New indomethacin analogs as selective COX-2 inhibitors: Synthesis, COX-1/2 inhibitory activity, anti-inflammatory, ulcerogenicity, histopathological, and docking studies

Article abstract of DOI:10.1002/ardp.202000328

New indomethacin analogs 4a–g, 5, 6, 8a, and 8b were synthesized to overcome the nonselectivity and ulcer liability of indomethacin. All newly synthesized compounds were more potent against cyclooxygenase 2 (COX-2; IC₅₀ value range: 0.09–0.4 μМ) as?compared with celecoxib (IC₅₀ = 0.89 μМ). Compounds 4a, 4b, 4d, 5, and 6 showed the highest COX-2 selectivity index (SI range = 4.07–6.33) as compared with indomethacin (SI = 1.14) and celecoxib (SI = 3.52). Additionally, 4a, 4b, 4d, 5, and 7 showed good anti-inflammatory activity with edema inhibition (79.36–88.8%), relative to celecoxib (78.96%) and indomethacin (90.43%), after 5 h. Also, ulcerogenic effects and histopathological examination were assessed for the most potent analogs,?4b, 4d, 5, and 6, to determine their safety. The results can shed light on indomethacin analog?5 as a remarkable anti-inflammatory lead compound with a good safety profile (ulcer index = 10.62) close to the nonulcerogenic drug celecoxib (ulcer index = 10.53) and better than indomethacin (ulcer index = 18.50). Docking studies were performed in the COX-2 active site for the most active compounds, to test their selectivity and to confirm their mechanism of action.

Full text of DOI:10.1002/ardp.202000328

Received: 29 August 2020
Revised: 14 November 2020
Accepted: 21 November 2020
|
|
DOI: 10.1002/ardp.202000328
F U L L P A P E R
New indomethacin analogs as selective COX‐2 inhibitors:
Synthesis, COX‐1/2 inhibitory activity, anti‐inflammatory,
ulcerogenicity, histopathological, and docking studies
Khaled R. A. Abdellatif1,2 | Eman K. A. Abdelall1 | Heba A. H. Elshemy1
|
El‐Shaymaa El‐Nahass3 | Maha M. Abdel‐Fattah4 | Yasmin Y. M. Abdelgawad1
1Department of Pharmaceutical Organic
Chemistry, Faculty of Pharmacy, Beni‐Suef
University, Beni‐Suef, Egypt
2Department of Pharmaceutical Science, Ibn
Sina National College for Medical Studies,
Jeddah, Kingdom of Saudi Arabia
Abstract
New indomethacin analogs 4a–g, 5, 6, 8a, and 8b were synthesized to overcome the
nonselectivity and ulcer liability of indomethacin. All newly synthesized compounds
were more potent against cyclooxygenase 2 (COX‐2; IC50 value range: 0.09–0.4 μМ)
as compared with celecoxib (IC50 = 0.89 μМ). Compounds 4a, 4b, 4d, 5, and 6
showed the highest COX‐2 selectivity index (SI range = 4.07–6.33) as compared
with indomethacin (SI = 1.14) and celecoxib (SI = 3.52). Additionally, 4a, 4b, 4d, 5,
and 7 showed good anti‐inflammatory activity with edema inhibition (79.36–88.8%),
relative to celecoxib (78.96%) and indomethacin (90.43%), after 5 h. Also, ulcero-
genic effects and histopathological examination were assessed for the most potent
analogs, 4b, 4d, 5, and 6, to determine their safety. The results can shed light on
indomethacin analog 5 as a remarkable anti‐inflammatory lead compound with a
good safety profile (ulcer index = 10.62) close to the nonulcerogenic drug celecoxib
(ulcer index = 10.53) and better than indomethacin (ulcer index = 18.50). Docking
studies were performed in the COX‐2 active site for the most active compounds, to
test their selectivity and to confirm their mechanism of action.
3Department of Pathology, Faculty of
Veterinary Medicine, Beni‐Suef University,
Beni‐Suef, Egypt
4Department of Pharmacology, Faculty of
Pharmacy, Beni‐Suef University, Beni‐Suef,
Egypt
Correspondence
Heba A. H. Elshemy, Department of
Pharmaceutical Organic Chemistry, Faculty of
Pharmacy, Beni‐Suef University, Beni‐Suef
62514, Egypt.
Email: ha_elshemy@yahoo.com
K E Y W O R D S
anti‐inflammatory activity, COX‐1, COX‐2, histopathology, indomethacin
|
1
INTRODUCTION
The structures of both COX‐1 and COX‐2 are closely similar to each
other. However, the COX‐2 active site is slightly larger, because
Nonsteroidal anti‐inflammatory drugs (NSAIDs) produce their phar-
macological effect through inhibiting the action of cyclooxygenase
enzymes (COX‐1 and COX‐2),[1] thus preventing the catalytic
biotransformation of arachidonic acid into the corresponding proin-
flammatory prostaglandins and thromboxanes (TXs).[2] It was
established that there are two distinct isoforms for cyclooxygenase
enzyme: COX‐1 isoenzyme, which is the constitutive form produced in
different tissues and responsible for maintenance of normal physio-
logical functions such as gastric cytoprotection, platelet aggregation,
vascular homeostasis[3]; and COX‐2 isoenzyme, which is the induced
form responsible for fever, pain, and other inflammatory symptoms.[4]
it contains a secondary internal side pocket that could accommodate
bigger structures and also its central channel is wider by
approximately 17%.[5]
The long‐term use of traditional nonsteroidal anti‐inflammatory
drugs, like aspirin, ibuprofen, indomethacin (1), and naproxen (I) as
free carboxylate derivatives, could produce gastroduodenal mucosal
injury, renal failure, and bleeding.[6] The ulcerogenicity for traditional
NSAIDs was attributed to the direct effect of the COOH moiety and
indirectly to nonselective inhibition of both COX enzymes. There-
fore, the researchers synthesized many amide derivatives of these
traditional NSAIDs that lacked a free COOH moiety. For example,
|
Arch Pharm. 2020;e2000328.
wileyonlinelibrary.com/journal/ardp © 2020 Deutsche Pharmazeutische Gesellschaft
1 of 13
https://doi.org/10.1002/ardp.202000328

2 of 13
ABDELLATIF ET AL.
|
transformation of the acidic COOH group in indomethacin and for-
mation of amide linkage resulted in producing compound II
(Figure 1), which was 10 times more potent as a COX‐2 inhibitor
derivatives 8a and 8b (Figure 2). The synthesized derivatives are free
of the acidic character responsible for the direct gastroduodenal
ulcer. Also, their bulkiness was increased, causing them to be too
large to fit the small COX‐1 active site, increasing the selectivity for
the COX‐2 enzyme,[22] and consequently decreasing the gastro-
duodenal ulcer. The synthesized derivatives were evaluated for their
COX‐1/COX‐2 inhibitory activity. Furthermore, the carrageenan‐
induced rat paw edema model and histopathological study were
accomplished to evaluate their anti‐inflammatory activity and their
gastric safety. The compounds were docked into the COX‐2 active
site to reveal their possible mechanism of action. The handling of
animals or tissues was preceded according to the institutional ethical
committee instructions.
than
fluorocoxibs.[7]
Also,
naproxen
derivatives
(III–VI)
(Figure 1)[8–10] were synthesized in which the ulcer effect was re-
duced as compared with the standard drug. Furthermore, selective
COX‐2 inhibitors (coxibs) as celecoxib, rofecoxib, and valdecoxib
were developed to overcome the nonselective drawbacks, but un-
fortunately, rofecoxib and valdecoxib were found to cause myo-
cardial infarction and high blood pressure; therefore, their clinical
use was terminated.[11,12] Regardless of the nonselectivity of the
traditional NSAIDs, they did not pose
a cardiac risk as
coxibs.[13] Researchers have reconsidered the use of NSAIDs in a
way aiming to improve their anti‐inflammatory activity and decrease
their ulcerogenic effect. According to these findings and as a con-
tinuation of our work on the development of anti‐inflammatory
drugs with a good safety profile,[14–21] we now report the synthesis
of new amide indomethacin derivatives in which the acidic COOH
group in position 3 of indomethacin was replaced with different
groups to afford different thiourea derivatives 4a–g, 2‐cyanoacetyl
derivative 5, 2‐pyridinone derivative 6, and 3‐cyanopyridone
|
2
RESULTS AND DISCUSSION
Chemistry
|
2.1
To synthesize the new indomethacin analogs, indomethacin 1 was
converted to the acid chloride 2 through reaction with thionyl
FIGURE 1 Chemical structures of traditional NSAIDs (indomethacin 1 and naproxen I) and some reported modified indomethacin and
naproxen amide derivatives (II–VI)[7–10] as selective COX‐2 inhibitors

ABDELLATIF ET AL.
3 of 13
|
FIGURE 2 The strategy for the design
of the new indomethacin analogs 4–8
chloride, as reported.[23] The active acid chloride 2 was reacted with
ammonium thiocyanate to give the intermediate indomethacin iso-
thiocyanate 3, which was not isolated and heated under reflux con-
dition with various aromatic amines to yield a series of new
thiourea–indomethacin hybrid analogs 4a–g (Scheme 1). However,
compound 2 was reacted with cyanoacetic acid hydrazide[24] to
afford 2‐(1‐(4‐chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl)‐N'‐
(2‐cyanoacetyl)acetohydrazide 5 (Scheme 2), which upon reaction
with acetylacetone furnished compound 6. The cyclized product 6
was postulated to be formed via a nucleophilic attack of 5 on acet-
ylacetone to form a Michael adduct, which spontaneously was cy-
clized by losing a molecule of water, giving 6. Finally, compound 5
was heated under reflux with the appropriate arylidenes 7a and
7b[25,26] in the presence of piperidine to afford the respective pyr-
idine derivatives 8a and 8b (Scheme 2). The infrared (IR) spectrum of
compounds 4a–g showed absorption bands at 3170.97– 3618.46,
referring to the NH group, and bands at 1666.50–1743.65, indicating
the presence of the C═O group. The 1H NMR (nuclear magnetic
resonance) spectrum revealed two singlet signals at
δ
10.35–12.38 ppm for two NH groups that are exchangeable with
D2O. The IR spectrum of compound 5 showed an absorption band at
3197.98, indicating the presence of two NH groups, and another
band at 2256.71 for the C≡N group. The 1H NMR analysis showed a
signal of two protons at δ 3.54 ppm, referring to CH2–CN, a singlet
signal at δ 10.28 ppm, indicating the presence of two NH groups
exchangeable with D2O. The 1H NMR spectrum of compound 6
SCHEME 1 The synthesis of compounds 4a–g. Reagents and conditions: (a) Thionyl chloride, benzene, reflux 2 h; (b) ammonium thiocyanate,
reflux 2 h; (c) suitable amine, reflux 3 h

4 of 13
ABDELLATIF ET AL.
|
SCHEME 2 The synthesis of compounds 5, 6, 8a, and 8b. Reagents and conditions: (a) Cyanoacetic acid hydrazide, benzene, RT 24 h; (b)
acetylacetone, absolute ethanol, piperidine, reflux 6 h; (c) absolute ethanol, piperidine, reflux 2 h
showed two singlet signals at δ 2.16 and 2.27 ppm, indicating the
presence of two CH3 groups of the oxopyridine ring. Unexpectedly,
during the reaction of 5 with 7a and 7b, the piperidine replaced the
Cl atom of the para‐chlorobenzoyl moiety of 5, which was confirmed
by the presence of a multiplet signal at δ 1.55–1.64 ppm and a triplet
signal at δ 2.99–3.01 ppm in the 1H NMR spectra and the presence of
three signals in 13C NMR spectra at δ 22.07–22.17 ppm (piperidine
C‐4), 22.69–22.83 ppm (piperidine C‐3, C‐5), and 44.26–44.34 ppm
(piperidine C‐2, C‐6).
4c showed a good SI = 3.33, which is closer to that of celecoxib
(SI = 3.52), whereas compounds 4e, 4f, 4g, 8a, and 8b showed a
moderate SI range (2.11–3.22). However, compounds 4a, 4b, 4d, 5,
and 6 showed the highest SI range (4.07–6.33), which appeared to
be more selective than celecoxib (SI = 3.52) and indomethacin
(SI = 1.14). After determination of SIs for all prepared com-
pounds, 4a–g, 5, 6, 8a, and 8b, the anti‐inflammatory (AI) activity
TABLE 1 In vitro IC50 values of COX‐1 and COX‐2 and selectivity
indices (SI) of the tested compounds, indomethacin, and celecoxib
|
2.2
Pharmacology/biology
Compounds
COX‐1 IC50 (μМ)a COX‐2 IC50 (μМ)a COX‐2 SIb
4a
4b
4c
4d
4e
4f
4g
5
1.10
0.57
0.30
0.63
0.29
0.27
0.22
1.82
1.54
0.39
0.19
0.27
0.09
0.09
0.11
0.09
0.09
0.09
0.4
4.07
6.33
3.33
5.72
3.22
3
The target indomethacin analogs were assayed against COX‐1 and
COX‐2 using N, N, N′,N′‐tetramethyl‐p‐phenylenediamine at 590 nm
and the enzyme immunoassay (EIA) kit.[27] The potency of the syn-
thesized compounds was determined as the concentration causing
50% inhibition (IC50). Also, the COX‐2 selectivity indexes (SI) were
calculated as IC50 (COX‐1)/IC50 (COX‐2) and then compared with the
SI of the standard drugs indomethacin as a nonselective COX
inhibitor and celecoxib as a selective COX‐2 inhibitor. The results
obtained from the evaluation of all compounds, 4a–g, 5, 6, 8a, and 8b,
are listed in Table 1.
2.44
4.55
4.96
2.6
6
0.31
0.15
0.09
0.087
0.89
All compounds, 4a–g, 5, 6, 8a, and 8b, inhibited the COX‐1
isozyme at higher doses with an IC50 range of 0.19–1.82 μМ. This
range is more than that of indomethacin (IC50 = 0.10 μМ), which
means that the prepared derivatives have lower potency against
COX‐1 and consequently would be safer. Regarding COX‐2, all
compounds, 4a–g, 5, 6, 8a, and 8b, showed an IC50 inhibitory range
of 0.08–0.4 μМ, which is more potent than celecoxib itself
(IC50 = 0.89 μМ). The SI calculation result showed that compound
8a
8b
2.11
1.14
3.52
Indomethacin 0.10
Celecoxib 3.14
aIC50: The concentration causing 50% COX inhibition.
bSelectivity index (IC50 COX‐1/IC50 COX‐2).

ABDELLATIF ET AL.
5 of 13
|
TABLE 2 In vivo anti‐inflammatory activity of the tested
compounds 4a, 4b, 4d, 5, and 6, indomethacin, and celecoxib against
carrageenan‐induced rat paw edema
safe drug, as it showed the highest ulcer index (UI = 24.87), which
exceeded that of indomethacin and celecoxib.
Examination of microscopical lesions was carried out to evaluate
the ulcerogenic effect of the in vivo active compounds on the rat's
stomach including both glandular and nonglandular portions. Also,
the severity of these lesions was compared with that induced by
reference drugs indomethacin, celecoxib, and the control negative
group (Table 4).
% of
% of inhibition
at 1 h
% of inhibition
at 3 h
inhibition
at 5 h
Compounds
4a
30.48 5.11b
21.36 4.64
76.76 6.07a,b
38.8 3.32a,b
23.57 3.2
73.03 7.1b
53.77 5.25
81.06 3.78a,b
74.47 11.19b
43.88 5.22
61.38 1.08
47.7 2.82
79.56 7.3
88.52 0.47
88.8 2.7
4b
In the control negative group, a normal histological structure of
both glandular and nonglandular stomach was observed. The gland-
ular portion showed normal mucosal lining, submucosa, and mucosal
layers (Figure 3a‐i), associated with the normal histological structure
of the nonglandular portion (Figure 3a‐ii). For the indomethacin
group, severe pathological lesions were observed in the form of
degenerative changes and necrotic changes in the glandular and
nonglandular stomach. The glandular stomach exhibited erosive and
ulcerative changes with the presence of lymphocytic infiltration
and congestive blood vessels accompanied by edema in the sub-
mucosal layer. Hyalinosis of the muscular layer and diffuse leukocytic
infiltration are observed (Figure 3b‐i). The nonglandular stomach
exhibited focal erosive and ulcerative lesions and hyperkeratosis
(Figure 3b‐ii). However, in the celecoxib group, mild lesions could be
detected in the glandular stomach. These lesions were in the form of
degenerative changes of the mucosal lining and mild leukocytic
infiltration in the submucosal layer (Figure 3c‐i). The nonglandular
stomach portion showed mild hyperkeratosis (Figure 3c‐ii).
4d
5
81.16 2.45
79.36 2.34
90.43 4.88
78.96 2.42
6
Indomethacin
Celecoxib
25.9 1.12
18.37 1.17
Note: Data are expressed as mean SEM (n = 6 rats). Statistical analysis
was carried out using the one‐way analysis of variance test, followed by
the least significant difference post‐hoc test.
a
Significantly different from the indomethacin group at p < .05.
b
Significantly different from the celecoxib group at p < .05.
of the most selective ones, 4a, 4b, 4d, 5, and 6, against COX‐2 was
evaluated using the carrageenan‐induced rat paw edema
method.[28] The AI activity was calculated at 1, 3, and 5 h after
carrageenan injection, as presented in Table 2.
It was noted that all compounds significantly decreased in-
flammation as compared with carrageenan at all time intervals. A
comparative study of the anti‐inflammatory activity of the test
compounds, relative to indomethacin and celecoxib as reference
drugs, at the different time intervals, showed that after 1 h, they had
poor to good anti‐inflammatory activity (AI = 21.36–76.76%) in
comparison with indomethacin (AI = 25.9%) and celecoxib (AI =
18.37%). After 3 and 5 h, the anti‐inflammatory activity was ex-
tremely increased and the target compounds showed
AI = 43.88–81.06% and 79.36–88.8%, respectively, in comparison
with indomethacin (AI = 61.38% and 90.43%, respectively) and cel-
ecoxib (AI = 47.7% and 78.96%, respectively). The methoxy deriva-
tive 4d showed the highest AI activities at all time intervals (after 1 h,
AI = 76.76%, after 3 h, AI = 81.06%, and after 5 h, AI = 88.8%). The
ulcerogenic effect of the most COX‐2 selective compounds, 4b, 4d, 5,
and 6, which also showed a high percentage of edema inhibition, was
evaluated according to a previous study (Table 3).[29]
Regarding 4b and 4d, moderate pathological lesions in the form
of multifocal areas of degeneration in the lining epithelium were
observed. These lesions were associated with moderate necrotic
changes. Also, moderate‐to‐severe submucosal congestion and leu-
kocytic infiltration were found to be associated with mild hyalinosis
in the muscular layer (Figure 4: 4b‐I and 4d‐I), and mild‐to‐moderate
hyperkeratosis of the nonglandular stomach could be found without
the presence of any erosive or ulcerative lesions (Figure 4: 4b‐II and
4d‐II). Administrations of 5 showed mild degenerative changes of the
glandular mucosal lining, mild submucosal lymphocytic infiltration,
and minimal necrotic changes (Figure 3: 5‐I), and the more or less
TABLE 3 Ulcerogenic liability for the most active derivatives, 4b,
4d, 5, and 6, and reference drugs indomethacin and celecoxib
Average
% of
number of
ulcers
Average
severity
Ulcer
index
The normal control group received only vehicles, the celecoxib
group received celecoxib in a dose of 50 mg/kg, the indomethacin
group received the drug at a dose of 50 mg/kg, and the other
four groups received the prepared compounds 4b, 4d, 5, and 6,
respectively, in a dose of 50 mg/kg for each. Compound 5 was the
safest within all derivatives and showed the least ulcer index (UI =
10.62) when compared with indomethacin (UI = 18.50) and it was
close to that of the nonulcerogenic reference drug celecoxib (UI =
10.53), whereas compounds 4b and 4d showed a lower ulcerogenic
activity with ulcer index 14 and 17.25, respectively, as compared
with indomethacin (UI = 18.50). However, compound 7 was the least
Compounds
incidence
4b
100
100
75
3
1
14
4d
6.25
2
1
17.25
10.62
24.87
8.90
5
1.12
1.12
1
6
100
66
13.75
1.30
6.30
1.75
Control
Indomethacin 100
Celecoxib 75
2.20
1.28
18.50
10.53

6 of 13
ABDELLATIF ET AL.
|
TABLE 4 Scoring of different
pathological lesions caused by the tested
compounds (4b, 4d, 5, 6) on both glandular
and nonglandular portions of the
stomach as compared with those induced
by celecoxib and indomethacin as standard
drugs
Lesion
4b
4d
5
6
Celecoxib
Indomethacin
Negative
Glandular stomach
Mucosa
Degenerative changes
Nuclear pyknosis
Erosion
+
++
+
++
−/+
+
+++
+++
++
++
+
+++
+++
+++
+++
−/+
−
+
++
+
++
+
+
−
Ulcer
−
+++
−
−
Submucosa
Congestion
++
++
+
++
++
+
+
+++
+++
+++
++
+
+++
+++
+++
−
−
−
Leukocytic infiltration
Edema
+
−/+
+
Musculosa
Degenerative changes
Hyalinosis
+
−
−
+
+
−
+
+
−
+++
++
+
+
+
+++
+++
+++
−
−
−
Leukocytic infiltration
++
Nonglandular stomach
Erosion
−
−
+
−
−
+
−
−
−
+
−
−
+
++
−
−
−
Ulcer
−
−/+
+++
Hyperkeratosis
++
Abbreviations: −/+, minimal, +, mild, ++, moderate, +++, severe.
|
Molecular docking
normal histological structure could be found in the nonglandular
stomach (Figure 3: 5‐II). Finally, administrations of 6 revealed the
presence of severe degenerative changes and necrotic changes of
the glandular stomach (Figure 4: 6‐I). Additionally, massive lympho-
cytic infiltration and congestion of the submucosal and muscular
layers were observed. The latter showed moderate hyalinosis,
moderate hyperkeratosis, and erosive and ulcerative changes, which
could be detected in the glandular layer. The nonglandular portion
showed focal areas of hyperkeratosis (Figure 4: 6‐II).
2.3
A docking study was performed for the most COX‐2 selective com-
pounds, 4a, 4b, 4d, 5, and 6, using the MOE modeling version
(2015.10) to illustrate the possible conformations that bound to the
COX‐2 receptor, which may reveal the possible mechanism of action
for their anti‐inflammatory activity toward the COX‐2 active site.
The X‐ray coordinates were downloaded from the Protein Data Bank
with code (PDB: 6BL4).[10] The dansyl moiety of the ligand was
FIGURE 3 Histopathological alterations of the glandular stomach (first row) and nonglandular stomach (second row) in the control negative
group (a‐i, a‐ii), indomethacin (b‐i, b‐ii), celecoxib (c‐i, c‐ii), and compound 5 (5‐I, 5‐II)

ABDELLATIF ET AL.
7 of 13
|
FIGURE 4 Histopathological alterations of the glandular stomach (first row, I) and nonglandular stomach (second row, II) in compound 6
(6‐I, 6‐II), compound 4b (4b‐I, 4b‐II), and compound 4d (4d‐I, 4d‐II). C, congestion; D, degenerative changes; E, erosion; Ed, edema; H, hyalinosis;
K, hyperkeratosis; L, leukocytic infiltration; U, ulcer
recognized to bind in the lobby region of the COX‐2 active site. It
was observed that a hydrogen bond was formed through the car-
bonyl oxygen between the phenyl and indole ring and Ser530 at a
distance of 3.0 Å.[10] Tyr355 formed a hydrogen bond with the
nitrogen atom of the amide at a distance of 2.8 Å. Arg120 formed a
hydrogen bond with the carbonyl oxygen of the amide. The binding
mode of the indomethacin–ethylenediamine–dansyl conjugate, a
highly COX‐2 selective inhibitor, inside the active site of the COX‐2
isoenzyme is shown in Figure 5.
COX‐2 active site with a binding energy score range of −12.82 to
−16.08 kcal/mol, which is close to the ligand (−17.43 kcal/mol). It was
observed that compound 4b formed two hydrogen bond interac-
tions: One of them was the interaction of C═O linked to the indole
ring with Val116 amino acid with a distance of 3.20 Å and the other
between C═O of the amide linkage and Arg120 amino acid at a
distance of 2.97 Å with an energy score of −16.08 kcal/mol (Figure 6).
Also, compound 4d formed one hydrogen bond interaction of
benzoyl C═O with Ser530 amino acid with a distance of 2.95 Å with
an energy score of −15.56 kcal/mol. Fortunately, compound 5 formed
three hydrogen bonds of NH, amide C═O, and N of C≡N with
Glu524, Arg120, and Lys83 at a distance of 2.88, 2.74, and 3.51 Å,
respectively, with an energy score of −14.79 kcal/mol. All these in-
teractions in addition to the presence of the C≡N group proved the
The ligand formed two hydrogen bonds of C═O linked to the
indole ring and C═O of the amide group with Arg120 and Ser530
amino acids with a distance of 3.00 and 2.91 Å, respectively. The
energy score for the ligand inside the COX‐2 isoform was 17.43 kcal/
mol. The most active compounds 4a, 4b, 4d, 5, and 6 fitted within the
FIGURE 5 Binding of the ligand (indomethacin–ethylenediamine–dansyl conjugate) inside the COX‐2 active site. (a) Two‐dimensional
interaction of the proposed binding mode of the ligand inside the active site of COX‐2, which forms two H‐bonds with Arg120 and Ser530
amino acids; (b) three‐dimensional interaction of the ligand

8 of 13
ABDELLATIF ET AL.
|
FIGURE 6 Binding of one of the most active compounds, 4b, inside the COX‐2 active site. (a) Two‐dimensional interaction of the proposed
binding mode of 4b, which forms two hydrogen bonds with Val116 and Arg120 amino acids; (b) three‐dimensional interaction of 4b (cyan)
high selectivity of compound 5 toward the COX‐2 isoenzyme
(Figure 7).
Thiourea derivatives 4a–g have higher selectivity than in-
domethacin toward the COX‐2 enzyme. From these derivatives,
compound 4a bearing the phenyl group has a higher SI and anti‐
inflammatory activity than celecoxib; however, the presence of an
electron‐donating group (CH3 (4b) and OCH3 (4d)) increases the SI
and gives the highest anti‐inflammatory activity at all time intervals
and also decreases the ulcerogenic activity. On the contrary, the
introduction of an electron‐withdrawing group (F (4c) and Cl (4e),
COOH (4f), COOC2H5 (4g)) decreases the SI.
Compound 6 showed two hydrogen bonds with Ser530 amino
acid with a distance of 2.93 and 3.39 Å, respectively. Also, compound
4a showed one hydrogen bond of C═S with Arg120 amino acid with
a distance of 3.21 and formed a hydrophobic interaction with Tyr115
with a distance of 4.10 Å. Docking scores, number of hydrogen
bonds, amino acid residues responsible for hydrogen bond formation,
distances, and functional groups are listed in Table 5.
Furthermore, 2‐cyanoacetyl derivative 5 has a higher SI than
celecoxib and promising anti‐inflammatory activity with the least
ulcer index (UI = 10.62), compared with indomethacin (UI = 18.50),
which is very close to that of the nonulcerogenic reference drug
celecoxib (UI = 10.53). For compound 6, the presence of 3‐cyano‐4,6‐
dimethyl‐2‐oxopyridine moiety increases both the anti‐inflammatory
activity and the selectivity toward the COX‐2 enzyme as compared
|
2.4
Structure–activity correlation
Four series of indomethacin amide derivatives 4–8 were synthesized and
evaluated as selective COX‐2 inhibitors. It was found that the amidation
of indomethacin increases the anti‐inflammatory activity and selectivity
toward the COX‐2 enzyme and diminishes the ulcerogenic effect.
to indomethacin and celecoxib. Finally, the introduction of
a
FIGURE 7 Binding of one of the most active compounds, 5, inside the COX‐2 active site. (a) Two‐dimensional interaction of the proposed
binding mode of 5, which forms three hydrogen bonds with Glu524, Arg120, and Lys83 amino acids; (b) three‐dimensional interaction of
5 (magenta)

ABDELLATIF ET AL.
9 of 13
|
TABLE 5 The molecular docking scores and binding interaction of compounds 4a, 4b, 4d, 5, 6, and 8b, and the indomethacin–dansyl
conjugate (ligand) inside the COX‐2 active site
E‐score
(kcal/mol)
Number of
Hydrogen
Functional
groups
Compounds
hydrogen bonds
bonding residues
Distance (Å)
4a
4b
−12.82
−16.08
1
1
Arg120
3.21
C═S
Val116
Arg120
3.20
2.97
C═O indole
C═O amide
4d
5
−15.56
−14.79
1
3
Ser530
2.95
C═O amide
Glu524
Arg120
Lys83
2.88
2.74
3.51
NH
C═O amide
C≡N
6
−13.93
−17.43
−11.27
−11.52
2
2
2
4
Ser530
Ser530
2.93
3.39
O
C≡N
Indomethacin–dansyl
Arg120
Ser530
3.00
2.91
C═O amide
C═O indole
conjugate (ligand)
Indomethacin
Arg120
Arg513
2.71
3.14
C═O indole
C═O COOH
Celecoxib
Tyr355
Arg513
Arg513
Arg513
3.10
3.11
3.03
2.81
S═O
S═O
S═O
SO2NH2
6‐amino‐3,5‐dicyano‐2‐oxo‐4‐phenylpyridine moiety at position 3 in
compounds 8a,b does not improve the SI as compared with celecoxib.
anti‐inflammatory agents as selective COX‐2 inhibitors with minimal
adverse effects on the gastric mucosa.
|
|
3
CONCLUSION
4
EXPERIMENTAL
Chemistry
|
The goal of our study was to design a novel series of amide in-
domethacin derivatives as selective COX‐2 inhibitors with minimal
adverse effects on the gastric mucosa than indomethacin (one of the
most ulcerogenic NSAIDs). Eleven indomethacin analogs, 4a–g, 5, 6,
8a, and 8b, were synthesized and evaluated for their COX inhibitory
activity, anti‐inflammatory activity, and ulcerogenic liability. For
these derivatives, the COOH group of indomethacin was replaced
with bulky moieties via an amide linkage. All synthesized compounds
showed 1.85‐ to 5.5‐fold higher COX‐2 selectivity index values than
indomethacin, and most of them were even more COX‐2 selective
than the COX‐2 selective reference drug celecoxib. Four derivatives,
4b, 4d, 5, and 6, showed good anti‐inflammatory activity with a
smaller number of ulcers than indomethacin. The histopathological
study revealed that compound 5 was the best drug, showing mild
lesions in the stomach, whereas compounds 4b and 4d showed
moderate lesions. Molecular docking revealed that the selectivity
could be attributed to the presence of C═O linked with the indole
ring and C═O of the amide group that formed hydrogen bonds with
Ser530 and Arg120 amino acids. Finally, we can conclude that
modifications operated on the indomethacin structure to yield
these novel derivatives achieved our main target to prepare
4.1
|
4.1.1
General
Melting points were determined by using the Griffin apparatus and
were not corrected. IR spectra were recorded with a Shimadzu
IR‐435 spectrophotometer by using KBr disks and the data were
represented in cm−1 1H NMR spectra (at 400 MHz) and 13C NMR
.
spectra (at 100 MHz) were carried out by using a Bruker spectro-
photometer at the Faculty of Pharmacy (Beni‐Suef University,
Egypt) in deuterated dimethyl sulfoxide (DMSO‑d6) and D2O with
tetramethylsilane as an internal standard. The coupling constant (J)
values were estimated in Hertz (Hz), whereas the chemical shift was
recorded in ppm on the δ scale. Splitting patterns are designated as
follows: s, singlet; d, doublet; t, triplet; m, multiplet; dd, doublet of
doublet. Mass spectra (MS) were determined by using a Hewlett
Packard 5988 spectrophotometer. Microanalyses for C, H, and N
(within 0.4% of the theoretical value) were carried out at the
Regional Center for Mycology and Biotechnology (Al‐Azhar
University, Egypt). All reactions were observed with thin‐layer
chromatography using a UV lamp. All other reagents and solvents

10 of 13
ABDELLATIF ET AL.
|
were purchased from Aldrich Chemical Company and used without
further purification. Indomethacin was purchased from Acros Che-
mical Company. A brown crystalline product 2 was obtained, as
reported.[23,30]
C27H24ClN3O3S: C, 64.09; H, 4.78; N, 8.30. Found: C, 63.79; H, 4.68;
N, 8.10.
2‐[1‐(4‐Chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl]‐N‐(4‐
fluorophenyl‐carbamothioyl)acetamide (4c)
The InChI codes of the investigated compounds, together with
some biological activity data, are provided as Supporting
Information.
Brown powder; yield, 53%; mp: 170−172°C; IR (KBr, cm−1): 3271.27
(2NH), 3047.53 (C–H aromatic), 2993.52, 2931.80 (C–H aliphatic),
1743.65, 1666.50 (C═O), and 1149.57 (C═S); 1H NMR (DMSO‐d6):
δH 2.29 (s, 3H, CH3), 3.75 (s, 5H, CH2, OCH3), 6.72 (dd, J = 8.8 Hz,
2 Hz, 1H, indole H‐6), 6.94 (d, J = 8.8 Hz, 1H, indole H‐7), 7.12–7.19
(m, 3H, indole H‐4, fluorobenzene H‐3, H‐5), 7.60–7.71 (m, 6H,
fluorobenzene H‐2, H‐6, benzoyl H‐2, H‐3, H‐5, H‐6), and 10.26
(s, 1H, NH exchange with D2O); 13C NMR (DMSO‐d6): δC 13.89
(CH3), 32.38 (CH2), 55.89 (OCH3), 102.38, 111.64, 114.47, 115.07,
115.67, 115.89 121.43, 121.51, 129.54, 130.71, 131.33, 131.65,
134.65, 135.94, 138.07, 156.02, 157.27, 159.65, 168.35, and 168.87;
MS (m/z, %): 509.98 (M+, 3.37%) and 111.11 (100%); Anal. calcd for
|
4.1.2
General procedure for the synthesis of
compounds 4a–g
To a solution of 2 (0.01 mol, 3.80 g) in benzene (30 ml), ammonium
thiocyanate (0.01 mol, 0.60 g) was added. The reaction mixture was
refluxed for 2 h, cooled, and filtered. Then suitable aromatic amine
(0.01 mol) was added to the filtrate. The mixture was heated under
reflux for 3 h and then cooled and filtered. Next, the solid product
was washed and crystallized from methanol to afford compounds
4a–g. The physical and spectral data are listed below.
C26H21ClFN3O3S: C, 61.23; H, 4.15; N, 8.24. Found: C, 61.25; H, 4.11;
N, 8.14.
2‐[1‐(4‐Chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl]‐N‐
(phenyl‐carbamothioyl)acetamide (4a)
2‐[1‐(4‐Chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl]‐N‐(4‐
methoxyphenyl‐carbamothioyl)acetamide (4d)
Buff powder; yield, 75%; mp: 170–172°C; IR (KBr, cm−1): 3178.69
(NH), 3032.10 (C–H aromatic), 2997.38, 2951.09, 2924.09 (C–H
aliphatic), 1689.64 (C═O), and 1149.57 (C═S); 1H NMR (DMSO‐
d6): δH 2.29 (s, 3H, CH3), 3.78 (s, 3H, OCH3), 3.97 (s, 2H, CH2),
6.73 (dd, J = 8.8 Hz, 2 Hz, 1H, indole H‐6), 6.92 (d, J = 8.8 Hz, 1H,
indole H‐7), 7.21–7.26 (m, 2H, indole H‐4, phenyl H‐4), 7.38
(t, J = 8 Hz, 2H, phenyl H‐3, H‐5), 7.60 (d, J = 8 Hz, 2H, phenyl H‐2,
H‐6), 7.67 (qAB, JAB = 8.4 Hz, 4H, benzoyl H‐2, H‐3, H‐5, H‐6),
11.76 (s, 1H, NH exchange with D2O), and 12.38 (s, 1H, NH ex-
change with D2O); 13C NMR (DMSO‐d6): δC 13.80 (CH3), 31.72
(CH2), 55.97 (OCH3), 102.44, 111.82, 113.02, 115.06, 124.68,
126.82, 129.14, 129.57, 130.69, 131.16, 131.66, 134.50, 136.43,
138.16, 138.20, 156.10, 168.38, 173.00, and 179.21; MS (m/z, %):
491.99 (M+, 52.89%) and 493.93 (M+2, 20.09%), and 335.09
(100%); Anal. calcd for C26H22ClN3O3S: C, 63.47; H, 4.51; N,
8.54. Found: C, 63.69; H, 4.73; N, 8.70.
Buff powder; yield, 76%; mp: 168–170°C; IR (KBr, cm−1): 3170.97
(NH), 3032.10 (C–H aromatic), 2951.09, 2924.09, 2897.08 (C–H
aliphatic), 1689.64, 1670.35 (C═O), and 1149.57 (C═S); 1H NMR
(DMSO‐d6): δH 2.28 (s, 3H, CH3), 3.75 (s, 3H, OCH3), 3.78 (s, 3H,
OCH3), 3.95 (s, 2H, CH2), 6.73 (dd, J = 8.8 Hz, 2 Hz, 1H, indole
H‐6), 6.91–6.94 (m, 3H, methoxyphenyl H‐3, H‐5, indole H‐7),
7.22 (s, 1H, indole H‐4), 7.48 (d, J = 8.8 Hz, 1H, methoxyphenyl
H‐2, H‐6), 7.67 (qAB, JAB = 8.4 Hz, 4H, benzoyl H‐2, H‐3, H‐5, H‐6),
11.71 (s, 1H, NH exchange with D2O), and 12.21 (s, 1H, NH
exchange with D2O); 13C NMR (DMSO‐d6): δC 13.80 (CH3), 31.70
(CH2), 55.74 (OCH3), 55.96 (OCH3), 102.45, 111.81, 113.08,
114.24, 115.06, 126.32, 129.57, 130.69, 131.04, 131.17, 131.66,
134.51, 136.40, 138.19, 156.10, 157.90, 168.37, 172.90, and
179.26; MS (m/z, %): 522.02 (M+, 33.99%) and 357.15 (100%);
Anal. calcd for C27H24ClN3O4S: C, 62.12; H, 4.63; N, 8.05. Found:
C, 61.87; H, 4.89; N, 8.27.
2‐[1‐(4‐Chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl]‐N‐(p‐
tolyl‐carbamothioyl)acetamide (4b)
2‐[1‐(4‐Chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl]‐N‐(4‐
chlorophenyl‐carbamothioyl)acetamide (4e)
Brown powder; yield, 68%; mp: 162–164°C; IR KBr, cm−1): 3178.69
(NH), 3032.10 (C–H aromatic), 2924.09 (C–H aliphatic), 1689.64
(C═O), and 1165.00 (C═S); 1H NMR (DMSO‐d6): δH 2.29 (s, 6H, 2
CH3), 3.86 (s, 3H, OCH3), 3.96 (s, 2H, CH2), 6.73 (dd, J = 8.8 Hz,
1.6 Hz, 1H, indole H‐6), 6.92 (d, J = 8.8 Hz, 1H, indole H‐7), 7.17–7.22
(m, 3H, indole H‐4, tolyl H‐2, H‐6), 7.48 (d, J = 8 Hz, 2H, tolyl H‐3,
H‐5), 7.67 (qAB, JAB = 8.4 Hz, 4H, benzoyl H‐2, H‐3, H‐5, H‐6), 11.73
(s, 1H, NH exchange with D2O), and 12.31 (s, 1H, NH exchange with
D2O); 13C NMR (DMSO‐d6): δC 13.79 (CH3 indole), 21.02 (CH3 tolyl),
31.71 (CH2), 55.96 (OCH3), 102.43, 111.81, 113.04, 115.05, 124.56,
129.56, 130.68, 131.16, 131.66, 134.49, 135.61, 136.18, 136.41,
138.20, 156.09, 168.38, 172.97, and 179.08; Anal. calcd for
Buff powder; yield, 83%; mp: 167–169°C; IR (KBr, cm−1): 3271.27
(NH), 3109.25, 3039.81 (C–H aromatic), 2924.09 (C–H aliphatic),
1666.50 (C═O), and 1149.57 (C═S); 1H NMR (DMSO‐d6): δH 2.28
(s, 3H, CH3), 3.75 (s, 3H, OCH3), 3.80 (s, 2H, CH2), 6.72 (dd, J = 9.2 Hz,
2.4 Hz, 1H, indole H‐6), 6.94 (d, J = 9.2 Hz, 1H, indole H‐7), 7.18
(d, J = 2.4 Hz, 1H, indole H‐4), 7.36 (d, J = 8.8 Hz, 2H, chlorophenyl H‐
2, H‐6), 7.63–7.71 (m, 6H, chlorophenyl H‐3, H‐5, benzoyl H‐2, H‐3,
H‐5, H‐6), and 10.35 (s, 1H, NH exchange with D2O); 13C NMR
(DMSO‐d6): δC 13.85 (CH3), 32.44 (CH2), 55.91 (OCH3), 102.39,
111.65, 114.35, 115.06, 121.29, 127.36, 129.12, 129.54, 130.73,
131.31, 131.62, 134.64, 135.93, 138.10, 138.48, 156.04, 168.37, and
169.15; MS (m/z, %): 526.43 (M+, 18.58%) and 139.13 (100%); Anal.

ABDELLATIF ET AL.
11 of 13
|
calcd for C26H21Cl2N3O3S: C, 59.32; H, 4.02; N, 7.98. Found: C,
59.42; H, 4.03; N, 7.94.
OCH3), 6.71 (dd, J = 8.8 Hz, 2.4 Hz, 1H, indole H‐6), 6.92 (d, J = 8.8 Hz, 1H,
indole H‐7), 7.16 (d, J = 2.4 Hz, 1H, indole H‐4), 7.66 (qAB, JAB = 8.8 Hz,
4H, benzoyl H‐2, H‐3, H‐5, H‐6), and 10.28 (s, 2H, 2NH exchange with
D2O); 13C NMR (DMSO‐d6): δC 13.79 (CH3), 24.17 (CH2–CN), 29.46
(CH2–C═O), 55.88 (OCH3), 102.34, 111.91, 113.88, 114.97, 116.07,
129.53, 130.69, 131.10, 131.64, 134.61, 135.85, 138.12, 156.03, 161.83,
168.36, and 168.92; MS (m/z, %): 438.86 (M+, 50.42%) and 139.25
(100%); Anal. calcd for C22H19ClN4O4: C, 60.21; H, 4.36; N, 12.77. Found:
C, 60.46; H, 4.49; N, 12.95.
4‐3‐{2‐[1‐(4‐Chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl]‐
acetyl}thioureido)benzoic acid (4f)
Brown powder; yield, 69%; mp: 175–177°C; IR (KBr, cm−1): 3618.46
(NH), 3248.13 (O–H acidic), 3008.95 (C–H aromatic), 2931.80 (C–H
aliphatic), 1681.93 (C═O), and 1172.72 (C═S); 1H NMR (DMSO‐d6):
δH2.29 (s, 3H, CH3), 3.75 (s, 3H, OCH3), 3.81 (s, 2H, CH2), 6.72
(d, J = 8.8 Hz, 1H, indole H‐6), 6.94 (d, J = 8.8 Hz, 1H, indole H‐7), 7.20
(s, 1H, indole H‐4), 7.65 (d, J = 8.4 Hz, 2H, benzoic acid H‐3, H‐5), 7.70
(d, J = 8.8 Hz, 2H, benzoyl, H‐3, H‐5), 7.74 (d, J = 8.8 Hz, 2H, benzoyl
H‐2, H‐6), 7.90 (d, J = 8.4 Hz, 2H, benzoic acid H‐2, H‐6), and 10.57
(s, 1H, NH exchange with D2O), OH acidic is not detected; 13C NMR
(DMSO‐d6): δC 13.89 (CH3), 32.52 (CH2), 55.90 (OCH3), 102.37,
111.67, 114.27, 115.08, 118.93, 125.63, 129.54, 130.70, 130.90,
131.31, 131.66, 134.65, 135.98, 138.08, 143.61, 156.03, 167.37,
168.35, and 169.50; Anal. calcd for C27H22ClN3O5S: C, 60.50; H,
4.14; N, 7.84. Found: C, 60.6; H, 4.34; N, 7.54.
|
4.1.4
General procedure for the synthesis of
compound 6
To a solution of 5 (0.01 mol, 4.38 g) in absolute ethanol (20 ml),
acetylacetone (0.01 mol, 1.00 g) and three drops of piperidine were
added, and the reaction was refluxed for 6 h. The reaction mixture
was cooled, and the formed solid was filtered, dried, and crystallized
from ethanol.
Ethyl‐4‐(3‐{2‐[1‐(4‐chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐
yl]acetyl}thioureido)benzoate (4g)
2‐[1‐(4‐Chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl]‐N‐(3‐
cyano‐4,6‐dimethyl‐2‐oxopyridin‐1(2H)‐yl)acetamide (6)
Buff powder; yield, 81%; mp: 150–152°C; IR (KBr, cm−1): 3294.42
(NH), 3070.68 (C–H aromatic), 2931.80 (C–H aliphatic), 1712.79,
1666.50 (C═O), and 1149.57 (C═S); 1H NMR (DMSO‐d6): δH 1.30
(t, J = 7.2 Hz, 3H, CH3 ethyl), 2.29 (s, 3H, CH3), 3.75 (s, 3H, OCH3),
3.81 (s, 2H, CH2), 4.28 (q, J = 7.2 Hz, 2H, CH2 ethyl), 6.73 (dd,
J = 8.8 Hz, 2.4 Hz, 1H, indole H‐6), 6.94 (d, J = 8.8 Hz, 1H, indole H‐7),
7.18 (d, J = 2.4 Hz, 1H, indole H‐4), 7.67 (qAB, JAB = 8.4 Hz, 4H, ben-
zoyl H‐2, H‐3, H‐5, H‐6), 7.75 (d, J = 8.4 Hz, 2H, ethylbenzoate H‐3,
H‐5), 7.92 (d, J = 8.8 Hz, 2H, ethylbenzoate H‐2, H‐6), and 10.63
(s, 1H, NH exchange with D2O); 13C NMR (DMSO‐d6): δC 13.85 (CH3
indole), 14.66 (CH3 ethyl), 32.51 (CH2 indole), 55.91 (OCH3), 60.93
(CH2 ethyl), 102.40, 111.66, 114.21, 115.07, 119.06, 124.78, 129.54,
130.72, 131.30, 131.63, 134.63, 136.00, 138.11, 143.88, 156.05,
165.80, 168.37, and 169.59; Anal. calcd for C29H26ClN3O5S: C,
61.75; H, 4.65; N, 7.45. Found: C, 61.95; H, 4.75; N, 7.35.
Buff powder; yield, 55%; mp: 242–244°C; IR (KBr, cm‐1): 3221.12 (NH),
3005.10 (C–H aromatic), 2924.09 (C–H aliphatic), 2368.59 (CN), and
1597.06 (C═O); 1H NMR (DMSO‐d6): δH 2.16 (s, 3H, CH3 oxopyridine),
2.27 (s, 3H, CH3 oxopyridine), 2.36 (s, 3H, CH3 indole), 3.80 (s, 3H,
OCH3), 3.85 (s, 2H, CH2), 6.36 (s, 1H, oxopyridine H‐5), 6.73
(dd, J = 8.8 Hz, 2.4 Hz, 1H, indole H‐6), 6.98 (d, J = 8.8 Hz, 1H, indole H‐7),
7.22 (d, J = 2.4 Hz, 1H, indole H‐4), 7.68 (qAB, JAB = 8.4 Hz, 4H, benzoyl H‐
2, H‐3, H‐5, H‐6), and 11.18 (s, 1H, NH exchange with D2O); 13C NMR
(DMSO‐d6): δC 13.83 (CH3 indole), 19.08 (CH3 oxopyridine C‐5), 21.22
(CH3 oxopyridine C‐3), 29.41 (CH3), 55.94 (OCH3), 100.41, 102.13,
108.66, 112.19, 113.10, 115.04, 115.84, 129.53, 130.74, 131.01, 131.70,
134.59, 136.17, 138.15, 154.64, 156.11, 158.65, 160.46, 168.36, and
169.92; MS (m/z, %): 502.96 (M+, 4.27%), 504.50 (M+2, 1.59%), and
139.15 (100%); Anal. calcd for C27H23ClN4O4: C, 64.48; H, 4.61; N,
11.14. Found: C, 64.16; H, 4.75; N, 11.43.
|
|
General procedure for the synthesis of
4.1.3
General procedure for the synthesis of
4.1.5
compound 5
compounds 8a,b
To a solution of 2 (0.01 mol, 3.80 g) in benzene (20 ml), cyanoacetic
acid hydrazide (0.01 mol, 0.99 g) was added and stirred for 24 h at
room temperature. Then the reaction mixture was filtered and the
precipitate was collected, washed, and crystallized from ethanol.
A solution of 5 (0.01 mol, 4.38 g) in absolute ethanol (20 ml) was
treated with 0.01 mol (1.54 g) of the appropriate arylidene 7a,b[25,26]
and five drops of piperidine; the mixture was heated under reflux for
2 h. The reaction mixture was filtered while hot, and the formed
precipitate was dried and crystallized from ethanol.
2‐[1‐(4‐Chlorobenzoyl)‐5‐methoxy‐2‐methyl‐1H‐indol‐3‐yl]‐N′‐(2‐
cyano‐acetyl)acetohydrazide (5)
Yellowish green powder; yield, 82.5%; mp: 250–252°C; IR (KBr, cm−1):
3197.98 (NH), 3051.39, 3008.95 (C–H aromatic), 2924.09 (C–H alipha-
tic), 2256.74 (CN), and 1678.07 (C═O); 1H NMR (DMSO‐d6): δH 2.25
(s, 3H, CH3), 3.54 (s, 2H, CH2–CN), 3.73 (s, 2H, CH2 indole), 3.78 (s, 3H,
N‐[6‐Amino‐3,5‐dicyano‐2‐oxo‐4‐phenylpyridin‐1(2H)‐yl]‐2‐{5‐
methoxy‐2‐methyl‐1‐[4‐(piperidin‐1‐yl)benzoyl]‐1H‐indol‐3‐yl}‐
acetamide (8a)
Yellowish white powder; yield, 79%; mp: 275–277°C; IR (KBr, cm−1):
3441.01 (NH, NH2), 3101.54 (C–H aromatic), 2970.36, 2931.80,

12 of 13
ABDELLATIF ET AL.
|
2854.65 (C–H aliphatic), 2214.28 (CN), 1689.64 (C═O); 1H NMR
(DMSO‐d6): δH 1.55–1.64 (m, 6H, piperidine H‐3, H‐4, H‐5), 2.33
(s, 3H, CH3), 2.99 (t, J = 5.2 Hz, 4H, piperidine H‐2, H‐6), 3.75 (s, 3H,
OCH3), 4.17 (s, 2H, CH2), 6.71 (dd, J = 8.8 Hz, 2.0 Hz, 1H, indole H‐6),
6.94 (d, J = 8.8 Hz, 1H, indole H‐7), 7.21 (d, J = 2.0 Hz, 1H, indole H‐4),
The anti‐inflammatory activity of five compounds was evaluated
by using the carrageenan‐induced rat paw edema model.[28] First,
five test compounds, 4a, 4b, 4d, 5, 6, indomethacin, and celecoxib
were administrated orally in a dose of 50 mg/kg dissolved in 1%
Tween in saline, and the control group received vehicle only 1 h
before subcutaneous injection of carrageenan. Paw edema was
induced by subcutaneous injection of 1% carrageenan in saline
(0.02 ml/rat) into the left hind paw of each rat. The hind paw
thickness of each rat was measured after 1, 3, and 5 h of
carrageenan injection, and then the change in thickness and %
inhibition of paw edema were calculated according to the formula
of Ratheesh and Helen[31] and Adedapo and Sofidiya[32] with two
standard drugs celecoxib and indomethacin:
7.46–7.50 (m, 5H, phenyl H‐2, H‐3, H‐4, H‐5, H‐6), 7.65 (dAB, JAB
=
8.4 Hz, 2H, benzoyl H‐3, H‐5), 7.71 (dAB, JAB = 8.4 Hz, 2H, benzoyl H‐
2, H‐6); 13C NMR (DMSO‐d6): δC 13.90 (CH3), 22.17 (piperidine C‐4),
22.83 (piperidine C‐3, C‐5), 24.04 (CH2), 44.34 (piperidine C‐2, C‐6),
55.78 (OCH3), 102.67, 111.69, 114.98, 116, 117, 118.86, 128.77,
129.03, 129.51, 129.75, 130.80, 131.17, 131.66, 134.77, 135.02,
136.48, 137.97, 155.12, 155.89, 156.65, 164.29, and 168.36; MS
(m/z, %): 639.72 (M+, 16.28%) and 110.68 (100%); Anal. calcd for
C
37H33N7O4: C, 69.47; H, 5.20; N, 15.33. Found: C, 69.57; H, 5.1;
(V0 – V)
t
% of inhibition =
× 100,
N, 15.36.
V0
N‐[6‐Amino‐4‐(4‐chlorophenyl)‐3,5‐dicyano‐2‐oxopyridin‐1(2 H)‐
yl]‐2‐{5‐methoxy‐2‐methyl‐1‐[4‐(piperidin‐1‐yl)benzoyl]‐1 H‐indol‐
3‐yl}acetamide (8b)
where V0 is the volume of the paw of the control at time 0 and Vt is
the volume of the paw of the drug‐treated animal at time t.
The ulcerogenic liability of the most active compounds that
showed an effective in vivo anti‐inflammatory activity was evaluated
according to a previous study.[29] Rats were fasted one day before
drug administration and divided into eight separate groups (each
group with six rats). All compounds were dissolved in 1% Tween in
saline and administrated orally. The normal control group received
only vehicle, the celecoxib group received celecoxib in a dose of
50 mg/kg, the indomethacin group received the drug in a dose of
50 mg/kg, and the other five groups received the tested compounds,
respectively, in a dose of 50 mg/kg. Then animals were fed after 2 h.
Rats were given the specified dose orally for three successive days.
Rats were killed after 2 h of the last dose, and then the stomach of
each rat was removed and opened along the greater curvature for
determination of the ulcer number and ulcer index according to a
previous study.[29] To examine the stomach, it was stretched by pins
on a corkboard. The gastric mucosa was carefully inspected for the
occurrence of ulcers with the aid of an illuminated magnifying lens
(×10), and the ulcer index was calculated.
Yellow crystals; yield, 84%; mp: 280–282°C; IR (KBr, cm−1): 3425.58
(NH, NH2), 3101.54 (C–H aromatic), 2924.09, 2854.65 (C–H ali-
phatic), 2214.28 (CN), and 1689.64 (C═O); 1H NMR (DMSO‐d6):
δH1.55–1.64 (m, 6H, piperidine H‐3, H‐4, H‐5), 2.33 (s, 3H, CH3), 3.01
(t, J = 5.6 Hz, 4H, piperidine H‐2, H‐6), 3.73 (s, 3H, OCH3), 4.18 (s, 2H,
CH2), 6.70 (dd, J = 8.8 Hz, 2 Hz, 1H, indole H‐6), 6.94 (d, J = 8.8 Hz,
1H, indole H‐7), 7.21 (d, J = 1.6 Hz, 1H, indole H‐4), 7.52 (d, J = 8.4 Hz,
2H, chlorophenyl H‐2, H‐6), 7.60 (d, J = 8.4 Hz, 2H, chlorophenyl H‐3,
H‐5), 7.65 (dAB, J = 8.4 Hz, 2H, benzoyl H‐3, H‐5), 7.71 (dAB, JAB = 8
Hz, benzoyl H‐2, H‐6), and 8.23 (s, 1H, NH exchange with D2O); 13C
NMR (DMSO‐d6): δC 13.89 (CH3), 22.07 (piperidine C‐4), 22.69
(piperidine C‐3, C‐5), 24.03 (CH2), 44.26 (piperidine C‐2, C‐6), 55.79
(OCH3), 102.66, 111.68, 114.98, 116.26, 117.31, 118.74, 128.94,
129.52, 130.80, 130.99, 131.15, 131.65, 134.65, 134.74, 135.05,
135.33, 138.00, 153.49, 153.86, 155.89, 156.55, 164.37, and 168.37;
MS (m/z, %): 674.16 (M+, 27.77%) and 359.93 (100%); Anal. calcd. for
C37H32ClN7O4: C,65.92; H, 4.78; N, 14.54. Found: C, 65.91; H, 4.81;
N, 14.58.
The number of mucosal damage that appeared as red spots was
counted, and their severity was determined and graded from 0 to 4.
The following parameters were calculated:
|
4.2
Pharmacological/biological assays
1. Percentage of incidence/10 = [number of rats showing ulcer of
any grade/total number of rats in group × 100]/10
2. The average number of ulcers = number of ulcers in the group/
total number of rats in the group
The ability of the tested compounds listed in Table 1 to inhibit ovine
COX‐1 and COX‐2 was measured with EIA[27] in which N′, N′, N′,
N′‐tetramethyl‐p‐phenylenediamine at 590 nm was used, as pre-
viously mentioned. Adult male Wistar albino rats, weighing 90–110 g,
were housed in groups (each group of six rats) in the animal house of
the Faculty of Pharmacy, Beni‐Suef University, under controlled
conditions (light period of 12 h/day and temperature 25 2°C). An-
imals were fasted overnight with free access to water before the
experiment. Before any drug administration, the thickness of the
left hind paw of each rat was measured in millimeters. The anti‐
inflammatory activity of the synthesized compounds was
determined in vivo by carrageenan‐induced paw edema in rats.
3. Average severity = ∑ [each ulcer multiplied by its score of
severity/number of ulcers in the group
4. Ulcer index = the sum of the above three parameters
The stomach was collected and fixed in 10% buffered formalin
for 72 h. Routine histological processing and paraffin embedding
were carried out according to the method of Bancroft and
Gamble.[33] Tissue sections (4–5 µm) were stained with hematoxylin
and eosin stain.

ABDELLATIF ET AL.
13 of 13
|
|
4.3
Molecular docking
[12] K. R. Abdellatif, Z. Huang, M. A. Chowdhury, S. Kaufman,
E. E. Knaus, Bioorg. Med. Chem. Lett. 2011, 21, 3951. https://doi.org/
10.1016/j.bmcl.2011.05.017
In our docking study, Molecular Operating Environment (MOE) version
2015.10 modeling software was used to apply molecular docking for the
most active compounds. First, we downloaded the crystal structure of
indomethacin–ethylenediamine–dansyl conjugate bound to the COX‐2
active site encoded (PDB: 6BL4)[10] from the Protein Data Bank. The
ligand and the tested compounds must follow some steps before being
docked including three‐dimensional protonation and energy minimizing
to get the lowest energy conformers. London dG and force field energy
were used to perform docking study and refinement of results. Only one
conformer was chosen according to its superposition with our ligand.
Finally, we obtained results indicating energy scores of binding with the
active site, amino acid interactions, and the length of the formed
hydrogen bonds (Table 5).
[13] Y. A. Ammar, E. A. Fayed, A. H. Bayoumi, M. A. Saleh, M. E. El‐Araby,
Am. J. Pharm. Technol. Res. 2015, 5, 245.
[14] E. K. Abdelall, Bioorg. Chem. 2020, 94, 103441. https://doi.org/10.
1016/j.bioorg.2019.103441
[15] H. A. Elshemy, E. K. Abdelall, A. A. Azouz, A. Moawad, W. A. Ali,
N. M. Safwat, Eur. J. Med. Chem. 2017, 127, 10. https://doi.org/10.
1016/j.ejmech.2016.12.030
[16] K. R. Abdellatif, E. K. Abdelall, P. F. Lamie, M. B. Labib, E.‐S. El‐
Nahaas, M. M. Abdelhakeem, Bioorg. Chem. 2020, 95, 103540.
https://doi.org/10.1016/j.bioorg.2019.103540
[17] K. R. Abdellatif, W. A. Fadaly, G. M. Kamel, Y. A. Elshaier, M. A. El‐
Magd, Bioorg. Chem. 2019, 82, 86. https://doi.org/10.1016/j.bioorg.
2018.09.034
[18] K. R. Abdellatif, M. A. Abdelgawad, H. A. Elshemy, S. S. Alsayed, Bioorg.
Chem. 2016, 64, 1. https://doi.org/10.1016/j.bioorg.2015.11.001
[19] K. R. Abdellatif, H. A. Elshemy, A. A. Azoz, Bioorg. Chem. 2015, 63,
13. https://doi.org/10.1016/j.bioorg.2015.09.002
ACKNOWLEDGMENTS
[20] K. R. Abdellatif, H. A. Elshemy, S. A. Salama, H. A. Omar, J. Enzyme
Inhib. Med. Chem. 2015, 30, 484. https://doi.org/10.3109/14756366.
2014.949255
The authors are grateful to all members of Pharmaceutical Organic
Chemistry, Faculty of Pharmacy, Beni‐Suef University, for all
support during the proceeding of research.
[21] N. H. Amin, M. T. El‐Saadi, A. A. Hefny, A. H. Abdelazeem,
H. A. Elshemy, K. R. Abdellatif, Future Med. Chem. 2018, 10, 2521.
https://doi.org/10.4155/fmc-2018-0224
CONFLICTS OF INTERESTS
[22] K. R. Abdellatif, P. F. Lamie, H. A. Omar, J. Enzyme Inhib. Med. Chem.
2016, 31, 318. https://doi.org/10.3109/14756366.2015.1022174
[23] S. Gupta, N. Moorthy, Trends Appl. Sci. Res. 2007, 2, 165.
[24] S. Bondock, A. E.‐G. Tarhoni, A. A. Fadda, ARKIVOC 2006, 9, 113.
[25] G. Vasuki, K. Kumaravel, Tetrahedron Lett. 2008, 49, 5636. https://
doi.org/10.1016/j.tetlet.2008.07.055
The authors declare that there are no conflicts of interests.
ETHICS STATEMENT
All applicable international, national, and/or institutional guidelines
for the care and use of animals were followed.
[26] A. M. Shestopalov, Y. M. Emeliyanova, A. A. Shestopalov,
L. A. Rodinovskaya, Z. I. Niazimbetova, D. H. Evans, Org. Lett. 2002,
4, 423.
ORCID
[27] E. K. Abdelall, G. M. Kamel, Eur. J. Med. Chem. 2016, 118, 250.
https://doi.org/10.1016/j.ejmech.2016.04.049
Heba A. H. Elshemy
https://orcid.org/0000-0003-0699-0650
[28] C. J. Morris, in Inflammation Protocols, Springer, Berlin/Heidelberg,
Germany 2003, pp. 115.
REFERENCES
[29] R. B. Bakr, A. A. Azouz, K. R. Abdellatif, J. Enzyme Inhib. Med. Chem.
2016, 31, 6. https://doi.org/10.1080/14756366.2016.1186018
[30] L. Shi, A. Hu, J. Xu, Y. Jiang, Chin. J. Chem. 2012, 30, 1339.
[31] M. Ratheesh, A. Helen, Afr. J. Biotechnol. 2007, 6, 1209.
[32] A. A. Adedapo, M. O. Sofidiya, V. Maphosa, B. Moyo, P. J. Masika,
A. J. Afolayan, Rec. Nat. Prod. 2008, 2, 46.
[1] C. P. Cannon, P. J. Cannon, Science 2012, 336, 1386. https://doi.org/
10.1126/science.1224398
[2] K. R. Abdellatif, E. K. Abdelall, W. A. Fadaly, G. M. Kamel, Med. Chem.
Res. 2015, 24, 2632. https://doi.org/10.1007/s00044-015-1327-7
[3] K. R. Abdellatif, A. Moawad, E. E. Knaus, Bioorg. Med. Chem. Lett.
2014, 24, 5015. https://doi.org/10.1016/j.bmcl.2014.09.024
[4] E. K. Abdelall, P. F. Lamie, A. K. Ahmed, E.‐N. EL‐Shaymaa, Bioorg.
Chem. 2019, 86, 235. https://doi.org/10.1016/j.bmcl.2014.09.024
[5] R. Botting, J. Physiol. Pharmacol. 2006, 57, 113.
[33] J. D. Bancroft, M. Gamble, Theory and Practice of Histological Tech-
niques, Elsevier Health Sciences, Amsterdam 2008.
SUPPORTING INFORMATION
[6] M. Takahashi, T. Ogawa, H. Kashiwagi, F. Fukushima, M. Yoshitsugu,
M. Haba, M. Hosokawa, Bioorg. Med. Chem. Lett. 2018, 28, 997. https://
doi.org/10.1016/j.bmcl.2018.02.035
Additional Supporting Information may be found online in the
supporting information tab for this article.
[7] S. Xu, M. J. Uddin, S. Banerjee, K. Duggan, J. Musee, J. R. Kiefer,
K. Ghebreselasie, C. A. Rouzer, L. J. Marnett, J. Biol. Chem. 2019,
294, 8690. https://doi.org/10.1074/jbc.RA119.007405
[8] N. E. El‐Ashmawy, E. G. Khedr, H. A. El‐Bahrawy, H. M. Selim,
Nutrition 2016, 32, 849. https://doi.org/10.1016/j.nut.2016.01.010
[9] K. Olugemo, D. Solorio, C. Sheridan, C. L. Young, Postgrad. Med.
2015, 127, 223. https://doi.org/10.1080/00325481.2015.1000231
[10] Y. Ammar, M. Salem, E. A. Fayed, M. Helal, M. El‐Gaby, H. K. Thabet,
Synth. Commun. 2017, 47, 1341. https://doi.org/10.1080/00397911.
2017.1328066
How to cite this article: Abdellatif KRA, Abdelall EKA,
Elshemy HAH, El‐Nahass E‐N, Abdel‐Fattah MM, Abdelgawad
YYM. New indomethacin analogs as selective COX‐2
inhibitors: Synthesis, COX‐1/2 inhibitory activity,
anti‐inflammatory, ulcerogenicity, histopathological, and
docking studies. Arch Pharm. 2020;e2000328.
[11] K. R. Abdellatif, M. A. Chowdhury, C. A. Velázquez, Z. Huang,
Y. Dong, D. Das, G. Yu, M. R. Suresh, E. E. Knaus, Bioorg. Med. Chem.
Lett. 2010, 20, 4544. https://doi.org/10.1016/j.bmcl.2011.05.017
https://doi.org/10.1002/ardp.202000328