (400MHz, DMSO-d6) δ: 117.76 (2C), 120.51
(2C), 123.11 (2C), 125.92 (2C), 128.56 (2C), 129.12 (2C), 129.50 (2C), 131.39 (2C), 137.61
(2C), 137.95 (2C), 141.87 (2C), 155.11 (2C), 170.42 (2C)
5,5'-(6,7-dimethylquinoxaline-2,3-diyl)bis(N4,N4-dimethylthiazole-2,4-diamine) (8c)
Final compound 8c was synthesized as per the procedure described in 8a using starting materials
2,3-bis(bromomethyl)-6,7-dimethylquinoxaline 2c and 1-(bis(dimethylamino)methylene)-3-
tritylthiourea 4c. 8c was purified by column chromatography (Chloroform: Methanol = 9.5:0.5).
Yellow solid; Yield: 74%; Purity: 99.5%; Melting point: 190-192 ºC; Molecular formula:
C20H24N8S2; LC-ESI-MS (m/z): 441.1 [M+H]+; IR (KBr, cm-1): 3474.91, 3256.16, 2925.17,

1585.55, 760.95; 1H NMR (400MHz, DMSO-d6) δ: 2.367 (s, 6H), 3.365 (s, 12H), 7.254 (s, D2O
13
exchangeable, 4H), 7.539 (s, 2H); C NMR (400MHz, DMSO-d6) δ: 19.64 (2C), 41.30 (4C),
127.49 (2C), 137.19 (2C), 140.42 (2C), 147.74 (2C), 157.92 (2C), 168.43 (2C), 179.38 (2C)
5,5'-(quinoxaline-2,3-diyl)bis(N4,N4-dimethylthiazole-2,4-diamine) (8d)
Final compound 8d was synthesized as per the procedure described in 8a using starting materials
2,3-bis(bromomethyl)quinoxaline 2a and 1-(bis(dimethylamino)methylene)-3-tritylthiourea 4c.
8d was purified by column chromatography (Chloroform: Methanol = 9.5:0.5). Yellow solid;
Yield: 67%; Purity: 99.6%; Melting point: 210-212 ºC; Molecular formula: C18H20N8S2; LC-ESI-
MS (m/z): 413.2 [M+H]+; IR (KBr, cm-1): 3477.66, 3251.46, 3062.96, 1597.06, 1332.81, 759.95;
1H NMR (400MHz, DMSO-d6) δ: 3.365 (s, 12H), 6.966 (s, D2O exchangeable, 4H), 7.221 (t, J=
13
6.8 Hz, 2H), 7.308 (d, J = 6.8 Hz, 2H); C NMR (400MHz, DMSO-d6) δ: 36.16 (4C), 125.55
(2C), 128.47 (2C), 129.05 (2C), 139.10 (2C), 142.44 (2C), 144.93 (2C), 170.24 (2C)
5,5'-(6-chloroquinoxaline-2,3-diyl)bis(N4,N4-dimethylthiazole-2,4-diamine) (8e)
Final compound 8e was synthesized as per the procedure described in 8a using starting materials
2,3-bis(bromomethyl)-6-chloroquinoxaline
2b
and
1-(bis(dimethylamino)methylene)-3-
tritylthiourea 4c. 8e was purified by column chromatography (Chloroform: Methanol = 9.5:0.5).
Yellow solid; Yield: 72%; Purity: 99.1%; Melting point: >250 °C; Molecular formula:
C18H19ClN8S2; LC-ESI-MS (m/z): 446.5 [M]+, 448.5 [M+2]+; IR (KBr, cm-1): 3455.62, 3294.56,
1
2927.10, 1562.41, 767.70; H NMR (400MHz, DMSO-d6) δ: 3.451 (s, 12H), 7.234 (s, D2O
exchangeable, 4H), 7.402 (d, J = 9.0 Hz, 1H), 7.621 (d, J = 9.0 Hz, 1H), 7.789 (s, 1H); 13C NMR
(400MHz, DMSO-d6) δ: 36.41 (4C), 117.81 (2C), 127.26, 128.13 (2C), 128.72 (2C), 129.25
(2C), 130.34, 139.12, 140.50, 141.34, 169.72
5,5'-(6-chloroquinoxaline-2,3-diyl)bis(4-methylthiazol-2-amine) (8f)
Final compound 8f was synthesized as per the procedure described in 8a using starting materials
2,3-bis(bromomethyl)-6-chloroquinoxaline
2b
and
N,N-diethyl-N'-
(tritylcarbamothioyl)acetimidamide 4b. 8f was purified by column chromatography
(Chloroform: Methanol = 9.5:0.5). Yellow solid; Yield: 76%; Purity: 99.7%; Melting point: 223-
225 ºC; Molecular formula: C16H13ClN6S2; LC-ESI-MS (m/z): 388.5 [M]+, 390.5 [M+2]+; IR
1
(KBr, cm-1): 3468.61, 3305.99, 3062.96, 1598.99, 700.16; H NMR (400MHz, DMSO-d6) δ:
2.457 (s, 6H), 7.240 (s, D2O exchangeable, 4H), 7.402 (d, J = 9.0 Hz, 1H), 7.620 (d, J = 9.0 Hz,

1H), 7.789 (s, 1H); 13C NMR (400MHz, DMSO-d6) δ: 14.21 (2C), 117.18 (2C), 126.31, 127.14,
127.70, 140.12, 141.43, 142.55, 145.05, 146.32, 146.78, 147.56 (2C), 171.72
4.3.2.5. Synthesis and characterization of final compounds: Substituted 5,5'-(quinoxaline-
2,3-diyl)bis(thiazol-2-amine) (8g-n)
5,5'-(quinoxaline-2,3-diyl)bis(N,4-diphenylthiazol-2-amine) (8g)
To a solution of 2,3-bis(bromomethyl)quinoxaline (1.0 eq) 2a in THF, addition of N,N-diethyl-
N'-(phenylcarbamothioyl)benzimidamide 4a (2.0 eq) was done. Drop wise addition of
trimethylamine (1.1 eq) was carried out. Reaction was allowed to proceed at room temperature
for 3 hr. After completion of reaction, reaction mixture was poured into ice cold water to get the
precipitates of compound 8g. 8g was purified by column chromatography (Ethyl acetate: Hexane
= 1:4). Yellow solid; Yield: 89%; Purity: 99.4%; Melting point: >250 ºC; Molecular formula:
C38H26N6S2; LC-ESI-MS (m/z): 631.1 [M+H]+; IR (KBr, cm-1): 3250.5, 3067.3, 1539.26,
1137.09; 1H NMR (400MHz, DMSO-d6) δ: 6.979-7.054 (m, 6H), 7.152-7.227 (m, 6H), 7.351 (t,
J = 7.6 Hz, 4H), 7.548 (d, J = 8.0 Hz, 4H), 7.816-7.840 (m, 2H), 7.981-8.001 (m, 2H), 10.255 (s,
13
D2O exchangeable, 2H); C NMR (400MHz, DMSO-d6) δ: 117.23 (2C), 118.54 (2C), 121.45
(2C), 127.17 (2C), 127.84 (2C), 127.97 (2C), 128.10 (2C), 128.25 (2C), 128.81 (2C), 129.01
(2C), 130.06 (2C), 135.13 (2C), 139.91 (2C), 139.99 (2C), 140.81 (2C), 147.11 (2C), 148.73
(2C), 162.08 (2C), 164.12 (2C)
5,5'-(quinoxaline-2,3-diyl)bis(thiazol-2-amine) (8h)
Final compound 8h was synthesized as per the procedure described in 8g using starting materials
2,3-bis(bromomethyl)quinoxaline 2a and N'-carbamothioyl-N,N-dimethylformimidamide 6a. 8h
was purified by column chromatography (Chloroform: Methanol = 9.5:0.5). Yellow solid;
Yield: 81%; Purity: 98.7%; Melting point: 160-162 ºC; Molecular formula: C14H10N6S2; LC-ESI-
1
MS (m/z): 327.2 [M+H]+; IR (KBr, cm-1): 3452.42, 3356.9, 3014.8, 1469.76, 700.12; H NMR
(400MHz, DMSO-d6) δ: 7.510 (s, 2H), 7.588 (s, D2O exchangeable, 4H), 7.684 (t, J = 6.8, 2H),
13
7.848 (d, J = 6.8 Hz, 2H); C NMR (400MHz, DMSO-d6) δ: 123.26 (2C), 127.73 (2C), 129.45
(2C), 139.17 (2C), 141.87 (2C), 144.86 (2C), 171.18 (2C)
5,5'-(quinoxaline-2,3-diyl)bis(N-phenylthiazol-2-amine) (8i)
Final compound 8i was synthesized as per the procedure described in 8g using starting materials
2,3-bis(bromomethyl)quinoxaline
2a
and
N,N-dimethyl-N'-
(phenylcarbamothioyl)formimidamide 6b. 8i was purified by column chromatography (Ethyl

acetate: Hexane = 2:3). Yellow solid; Yield: 68%; Purity: 99.7%; Melting point: 155-157 ºC;
Molecular formula: C26H18N6S2; LC-ESI-MS (m/z): 479.1 [M+H]+; HRMS (TOF) m/z calcd for
C26H18N6S2 [M+H]+ 479.1034, found: 479.1081; IR (KBr, cm-1): 3291.51, 3018.9, 1442.66,
1
756.23; H NMR (400MHz, DMSO-d6) δ: 6.9940 (t, J = 7.36 Hz, 2H), 7.2127 (t, J = 7.4 Hz,
4H), 7.6726-7.7387 (m, 6H), 7.7560 (s, 2H), 7.9375 (d, J = 6.4 Hz, 2H), 10.5465 (s, D2O
13
exchangeable, 2H); C NMR (400MHz, DMSO-d6) δ: 117.54 (4C), 121.80 (2C), 124.55 (2C),
127.92 (2C), 128.69 (2C), 128.79 (2C), 129.65 (2C), 139.52 (2C), 140.48 (2C), 141.21 (2C),
144.48 (2C), 165.89 (2C)
5,5'-(quinoxaline-2,3-diyl)bis(N-(4-fluorophenyl)thiazol-2-amine) (8j)
Final compound 8j was synthesized as per the procedure described in 8g using starting materials
2,3-bis(bromomethyl)quinoxaline
2a
and
N'-((4-fluorophenyl)carbamothioyl)-N,N-
dimethylformimidamide 6c. 8j was purified by column chromatography (Ethyl acetate: Hexane =
3:7). Yellow solid; Yield: 72%; Purity: 99.8%; Melting point: >250 ºC; Molecular formula:
C26H16F2N6S2; LC-ESI-MS (m/z): 515.1 [M+H]+; IR (KBr, cm-1): 3207.76, 3045.69, 1507.43,
1
828.48; H NMR (400MHz, DMSO-d6) δ: 7.752 (s, 2H), 7.761-7.777 (m, 4H), 7.786-7.872 (m,
13
4H), 7.917 (t, J = 8 Hz, 2H), 8.042 (d, J = 6.4 Hz, 2H), 10.684 (s, D2O exchangeable, 2H); C
NMR (400MHz, DMSO-d6) δ: 115.09, 115.16, 115.31, 119.09, 119.17, 119.39, 119.46, 124.64,
125.69, 127.88, 129.46, 129.63, 130.68, 136.90, 136.92 (2C), 137.35, 139.47, 140.42, 140.97,
143.45, 144.39, 144.44, 156.04, 158.43, 165.88, 166.56
5,5'-(quinoxaline-2,3-diyl)bis(N-(4-chlorophenyl)thiazol-2-amine) (8k)
Final compound 8k was synthesized as per the procedure described in 8g using starting materials
2,3-bis(bromomethyl)quinoxaline
2a
and
N'-((4-chlorophenyl)carbamothioyl)-N,N-
dimethylformimidamide 6d. 8k was purified by column chromatography (Ethyl acetate: Hexane
= 3:7). Yellow solid; Yield: 81%; Purity: 99.9%; Melting point: >250 ºC; Molecular formula:
C26H16Cl2N6S2; LC-ESI-MS (m/z): 546.1 [M]+, 548.1 [M+2]+, 550.1 [M+4]+; HRMS (TOF) m/z
calcd for C26H16Cl2N6S2 [M]+ 546.0255, found: 546.0235 [M]+, 548.0188 [M+2]+, 550.0231
1
[M+4]+; IR (KBr, cm-1): 3375.57, 2914.57, 1532.51, 763.84; H NMR (400MHz, DMSO-d6) δ:
7.326 (s, 2H), 7.342 (d, J = 6.9 Hz, 4H), 7.701-7.764 (m, 6H), 7.946 (d, J = 6.4 Hz, 2H), 10.705
(s, D2O exchangeable, 2H); 13C NMR (400MHz, DMSO-d6) δ: 118.90 (4C), 124.97 (2C), 125.34
(4C), 128.01 (2C), 128.68 (2C), 129.97 (2C), 139.36 (2C), 139.54 (2C), 141.16 (2C), 144.44
(2C), 165.38 (2C)

5,5'-(6-chloroquinoxaline-2,3-diyl)bis(thiazol-2-amine) (8l)
Final compound 8l was synthesized as per the procedure described in 8g using starting materials
2,3-bis(bromomethyl)-6-chloroquinoxaline
2b
and
N'-carbamothioyl-N,N-
dimethylformimidamide 6a. 8l was purified by column chromatography (Ethyl acetate: Hexane =
1:1). Yellow solid; Yield: 86%; Purity: 99.0%; Melting point: 158-160 ºC; Molecular formula:
C14H9ClN6S2; LC-ESI-MS (m/z): 360.1 [M]+, 362.1 [M+2]+; HRMS (TOF) m/z calcd for
C14H9ClN6S2 [M]+ 360.0019, found: 360.0045 [M]+, 362.0024 [M+2]+; IR (KBr, cm-1): 3310.95,
1
3224.16, 1622.20, 1486.22, 826.53; H NMR (400MHz, DMSO-d6) δ: 7.6103 (s, 2H), 7.796 (s,
D2O exchangeable, 4H), 7.822 (d, J = 2.4 Hz, 1H), 7.8634 (s, 1H), 7.882 (d, J = 2.4 Hz, 1H); 13C
NMR (400MHz, DMSO-d6) δ: 123.18, 123.29, 126.07, 128.68, 129.31, 133.34, 136.80, 142.09,
144.96, 148.04, 150.59, 171.41, 171.63
5,5'-(quinoxaline-2,3-diyl)bis(N-(p-tolyl)thiazol-2-amine) (8m)
Final compound 8m was synthesized as per the procedure described in 8g using starting
materials
2,3-bis(bromomethyl)quinoxaline
2a
and
N,N-dimethyl-N'-(p-
tolylcarbamothioyl)formimidamide 6e. 8m was purified by column chromatography (Ethyl
acetate: Hexane = 2:3). Yellow solid; Yield: 72%; Purity: 99.8%; Melting point: >250 ºC;
Molecular formula: C28H22N6S2; LC-ESI-MS (m/z): 507.1 [M+H]+; HRMS (TOF) m/z calcd for
C28H22N6S2 [M+H]+ 507.1347, found: 507.1272; IR (KBr, cm-1): 3388.11, 2923.25, 1533.47,
1
1217.14, 815.92; H NMR (400MHz, DMSO-d6) δ: 2.277 (s, 6H), 7.148 (d, J = 8.4 Hz, 4H),
7.349 (s, 2H), 7.551 (d, J = 8.4 Hz, 4H), 7.748 (t, J = 6.4 Hz, 2H), 7.940 (d, J = 7.32 Hz, 2H),
13
10.53 (s, 2H); C NMR (400MHz, DMSO-d6) δ: 20.36 (2C), 117.75 (4C), 124.18 (2C), 127.96
(2C), 129.43 (2C), 129.98 (4C), 131.03 (2C), 138.07 (2C), 139.42 (2C), 141.47 (2C), 144.62
(2C), 166.06 (2C)
5,5'-(6,7-dimethylquinoxaline-2,3-diyl)bis(thiazol-2-amine) (8n)
Final compound 8n was synthesized as per the procedure described in 8g using starting materials
2,3-bis(bromomethyl)-6,7-dimethylquinoxaline
2c
and
N'-carbamothioyl-N,N-
dimethylformimidamide 6a. 8n was purified by column chromatography (Ethyl acetate: Hexane
= 1:1). Yellow solid; Yield: 71%; Purity: 99.6%; Melting point: 110-112 ºC; Molecular formula:
C16H14N6S2; LC-ESI-MS (m/z): 355.1 [M+H]+; HRMS (TOF) m/z calcd for C16H14N6S2 [M+H]+
1
355.0721, found: 355.0789; IR (KBr, cm-1): 3298.42, 3150.23, 1493.93, 12119.06, 871.88; H
NMR (400MHz, DMSO-d6) δ: 2.436 (s, 6H), 7.442 (s, 2H), 7.639 (s, D2O exchangeable, 4H),

13
7.675 (s, 2H); C NMR (400MHz, DMSO-d6) δ: 19.71 (2C), 123.47 (2C), 127.08 (2C), 141.14
(2C), 143.87 (2C), 144.81 (2C), 148.05 (2C), 171.25 (2C)
4.4. Biological activity
4.4.1 In Vitro BACE-1 enzyme inhibition assay
BACE-1 inhibitory trends of test compounds 8a-8n were checked through in vitro enzyme
inhibition assays. For this purpose, Corning 96 Well Black Polystyrene well-plates were used
and a total volume of 100 µL, per well was added to facilitate enzyme-substrate reaction. The
other enzymatic reaction contents included 4.0 ng/µL of human recombinant BACE-1 enzyme
(procured form BioVision Inc., India), 50 µM substrate (MCA-SEVNLDAEFR-Ednp-KRR-
.
NH2 3TFA), various concentrations of the test compound (100 µM, 10 µM and 1 µM, dissolved
in DMSO) and 0.1 M sodium-acetate buffer (pH 4.5). After addition of all these reagents,
fluorescence readings were recorded by Varioskan Flash multimode plate reader under excitation
and emission wavelength of 320 nm and 405 nm, respectively and were termed as zero hour
reading. The plate was then incubated for 2 hr at 37 °C and fluorescence was recorded again. %
inhibition of BACE-1 enzyme was determined through calculations and IC50 values [denoted as
µM ± SEM (n=3)] of all test compounds were obtained. These data are denoted as mean of 3
determinations ± SEM and compared with standard compound’s (AZD3839) reported Ki value
of 26.1 nmol/L[40], as enumerated in Table 3.
4.4.2. Animals and treatments
Animal care and all in vivo experiments were conducted after sanctioned from Institutional
Animal Ethics Committee (Approval No: PERD/IAEC/2016/049 & PERD/IAEC/2016/050). The
experiments were carried out in harmony with internationally approved good laboratory practices
(GLP) for housing and for the handling of animals CPCSEA guidelines (Committee for the
Purpose of Control and Supervision of Experiments on Animals, Govt. of India) were strictly
followed. Sprague-Dawley rats (female, 4-6 months old, 150-200 g body weight) were housed in
well ventilated cages at standard conditions of relative humidity (60 ± 5 %), temperature (25 ± 3
°C), 12 : 12 hr light/dark cycle and 10 % air exhaust conditioning unit were maintained. Animals
were housed in group of 3 rats per cage in animal house (at B. V. Patel PERD Centre,
Ahmedabad) with unrestricted access to food and water. All behavioural experiments were
conducted in a semi-soundproof and semi-dark laboratory between 10 am and 6 pm being the

active phase of rodents. Animals were deprived of food overnight prior to the surgical
procedures and sacrificed.
4.4.3. In vivo analysis
4.4.3.1. Acute anti-inflammatory activity: Carrageenan induced rat paw edema study
Carrageenan induced rat paw edema model is widely used model for determination of anti-
inflammatory activity.[50] A series of 14 test compounds were screened to determine their acute
inflammatory potentials using a carrageenan induced rat paw edema model. 17 groups (n=6)
denoted as normal control (NC), disease control (DC), positive control (PC) and 14 test groups
(8a-8n) were randomly grouped and used for the study. For pre-dosing, all test compounds (50
mg·kg-1) and PC (7.5 mg·kg-1of diclofenac, Sigma Aldrich) were administered orally as 0.2%
agar suspension. The NC and DC groups were fed with vehicle (0.2 % agar suspension) only.
Edema was induced by subcutaneously injecting 0.1 mL of 1% carrageenan (Spectrochem;
solution in normal saline) into left hind paw of rats, after 1 hr of oral dosing. The alteration in
their paw volume before and after 3 hr of carrageenan injection was measured using a mercury
plethesmometer.[50] The % edema inhibition of paw was calculated by using the following
equation:
% edema inhibition = [{(Vt/V0) DC – (Vt/V0) treated} / (Vt/V0) DC] × 100
where, Vt = rat paw volume at 3rd hr
V0 = rat paw volume at 0th hr
% edema inhibition data are represented as mean ± SEM (n=6 animals/group) and statistical
analysis was performed utilizing GraphPad Prism software (GraphPad Software, San Diego, CA,
USA) by one-way analysis of variance (ANOVA) was used to detect significant effect of
treatment, followed by Tukey’s multiple comparison test (*p< 0.05).
4.4.3.2. Chronic anti-inflammatory activity: Formalin induced rat paw edema study
Out of 14 compounds, 4 compounds (8c, 8g, 8l and 8n) were taken forward to check their
potential activity on chronic inflammation model.[50] Animals were divided into random groups
of seven (n=6) designated as NC, DC, PC and 4 test groups (8c, 8g, 8l and 8n). In this model,
rats were orally pre-dosed with celecoxib (≥ 98%, Sigma Aldrich, 40 mg·kg-1) for PC group and
with test compounds (50 mg·kg-1) as 0.2% agar (HiMedia) suspension. NC and DC groups
received vehicle only (0.2% agar suspension). Posterior to 1 hr of oral dosing, animals were
subcutaneously injected 0.1 mL of 2% formalin (SD Fine Chemicals; in normal saline) in the sub

plantar region of their left hind paw. Oral dosing and subcutaneous injections were continued for
5 consecutive days. Paw volume measurements were recorded before and after 5 hr of formalin
administration using mercury plethesmometer.[50] % inhibition of edema in rat paw was
calculated using following equation:
% edema inhibition = [{(Vt/V0) DC – (Vt/V0) treated} / (Vt/V0) DC] × 100
where, Vt = rat paw volume at 5th hr
V0 = rat paw volume at 0th hr
The data are represented as mean % edema inhibition ± SEM (n=6 animals/group) and statistical
analysis was performed by one-way ANOVA followed by Tukey’s multiple comparison test
(*p< 0.05).
4.4.3.3. Aluminium chloride induced Alzheimer’s disease model
The highest active test compound 8n apparent from BACE-1 inhibition assay, acute and chronic
inflammation models was further assessed for its anti-Alzheimer’s activity. Compound 8n was
screened through behavioural parameters, antioxidant property and histopathological studies to
shed light onto its anti-Alzheimer’s propensity. Animals were divided into 4 groups (n=6)
designated as NC, DC, PC and test groups (8n). The PC group was pre-dosed orally with
meloxicam (as meloxicam sodium salt hydrate, Sigma Aldrich, 5 mg·kg-1) whereas the test group
with test compound 8n (50 mg·kg-1) prepared in 0.2 % agar suspension for 28 days. NC and DC
groups received vehicle only. After oral dosing of compounds, animals were injected with
aluminium chloride (Sigma Aldrich; 4.2 mg·kg-1 i.p.) for 28 days. This model has been in use for
several years.[41] The described procedures were repeated for 28 days and during this period
following studies were carried out.
4.4.3.3.1. Estimation of behavioural parameters
Behavioural parameters were scrutinized using Y-maze test, pole climbing apparatus and
elevated plus maze test. Animals were trained for 5 days ahead of the intra-peritonial AlCl3
administration. Thereafter behavioural parameters were recorded during training and retention
sessions on the 7th, 14th, 21st and 28th day of AlCl3 induced rat model.
4.4.3.3.1.1. Y-maze test
Y-maze test works on the principle of immediate working memory of rodents. It assesses the
rodent’s tendency to explore novel environment as they preferably choose to visit the unexplored
region over previously visited region and this peculiar behavior is scientifically measured in

terms of ‘simultaneous alteration performance’ (SAP). SAP was observed using Y maze
apparatus.[37] Y-maze was made up of 3 identical wooden arms (40 × 9 × 16 cm) and placed on
an evenly planar area at an angle of 120° with respect to each other. In our experiment, each arm
was labeled as ‘A’, ‘B’, and ‘C’, respectively. Experimental rats were placed at the end of one
arm and allowed to explore the maze environment for 5 min. Arm entries were recorded visually
like A, B and C. SAP was calculated as total number of entries minus two, and the % alteration is
calculated as (actual correct alteration ÷ maximum alteration) × 100. Data are given as mean %
alteration ± SEM and statistical analysis was performed by two-way ANOVA test followed by
multiple comparisons of Tukey’s test. Statistical significance was measured as *p<0.05
compared to NC and as #p<0.05 compared to DC.
4.4.3.3.1.2. Conditioned avoidance response test
In rats, the conditioned avoidance response (CAR) was assessed by using Cook’s pole climbing
apparatus made up of a sound proof wooden chamber having dimensions 25 × 25 × 25 cm.[51]
Electric shock was delivered by a metallic bar located on the chamber’s floor. A shock-proof
pole of diameter 2.5 cm was equipped at the core of chamber. Study was initiated by placing a
rat inside the apparatus and 45 sec time was stipulated to each rat in-order to explore the
chamber’s internal environment. After that, conditioned stimulus was delivered using buzzer
signal for 10 sec and unconditioned stimulus was delivered by giving electric shock for 45 sec.
The rats were trained for 3 consecutive days with 1st day 5 trials, 2nd day 3 trials and 3rd day 1
trial were conducted. During training session rats learned to escape the floor shock by climbing
up the pole after the buzzer signal and the time duration taken by them is termed as conditioned
avoidance response (CAR). CAR was recorded for each rat during retention session, which had 1
trial of experiment on 7th, 14th, 21st and 28th day in AlCl3 model. Data were represented as (mean
± SEM) and statistical analysis was performed by two-way ANOVA test followed by multiple
comparisons of Tukey’s test. Statistical significance was measured as *p<0.05 compared to NC
and #p<0.05 compared to DC.
4.4.3.3.1.3. Elevated plus maze test
Elevated plus maze (EPM) test works on the principle of aversion of rodents towards heights and
open space environment. Animals naturally tend to escape from open spacious region and high
altitudes and prefer to adobe in closed or rather safe environment.[38] In our experiment, EPM
test was availed to check the spatial working memory of rats and time taken to find the closed

arm location was measured. The EPM assembly was framed up by wooden arms; two open arms
(29 × 5 cm) and two enclosed arms (29 × 5 × 15 cm) were connected by a central platform (5 × 5
cm) elevated at a height of 40 cm with respect to the floor. During the training session which
lasted for 5 days each rat was kept at one open arm facing away from central platform and
allowed to explore the maze environment for 5 minutes. Open arm entries, enclosed arm entries,
total entries, % time spent in open arm and % time spent in enclosed arm were noted down.
Transfer latencies were recorded during retention session on 7th, 14th, 21st and 28th day of the
AlCl3 model. Data were calculated by two-way ANOVA test followed by multiple comparisons
of Tukey’s test (mean ± SEM). Statistical significance was measured as *p<0.05 when compared
to NC and (#p<0.05) compared to DC.
4.4.3.3.2. Haematological parameters estimation
Haematological parameters were measured after 28 days of treatment. On 29th day, rats were
anaesthetized by isoflurane. Blood samples were collected in 1 mL heparinized micro centrifuge
tubes from the retro orbital sinus of rats. Automated haematology analyser (VetScan HM-5;
Abaxis Inc., Union City, CA, USA) was used for estimation of haemoglobin and haematocrit
level.[39]
4.4.3.3.3. Estimation of serum biochemical parameters
Blood samples were collected from retro orbital sinus in 1 mL non-heparinized micro centrifuge
tubes. Serum was separated from the blood samples and biochemical parameters like % albumin
and total protein was estimated by an automated analyser (Transasia EM 360).[39]
4.4.3.3.4. Antioxidant activity
After 28 days of treatment, on the 29th day animals were euthanized, their intact brains were
removed by cervical dislocation and washed with saline to remove traces of blood. The brain
samples were divided into two hemispheres and used for lipid peroxidase and superoxide
dismutase assays.
4.4.3.3.4.1. Lipid peroxidase (LPO) assay
Brain tissues from all rats were taken into Hank’s balanced salt solution (HBSS) of pH 7.4 and
homogenized at 5000 rpm for 3 cycles (30 sec each) using Polytron homogenizer (Kinematica,
Switzerland). The homogenized samples were centrifuged at 35,000 rpm for 10 min using
Sorvall, legend X1R centrifuge. From each centrifuged sample, its supernatant was discarded
and pellet was re-suspended in 0.1 mL of HBSS which was further used for LPO assay. The LPO

activity was measured by reaction of malonaldehyde (MDA): thiobarbituric acid (TBA). The
assay reaction contained 0.1 mL of tissue homogenate (as processed earlier), 2 mL of 8.1%
sodium dodecyl sulfate, 1.5 mL of 20% acetic acid (pH was adjusted to 3.5 using 1 M NaOH),
and 1.5 mL of 0.8% aqueous TBA solution. Further, 0.7 mL of double distilled water was added
to the reaction mixture and volume was adjusted up to 4 mL. The reaction mixtures were heated
up to 95 °C for 1hr in water bath. 5 mL of n-butanol: pyridine (15:1 v/v) mixture and 1 mL of
double distilled water was added to it. The contents were first vortexed and then centrifuged at
3000 rpm for 7 min. The supernatant obtained after centrifugation was collected and processed
further for MDA detection. Formation of MDA in the supernatant was measured at a wavelength
of 532 nm in UV/visible spectrophotometer (Shimadzu UV-1800). Extinction coefficient of
MDA (1.45 × 10-5/min/cm) was used for calculation.[39,41]
LPO was calculated using following formula:
LPO = [(Sample - Blank) ×145] ×10-5/weight of organ (gram)
4.4.3.3.4.2. Superoxide dismutase (SOD) assay
The brain tissues were further processed for SOD activity assay. Tissues were taken in 2 mL of
chilled Tris buffer (50 mM, pH 8.2 was adjusted using 2 mM EDTA) homogenized for 3 cycles
(30 sec). The homogenates were treated with 0.1% of Triton X 100 (1 mL) at 4 °C for 30 min.
Each sample was again subjected to homogenization at 5600 rpm for 30 min at 4 °C. The
supernatant obtained was separated and used for SOD activity. Further, the supernatant was
divided into two parts one stored at 4-8 °C and the other heated in water bath at 95 °C for 1 hr.
Absorbance readings were recorded after 0th and 9th min of pyrogallol (HiMedia) addition to
each sample.[41,42] The calculations were done using per gram fresh tissue weight. SOD enzyme
activity was evaluated by the following equation:
Enzyme activity (units mg-1of fresh tissue weight) = [(ΔB – ΔE)/ΔP] ×120
Where, ΔB = boiled sample absorbance per min (9th min. reading – 0th min. reading)
ΔE = cold sample absorbance per min (9th min. reading – 0th min. reading)
ΔP = pyrogallol control per min (9th min. reading – 0th min. reading)
4.4.3.4. Histopathological studies of rat brain
For histopathological studies, brain samples were collected and stored in formalin (in 10%
neutral buffer) solution for 24 hr. Brain sections were stained using Congo-red and
Haematoxylin-Eosin (HE) stains and transferred to transparent glass slides. These were observed