Design, synthesis, and biological evaluation of some novel indolizine derivatives as dual cyclooxygenase and lipoxygenase inhibitor for anti-inflammatory activity

Article abstract of DOI:10.1016/j.bmc.2017.06.027

Some novel indolizine derivatives were synthesized by bioisosteric modification of imidazo[1,2-a]pyridine for anti-inflammatory activity. The physicochemical characterization and structure of compounds were elucidated by state of the art spectroscopic technique. Induced fit docking was performed for initial screening to elucidate the interactions with corresponding amino acids of cyclooxygenase (COX-1, COX-2) and lipoxygenase (LOX) enzymes. The target compounds 53–60 were then evaluated against in vivo carrageenan and arachidonic acid induced rat paw edema models for anti-inflammatory activity. Amongst all the synthesized derivatives, compound 56 showed the significant anti-inflammatory activity in both rat paw edema models with very less ulcerogenic liability in comparison to standard diclofenac, celecoxib, and zileuton. The compounds 56 was further assessed to observe in vitro enzyme inhibition assay on both cyclooxygenase and lipoxygenase enzyme where it showed a preferential and selective non-competitive enzyme inhibition towards the COX-2 (IC₅₀?=?14.91?μM, Ki?=?0.72?μM) over COX-1 (IC₅₀?>?50?μM) and a significant non-competitive inhibition of soybean lipoxygenase enzyme (IC₅₀?=?13.09?μM, Ki?=?0.92?μM). Thus, in silico, in vivo, and in vitro findings suggested that the synthesized indolizine compound 56 has a dual COX-2 and LOX inhibition characteristic and parallel in vivo anti-inflammatory activity in comparison to the standard drugs.

Full text of DOI:10.1016/j.bmc.2017.06.027

Accepted Manuscript
Design, synthesis, and biological evaluation of some novel indolizine deriva-
tives as dual cyclooxygenase and lipoxygenase inhibitor for anti-inflammatory
activity
Sushant K Shrivastava, Pavan Srivastava, Robin Bendresh, Prabhash Nath
Tripathi, Avanish Tripathi
PII:
S0968-0896(17)30566-7
DOI:
http://dx.doi.org/10.1016/j.bmc.2017.06.027
BMC 13810
Reference:
To appear in:
Bioorganic & Medicinal Chemistry
Received Date:
Revised Date:
Accepted Date:
20 March 2017
14 June 2017
15 June 2017
Please cite this article as: Shrivastava, S.K., Srivastava, P., Bendresh, R., Tripathi, P.N., Tripathi, A., Design,
synthesis, and biological evaluation of some novel indolizine derivatives as dual cyclooxygenase and lipoxygenase
inhibitor for anti-inflammatory activity, Bioorganic & Medicinal Chemistry (2017), doi: http://dx.doi.org/10.1016/
j.bmc.2017.06.027
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.

Graphical Abstract
To create your abstract, type over the instructions in the template box below.
Fonts or abstract dimensions should not be changed or altered.
Design, synthesis, and biological evaluation of some novel
indolizine derivatives as dual cyclooxygenase and
lipoxygenase inhibitor for anti-inflammatory activity
Leave this area blank for abstract info.
a
a
a
Sushant K Shrivastava ∗, Pavan Srivastava , Robin Bendresh ,
a
a
Prabhash Nath Tripathi , and Avanish Tripathi
a
Pharmaceutical Chemistry Research Laboratory, Department of Pharmaceutics, Indian Institute of Technology
Banaras Hindu University), Varanasi-221005, U.P., India
(
1

Bioorganic & Medicinal Chemistry
journal homepage: www.els evier.com
Design, synthesis, and biological evaluation of some novel indolizine derivatives as
dual cyclooxygenase and lipoxygenase inhibitor for anti-inflammatory activity
a
a
a
a
Sushant K Shrivastava ∗, Pavan Srivastava , Robin Bendresh , Prabhash Nath Tripathi , and Avanish
Tripathi
a
a
Pharmaceutical Chemistry Research Laboratory, Department of Pharmaceutics, Indian Institute of Technology (Banaras Hindu University), Varanasi-221005,
U.P., India.
ARTICLE INFO
ABSTRACT
Article history:
Received
Some novel indolizine derivatives were synthesized by bioisosteric modification of imidazo[1,2-
a]pyridine for anti-inflammatory activity. The physicochemical characterization and structure of
compounds were elucidated by state of the art spectroscopic technique. Induced fit docking was
performed for initial screening to elucidate the interactions with corresponding amino acids of
cyclooxygenase (COX-1, COX-2) and lipoxygenase (LOX) enzymes. The target compounds 53–
Received in revised form
Accepted
6
0 were then evaluated against in vivo carrageenan and arachidonic acid induced rat paw edema
Available online
models for anti-inflammatory activity. Amongst all the synthesized derivatives, compound 56
showed the significant anti-inflammatory activity in both rat paw edema models with very less
ulcerogenic liability in comparison to standard diclofenac, celecoxib, and zileuton. The
compounds 56 was further assessed to observe in vitro enzyme inhibition assay on both
cyclooxygenase and lipoxygenase enzyme where it showed a preferential and selective non-
competitive enzyme inhibition towards the COX-2 (IC50 = 14.91 µM, Ki = 0.72 µM) over COX-
Keywords:
Indolizine
Non-classical bioisostere
Dual COX-2/LOX inhibitior
Induced fit docking
Non-competitive
1
(IC50 > 50 µM) and a significant non-competitive inhibition of soybean lipoxygenase enzyme
(
IC50 = 13.09 µM, Ki = 0.92 µM). Thus, in silico, in vivo, and in vitro findings suggested that the
synthesized indolizine compound 56 has a dual COX-2 and LOX inhibition characteristic and
parallel in vivo anti-inflammatory activity in comparison to the standard drugs.
ꢀ
ꢀ ꢀ2017 Elsevier Ltd. All rights reserved.
—
——
∗
Corresponding author. Tel.: +919452156527; fax: +91-542-2368428; e-mail: skshrivastava.phe@itbhu.ac.in
2

1
. Introduction
The prostaglandins and leukotrienes are naturally occurring twenty carbon fatty acid derivatives produced through biochemical
1
oxidation of arachidonic acid (AA), which play an essential modulatory role in many normal and disease-related cellular processes. In
fact, much of the inflammation, pain, fever, nausea, asthmatic and allergic reaction occurs due to excessive production of prostaglandins
2
and leukotrienes.
The primary enzyme involved in the first step of the AA cascade is cyclooxygenase (COX), which exists in three isoforms as COX-
3
1
, COX-2, and COX-3. COX-1 and COX-2 are structurally 63% identical and 77% similar at the amino acid level. The COX-1 is
ubiquitous form typically produced in normal, quiescent condition and remains as a constitutive protein of normal cell. It is also
important in the production of prostaglandins that regulates cellular homeostasis, such as renal blood flow, and in circumstances where
4
prostaglandins have a protective function such as gastric mucous production. COX-2 is the inducible form of the enzyme, expressed in
the endothelial cell, chondrocytes, and osteoblast of traumatic tissue after tissue trauma and therefore plays a major role in
inflammation. COX-3 is an enzyme mostly present in the brain, expressed under the influence of COX-1 gene, but not functional in the
5
human being.
There are some other important AA metabolites like leukotriene produced by Lipoxygenase (LOX) enzyme activity. LOXs are the
member of non-iron containing dioxygenases family and be available for animals, plants, and fungi. In humans, three functional
6
isoforms of LOX exist as 5-, 12-, and 15-LOXs, whereas two isoforms 9- and 13-LOX exist in plants.
The crystal structures of two COXs suggested that the active site has a narrow hydrophobic channel extending from the membrane-
binding region to the protein core. Initially, the binding of the substrate at COX enzyme occurs at the channel opening pocket lined with
7
Arg120 and Ile345 in both the enzymes. The key difference between the COX-1 and COX-2 isozyme active site is the exchange of
8
isoleucine in COX-1 for valine in COX-2 at positions 434 and 523. The differences in the amino acid sequence make the COX-2
substrate-binding site more flexible and somewhat larger by creating a secondary pocket. The COX-2 selective inhibitors explicitly bind
to this secondary binding pocket (lined by His90, Arg513, and Val523) resulting in specific inhibition of COX-2 activity. Apart from
this secondary pocket, another critical region in the COX-2 active site lined by Trp387, Tyr385, Phe518, Phe381, Met522, and Leu352
is known as the hydrophobic pocket. The selective COX-2 inhibitors acquire a pharmacophore which can selectively bind to the
8
secondary pocket and bring enough steric bulk to block the hydrophobic channel of COX-2. The active site of 5-LOX is an elongated
cavity, with no clear access to bulk solvent, lined with both invariant (Leucines 368, 373, 414, and 607 and Ile406) and 5-LOX-specific
9
amino acids (Tyr181, Ala603, Ala606, His600, and Thr364). Further, alignment studies of five isoforms of LOX and two isoforms of
COX suggested that pharmacophoric interaction with amino acid Tyr181, Phe359, Phe421, and Trp599 at 5-LOX binding site may
1
0
increase specificity towards COX-2 and 5-LOX.
Non-steroidal anti-inflammatory drugs (NSAIDs) as COX inhibitors are the leading prescription medicine worldwide for the cure
of inflammation, but their long-term use is restricted due to gastrointestinal, bronchoconstriction and hepatotoxic side effects.
Indeed significant NSAIDs are fluxed in the world market, but the blockade of arachidonate cascade at the COX level diverts the
substrate towards increased production of LOX-derived eicosanoids such as leukotrienes (LTs) that cause bronchoconstriction,
ulceration, and inflammation, which exists as a big challenge for medicinal chemists.
Considering the pro-inflammatory properties of LTs and prostanoids, the drugs able to block the synthesis of both eicosanoids,
should prove itself as a better anti-inflammatory drug molecule with fewer side effects in comparison to established classical NSAIDs
11
and selective COX-2 inhibitors. The COX and LOX drug inhibitors are expected to enhance anti-inflammatory potency without risks
of serious side effects. Hence the discovery of dual COX and LOX enzymes inhibitors with reduced toxicity and side effects is the need
1
2
of pharmacotherapeutics in the modern age.
1
.1. Designing considerations
In the field of drug development and therapeutics, bioisosterism is a successful analog designing strategy over the years and
1
3
translated into the development of drugs like Alloxanthine and Procainamide, etc. Taking a cue from this, we have studied some
th
th
Imidazo[1,2-a]pyridine derivatives having an aromatic ring at 4 position, and cyclohexanamine or cyclopentanamine at 5 position
1
4
reported for LOX inhibition (IC50 = 0.21 µM) at micromolar concentration. In the designing, the Imidazo[1,2-a]pyridine nucleus was
chosen, where bioisosteric replacement of =N- with =CH- (ring equivalent) leads to flanged bicyclic nucleus ‘indolizine’ that could be
considered as a novel class under non-classical bioisostere design.
Compounds with indolizine ring have received attention in recent years like Curindolizine, a chemical generated from Curvularia
(
species: IFB-Z10) reported for anti-inflammatory activity in lipopolysaccharide (LPS)-induced RAW 264.7 macrophages (IC50 = 5.31
15
µM). Licofelone is another molecule having 3H-pyrrolizine fragment like Curindolizine, reported for 5-LOX (IC50 = 0.21 µM), COX-
(IC50 = 0.16 µM) and COX-2 (IC50 = 0.37 µM) inhibitory activity and had been passed the safety level in the clinical trial, but the
1
16
challenge of the ulceration persists in licofelone at doses of 30 and 100 mg/kg. It may be due to the blocking of COX-1 enzyme and
the strategy to block COX-2, and LOX enzyme selectively may produce a new dual COX-2 and LOX inhibitor to treat inflammatory
disorders.
The primary objective of this study was to design and synthesize some novel 3-(aminomethyl)indolizine-1-carboxylic acid
derivatives using the bioisosteric modification of imidazo[1,2-a]pyridine to indolizine (Figure 1) and further evaluated for their
particular COX and LOX enzyme inhibition, anti-inflammatory activity, and ulcerogenic liability.
2
. Results and discussion
2
.1. Chemistry
The targeted compounds were synthesized as per Schemes 1. The known malonate derivatives (10-17) were synthesized through the
Knoevenagel condensation by reacting substituted aldehydes (1-8) with diethyl malonate (9) in the presence of catalytic amount of
3

17
piperidine. The reported pyridinium bromide salt (20) was then prepared by the addition of Bromo acetonitrile (19) in pyridine (18).
Pyridinium bromide salt (20) then conjugated with malonate derivatives (10-17) through Huisgen [3+2]-cycloaddition reaction in
dichloromethane to give unstable intermediate compounds (21-28) which than cyclized to five-membered heterocyclic indolizine
compounds (29-36). Intermediate compounds (21-28) are unstable, so we made no attempts to isolate them. Also, indolizine compounds
(
29-36) were carried forward in next step without further purification and reacted with chloranil or chromium trioxide in the presence of
18, 19
the strong base and resulted in the corresponding aromatized indolizine derivatives (37-44),
positive Dragendorff test on TLC. Further, the nitrile was reduced by nickel boride,
which was initially confirmed by
2
0
Cl
N
OH
N
O
N
O
N
HOOC
H
NH
H
N
N
N
Licofelone
Curindolizine
imidazo[1,2-a]pyridine
HO
O
R
N
NH
2
Designed compound
Figure 1: Design of a novel series of compounds as multitarget-directed potential anti-inflammatory agents.
4

O
O
O
Et
O
O
Et
O
a
+
+
Et
R
H
Et
O
O
O
9
R
1
0-17
1-8
b
Br-
Br
N
N
N+
N
18
19
20
2
EtO C
Et
EtO
N+
2
C
CO Et
2
O
O
R
2
CO Et
Br-
N+
c
C-
Et
N
R
O
O
N
+
R
CN
CN
21-28 (Intermediate)
20
10-17
29-36 (Intermediate)
d
CO
2
Et
R
N
R =
CN
3
3
3
4
4
4
4
4
7; 45; 53 = Methyl
8; 46; 54 = Ethyl
9; 47; 55 = Butyl
0; 48; 56 = Phenyl
37-44
e
HO
R
2
C
5
2
EtO
R
2
C
6
7
4
3
f
1; 49; 57 = 1,3-benzodioxole-5-yl
2; 50; 58 = 4-(trifluoromethyl)phenyl
3; 51; 59 = 4-Florophenyl
8
N
1
9
N
H
2
N
CH
2
H N CH2
2
5
3-60
45-52
4; 52; 60 = 3-methoxy-4-hydroxyphenyl
Scheme 1: Reagents and conditions: (a) piperidine, EtOH, refluxed, 5- 6 h (b) rt, overnight (c) NaOH (32%), DCM (d) in situ, Chromium
trioxide/Chloranil, rt, 3- 4 h (e) NiCl , NaBH , anhydrous EtOH, 0-5°C, 3- 4 h (f) NaOH (12%), THF, rt, overnight.
2
4
Figure 2. Induced fit docking study of best pose generated on COX-2 (PDB: 1CX2) for (A) celecoxib (purple) and compound 56 (sky blue); ligand binding
site for docked compound 56 represented in blue, (B) 2- Dimensional representation of compound 56 at the active site.
5

generated from in-situ sodium borohydride and nickel (II) chloride reaction in dry ethanol, resulted in the corresponding aminomethyl
2
1
22
carboxylate (45-52), which was confirmed by positive Ninhydrin Test on TLC. The novel target 3-(aminomethyl)indolizine-1-
carboxylic acid derivatives (53-60) were generated by base-catalysed hydrolysis of the ester. Preliminary identification of acid was
2
3
confirmed by positive Bromo Cresol Green test on TLC.
The FT-IR spectra of nitrile compounds (37-44) showed the characteristic medium C-N stretching of nitrile peak in the range of
-
1
-1
2
300-2350 cm and C-O stretching of ester peak in the range of 1735-1755 cm . Structures of compounds (37-60) were also confirmed
1
by NMR spectroscopy. In all compounds, H NMR spectra showed a peak of acid at down field (at 11 δ value). The characteristic
amine peak was observed at 3-5 δ value with integration value of 2. Another peak of the CH proton with integration value of 2 as triplet
was observed at near to 3 δ value. Further, C NMR spectra also confirmed the presence of carboxylate group by showing a peak near
2
13
to 200 δ at down field.
2
.2. Log P determination and prediction of drug-like properties
The lipophilicity (log P) of a molecule is considered as a characteristic to establish a relationship between the pharmacodynamic and
24
pharmacokinetic property of drugs. We have determined the partition coefficient of the synthesized compounds and standards by
octanol/water system (Supplementary Table 1). Almost, the log P values of compounds were observed under suitable lipophilic range
(
2-4). The drug-likeness of all the compounds and standard were also calculated through QikProp software tool (Maestro 10.5.014,
Schrödinger, LLC, and New York-1), in silico ADME/Tox Predictions ensured that all compounds follows Lipinski's rule of five and
could be considered as a drug-like molecule. (Table 1)
2
.3. Docking study
2
5
The synthesized compounds (37-60) were subject for Rigid Docking on COX-1, COX-2, and LOX enzymes (PDB Code: 1CX2,
2
6
27
3
N8Y, and 3V99 respectively) using Glide-XP protocol in the Schrödinger software suite. The Glide Score (GScore) and the
interactions at the active site were recorded. The compounds 53-60 were observed with top GScore and selected for Induced Fit
2
8
Docking (IFD) protocol to obtain more accurate ligand placements under fine flexibility treatment of the protein. The GScore and IFD
scores of compounds (53-60) are depicted in Table 1.
The docking poses of most potential compound 56 on COX-2 enzyme displayed poses similar to that of standard celecoxib (Figure
2
A). The indolizine ring of compound 56 was interacting at the hydrophobic channel of COX-2 lined by Leu352, Tyr385, Trp387,
Phe518, and Met522 along with contributions from the backbone atoms of Ala527 (via hydrogen bonding). Beyond this hydrophobic
pocket, the carboxylic group showed the penetration into secondary pocket lined by His90, Phe518, Trp355, and Val523. One of the
oxygen atoms of the carboxylic acid of compound 56 formed a hydrogen bond to Trp355 and linked by a salt bridge with His90 at the
secondary pocket (Figure 2B). The interaction of celecoxib (Supplementary Figure 1) and compound 56 at His90 of the secondary
2 2
pocket of COX-2 enzyme showed the same binding owing to the bioisosteric similarity (SO NH vs. COOH).
A similar method was used to compare ligand interactions in COX-1 (PDB code: 3N8Y) enzyme active site. All compounds showed
a different conformation in COX-1 with respect to the conformation of the COX-2 active site. (Figure 3)
The docking study of licofelone was performed on 5-LOX enzyme (PDB: 3V99), it interacts with His372 via hydrogen bonding and
also showed hydrophobic interaction with Phe359 (Supplementary Figure 2). The binding pose of compound 56 and licofelone was
superimposed, and the interactions were compared with respect to the LOX enzyme (Figure 4A). It was found that indolizine rings of
compound 56 and pyrrolizine nucleus of licofelone oriented toward the same region and exhibited hydrophobic interaction with
surrounding active site amino acids Leu414, Leu607, and Ala603 in the hydrophobic region of LOX. Compound 56 was stabilized in
the active site of LOX enzyme by the forming hydrogen bonds with Gln363 and showed π-cation interaction with Phe177. Compound
5
6 also showed hydrophobic interaction with Phe359 responsible for specificity towards COX-2 and LOX (Figure 4B).
6

Figure 3. Induced fit docking study of best pose generated on COX-1 (PDB: 3N8Y) of (A) compound 56, ligand binding site for docked compound 56
represented in blue, (B) 2- Dimensional representation of compound 56 at the active site.
Figure 4. Induced fit docking study of best pose generated on LOX (PDB: 3V99) of (A) licofelone (purple) and compound 56 (sky blue), ligand binding site for
docked compound 56 represented in blue, (B) 2- Dimensional representation of compound 56 at the active site.
2
2
.4. Pharmacology
.4.1. Carrageenan-induced paw edema
Carrageenan-induced rat paw edema model is considered as standard screening experiment to observe acute inflammation. The paw
edema was induced by the subplantar injection of carrageenan and showed a biphasic response. It is reported that early phase releases of
pro-inflammatory agents such as serotonin, histamine, whereas kinins and, the late phase (3 h post treatment) mediated by
2
9
prostaglandins. The outcomes of the anti-inflammatory activity against carrageenan-induced hind paw edema of the target compounds
53-60) is summarized in Table 2. Mean changes in the paw edema thickness at two hours intervals from the induction of inflammation
(
up to six hours along with inhibition percentage of edema were calculated and recorded. The compound 56 exhibited parallel inhibition
compared to the standard diclofenac which exhibited an inhibitory activity of 90.9 percent at six hours. In all eight evaluated
derivatives, compounds 56, 57, 58, 59 and 60 were found to possess good anti-inflammatory potential in the range of 88.6, 78.1, 67.7,
8
6.4, and 59.4 percent respectively. Compound 56, bearing a phenyl group on indolizine moiety, exhibited activity (88.6 percent)
parallel to diclofenac (90.9 percent). The aliphatic nucleus bearing Methyl (53), Ethyl (54) and Butyl (55), exhibited a decreasing order
of anti-inflammatory activity (38.5, 34.3, and 23.9 percent). Structure-activity relationship inferred that substitution pattern at position 8
of indolizine ring is supposed to be an active site for determining the anti-inflammatory activity. The result indicated that substitution of
th
the aromatic ring system at position 8 (56-60) resulted in more active compound than the aliphatic chain substituted compounds (53-
5
5). Compounds at position 8 with the electronegative groups like CF , fluorine and hydroxyl group (57-60) showed less anti-
3
th
inflammatory protection than the other compound (53-55). Overall unsubstituted phenyl ring at 8 position showed best anti-
inflammatory activity among synthesized compounds. Based on this observation and the biphasic nature of carrageenan-induced paw
edema, it is possible to propose that the significant activity
7

Table 1. Docking and QikProp Analysis
b
c
d
e
f
g
h
Code
COX-2 COX-2
LOX
LOX
SASA
HB
HB
QPlogPC16
QPlogPoct
QPlogPo/w
QplogS
GScore IFD
score
GScore IFD
score
5
5
5
5
5
5
5
6
3
4
5
6
7
8
9
0
-8.75
-1100.6 -8.60
-1510.6
428.7
435.4
502.8
500
3
3
3
3
3
3
3
4
2
1
3
3
7.6
13.2
13.3
14.4
16.1
17.3
18.4
16.8
18.9
21.2
17.7
14.6
-0.9
-0.6
0.3
0.5
0.1
1.5
0.7
0.05
3.4
6.4
0.9
-1.4
-1.4
-2.4
-2.9
-2.7
-4.3
-3.2
-2.8
-6.4
-7.3
-1.6
-9.19
-1110.9 -8.53
-1118.5 -6.55
-1122.7 -8.26
-1125.9 -8.02
-1124.1 -8.06
-1123.2 -8.41
-1511.2
-1508.8
-1512.3
-1509.9
-1511.3
-1512.3
-1513.0
3
7.8
-9.11
3
8.9
-9.80
3
10.1
10.1
9.8
-10.26
-11.12
-10.31
-10.70
-12.69
503.7
551.8
509.7
546.2
661.6
650.4
447.7
4.5
3
3
9.7
-1124.7 -8.66
4.5
5.5
2
10.9
11.4
12.5
8.4
a
a
celecoxib
-1125.1
nd
nd
licofelone -10.98
-1120.7 -8.07
-1496.1
-1500.7
a
a
zileuton
nd
nd
-7.66
3.7
a
Not determined.
b
c
d
e
f
SASA, total solvent accessible surface area (range, 300-1000 Å2).
H-bond donors.
H-bond acceptors.
QPlogPC16, predicted log of hexadecane/gas partition coefficient (range, 4-18).
QPlogPoct, predicted log of octanol/gas partition coefficient (range, 8-43).
QPlogPo/w predicted log of octanol/water partition coefficient (range 2 to 6).
QplogS, predicted log of aqueous solubility (range, 6 to 0.5 M).
g
h
observed in the late phase of inflammation is attributed to the ability of the compounds to inhibit the release of late mediators. Thus, it is
proposed that the evaluated compounds mitigate the inflammatory effects of carrageenan by inhibiting the release of prostaglandins
2
.4.2. Arachidonic acid induced paw edema
AA-induced paw edema is used to distinguish between COX and lipoxygenase (LOX) inhibitors, as the model is known to be more
30
sensitive to LOX inhibitors. A subplantar injection of AA produced significant edema within 30 to 60 minute. The paw edema
induced by AA is perceptibly reduced by LOX inhibitors. The results illustrated in Table 3 indicating that zileuton (a selective LOX
inhibitor) exhibited an inhibition (72.3 percent) against AA-induced paw edema. Compounds (56, 58, 59 and 60) were effective in
inhibiting the inflammation induced by AA, with their mean percentage protection as 66.2, 40, 61.5, and 36.9 percent respectively. The
outcome of this study proposes that the most active compound 56 mediate their effects by inhibiting the COX pathway as well as LOX
pathway.
2
.4.3. Ulcerogenic risk evaluation
Compound 56 exhibiting promising anti-inflammatory profiles on in vivo model was further assessed for its ulcerogenic liability in
terms of the ulcer index (UI). The results obtained from the postmortem studies of animals sacrificed on the 21 days of post adjuvant-
induced arthritis. The study revealed that compounds 56 were found safer with respect to gastrointestinal (GI) toxicity (UI 10 ± 1.4) in
comparison with standard diclofenac, celecoxib, and zileuton (UI 46 ± 4.1, 53 ± 5.5, and 52 ± 4.9 respectively) (Table 4). These
findings, corroborated with the outcome of the histological evaluation of the gastric mucosa that support the high gastric tolerability of
the compounds relative to standard compounds (Figure 5).
Table 2. Carrageenan-induced paw edema.
Compound
a
Mean protection (%)
2
h
4h
6h
control
0
0
0
diclofenac
74.6
41.3
30.2
23.8
68.2
61.9
42.8
69.8
91.6
44.2
32.5
24.4
90.5
74.4
62.8
89.5
90.9
38.5
34.3
23.9
88.6
78.1
67.7
86.4
5
5
5
5
5
5
5
3
4
5
6
7
8
9
8

6
0
50.8
63.9
59.4
Control: 0.3% carboxymethylcellulose sodium (CMC) solution in distilled water (10 ml/kg/p.o.).
Standard: diclofenac (10 mg/kg/p.o) in 0.3% CMC solution.
All other compounds were administered at an equimolar oral dose relative to 10 mg/kg diclofenac
a
Mean protection (%) was expressed as ( %) edema inhibition of the tested compounds relative to (%) edema inhibition of standard.
2
.5. COX and LOX inhibition assay
Based on the results of the preliminary in vivo screening, the compound 56 showed significant anti-inflammatory activity, and
further assessed in vitro COX and LOX enzyme inhibition assay. (Table 5)
9

Table 3. Arachidonic acid induced paw edema.
Compound
%Protection
control
0
zileuton
72.3
23.1
20.2
15.4
66.2
23.1
40.0
61.5
36.9
5
5
5
5
5
5
5
6
3
4
5
6
7
8
9
0
Standard: zileuton (10 mg/kg/p.o) in 0.3% CMC solution.
All other compounds were administered at an equimolar oral dose relative to 10 mg/kg standard
a
Protection (%) was expressed as % edema inhibition of the tested compounds relative to % edema inhibition of standard.
2
.5.1. In vitro COX inhibition assay
In vitro COX inhibition of compound 56, performed on ovine COX-1 and human COX-2 enzymes using colorimetric enzyme
31
immunoassay (EIA). (Supplementary Figure 3) Along with the compound 56, diclofenac and celecoxib were tested and the data
obtained are reported in Table 5. The result obtained infers a specific inhibition of COX-2 over COX-1 of compound 56, (COX-1; IC50
>
50 µM, COX-2 (IC50 = 14.91 µM) in comparison to standard diclofenac (COX-1; IC50 = 0.15 ± 0.009 µM, COX-2; IC50 = 0.05 ±
0
.001µM) and celecoxib (COX-1; IC50 = 6.7 ± 0.029 µM, COX-2; IC50 = 0.87 ± 0.041 µM). Compound 56 exhibited a selective COX-2
inhibition very close to the standard celecoxib. It may be due to the proper size, orientation and binding of compound 56 into the
secondary pocket of the COX-2 enzyme as observed in IFD studies.
2.5.2. In vitro LOX inhibition assay
The in vitro analysis of lipoxygenase inhibitory activity was performed using Soybean lipoxygenase (linoleate 13S- lipoxygenase,
3
2, 33
linoleate: oxygen oxidoreductase, EC 1.13.11.12).
IC50 values were determined through Continuous Spectrophotometric Rate
Determination method. The LOX enzyme formed the 9- and 13-hydroperoxides by LOX enzyme with substrate linoleic acid and
resulted in increased absorbance at 234 nm. Compound 56 showed in vitro LOX inhibitory activity (LOX; IC50 = 13.09 µM) in
comparison with zileuton (LOX; IC50 = 3.429 µM) (Table 5). The in vitro studies demonstrated that compound 56 inhibits LOX as well
as COX-2 activity. The results of in vitro activity on both COX and LOX were corroborated with the outcome of earlier in vivo results,
hence confirmed the dual inhibition nature of compound 56 on both LOX and COX.
2
.6. In vitro kinetic study
The mechanism of inhibition of cyclooxygenase and lipoxygenase enzyme by compound 56 was determined by measuring the
enzyme activity at fixed concentrations of arachidonic acid in the presence of increasing concentrations of compound 56. The enzyme
3
4
kinetic study was observed and determined by Michaelis-Menten equation. The Y-axis on the Lineweaver-Burk Plot was graphed as
the 1/Vmax (reciprocal of reaction rate) and the X axis as the reciprocal of substrate concentration (Figure 6). The results obtained from
the kinetic inhibition pattern demonstrated a non-competitive inhibition of LOX enzyme (Ki = 0.72 ± 0.060 µM) and COX-2 enzyme
(
Ki = 0.92 ± 0.056 µM). Thus, the in vitro enzyme kinetics and IC50 of compound 56 reflected the dual inhibition of COX-2 and 5-LOX
enzyme.
Table 4. Ulcer index of the representative compound 56.
Table 5. IC50 values of the representative compound 56.
Compound Code
COX-2
Ulcer
Compound Code
COX-1
LOX
56
10 ± 1
53 ± 5
46 ± 4
52 ± 4
IC50 (µM) ± SEM
>50
IC50 (µM) ± SEM
IC50 (µM) ± SEM
13.09 ± 0.260
celecoxib
56
diclo f1e 4n . a9 c1 ± 0.203
a
celecoxib
diclofenac
zileuton
6.7 ± 0.029
0.87 ± 0.041
zileuton
nd
a
0.15 ± 0.009
0.05 ± 0.001
nd
a
a
nd
nd
3.429 ± 0.045
a
Not determined
3
. Materials and methods
3
.1. Chemistry
Chemicals were purchased from Sigma, Aldrich, SD-Fine, Spectrochem, Avara, and Merck. The synthesized compounds were
purified by column chromatography. The reaction was monitored by thin layer chromatography with ethyl acetate: hexane as the mobile
10

phase and performed on Merck silica gel 60 F254 aluminum sheets (Merck, Germany). FTIR spectra were recorded on Shimadzu
400S FTIR (Shimadzu Corporation, Japan) by KBr pellets. Elemental analyses for C, H, and N were performed on Exeter CE-440
8
1
(
Hewlett-Packard, USA) elemental analyzer. H NMR spectra were recorded at room temperature on a Bruker 500 MHz using TMS as
the internal standard, and the chemical shifts were reported in δ.
17
3
.1.1. Synthesis of diethyl 2-benzylidenemalonate (10-17)
The respective aldehyde (1-8) (5 mmol) and diethyl malonate (9) (5 mmol) were added in the presence of piperidine (4 mmol) under
an inert atmosphere and transformed into a viscous solution. It was refluxed for 5- 6 h and quenched with 6N HCl in cold conditions.
The product was extracted with ethyl acetate (3×10 mL). The combined organic extracts were dried and evaporated under reduced
pressure. Compounds (10-17) were recrystallized from 50% aqueous ethanol and carried forward without further purification.
19
3
.1.2.Synthesis of 1-(cyanomethyl)pyridinium Hydrogen Bromide salt (20)
The two-necked round bottom flask fitted with a thermometer, mechanical stirrer, and a dropping funnel was charged with Bromo
acetonitrile (19) (14.3 mmol) to solution of pyridine (18) (18.6
a
11

mmol) in tetrahydrofuran (THF). The reaction mixture was kept with vigorous stirring for overnight at room temperature. The reaction
was monitored by TLC and recrystallization was carried out by with 95% ethanol. The pure compound was separated out as the white
powder. The pyridinium hydrobromide salt (20) was further used in next step of the reaction.
19
3.1.3. Synthesis of ethyl 3-cyano-2-phenylindolizine-1-carboxylate (37-44)
Malonate derivatives (10-17) (500 µmol) and pyridinium hydrobromide salt (20) (750 µmol) were dissolved in dichloromethane (15
mL), further 1 mL of NaOH (37%) was added at room temperature. After completion, the reaction was poured into water and then
extracted with the dichloromethane. The organic layer was washed with brine solution and dried over anhydrous Na SO . The reaction
2 4
mixture was subjected to oxidation by Chromium trioxide/Chloranil (1 equiv.), and the solution was stirred at room temperature for 3- 4
h. The solvent was evaporated, and the residue was subjected to column chromatography (n-hexane/EtOAc). The resulting compounds
(
37-44) were dissolved in chloroform, and the insoluble precipitate was removed by filtration. After evaporation, the product was
further purified by recrystallization using diethyl ether.
2
1
3
.1.4. Synthesis of ethyl 3-(aminomethyl)-2-phenylindolizine-1-carboxylate (45-52)
A mixture of substituted nitrile (37-44) (9.8 mmol) and anhydrous nickel (II) chloride (9.8 mmol) was added into pre-cooled dry
ethanol (20 ml) at 0-5°C. Sodium borohydride (29.4 mmol) was added slowly to the reaction mixture. The reaction mixture was stirred
at 0-5°C and monitored by TLC. Water (6 ml) was added, and inorganic impurities were separated out by filtering the reaction mixture.
The resulted compounds (45-52) were evaporated under vacuum and purified by column chromatography over silica gel using hexane:
ethyl acetate as the eluent.
3
.1.5. Synthesis of Substituted 3-(aminomethyl)-2-phenylindolizine-1-carboxylic acid (53-60).
NaOH (14mmol) was added to the ethyl 3-(aminomethyl)-2-phenylindolizine-1-carboxylate (45-52) in THF and stirred for
overnight. After completion of the reaction, THF was removed under high vacuum, diluted with water and then extracted with the ethyl
acetate layer. The organic layer was washed with water and 1N HCl. The crude mixture was purified by column chromatography over
silica gel (120 mesh) and eluted with 25% ethyl acetate in hexane furnished the target compounds (53-60).
3
3
.2. Molecular docking study
.2.1 Rigid Docking using Glide-XP
The synthesized compounds and substrate were computationally docked into enzymes using Glide to gain some structural insights
into the binding mode of the ligand with COX-1, COX-2, and LOX enzymes. The crystal structure of enzymes in complex with
substrate retrieved from the protein data bank and used for molecular docking. Enzymes were subsequently optimized with the “protein
preparation wizard” workflow. The ligand was removed, and the ligand binding site was defined. The ligands were built using Maestro
1
0.5.014 build panel and prepared by Lig Prep (Schrödinger, LLC, USA, 2016-1) application through OPLS 2005 force field. OPLS
stands for
Figure 5. Photomicrographs (10x magnification) of (a) diclofenac (b) control and (c) compound 56 treated groups in rat stomach tissues (hematoxylin and
eosin staining). The arrow indicates severe detachment of surface epithelium, resulting in the formation of lesions.
12

Figure 6. Lineweaver-Burk plot of in vitro COX-2 inhibition by compound 56 and in vitro LOX inhibition by compound 56.
optimized potential liquid simulations, and it gave the corresponding energy minima 25 conformers of the ligands. The default settings
were used for all other parameters. The extra precision (GLIDE-XP) protocol implemented in GLIDE was used for docking (Glide 5.8,
Schrödinger Inc., USA).
3
.2.2. Induced Fit Docking
In IFD, softened-potential docking into a rigid receptor was performed to generate groups of poses, and top 20 poses were retained.
Further, the ligand was re-docked into low energy induced-fit structures obtained using Prime program in Schrödinger software suite,
and then IFD scoring was calculated (IFD score = GScore + 0.05 * Prime_Energy) that accounts for both docking energy (GScore),
2
8
and receptor strain and solvation terms (Prime_energy).
3
3
.3. Pharmacological screening
.3.1. Experimental animals
Healthy, adult, Swiss albino mice of either sex having weight 28 to 32 g or Wistar rats of either sex weighing 180-220 g were used
for the experimental protocols and procured from the Central Animal House, Institute of Medical Sciences, Banaras Hindu University.
The present biological study was approved by Banaras Hindu University Animal Ethical Committee (BHU; Dean/12-13/CAEC/19).
The entire experimental methods were in agreement with CPCSEA guidelines, Ministry of Environment and Forests, Government of
o
India. The animals were adapted to guarded laboratory surroundings (12 h light/dark cycle, 22-24 C at 40-60% R
H
) and were allowed to
access water and food ad libitum.
3
5
3
.3.2 Carrageenan-induced paw edema (acute inflammatory model)
Anti-inflammatory activity of synthesized compounds was screened using carrageenan-induced rat paw edema model. Paw edema
was induced by administration of 0.1 ml of a 1% w/v solution carrageenan into the sub-planter region in the left hind paw of the rats.
Briefly, all the animals were randomly divided into various groups and treated with standard diclofenac sodium (10 mg/kg, p.o.); test
compounds (53-60) (dose equimolar relative to 10 mg/kg diclofenac sodium) and the vehicle as a control was administered orally. The
paw size was measured in mL using Plethysmometer after two, four and six hour of the intoxication of carrageenan injection. The
subjects were pretreated with the selected derivatives and standard drug 1 h before the administration of carrageenan.
36
3
.3.3 Arachidonic acid induced rat paw edema
Paw edema was induced by injecting 0.1 ml of 0.5% w/v AA in 0.2 M carbonate buffer, pH 8.2 via sub-plantar region into right
hind paw of rats after 30 min post drug treatment. The 0.3% carboxymethylcellulose sodium (CMC) was used as control (10 ml/kg
oral), and the test compounds were given at an equimolar oral dose relative to standard 10 mg/kg zileuton. Changes in thickness (mm)
were measured by a digital Vernier caliper 1 h after the AA injection.
3
7
3
.3.4 Ulcerogenic potential
Ulcerogenic studies were performed on fasting animal via administering test compounds for 7 successive days and was forfeited
with cervical dislocation. The incision was done on the peritoneal cavity; stomach was separated and place in a saline plate. A
longitudinal incision along with greater curvature was made, and the stomach was cleaned by washing with cooled saline and inspected
with a 3X magnifying lens for any evidence of ulcer. Lesions area was counted, and ulcer index was calculated as per Szelenyl and
3
8
Thiemer. Further, histological studies were carried out (embedded in paraffin blocks, fixed in 10% v/v formalin and stained with
hematoxylin and eosin).
3
.4. In vitro inhibitory activity
.4.1. In vitro COX inhibition assays
The COX activity of the tested compounds to inhibit both COX-1 and COX-2 isozymes was evaluated using colorimetric COX
ovine) Inhibitor Screening Assay Kit (Cayman Chemicals; Item no. 560131). Different dilutions of celecoxib and tested compounds
3
1
3
(
were incubated with the enzymes for 5 min at 25°C. After the incubation, an addition of the colorimetric substrate and AA was done,
and the absorbance was measured at 412 nm using plate reader.
3
.4.2. Inhibition of LOX
13

Soybean lipoxygenase (linoleate 13S-lipoxygenase) is frequently used as a dependable screen for LOX inhibition. It has been studied
that AA binding sites in soybean lipoxygenases contribute to almost the similar resemblance with animal LOX. It has been reported that
amino acid in the binding sites of human LOX shares almost the same similarity with AA binding sites in soybean lipoxygenases. The
assay was performed according to Continuous Spectrophotometric Rate Determination method. The stock solution of lipoxygenase was
prepared by dissolving 5,000 - 10,000 units/ml of Lipoxidase in 200 mM Borate Buffer, pH 9.0. Six different concentrations of
inhibitors were selected and made for determining the enzyme inhibition activity. Increasing concentrations of inhibitors were added to
enzyme solution and kept for 5 min followed by substrate addition and further kept at room temperature with stirring for 20 mins. The
control contains the only vehicle resulted in 100% enzymatic reaction. Blank contains all the substances except the enzyme solution to
account for non-enzymatic reaction. The reaction rates were compared, and the percentage inhibition due to increasing concentration of
inhibitors was calculated. The concentration of each test compound was recorded in triplicate, and their IC50 values were determined
graphically from absorption v/s concentration curve.
3
9
3
.5. In vitro kinetic study
Continuous Spectrophotometric rate determination method was used to identify the type of inhibition. The substrate (linoleic acid
and AA) was used in different concentrations and prepared in tween-20 in 200 mM borate buffer, and pH 9 was maintained. And a
fixed concentration of enzyme i.e. 10,000 units/ml of Lipoxidase was prepared in 200 mM borate buffer at pH 9.0 in the presence and
absence of most active compound 56 (concluded from IC50 determination). The inhibitory kinetics was evaluated by Lineweaver and
Burk method.
4
. Conclusion
The biological activity with respect to the synthesized indolizine derivative inferred that substitution pattern at position 8 of
indolizine ring proved to be an active site for determining the anti-inflammatory activity. The result indicated that substitution of the
aromatic ring system at position 8th (56-60) resulted in more active compound than substituted aliphatic group (53-55). Compounds at
3
position 8 with the electronegative halogen groups like CF , fluorine and hydroxyl groups (57-60) showed less mean protection to
inflammation, in comparison with the other compound (53-55).The best compound 56 showed a safer gastric profile than diclofenac,
celecoxib, and zileuton as indicated by its low ulcerogenic indices and the histopathological studies. The molecular mechanism of the
best compounds 56 was investigated for COX-1/-2 enzymes and LOX enzyme inhibition. The compound 56 was found dual nature of
inhibition against both COX-2 and LOX enzymes. The in vitro kinetic study of 56 showed a non-competitive type of inhibition on both
LOX and COX-2 enzymes. In conclusion, considering the in vivo, in silico, and in vitro results, it can be suggested that the compound
5
6 is a promising dual COX-2 and LOX inhibitor, deserves further studies which can lead to a discovery of a new lead a potent anti-
inflammatory property.
Acknowledgment
The authors gratefully acknowledge to Indian Institute of Technology (BHU) for providing financial assistance through Institute
Research Project (IRP) scheme and Design Innovation Center (DIC) scheme. Grant No. IIT (BHU) / R&D/IRP/2015-16/3471/L &
Grant No DIC-IIT(BHU)/L-11 respectively.
Supplementary Material
1
13
Detailed experimental procedures with analytical data for compounds including FT-IR, H NMR, and C NMR. (PDF)
References and notes
1
2
3
.
.
.
Shimizu T, Wolfe LS. Journal of neurochemistry. 1990;55(1):1-15.
Henry D. Bailliere's clinical rheumatology. 1988;2(2):425-454.
Gierse JK, McDonald JJ, Hauser SD, Rangwala SH, Koboldt CM, Seibert K. Journal of Biological Chemistry. 1996;271(26):15810-
1
5814.
4
5
6
7
.
.
.
.
Simmons DL, Botting RM, Hla T. Pharmacological reviews. 2004;56(3):387-437.
Aronoff DM, Neilson EG. The American journal of medicine. 2001;111(4):304-315.
Reddy KK, Rajan VKV, Gupta A, Aparoy P, Reddanna P. BMC research notes. 2015;8(1):1.
Yamamoto Y, Hisa T, Arai J, Saito Y, Yamamoto F, Mukai T, Ohshima T, Maeda M, Ohkubo Y. Bioorganic & medicinal chemistry.
2
015;23(21):6807-6814.
8
9
1
1
1
1
1
.
.
0.
1.
2.
3.
4.
Bhardwaj A, Kaur J, Wuest M, Wuest F. Nature Communications. 2017;8(1):1.
Gilbert NC, Bartlett SG, Waight MT, Neau DB, Boeglin WE, Brash AR, Newcomer ME. Science. 2011;331(6014):217-219.
Reddy KK, Rajan VKV, Gupta A, Aparoy P, Reddanna P. BMC research notes. 2015;8(1):152.
Charlier C, Michaux C. European journal of medicinal chemistry. 2003;38(7):645-659.
Abdelall EK, Lamie PF, Ali WA. Bioorganic & medicinal chemistry letters. 2016;26(12):2893-2899.
Mannhold R, Kubinyi H, Folkers G, Brown N, vol. 54: John Wiley & Sons; 2012.
Hieke M, Rödl CB, Wisniewska JM, la Buscató E, Stark H, Schubert-Zsilavecz M, Steinhilber D, Hofmann B, Proschak E.
Bioorganic & medicinal chemistry letters. 2012;22(5):1969-1975.
1
1
1
1
1
5.
6.
7.
8.
9.
Han WB, Zhang AH, Deng XZ, Lei X, Tan RX. Organic letters. 2016;18(8):1816-1819.
Ding C, Cicuttini F. IDrugs. 2003;6(8):802-808.
Gu J-X, Holland HL. Synthetic communications. 1998;28(18):3305-3315.
Allgäuer DS, Mayer P, Mayr H. Journal of the American Chemical Society. 2013;135(40):15216-15224.
Allgäuer DS, Mayr H. European Journal of Organic Chemistry. 2013;2013(28):6379-6388.
14

2
2
2
2
2
2
0.
1.
2.
3.
4.
5.
Rubia LB, Gomez R. Journal of pharmaceutical sciences. 1977;66(11):1656-1657.
Khurana JM, Kukreja G. Synthetic communications. 2002;32(8):1265-1269.
Friedman M. Canadian Journal of Chemistry. 1967;45(19):2271-2275.
Waldi D. In: Thin-layer chromatography. Springer; 1965: 483-502.
Veber DF, Johnson SR, Cheng H-Y, Smith BR, Ward KW, Kopple KD. Journal of medicinal chemistry. 2002;45(12):2615-2623.
Kurumbail RG, Stevens AM, Gierse JK, McDonald JJ, Stegeman RA, Pak JY, Gildehaus D, Miyashiro JM, Penning TD, Seibert K et
al. Nature. 1996;384(6610):644-648.
2
2
6.
7.
Sidhu RS, Lee JY, Yuan C, Smith WL. Biochemistry. 2010;49(33):7069-7079.
Gilbert NC, Rui Z, Neau DB, Waight MT, Bartlett SG, Boeglin WE, Brash AR, Newcomer ME. FASEB journal : official publication
of the Federation of American Societies for Experimental Biology. 2012;26(8):3222-3229.
Sherman W, Day T, Jacobson MP, Friesner RA, Farid R. Journal of medicinal chemistry. 2006;49(2):534-553.
Vinegar R, Schreiber W, Hugo R. Journal of pharmacology and experimental therapeutics. 1969;166(1):96-103.
Tsurumi K, Kyuki K, Niwa M, Kokuba S, Fujimura H. Arzneimittel-Forschung. 1986;36(12):1796-1800.
Pradelles P, Grassi J, Maclouf J. Analytical Chemistry. 1985;57(7):1170-1173.
Taraporewala IB, Kauffman JM. Journal of pharmaceutical sciences. 1990;79(2):173-178.
Srivastava P, Vyas VK, Variya B, Patel P, Qureshi G, Ghate M. Bioorganic Chemistry. 2016;67:130-138.
Michaelis L, Menten ML. Biochem z. 1913;49(333-369):352.
Winter CA, Risley EA, Nuss GW. Experimental Biology and Medicine. 1962;111(3):544-547.
DiMartino M, Campbell Jr G, Wolff C, Hanna N. Agents and Actions. 1987;21(3-4):303-305.
Banerjee AG, Das N, Shengule SA, Srivastava RS, Shrivastava SK. European journal of medicinal chemistry. 2015;101:81-95.
Szelenyi I, Thiemer K. Archives of Toxicology. 1978;41(1):99-105.
2
2
3
3
3
3
3
3
3
3
3
3
8.
9.
0.
1.
2.
3.
4.
5.
6.
7.
8.
9.
Lineweaver H, Burk D. Journal of the American Chemical Society. 1934;56(3):658-666.
15