Synthesis, evaluation, and molecular docking studies of cycloalkyl/aryl-3,4,5-trimethylgallates as potent non-ulcerogenic and gastroprotective anti-inflammatory agents

Article abstract of DOI:10.1007/s00044-013-0620-6

In our effort to identify the effective gastric sparing and protective anti-inflammatory agents, a series of cycloalkyl/aryl-3,4,5-trimethylgallates were synthesized and characterized. The physicochemical properties were studied to assess the lipophilicity and chemical stability. Subsequently, the compounds were evaluated for their anti-inflammatory activity and effect on gastric mucosa by most active compounds. All the compounds exhibited promising anti-inflammatory activity. In particular, 4a, 4b, 4g, and 4h emerged as the most active compounds in the series. The results of gastric mucosal studies and biochemical estimations suggested that these compounds are non-ulcerogenic and gastroprotective. The molecular docking analysis was performed to understand the binding interactions of these compounds to cyclooxygenase isoenzyme (COX-1 and COX-2). The results from this investigation suggests cycloalkyl/aryl-3,4,5- trimethylgallates as potent safer gastrosparing and protective anti-inflammatory agents.

Full text of DOI:10.1007/s00044-013-0620-6

MEDICINAL
CHEMISTRY
RESEARCH
Med Chem Res
DOI 10.1007/s00044-013-0620-6
ORIGINAL RESEARCH
Synthesis, evaluation, and molecular docking studies of cycloalkyl/
aryl-3,4,5-trimethylgallates as potent non-ulcerogenic
and gastroprotective anti-inflammatory agents
•
Mamta Sachdeva Dhingra Pran Kishore Deb
•
•
Renu Chadha Tejvir Singh Maninder Karan
•
Received: 20 January 2013 / Accepted: 4 May 2013
Ó Springer Science+Business Media New York 2013
Abstract In our effort to identify the effective gastric
sparing and protective anti-inflammatory agents, a series of
cycloalkyl/aryl-3,4,5-trimethylgallates were synthesized
and characterized. The physicochemical properties were
studied to assess the lipophilicity and chemical stability.
Subsequently, the compounds were evaluated for their anti-
inflammatory activity and effect on gastric mucosa by most
active compounds. All the compounds exhibited promising
anti-inflammatory activity. In particular, 4a, 4b, 4g, and 4h
emerged as the most active compounds in the series. The
results of gastric mucosal studies and biochemical esti-
mations suggested that these compounds are non-ulcero-
genic and gastroprotective. The molecular docking analysis
was performed to understand the binding interactions of
these compounds to cyclooxygenase isoenzyme (COX-1
and COX-2). The results from this investigation suggests
cycloalkyl/aryl-3,4,5-trimethylgallates as potent safer gas-
trosparing and protective anti-inflammatory agents.
Introduction
Nonsteroidal anti-inflammatory drugs (NSAIDs) belong to
one of the most common therapeutic agents used world-
wide for the treatment of pain, fever, and inflammation.
However, usefulness of these agents is limited due to
higher incidence of gastrointestinal (GI) damage including
gastric ulceration, perforation, and other associated com-
plications. These side effects are related to the intrinsic
mechanism responsible for the desired activity (Gibson,
1988; Vane and Botting, 1998). NSAIDs exert therapeutic
effects by inhibiting the activity of the enzyme cyclooxy-
genase (COX), resulting in the prevention of prostaglan-
dins synthesis (Vane, 1971). It is now well known that
COX exists in three isoforms, COX-1, COX-2, and COX-3.
COX-1 is constitutive and provides cytoprotection in the
GI tract; COX-2 is inducible and mediates inflammation,
whereas COX-3 is a variant of COX-1 and nonfunctional in
humans (Pontiki et al., 2011; Chandrasekharan et al.,
2002). The mucosal integrity in normal GI tract is pri-
marily maintained by the prostaglandins that are derived
from COX-1 and therefore, inhibition of COX-1 along with
COX-2 by NSAIDs is responsible for their ulcerogenic side
effects (Mc Carthy, 1989; Warner et al., 1999; Wolfe et al.,
1999). Based on these observations, it has been suggested
that selective COX-2 inhibitors may act as safer NSAIDs,
devoid of their ulcerogenic side effects and a number of
selective COX-2 inhibitors have been developed and
introduced in the market for clinical use (Hawkey, 1999;
Xie et al., 1992). However, long term use of these agents
has shown some potential limitation, including ulcer
exacerbation in high risk patients, delayed GI ulcer healing,
kidney toxicity, as well as, cardiovascular side effects
(Dogne et al., 2005). These observations indicated that
safety of such agents is questionable on their long-term use
Keywords Cycloalkyl/aryl-3,4,5-trimethylgallates ꢀ
Physicochemical ꢀ Anti-inflammatory activity ꢀ
Ulcerogenic activity ꢀ Gastroprotective ꢀ
Molecular docking
Electronic supplementary material The online version of this
article (doi:10.1007/s00044-013-0620-6) contains supplementary
material, which is available to authorized users.
M. S. Dhingra ꢀ P. K. Deb ꢀ R. Chadha ꢀ M. Karan (&)
University Institute of Pharmaceutical Sciences, UGC Center
of Advanced Study (UGC-CAS) in Pharmaceutical Sciences,
Panjab University, Chandigarh 160 014, India
e-mail: maninderkaran@hotmail.com
T. Singh
Department of Chemistry, Panjab University,
Chandigarh 160 014, India
123

Med Chem Res
and some of these agents have already been withdrawn
from the market (Schnitzer, 2001). Thus, the initial
enthusiasm of developing selective COX-2 inhibitors has
been faded away and need for the design and development
of safer agents still remains.
(TMGA) with naturally occurring phenolic/alcoholic anti-
oxidant compounds which can act synergistically with
enhanced therapeutic outcomes as non-ulcerogenic and
gastroprotective anti-inflammatory agents. Hypothetical
binding mode was proposed for synthesized compounds
with respect to the target enzyme COX (both COX-1 and
COX-2) based on in silico docking studies.
During recent years, it has been well known that
excessive free radical generation takes place in inflamma-
tory disorders and these observations indicate that antiox-
idants may be used to prevent free radicals-induced
inflammation (Bandyopadhyay et al., 1999). During the
past few decades, a large number of naturally occurring
phenolic compounds have been identified as antioxidants,
which are viewed as promising therapeutic agents for
treating free radical-mediated inflammatory diseases. Sev-
eral reviews have addressed the anti-inflammatory activity
of phenols, attributing their property not only to the anti-
oxidant capacity, but also to inflammatory mediators’
modulation, namely cytokines and proinflammatory pro-
teins, such as inducible nitric oxide synthase and COX-2
(Costa et al., 2012).
Results and discussion
Chemistry
For the syntheses of compounds (4a–4i), 3,4,5-trimeth-
ylgallolyl chloride (3) was treated with various naturally
occurring antioxidant compounds (Fig. 1) either alone or
with their potassium salts (Scheme 1). The characteristic
spectral features of synthesized compounds indicated that
the carbonyl group in all esters appeared in IR spectra at
an expected value of around tc=o 1,730 cm-1 except in
compounds 4e and 4i (Table 1). In compound 4e, the
presence of electron withdrawing aldehyde group at para
position strengthened the C=O bond resulting in slightly
higher frequency (1,737 cm-1). However, such increase
was not seen in compound 4f (1,730 cm-1), because the
presence of electron withdrawing aldehyde group at meta
position does not involve conjugation. Similarly, the 13C
chemical shift of ester carbonyl group was observed at d
163–165 while other peaks for 1H and 13C carbon
appeared at expected d values. All the compounds (4a–4i)
showed M??1 peaks in their mass spectra at expected
m/z values including M?? Na ion (Table 1). Compounds
4b, 4c, and 4e are reported in literature (Pepe, 1951;
Mathieu and Blum, 1981; Ruan et al., 2009), but none of
these compounds have been studied before for medicinal
use.
Gallic acid (GA, 3,4,5-trihydroxybenzoic acid) and its
derivatives are widely distributed in the plant kingdom and
represent a large family of plant secondary polyphenolic
metabolites. They are present in the form of either meth-
ylated GAs (syringic acid) or galloyl conjugates of catechin
derivatives, i.e., flavan-3-ols, or polygalloyl esters of glu-
cose, quinic acid or glycerol (Lu et al., 2006). GA and its
derivatives have been widely reported for various biolog-
ical and pharmacological activities including anti-inflam-
matory activity (Lu et al., 2006; Kroes et al., 1992). These
activities are possibly linked with their antioxidant poten-
tial due to their ability to prevent damage from free radicals
or to prevent generation of these free radicals (Perron and
Brumaghim, 2009).
Based on these observations and the previous studies
reported from our laboratory (Sawraj et al., 2012; Mad-
H3CO
H3CO
OCH3
O
O
H3CO
H3CO
H3CO
H
O
H3CO
H3CO
O
H
O
O
Solubility and partition coefficient
hukar et al., 2010), it is suggested that there are potential
advantages in giving naturally occurring phenolic/alcoholic
agents with complimentary pharmacological activities in
the form of a single chemical/drug entity. Such agents are
designed with improved physicochemical properties and
release the parent drugs at the site of action (Singh and
Sharma, 1994; Bhosle et al., 2006; Leppanen et al., 2002).
Hence, our intention was to synthesize cycloalkyl/aryl-
3,4,5-trimethylgallates by conjugating trimethylgallic acid
Lipophilicity is an important factor controlling the inter-
action of drugs with biological membranes. It is generally
accepted that good absorption of an orally administered
drug could be obtained when apparent partition coefficient
value (Log P) is more than 2 (Yalkowsky and Morozowich,
1980). The evaluation of Log P of compounds was per-
formed by saturation shake flask method (Hairsine, 1989;
123

Med Chem Res
Fig. 1 Structures of naturally
occurring antioxidants;
3-hydroxyphenol (1a),
OCH3
OCH3
HO
OH
HO
HO
2-methoxyphenol (1b), 4-allyl-
2-methoxyphenol (1c),
5-methyl-2-isopropylphenol
(1d), 4-hydroxy-3-
methoxybenzaldehyde (1e),
3-hydroxy-4-
methoxybenzaldehyde (1f),
benzo[d][1,3]dioxol-5-ol (1g),
7-hydroxy-2H-chromen-2-one
(1h), and (1a, 2b, 5a)-5-methyl-
2-isopropylcyclohexanol (1i)
(1c)
(1a)
(1d)
(1g)
(1b)
OCH3
OH
OH
OCH3
HO
CHO
CHO
(1f)
(1e)
HO
HO
O
O
HO
O
O
(1h)
(1i)
Scheme 1 Sequence of steps
involved in the synthesis of
cycloalkyl/aryl-3,4,5-
COOH
trimethylgallates (4a–4i).
Reagents and conditions:
(i) (a) (CH3)SO4, NaOH;
HO
OH
OH
Gallic acid
(b) H?; (ii) SOCl2, reflux, 3 h;
(iii) Chloroform/Toluene, NEt3,
ArOH (1a–1e, 1i), O °C (1 h),
room temperature (overnight);
and (iv) Tetrahydrofuran, KH,
ArOH (1f–1h), O °C (1 h),
room temperature (overnight)
(i)
COOH
H3CO
OCH3
OCH3
(2)
(ii)
COCl
H3CO
OCH3
OCH3
(3)
Methodology 2
Methodology 1
(iv)
(iii)
O
O
O
O
C
Ar
C
Ar
Et3NHCl
KCl
H3CO
OCH3
OCH3
H3CO
OCH3
OCH3
(4a-4e, 4i)
(4f-4h)
123

Med Chem Res
Table 1 Characteristic spectral features of test compounds
Compound
Structure
I.R.
(t, cm-1
13C-NMR
(CDCl3, d, ppm)
Mass
ESI-MS (m/z)
)
(C=O)
4a
1,734
165.28
305.1 (M??1)
O
OCH3
OCH3
HO
O
OCH3
4b
1,730
164.38
319.2 (M??1), 341.2 (M??Na?)
O
OCH3
OCH3
O
OCH3
OCH3
4c
1,727
164.49
381.39 (M??Na?)
O
OCH3
OCH3
O
OCH3
OCH3
CH3
4d
1,731
164.85
345.2 (M??1), 367.2 (M??Na?)
O
O
OCH3
OCH3
H3C
CH3
OCH3
347.1 (M??1)
OHC
4e
1,737
163.78
O
OCH3
O
OCH3
OCH3
OCH3
123

Med Chem Res
Table 1 continued
Compound
Structure
I.R.
(t, cm-1
13C-NMR
(CDCl3, d, ppm)
Mass
ESI-MS (m/z)
)
(C=O)
4f
1,730
164.04
369.11 (M??Na?)
CHO
O
OCH3
OCH3
O
OCH3
OCH3
4g
1,731
165.14
355.1 (M??Na?)
O
O
OCH3
O
O
OCH3
OCH3
4h
1,733
164.23
357.1 (M??1), 379.1 (M??Na?)
O
OCH3
OCH3
O
O
O
OCH3
4i
1,752
163.24
351.1 (M??1)
O
OCH3
OCH3
O
OCH3
Yalkowsky et al., 1992; Streng, 2001). The standard plots
of 4a–4i were constructed using UV spectrophotometer and
quantification was carried out accordingly. To assess the
lipophilicity, the Log P of TMGA and its esters was
determined in n-octanol and buffer layers (pH 7.4) and
calculated by correlating the absorbance with the concen-
tration in standard plot. The obtained values of Log P and
solubilities at pH 7.4 are shown in Table 2, indicating that
the synthesized derivatives meet the requirement for gas-
trointestinal absorption.
kinetics of chemical hydrolysis was studied at 37 °C
using buffer solutions of pH 1.2 and 7.4 by HPLC. The
reactivity to chemical hydrolysis was evaluated by first
order rate constants obtained from slopes of semi-loga-
rithmic plots of six concentrations against time. The half
life (t1/2) and rate constants (k) are shown in Table 2,
indicating the stability in buffer solutions. The deriva-
tives showed considerable chemical stability at pH 1.2
which implies that these passed unhydrolyzed through the
stomach after oral administration. At pH 7.4, the com-
pounds showed enough stability to be absorbed intact
from the intestine.
Chemical stability: in aqueous buffer solutions
Ester derivatives should be chemically stable so that they
can be formulated in an appropriate pharmaceutical
dosage form with optimum half life (Wadhwa and
Sharma, 1995). Therefore, the esters (4a–4i) were
assayed in vitro to evaluate their chemical stability. The
In vivo anti-inflammatory activity
All the compounds were screened for in vivo anti-inflam-
matory activity with carrageenan-induced rat paw edema
using indomethacin as reference drug. Carrageenan (1 %
123

Med Chem Res
Table 2 Evaluation of physicochemical properties of test compounds (4a–4i)
Compound
Partition
coefficient
(Log P)
Solubility
(lg/ml)
kmax (nm)
Extinction
coefficient
pH 7.4
pH 1.2
k (h-1
)
t1/2 (h)
k (h-1
)
t1/2 (h)
1 %
1cm
(E
)
TMGA (2)
1.18
2.62
3.07
3.94
4.68
2.57
2.59
2.23
2.73
4.45
10,600
170.59
143.70
155.30
111.33
190.70
168.20
182.31
162.70
111.70
236.0
268.3
271.2
273.0
256.0
267.0
263.6
270.8
267.2
242.6
510
520
380
383
450
480
680
390
890
560
–
–
–
–
4a
4b
4c
4d
4e
4f
0.0016
0.0019
0.0116
0.001
0.018
0.0051
0.0117
0.033
0.0015
425.172
364.737
59.741
580.124
38.287
135.882
59.231
20.873
451.041
0.005
0.001
0.006
0.001
0.003
0.013
0.061
0.001
0.006
143.241
579.251
113.607
578.514
247.500
55.000
11.379
577.500
102.347
4g
4h
4i
w/v) was used to produce paw edema. The paw edema
induced by subplantar injection of carrageenan was more
prominent in control group where indomethacin exerted
86 % anti-inflammatory effect after 2 h. Among the tested
compounds (Table 3); 4a, 4b, 4g, and 4h exhibited more
than 75 % edema inhibition and decreased the difference in
paw thickness comparable to that of indomethacin
(p \ 0.05). The comparable anti-inflammatory activity of
these compounds may be attributed to the combined effect
of improved physicochemical properties of esters and
contribution by their corresponding promoieties. Further-
more, equimolar physical mixtures of test compounds (1a–
1i) were also studied for their anti-inflammatory activity
(Table 3). However, these physical mixtures showed lower
anti-inflammatory activity as compared with their corre-
sponding test compounds.
reduction in ulcer index by the physical mixture of test
compounds was also less as compared to their test com-
pounds (Table 4). This may be due to the polar nature of
their antioxidant moieties resulting in poor bioavailability,
whereas reduction in ulcer index by the test compounds
was significant (p \ 0.001) which may be due to the
improved physicochemical properties and contribution by
the antioxidant promoieties after their cleavage. From the
results of these studies, it is evident that in all the groups
treated with test compounds, the structure of gastric
mucosa is quite normal.
Biochemical evaluation
Oxidative stress plays an important role in gastric mucosal
damage (Rastogi et al., 1998; Pal et al., 2010). In aerobic
cells, mitochondria are the major source of free radicals
and also a sensitive target for free radical-mediated damage
(Orrenius et al., 2007). Increased generation of free radi-
cals causes oxidative stress and is responsible for the
gastric mucosal injury (Rastogi et al., 1998). To find out
whether the antiulcer activity of said compounds is medi-
ated through its antioxidant action, pyloric ligation-induced
mitochondrial oxidative stress was measured in gastric
mucosal cells in both the presence and absence of these test
compounds. Treatment with pyloric ligation increased
peroxidation of mitochondrial lipids and depleted the
mitochondrial total glutathione, superoxide dismutase, and
catalase contents (Table 5). Test compounds (4a, 4b, 4g,
and 4h) at their anti-inflammatory doses significantly pre-
vented mitochondrial lipid peroxidation, depletion of total
glutathione, superoxide dismutase, and catalase contents.
However, the effects of physical mixtures on antioxidant
enzymes were lesser compared with their test compounds
(Table 5). The results suggest that the inhibition of PL-
induced mitochondrial oxidative stress by test compounds
Ulcerogenic and antiulcer activity
Although the anti-inflammatory activity of many synthe-
sized compounds was comparable to the reference drug, the
gastric studies were only performed with most active
compounds (4a, 4b, 4g, and 4h). These test compounds and
indomethacin were given orally to rats and animals were
sacrificed after treatment for determining the ulcerogenic
index of the said compounds. None of the test compounds
showed any ulcerogenic effects (Table 4) in the gastric
mucosa in fasted rats. However, gastric ulcers were
observed in all the animals treated with indomethacin.
Further, the same test compounds along with their physical
mixtures were screened for their gastroprotective effects in
rats using pyloric ligation (PL) method, which produced a
significant increase in ulcer index (5.23 0.34) as com-
pared to the control group (0.31 0.21). All four com-
pounds (4a, 4b, 4g, and 4h) showed significantly reduced
gastric damage ([64 % protection) (Table 4). The
123

Med Chem Res
123

Med Chem Res
Table 4 Effect of active test compounds and their physical mixtures on gastric mucosa
Ulcerogenic activity
Treatment
Antiulcer activity
Dose
(mg/kg)
Ulcer index*
mean SEM
Treatment
Dose
(mg/kg)
Ulcer index*
mean SEM
Control
Indomethacin
4a
0.5 % CMC
0.34 0.07
Control
Pyloric ligated
4a
0.5 % CMC
–
0.31 0.21
5.23 0.34#
1.82 0.27ab
3.6 0.34c
1.62 0.28ab
3.7 0.41c
1.72 0.32ab
3.4 0.28c
1.88 0.37ab
3.9 0.41c
48
6.68 0.6
114.82
0.0
28.71
2 ? 1a
4b
–
–
2 ? 1a
4b
20 ? 10.37
30.03
120.12
0.0
2 ? 1b
4g
–
–
2 ? 1b
4g
20 ? 11.7
31.35
125.39
0.16 0.53
2 ? 1g
4h
–
–
2 ? 1g
4h
20 ? 13.01
33.62
134.46
–
0.11 0.42
–
2 ? 1h
2 ? 1h
20 ? 15.28
* Values are expressed as mean SEM (n = 6) and analyzed by one-way ANOVA followed by Dunnett’s test
#
Significant as compared to control group (p \ 0.001)
a
Significant as compared to control group (p \ 0.01)
b
Significant as compared to pyloric ligated group (p \ 0.001)
c
Significant as compared to pyloric ligated group (p \ 0.05)
Table 5 Effect of test compounds (4a, 4b, 4g, and 4h) on various biomarkers of oxidative stress of gastric mucosa
Treatment
LPO (nmoles of MDA/
mg protein)
GSH (lg of GSH/
mg protein)
SOD (Units/
mg protein)
Catalase (U/
mg protein)
Control
Pyloric ligated
4a
3.56 0.42
11.2 0.31*
6.24 0.32a
9.12 0.21b
6.02 0.28a
9.57 0.37c
7.12 0.35a
10.54 0.52c
7.48 0.22a
10.18 0.43c
28.39 0.90
12.56 0.77*
22.35 0.81a
16.58 ? 0.74b
22.88 1.05a
16.42 0.82c
21.22 0.92a
16.21 0.71c
17.38 0.81b
16.34 0.68c
8.12 0.45
3.23 0.24*
6.04 0.39a
5.12 0.61c
6.38 0.64a
4.87 0.26c
5.34 0.42b
4.98 0.23c
5.42 0.28b
5.08 0.38c
13.57 0.57
5.61 0.27*
10.54 0.67a
7.98 0.57b
10.78 0.38a
8.01 0.37b
10.12 0.19a
7.94 0.42b
8.78 0.28b
7.85 0.63b
2 ? 1a
4b
2 ? 1b
4g
2 ? 1g
4h
2 ? 1h
Data are expressed as mean SEM (n = 6)
* p \ 0.001 as compared to control group
a
p \ 0.001 as compared to pyloric ligated group
b
p \ 0.01 as compared to pyloric ligated group
c
p \ 0.05 as compared to pyloric ligated group
may result from the scavenging of free radicals, the for-
mation of which is augmented by PL-induced gastric
mucosal injury.
paw edema assay to postulate a hypothetical binding model
for their interaction with COX-2 isoenzyme using the
X-ray crystal structure of COX-2 (PDB ID: 1CX2). To
investigate the ability of molecular docking to reproduce an
experimentally observed ligand-binding mode, the co-
crystallized ligand SC-558 (a highly selective COX-2
inhibitor) was used as reference ligand. SC-558 was
docked back into its binding site (Fig. 2a) of the crystal
structure of the COX-2 using Glide docking program
Virtual screening
In silico (Docking studies)
Docking (in silico) studies were performed on the com-
¨
(Glide, version 5.7, Schrodinger, 2011). The top docked
pounds 4a–4i, as well as indomethacin found active in rat
123

Med Chem Res
Fig. 2 a Superimposition of redocked SC-558 (cyan color) with its
original position (yellow color) as co-crystal in the binding site of the
crystal structure of COX-2 (PDB ID:1CX2) isoenzyme (RMSD-0.70)
showing H-bond interactions with the amino acid residues His90,
Arg120, Ser353, and Arg513, respectively. b Binding orientation of
indomethacin within the binding site of COX-2 showing salt bridge
formation between the carboxylate ion and Arg120. c Binding
orientation of TMGA (2) within the binding site of COX-2 showing
salt bridge formation between the carboxylate ion and Arg120.
d Hypothetical binding motif of compound 4a within the binding site
of COX-2 showing important H-bonding interaction with the amino
acid residues Arg120, Tyr355, Ser353, and Tyr385, respectively and
p-stacking interaction with Tyr355. e Hypothetical binding motif of
most active compound 4b within the binding site of COX-2 showing
important H-bonding interaction with the amino acid residues
Arg120, Tyr355, Tyr385, and Ser530, respectively and p-stacking
interaction with Tyr355, Trp387, and Tyr385, respectively. f Hypo-
thetical binding orientation of compound 4h within the binding site of
COX-2 showing important H-bonding interaction with the amino acid
residues Arg120, Tyr355, and Arg513, respectively, and p-stacking
interaction with Tyr355 and Tyr385, respectively (Color figure
online)
conformations (poses) closely resembled the co-crystal-
lized conformation with a root-mean square deviation
(RMSD) 0.70 of non-hydrogen atomic positions of the
ligand (SC-558). Crucial residues involved in the binding
were identified, which are perhaps contributing to the
selectivity of the ligand.
freedom of movement of the ligand. The docking studies
using the same protocol with compounds 4a–4i suggested
that the carboxylic group of the key intermediate trimeth-
ylgallic acid, TMGA (2) formed salt bridge with the gua-
nidine moiety of Arg120 in COX-2 isoenzyme (Fig. 2c).
The esterification of carboxylic group of TMGA in the target
compounds (4a–4i) resulted in different orientations at the
binding site of the receptor without the formation of any salt
bridge. For instance, in case of compound 4a (Fig. 2d), the
methoxy groups at 3rd and 4th positions of aromatic ring
were found to form significant hydrogen bonding with OH
group of Tyr355 and NH group of Arg120 at a distance of
Further, the docking studies performed for indomethacin
(a nonselective COX-1/2 inhibitor) showed the formation of
a salt bridge by carboxylate ion with the Arg120 in both
COX-1 and COX-2 isoenzymes (Figs. 2b, 3b), which pro-
duces a more generalized anchoring point for all the classical
NSAIDs thus limiting their selectivity due to restricted
123

Med Chem Res
Fig. 3 a Binding orientation of celecoxib in the binding site of the
crystal structure of COX-1 (PDB ID:3KK6) isoenzyme showing
H-bond interactions with the amino acid residues His90, Gln192,
Ser353, and Tyr355, respectively. b Binding orientation of indo-
methacin within the binding site of COX-1 showing salt bridge
formation between the carboxylate ion and Arg120 and H-bond
interaction with Tyr355. c Binding orientation of TMGA (2) within
the binding site of COX-1 showing salt bridge formation between the
carboxylate ion and Arg120. d Hypothetical binding motif of
compound 4a within the binding site of COX-1 showing important
H-bonding interaction with the amino acid residues Gln192, Tyr385,
Ile517, Ala527, and Ser530, respectively and p-stacking interaction
with Phe518. e Hypothetical binding motif of most active compound
4b within the binding site of COX-1 showing important H-bonding
interaction with the amino acid residues Arg120, Tyr355, and Gly526,
respectively and p-stacking interaction with Phe518. f Hypothetical
binding orientation of compound 4h within the binding site of COX-1
showing important H-bonding interaction with the amino acid
residues Gly526, His90, Ser516, and Ser530, respectively and
aromatic p-stacking interaction with Phe518
˚
2.203 and 2.093 A, respectively. The carbonyl group (C=O)
respectively. Surprisingly, the ester group showed different
orientation as compared with compound 4a where the car-
bonyl group (C=O) was facing toward Ser353. The oxygen
atom of the methoxy group attached at 2nd position of
phenyl ring showed significant hydrogen bonding interac-
of ester linkage and phenolic OH group showed strong
˚
hydrogen bonding interactions with NH of Ser353 (1.994 A)
˚
and OH of Tyr385 (1.850 A), respectively. As shown in
Fig. 2e, the most active compound 4b docked well into the
active site of COX-2 with interaction energy of DGbind
˚
˚
=
tions with Ser530 (1.987 A) and Tyr 385 (2.446 A),
respectively. The electron donating methoxy groups at 2nd
position reinforced the aromatic moiety (guaiacol) to form a
significant p-stacking interaction with the phenyl ring of
Tyr385 and Trp387, respectively. Stronger hydrogen
bonding interactions with the crucial amino acid residues
-70.44 kcal/mol (Table S1 in Supplementary Section). The
methoxy groups of compound 4b at 3rd and 4th positions of
aromatic ring were found to form strong hydrogen bonding
interactions with the OH group of Tyr355 and NH group of
˚
Arg120 at a distance of 1.566, 1.465, and 1.713 A,
123

Med Chem Res
˚
(1.526 A), and Tyr385 (1.531 A), respectively. The OH
˚
such as Arg120, Tyr355, Ser530, Tyr385 and significant
p-stacking interactions with Tyr355, Tyr385, and Trp387
could be the reason for better potency of 4b as compared
with 4a and other analogs of the series.
group of phenolic moiety formed strong hydrogen bond
˚
˚
with Gln192 (1.827 A) and Ile517 (1.989 A), whereas the
aromatic ring formed p-stacking interaction with the phe-
nyl ring of Phe518 (Fig. 3d). Interestingly, the most active
compound 4b was found to dock into the active site of
COX-1 with interaction energy of DGbind = -68 kcal/mol
(Fig. 3e and Table S1 in Supplementary Section). The
methoxy groups at 3rd and 4th positions of trimethoxy
phenolic ring formed significant hydrogen bonding with
In order to analyze the possible ligand–receptor recog-
nition mechanism in a more quantitative way, we calcu-
lated the contribution of individual electrostatic and Van
der Waals (vdW) interaction energy of each amino acid
˚
residue of receptor within 12-A area of the ligand. It was
observed that 4b formed stable electrostatic and vdW
interactions with all the crucial amino acid residues as
indicated from their negative electrostatic and vdW inter-
action energy (Fig. S1, S2, in Supplementary Section).
Further, the presence of electron withdrawing aldehydic
group (–HC=O) at both the para and meta position of the
phenyl ring of compounds 4e and 4f causes to decrease in
their affinity by probably reducing the possibility of aro-
matic p-stacking interaction with Tyr385. These com-
pounds were also found to exhibit unstable electrostatic
and vdW interactions (positive electrostatic and vdW
interaction energy) with certain crucial amino acid resi-
dues. Similarly, the introduction of allyl chain at para
position of compound 4c showed positive electrostatic and
vdW interaction energy with certain crucial amino acid
residues, which causes detrimental effect in their affinity.
Further, replacement of methoxy group from 2nd position
by electron donating isopropyl moiety with simultaneous
addition of methyl group at 5th position of the phenyl ring
of compound 4d enriched the p-electron system of aro-
matic ring, but resulted in the change in their orientations
and loss of hydrogen bonding interactions with the crucial
amino acid residues Ser530 and Tyr385, respectively.
Interestingly, compounds 4g and 4h with sesamol and
umbelliferone moieties showed moderate binding affinities,
probably due to the presence of electron withdrawing
methylenedioxy group and carbonyl group attached to the
phenyl ring. However, all the three methoxy (–OCH3)
groups of compound 4h showed good hydrogen bonding
interactions with crucial amino acid residues Arg120
˚
˚
Tyr355 (1.793 A) and Arg120 (2.318 A), respectively. The
O-methoxy phenyl ring formed strong p-stacking interac-
tion with the phenyl ring of Phe518 (Fig. 3e). Similarly, in
case of compound 4h, the methoxy groups at 3rd and 5th
positions of trimethoxy phenolic ring showed moderate
˚
hydrogen bonding interactions with Gly526 (2.827 A) and
˚
Ser530 (2.536 A), respectively. The aromatic ring of cou-
marin moiety showed significant p-stacking interaction
with the phenyl ring of Phe518 (Fig. 3f) whereas the
oxygen atom (–O–) and carbonyl group (C=O) formed
strong hydrogen bonding interactions with His90 and
˚
Ser516 at a distance of 1.726 and 1.806 A, respectively
(Fig. 3f). The electrostatic interaction (Coulomb) energy
(in kcal/mol) and vdW interaction energy (in kcal/mol)
between indomethacin, celecoxib, SC-558, 4b, 4d and each
single amino acid involved in ligand recognition obtained
after docking simulations inside the binding site COX-1
receptor is presented in Fig. S3 and S4 in Supplementary
Information Section.
Post-scoring with MM-GB/SA
The calculation of binding affinities using molecular
mechanics and a continuum solvation mode was introduced
by Kuhn and Kollman using PB/SA to represent the solvent
(Kuhn and Kollman, 2000). Since then, several modifica-
tions of this procedure have appeared in the literature
(Bernacki et al., 2005; Huang et al., 2006a; Huang et al.,
2006b; Lyne et al., 2006). In this study, a conformational
search for the molecules in the unbound state and energy
minimization for the complexes using OPLS-AA and GB/
SA (Lyne et al., 2006; Still et al., 1990) within Macro-
˚
˚
˚
(2.468 A), Tyr355 (2.468 A), and Arg513 (2.468 A),
respectively as shown in Fig. 2f.
Further, the docking studies performed for celecoxib to
COX-1 isoenzyme showed similar binding affinity to ori-
ginal co-crystallized conformation with heavy atom RMSD
of 0.70–1.20 (Fig. 3a). It was shown in previous section
that docking of indomethacin and TMGA within COX-1
includes a salt bridge formation with Arg120 (Fig. 3b, c).
However, no such salt bridge formation was observed
between the test compounds (4a–4i) and Arg120 (Fig. 3d–
f). For example, in case of compound 4a, significant
hydrogen bond formation was observed between all the
¨
Model (version 9.9, Schrodinger, 2011) was performed. As
more computationally demanding, the MM-GB/SA scoring
gives far superior correlations with experimental activity
data than standard docking scoring functions. The MMGB/
SA methodology has also been much more reliable than
docking for rank-ordering. As shown in Table S1 (Sup-
plementary Information Section), the MM-GB/SA scoring
yields a very good correlation with the anti-inflammatory
data for all the synthesized compounds as compared with
G_Score (Glide docking score).
˚
trimethoxy groups and Ala527 (1.980 A), Ser530
123

Med Chem Res
Table 6 Calculation of various molecular properties of test compounds (4a–4i)
Compound
Mol.
weight
(MW)a
Dipoleb Solvent
accessible
Donor Acceptors QPlogPo/ QPlogSg Human oral No. of
Wiener QPlogBBj PSAk
HBd HBe
wf
absorptionh violations index
surface area
(SASA)c
(%)
LRi
(WI)
4a
4b
4c
4d
4e
4f
304.299 6.492
318.326 4.071
358.390 3.934
344.407 1.631
346.336 2.631
346.336 7.599
332.309 4.469
356.331 3.727
350.454 3.279
514.557
564.374
630.209
633.328
540.358
599.16
1
0
0
0
0
0
0
0
0
1
0
2
5.5
2.406
3.186
3.514
4.351
1.59
-2.778 100
-3.055 100
-3.325 100
-4.755 100
0
0
0
0
0
0
0
0
0
0
0
0
1,090 -0.563
1,216 -0.065
1,763 -0.336
1,510 -0.079
1,556 -0.823
1,530 -0.852
1,395 -0.161
1,758 -0.823
1,510 -0.109
1,424 -0.651
1,061 -0.663
706 -0.195
80.708
58.652
68.578
50.54
5.5
5.5
4.75
7.5
-1.39
87.549
106.809
101.003
77.593
96.32
7.5
2.208
2.12
-2.505 92.716
-1.655 100
-2.936 91.62
-4.741 100
-5.111 92.221
-2.127 82.038
-5.160 100
4g
4h
4i
502.613
597.828
618.782
595.039
513.831
478.712
6.25
7.25
4.25
5.75
7
2.163
4.404
4.269
1.205
4.479
55.35
Indomethacin 357.793 4.935
82.039
77.141
53.222
Rofecoxib
314.355 5.552
296.152 4.128
Diclofenac
Sodium
2.5
a
b
c
d
e
f
g
h
Range 95 % of drugs: (130–725), (1–12.5), (300–1000), (0–6), (2–20), (-2 to 6.5), (-6.5 to 0.5), ([80 % high, \25 % poor),
(maximum 4), (-3 to 1.2), and (7–200)
i
j
k
Prediction of ADME properties
has been observed that the results of practically evaluated
Log P and Log S values are in consonance with the theo-
retically predicted Log P and Log S values as evident from
their good correlation coefficient of R2 = 0.975 and 0.946,
as shown in Fig. S5, S6 (in Supplementary Section). This
good correlation further reveals that other predicted ADME
properties of these compounds (4a–4i) could be a reliable
indicator supporting their druggable properties. The results
of predicted properties for all the compounds 4a–4i are in
the ranges predicted by QikProp for 95 % of known oral
drugs and also satisfy the Lipinski’s rule of five to be
considered as drug like. Moreover, the standard drugs also
do not show any violation. Theoretically, these compounds
should present good passive oral absorption and differences
in their bioactivity may not be attributed to this aspect.
Certain molecular properties which could influence the
metabolism, cell permeability. and bioavailability for the
synthesized compounds (4a–4i) along with three repre-
sentative anti-inflammatory drugs indomethacin, diclofe-
nac, and rofecoxib were evaluated using QikProp (version
¨
3.4) module of Schrodinger (Table 6). Some of the
parameters such as QPlogPo/w, QPlogS, and polar surface
area (PSA) are recognized parameters for prediction of
drug transport properties. Further, steric and molecular
surface descriptors, e.g., solvent accessible surface area
(SASA) and other parameters such as wiener index (WI)
were calculated. ADME prediction methods were used to
assess the bioavailability of the test compounds (4a–4i) and
the reference drugs. Herein, we calculated the compliance
of all the compounds to the Lipinski’s ‘rule of five’ (Li-
pinski et al., 2001) which has been widely used as a filter
for predicting the druggable properties of any molecule.
According to this rule, poor absorption or permeation is
more likely when there are more than five H-bond donors,
ten H-bond acceptors, the molecular weight (MW) is
greater than 500, and the calculated Log P (CLogP) is
greater than 5. Molecules violating more than one of these
rules may have problems with bioavailability. Further,
PSA, which is a measure of a molecule’s hydrogen bonding
capacity, is another key property that has been linked to
drug bioavailability. Passively absorbed molecules with a
PSA [ 200 are thought to have low-oral bioavailability
(Clark and Pickett, 2000). Predictions of ADME properties
for test compounds are given in Table 6. Interestingly, it
Conclusion
In conclusion, a number of cycloalkyl/aryl-3,4,5-trimeth-
ylgallates (4a–4i) have been synthesized, characterized,
and studied for physicochemical properties. All the com-
pounds exhibited promising anti-inflammatory activity. In
particular, 4a, 4b, 4g, and 4h emerged as the most active
compounds in the series. The results of ulcerogenic studies
and biochemical evaluation suggest that these compounds
are gastric safe. The results from the docking studies were
found to be in good agreement with the results from
pharmacological studies. Further, the compounds comply
with Lipinski’s rule of five which signifies a good
absorption and hence, good bioavailability. The present
123

Med Chem Res
Synthesis of compounds (4a–4i)
investigation suggests cycloalkyl/aryl-3,4,5-trimethylgal-
lates as potent new gastric safe anti-inflammatory agents,
which can be further explored for their therapeutic out-
comes. Further studies and related analogs will be reported
in due course.
Cycloalkyl/aryl-3,4,5-trimethylgallates (4a–4i) were syn-
thesized by conjugating TMGA with various naturally
occurring antioxidant compounds such as 3-hydoxyphenol
(resorcinol) (1a), 2-methoxyphenol (guaiacol) (1b), 4-allyl-
2-methoxyphenol (eugenol) (1c), 5-methyl-2-isopropyl-
phenol (thymol) (1d), 4-hydroxy-3-methoxybenzaldehyde
(vanillin) (1e), 3-hydroxy-4-methoxybenzaldehyde (isova-
nillin) (1f), benzo[d][1,3]dioxol-5-ol (sesamol) (1g),
7-hydroxy-2H-chromen-2-one (umbelliferone) (1h), and
(1a,2b,5a)-5-methyl-2-isopropylcyclohexanol (menthol)
(1i), respectively (Fig. 1). These agents are important
components of human diet and their safety profile is well
known (Cotelle, 2001; Martin et al., 1998).The sequence of
steps involved in the synthesis of cycloalkyl/aryl-3,4,5-
trimethylgallates is shown in Scheme 1.
Experimental
Materials
Chemicals and reagents viz: GA, nitroblue tetrazolium,
hydrogen peroxide, ethylenediaminetetracetic acid (EDTA),
thiobarbituric acid, and p-nitrosodimethyl aniline were pur-
chased from Sigma-Aldrich chemicals. Drug indomethacin
was procured from Ind-Swift laboratories limited. All the
solventswereofanalytical grade andweredistilledbeforeuse.
Procedure for the synthesis of 3,4,5-trimethylgallic acid
Methods
(2)
The melting points were recorded in open capillaries on
Casiaa Siamea (VMP-AM) melting point apparatus and are
uncorrected. Infrared (IR) spectra were recorded on a
Perkin-Elmer FT-IR 240-C spectrophotometer using KBr
disks. 1H NMR spectra were recorded on Bruker Avance II
400 MHz spectrometer in CDCl3 using tetramethylsilane
(TMS) as internal standard. The spin multiplicities are
indicated by the symbols, s (singlet), d (doublet), t (triplet),
q (quartlet), and m (multiplet). Mass spectra were obtained
with Waters Micromass Q-TOF Micro mass spectrometer
at 70 eV using electron ionization (EI) sources. The ele-
mental analyses were performed using Thermo EA 2110
series elemental analyzer. All the reactions were monitored
by thin layer chromatography (TLC) on precoated silica gel
60 F254 (aluminum base) (E. Merck, Mumbai) and spots
were visualized under UV light (254 nm). Silica gel from
E. Merck, Mumbai (100–200 mesh) was used for column
chromatography. UV–Vis spectra were obtained on a Per-
kin-Elmer 554 double beam spectrophotometer and on a
Hitachi U-2001 spectrophotometer. High-pressure liquid
chromatography (HPLC) separation was carried out using
reverse phase HPLC C-18 column at isocratic mode. The
Waters Associates fitted with pump (Model-510), injector
(Model 6UK), detector (Lambda-Max Model 481 LC
spectrophotometer), and interfaced with Winchrom were
used. The elution was carried out at ambient temperature.
Lichrosphere C-18 column (250 mm length 9 4.6 mm
diameter, E. Merck, India), was used for qualitative anal-
ysis. The samples were prepared by filtering the appropri-
ately diluted extract through 0.4 lm filter before injection.
Elution was carried out using methanol:water (80:20) at a
flow rate of 1.0 mL/min for the detection of parent
compound.
To a stirred solution of GA (1) (5 g, 29.4 mmol) in 50 mL
of 4 N sodium hydroxide, dimethyl sulfate was added
(8.9 g, 71 mmol) dropwise while keeping the reaction
temperature below 20 °C. The solution was allowed to stir
at 30–40 °C for 20 min; same quantity of GA in sodium
hydroxide was again added to the above reaction mixture.
The resulting mixture was slowly heated to 90 °C and
maintained at this temperature for 1 h. After refluxing for
2 h, the reaction mixture was allowed to cool to room
temperature. The solution was acidified with 2 N hydro-
chloric acid to pH 2, filtered, and the solid was washed with
water. The crude product was recrystallized with 50 %
ethanol to give 3,4,5-trimethylgallic acid (2) (40.8 g,
65 %) as white, needle shaped crystals; mp 170–171 °C
(Lit. 168–171 °C) (Li et al., 2006).
Procedure for the synthesis of 3,4,5-trimethylgallolyl
chloride (3)
Thionyl chloride (4 mL, 54.8 mmol) was added dropwise
to acid compound (2) (2.12 g, 10 mmol) and heated under
reflux for 3 h. Thionyl chloride was removed to give 3,4,5-
trimethylgallolyl chloride (3). Freshly prepared acid chlo-
ride 3 was used for the synthesis of compounds 4a–4i.
General procedure for the synthesis of compounds
4a–4i
A solution of acid chloride (3) in chloroform/toluene was
treated with phytophenolic/alcoholic antioxidant com-
pounds (1a–1e, 1i) in the presence of triethylamine (TEA)/
°
pyridine at 0 C. The other method involves the reaction of
123

Med Chem Res
less nucleophilic phenols (1f–1h) in tetrahydrofuran (THF)
with potassium hydride (KH) followed by reaction with
acid chloride (3). The reaction mixtures were stirred
overnight at room temperature and after usual work up, the
desired products (4a–4i) were obtained in yields ranging
from 48 to 69 %. The methods were standardized with
respect to various reaction conditions (solvent, tempera-
ture, and time) by monitoring their progress by TLC. The
structure formulae of the products were authenticated by
spectroscopic techniques such as UV, IR, NMR, MS, and
elemental analyses.
(400 MHz, CDCl3, d ppm): 7.46 (s, 2H, 2,6-Ar–H), 7.25
(m, 1H, 40-Ar–H), 7.15 (m, 1H, 60-Ar–H), 7.01 (m, 2H,
30,50-Ar–H), 3.94 (s, 9H, 3,4,5-OCH3), 3.83 (s, 3H, 20-
OCH3); 13C NMR (100 MHz, CDCl3, d ppm): 164.38
(C=O), 153.02 (C-3, C-5), 151.35 (C-20), 142.61 (C-4),
139.96 (C-10), 127.02 (C-40), 124.34 (C-1), 122.91 (C-60),
120.84 (C-50), 112.50 (C-30), 107.47 (C-2, C-6), 60.98
(OCH3), 56.30 (OCH3), 55.90 (OCH3); ESI–MS (m/z):
319.2 (M??1), 341.2 (M??Na). Anal. Calcd. %:
(C17H18O6): C, 64.18; H, 5.74. Found (%): C, 64.14; H,
5.70.
General procedure for the synthesis of compounds 4a–4d
20-Methoxy-40-(200-propenyl)phenyl-3,4,5-trimethylgallate
(4c)
To a stirred solution of 3,4,5-trimethylgallolyl chloride (3)
(2.3 g, 0.01 mol) in chloroform, a solution of different
phytophenolic/alcoholic antioxidant compounds (1a–1d)
(0.01 mol) were added dropwise over a period of 1 h at
0 °C in the presence of TEA. The reaction mixture was
stirred overnight at room temperature. The organic layer
was washed with 5 % hydrochloric acid, 10 % sodium
bicarbonate and brine, dried over magnesium sulfate, fil-
tered, and concentrated in vacuo. The residue was purified
by column chromatography on silica gel (hexane/ethyl
acetate = 92:8).
The compound 4c was prepared using phenolic antioxidant
compound 1c (1.63 g, 0.01 mol) by the method as descri-
bed above, that resulted in white solid with 55 % yield: mp
131–132 °C; IR (KBr): 2939, 2837 (aliphatic C–H st),
1727 (C=O st), 1592, 1510 (aromatic C=C st), 1270 (C–O–
C asym st), 1123 (C–O–C symm st) cm-1 1H NMR
;
(400 MHz, CDCl3, d ppm): 7.46 (s, 2H, 2,6-Ar–H), 7.06 (d,
1H, 60-Ar–H), 6.82 (m, 2H, 30,50-Ar–H), 5.98 (m, 1H, 200-
H), 5.12(m, 2H, 100-H), 3.42 (d, 2H, J = 6.68, 300-H), 3.81
(s, 3H, 20-OCH3), 3.93 (s, 9H, 3,4,5-OCH3); 13C NMR
(100 MHz, CDCl3, d ppm): 164.49 (C=O), 153.02 (C-3,
C-5), 151.11 (C-20), 142.61 (C-4), 139.12 (C-10), 138.19
(C-200), 137.10 (C-40), 124.42 (C-1), 122.60 (C-60), 120.78
(C-50), 116.18 (C-100), 112.86 (C-30), 107.55 (C-2, C-6),
60.97 (OCH3), 56.31 (3,5-OCH3), 55.89 (OCH3), 40.13 (C-
300); ESI–MS (m/z): 381.39 (M??Na); Anal. Calcd. %:
(C20H22O6): C, 67.05; H, 6.22. Found (%): C, 67.03; H,
6.19.
30-Hydroxyphenyl-3,4,5-trimethylgallate (4a)
The compound 4a was prepared using phenolic antioxidant
compound 1a (1.10 g, 0.01 mol) by the method as descri-
bed above, that resulted in brown semisolid with 48 %
yield: IR (KBr): 2935, 2840 (aliphatic C–H st), 1734
(C=O), 1592, 1503 (aromatic C=C str), 1238 (C–O–C asym
1
st), 1127 (C–O–C symm st) cm-1; H NMR (400 MHz,
CDCl3,d ppm): 7.41 (m, 1H, 50-Ar–H), 7.36 (s, 2H, 2,6-Ar–
H), 7.06 (m, 1H, 60-H), 6.92 (m, 2H, 20,40-Ar–H), 4.8 (s,
1H, 2-OH), 3.89 (s, 9H, 3,4,5-OCH3); 13C NMR
(100 MHz, CDCl3, d ppm): 165.28 (C=O), 156.50 (C-30),
153.96 (C-10), 152.02 (C-3, C-5), 144.61 (C-4), 130.84 (C-
50), 124.34 (C-1), 114.91 (C-60), 113.02 (C-40), 110.35 (C-
20), 107.47 (C-2, C-6), 56.38 (OCH3), 56.30 (OCH3), 55.90
(OCH3); ESI–MS (m/z): 305.1 (M??1), 327.1 (M??Na);
Anal. Calcd %: (C16H16O6): C, 63.24; H, 5.34. Found (%):
C, 63.15; H, 5.30.
50-Methyl-20-(100-methylethyl)phenyl-3,4,5-trimethylgallate
(4d)
The compound 4d was prepared using phenolic antioxidant
compound 1d (1.50 g, 0.01 mol) by the method as
described above, that resulted in white solid with 53 %
yield: mp 63 °C; IR (KBr): 2961, 2839 (aliphatic C–H st),
1731 (C=O), 1589, 1502 (aromatic C=C st), 1225 (C–O–C
asym st), 1124 (C–O–C symm st)cm-1
;
1H NMR
(400 MHz, CDCl3, d ppm): 7.48 (s, 2H, 2,6-Ar–H), 7.24 (d,
1H, J = 7.9, 40-H), 7.07 (d, 1H, J = 7.9, 30-H), 6.93 (s, 1H,
60-H), 3.95 (s, 3H, OCH3), 3.94 (s, 6H, OCH3), 3.04 (sept,
1H, J = 6.9, CH(CH3)2), 2.34 (s, 3H, 100-H), 1.21 (d, 6H,
J = 6.9, 80-CH(CH3)2); 13C NMR (100 MHz, CDCl3, d
ppm): 164.85 (C=O), 153.45 (C-3, C-5), 151.35 (C-
20),143.56 (C-4), 139.96 (C-10), 127.02 (C-40), 124.34 (C-
1), 122.91 (C-60), 120.84 (C-50), 112.50 (C30), 107.47 (C-2,
C-6), 60.98 (OCH3), 56.30 (OCH3), 55.90 (OCH3), 26 (C-
70), 24 (C-100), 23 (C-80, C-90); ESI–MS (m/z):
20-Methoxyphenyl-3,4,5-trimethylgallate (4b)
The compound 4b was prepared using phenolic antioxidant
compound 1b (1.24 g, 0.01 mol) by the method as
described above, that resulted in white solid with 61 %
yield: mp 113 °C; IR (KBr): 2937, 2839 (aliphatic C–H st),
1730 (C=O), 1591, 1506 (aromatic C=C st), 1257 (C–O–C
asym st), 1128 (C–O–C symm st) cm-1
;
1H NMR
123

Med Chem Res
345.2(M??1), 367.2 (M??Na); Anal. Calcd %:
(C20H24O5): C, 69.69; H, 7.06. Found (%): C, 69.75; H,
7.02.
100-H, CH3); 13C NMR (100 MHz, CDCl3, d ppm): 163.24
(C=O), 152.23 (C-3, C-5), 142.61 (C-4), 123.47 (C-1),
107.47 (C-2, C-6), 75.4 (C-10), 60.98 (OCH3), 56.30
(OCH3), 55.90 (OCH3), 46.24 (C-20), 38.42 (C-60), 32.14
(C-40), 27.52 (C-50), 25.10 (C-70) 22.74 (C-30), 21.72 (C-80,
C-90), 20.45 (C-100); ESI–MS (m/z): 351.1 (M??1); Anal.
Calcd %: (C20H30O5): C, 68.42; H, 8.58. Found (%): C,
68.54; H, 8.63.
General procedure for the synthesis of compounds 4e
and 4i
Pyridine (6.95 mL, 0.075 mol) was added to a solution of
phenolic antioxidants 1e and 1i (0.01 mol) in three times
by volume of dry toluene. To this, a solution of compound
(3) (2.3 g, 0.01 mol) in three times by volume of dry tol-
uene was added. Reaction mixture was stirred overnight in
tightly sealed container. Reaction mixture was mixed with
water to dissolve pyridine hydrochloride precipitate. The
organic layer was washed with 5 % hydrochloric acid,
10 % sodium bicarbonate and brine, dried over magnesium
sulfate, filtered, and concentrated in vacuo. The residue
was purified by crystallization with hexane/ethyl acetate
(90:10) to obtain the desired products.
General procedure for the synthesis of potassium salt
of compounds 1f–1h
To a stirred solution of compounds 1f–1h (0.01 mol) in
dried tetrahydrofuran, potassium hydride (0.01 mol) was
added to form the potassium salts of said compounds.
General procedure for the synthesis of compounds
4f–4h
To a stirred solution of acid chloride (3) (2.3 g, 0.01 mol),
the solution of potassium salt of 1f–1h phenols (0.01 mol)
in dried tetrahydrofuran was added dropwise at 0 °C. The
reaction mixture was stirred at room temperature over-
night. Solvent was removed under vacuum and slight
water was added and extracted three times with ethyl
acetate. The organic layer was washed with 5 % hydro-
chloric acid, 10 % sodium bicarbonate and brine, dried
over magnesium sulfate, filtered, and concentrated in
vacuo. The residue was purified by column chromatog-
raphy on silica gel (hexane/ethyl acetate) to afford the
desired products.
40-Formyl-20-methoxyphenyl-3,4,5-trimethylgallate (4e)
The compound 4e was prepared using phenolic antioxidant
compound 1e (1.52 g, 0.01 mol) by the method as descri-
bed above, that resulted in white solid with 59 % yield: mp
131 °C; IR (KBr): 2931, 2844 (aliphatic C–H st), 1737
(C=O), 1594, 1508 (aromatic C=C st), 1277 (C–O–C asym
1
st), 1128 (C–O–C symm st), cm-1; H NMR (400 MHz,
CDCl3, d ppm): 9.91 (s, 1H, CHO), 7.53 (m, 2H, 30,50-Ar–
H), 7.46 (s, 2H, 2,6-Ar–H), 7.34 (m, 1H, 60-Ar–H), 3.88 (s,
3H, OCH3), 3.87 (s, 6H, OCH3), 3.85 (s, 3H, OCH3); 13C
NMR (100 MHz, CDCl3, d ppm): 191.10 (C=O, aldehyde),
163.78 (C=O, ester), 153.11 (C-3, C-5), 152.24 (C-20),
145.25 (C-10), 143.02 (C-4), 135.33 (C-40), 124.81 (C-1),
123.62 (C-60), 123.58 (C-50), 110.91 (C-3), 107.84 (C-2,
C-6), 61 (OCH3), 56.15 (OCH3); ESI–MS (m/z): 369.11
(M??Na); Anal. Calcd. %: (C18H18O7): C, 62.41; H, 5.21.
Found (%): C, 62.42; H, 5.24.
30-Formyl-60-methoxyphenyl-3,4,5-trimethylgallate (4f)
The compound 4f was prepared using potassium salt of
antioxidant compound 1f (1.52 g, 0.01 mol) by the above
described method, that resulted in white solid with 69 %
yield: mp 167 °C; IR (KBr): 2940, 2842 (aliphatic C–H st),
1730 (C=O), 1598, 1507 (aromatic C=C st), 1270 (C–O–C
(1a,2b,5a)-50-Methyl-20-(100-methylethyl)cyclohexyl-3,4,5-
trimethylgallate (4i)
asym st), 1123 (C–O–C symm st) cm-1
;
1H NMR
(400 MHz, CDCl3, d ppm): 9.84 (s, 1H, CHO), 7.74 (dd,
1H, 40-Ar–H, J = 2.0, J = 8.4), 7.64 (d, 1H, 20-Ar–H,
J = 2.0), 7.30 (s, 2H, 2,6-Ar–H), 7.05 (d, 1H, 50-Ar–H,
J = 8.4), 3.88 (s, 3H, –OCH3), 3.87 (s, 6H, –OCH3), 3.85
(s, 3H, –OCH3); 13C NMR (100 MHz, CDCl3, d ppm):
190.04 (C=O, aldehyde), 164.04 (C=O, ester), 156.61 (C-
60), 153.11 (C-3, C-5), 142.97 (C-4), 140.12 (C-10), 130.25
(C-30), 130.09 (C-40), 123.72 (C-1), 123.66 (C-20), 112.11
(C-50), 107.59 (C-2, C-6), 61 (OCH3), 56.31 (OCH3), 56.33
(OCH3); ESI–MS (m/z): 369.11 (M??Na); Anal. Calcd. %:
(C18H18O7): C, 62.41; H, 5.21. Found (%): C, 62.42;
H, 5.24.
The compound 4i was prepared using phenolic antioxidant
compound 1i (1.56 g, 0.01 mol) by the method as descri-
bed above, that resulted in white solid with 48 % yield: mp
198 °C; IR (KBr): 2951, 2842 (aliphatic C–H st), 1752
(C=O), 1589, 1502 (aromatic C=C st), 1229 (C–O–C asym
1
st), 1129 (C–O–C symm st) cm-1; H NMR (400 MHz,
CDCl3, d ppm): 7.38 (s, 2H, 2,6-Ar–H), 3.95 (s, 3H,
OCH3), 3.92 (s, 6H, OCH3), 3.39 (dt, 1H, 10-H, OCH), 2.15
(m, 1H, 70-H, CHC(CH3)2), 1.94 (m, 2H, CH2), 1.67 (m,
1H, CH), 1.40 (m, 2H, CH2), 1.08 (m, 2H, CH2), 0.96 (m,
6H, 80,90-H, CHC(CH3)2), 0.85 (d, 1H, CH), 0.80 (d, 3H,
123

Med Chem Res
30,40-(Methylenedioxy)phenyl-3,4,5-trimethylgallate (4g)
TMGA was diluted to obtain solutions of concentration in
the range of 2–10 lg/mL. The spectrophotometric absor-
bances were recorded at 236 nm using methanol as blank.
The compound 4g was prepared using potassium salt of
antioxidant compound 1g (1.38 g, 0.01 mol) by the above
described method, which resulted in white solid with 56 %
yield: mp 137 °C; IR (KBr): 2942, 2839 (aliphatic C–H st),
1731 (C=O), 1594, 1497 (aromatic C=C st), 1218 (C–O–C
Linear calibration curve was obtained with slope as 0.0510
1 %
and E
510 (Table 2). Similarly, calibration plots of
1cm
synthesized ester derivatives (4a–4i) were established. The
solubility studies of TMGA and its ester derivatives (4a–4i)
were carried out in phosphate buffer (pH 7.4) (Table 2).
Excess amount of each compound was added to 10 mL of
buffer and shaken for 24 h at 37 2 °C using water bath
shaker. Solutions were filtered and analyzed spectropho-
tometrically for determining the amount of compounds.
asym st), 1131 (C–O–C symm st) cm-1
;
1H NMR
(400 MHz, CDCl3, d ppm): 7.42 (s, 2H, 2,6-Ar–H), 6.82
(d, 1H, J = 8.36, 50-Ar–H), 6.71 (d, 1H, J = 2.3, 20-Ar–H),
6.63 (dd, 1H, 60-Ar–H, J = 2.3, J = 8.36), 6.02 (s, 2H, O–
CH2–O), 3.94 (s, 9H, 3,4,5-OCH3); 13C NMR (CDCl3, d
ppm): 165.14 (C=O), 153.06 (C-3, C-5), 148.12 (C-30),
145.47 (C-40), 145.25 (C-10), 142.80 (C-4), 124.30 (C-1),
114.10 (C-50), 108.06 (C-60), 107.38 (C-2,6), 103.86 (C-20),
101.77 (O–CH2–O), 61 (OCH3), 56.34 (OCH3); ESI–MS
(m/z): 355.1 (M??Na?); Anal. Calcd. %: (C17H16O7): C,
61.42; H, 4.86. Found (%): C, 61.44; H, 4.85.
Partition coefficient determination
Partition coefficients of TMGA and its esters (4a–4i) were
determined in octanol/phosphate buffer (pH 7.4) system
using shake flask method (Table 2). Saturated solutions of
all the compounds were prepared in n-octanol (5 mL) and
equal volumes of phosphate buffer (pH 7.4) were added to
the solutions in conical flasks. The sealed flasks were kept
for shaking in a water bath shaker maintained at 37 2 °C
for 24 h and then allowed to stand for 30 min for both the
phases to fully separate. Thereafter, the respective phases
were analyzed spectrophotometrically.
20-Oxo-20H-chromene-70-yl-3,4,5-trimethylgallate (4h)
The compound 4h was prepared using potassium salt of
antioxidant compound 1h (1.62 g, 0.01 mol) by the above
described method, which resulted in white solid with 51 %
yield: mp 169 °C; IR (KBr): 2932, 2840 (aliphatic C–H st),
1733 (C=O), 1592, 1504 (aromatic C=C st), 1215 (C–O–C
asym st), 1124 (C–O–C symm st) cm-1
;
1H NMR
Chemical stability in aqueous buffer solutions
(400 MHz, CDCl3, d ppm): 7.65 (1H, d, 40-H, lactone ring,
J = 9.6), 7.47 (1H, d, 50-H, J = 8.44), 7.38 (s, 2H, 2,6-Ar–
H), 7.17 (1H, d, 80-H, J = 2.2), 7.11 (1H, dd, 60-H,
J = 8.4, J = 2.2), 6.35 (d, 1H, 30-H, J = 9.6), 3.89 (s, 6H,
OCH3), 3.88 (s, 3H, OCH3); 13C NMR (CDCl3, d ppm):
164.23 (C=O, ester), 160.36 (C=O, lactone ring), 154.80
(C-70), 153.54 (C-80a), 153.16 (C-3, C-5), 142.97 (C-4),
142.88 (C-40), 128.64 (C-50), 123.51 (C-1), 118.65 (C-40a),
116.79 (C-60), 116.18 (C-80), 110.73 (C-30), 107.59 (C-2,
C-6), 61.05 (OCH3), 56.39 (OCH3, C-3, C-5); ESI–MS (m/
z): 357.1 (M??1), 379.1 (M??Na); Anal. Calcd. %:
(C19H16O7): C, 64.42; H, 4.59. Found (%): C, 64.04; H,
4.53.
Degradation rate of the synthesized derivatives (4a–4i) in
aqueous solutions (containing 0.02 % w/v Tween 80) of
pH 1.2 (nonenzymatic simulated gastric fluid, SGF) and
isotonic phosphate buffer of pH 7.4 was determined at
37 °C. The ionic strength of the buffer solutions was
adjusted to 0.5 by addition of a calculated amount of
potassium chloride. The reactions were initiated by adding
250 lL of a methanolic solution (2 9 10-3 M) of the ester
derivatives to 2.5 mL of preheated buffer solutions in
screw-capped test tubes and at appropriate intervals, ali-
quots of 20 lL were withdrawn and analyzed by HPLC for
the residuals.
Physicochemical studies
Pharmacological studies
Solubility studies
Animals
Solutions of TMGA (2) and its ester derivatives were
prepared in methanol. The kmax was determined by scan-
ning solutions containing 20–50 lg/mL of TMGA between
200 and 400 nm using UV–visible spectrophotometer. The
Wistar rats (8–10 weeks old; 200–250 g) of both sexes
procured from central animal house, Panjab University,
Chandigarh, India were used. The animals were housed in
plastic cages under standard laboratory conditions and
maintained on rat chow and water. The experimental pro-
tocols were approved by the Institutional Animal Ethics
Committee (IAEC) and conducted according to the
guidelines of Committee for the Purpose of Control and
k
max of TMGA was found to be 236 nm. Similarly, kmax of
synthesized compounds (4a–4i) was determined and shown
in Table 2.The standard plot for TMGA was constructed in
methanol. The stock solution containing 1 mg/mL of
123

Med Chem Res
Biochemical studies
Supervision of Experiments on Animals (CPCSEA), New
Delhi, India. Unless otherwise stated, the following con-
ditions were employed in all experiments.
Preparation of tissue homogenate: The glandular parts of
excised stomachs from ulcer studies were homogenized in
ice cold phosphate buffer (pH 7.4) with a Potter–Elvehjerr
glass homogenizer for 30 s. The homogenate was centri-
fuged at 800g for 10 min. The supernatant was again
centrifuged at 12,000g for 15 min and the obtained post
mitochondrial fraction (PMF) was used for following
estimations.
In vivo anti-inflammatory activity using carrageenan-
induced rat paw edema assay method
Anti-inflammatory activity was determined using carra-
geenan-induced foot paw edema assay method in rats using
indomethacin as standard drug (Winter et al., 1962). The
test compounds were administered orally at molar equiv-
alent doses of TMBA (20 mg/kg, p.o.). Acute edema was
induced in the left hind paw of rats by injecting freshly
prepared solution of carrageenan (Type IV, 0.1 mL, 1 %)
under plantar region of the left hind paw. In the right paw,
saline (1 mL, 0.9 %) was injected, which served as control
for comparison. The increase in paw volume was measured
by using plethysmometer (water displacement, UGO BA-
SILE, Italy) at 2 and 4 h after carrageenan challenge.
Percentage change in paw volume was calculated and
expressed as the amount of inflammation.
Estimation of lipid peroxidation
The malondialdehyde (MDA) content, a measure of lipid
peroxidation, was assayed in the form of thiobarbituric
acid-reactive substances by the method of Wills (1965). In
brief, 0.5 mL of postmitochondrial supernatant and 0.5 mL
of Tris–hydrochloric acid were incubated at 37 °C for 2 h.
After incubation, 1 mL of 10 % trichloroacetic acid was
added and centrifuged at 1000g for 10 min. To 1 mL of
supernatant, 1 mL of 0.67 % thiobarbituric acid was added
and the tubes were kept in boiling water for 10 min. After
cooling, 1 mL of double distilled water was added and
absorbance was measured at 532 nm. Thiobarbituric acid-
reactive substances were quantified using an extinction
coefficient of 1.56 9 105 M-1 cm-1 and expressed as
nanomole of MDA per milligram of protein. Tissue protein
was estimated using the Biuret method, and the gastric
MDA content expressed as nanomole of MDA per milli-
gram of protein.
Ulcerogenic and antiulcer activity
Wistar rats of either sex were distributed at random in
different groups of six animals each. Animals were treated
with indomethacin (48 mg/kg, p.o.), equimolar doses of
test compounds to indomethacin once daily for four con-
secutive days. The animals were killed under deep ether
anesthesia and stomachs were removed. The abdomen of
each rat was opened through great curvature and examined
under dissecting microscope for lesions or bleedings. The
severity of the mucosal damage (ulcerogenic index) was
calculated by means of scores (Milanino et al., 1988).
Separate groups of rats (n = 6) were used to see the
effect of test compounds and their physical mixtures on
pyloric ligation (PL)-induced gastric mucosal injury. Ani-
mals were divided into different groups: control, pyloric
ligated, PL-induced plus test compounds (4a, 4b, 4g, and
4h; separate group for each test compound), and PL-
induced plus physical mixture of same test compounds
(2 ? 1a, 2 ? 1b, 2 ? 1g, and 2 ? 1h; separate group for
each physical mixture). Gastric ulcers were produced by
ligation of the pyloric end of the rat stomach (Shay et al.,
1945). The abdomen was opened under ether anesthesia
below the xiphoid process; the pyloric portion of the
stomach was slightly lifted and ligated avoiding any
damage to the adjacent blood vessels. Test compounds and
their physical mixtures were administered orally at their
anti-inflammatory doses 1 h prior to pyloric ligation.
Animals were sacrificed 8 h after pyloric ligation and the
stomachs were collected for ulcer scores and biochemical
studies.
Estimation of reduced glutathione (GSH)
Reduced glutathione was assayed by the method of Jollow
et al. (1974). In brief, 1 mL of postmitochondrial super-
natant (10 %) was precipitated with 1 mL of sulfosalicylic
acid (4 %). The samples were kept at 4 °C for at least 1 h
and then subjected to centrifugation at 1,200g for 15 min at
4 °C. The assay mixture contained 0.1 mL supernatant,
2.7 mL phosphate buffer (0.1 M, pH 7.4), and 0.2 mL 5,5-
dithiobis-(2-nitro benzoic acid) (Ellman’s reagent,
0.1 mM, pH 8) in a total volume of 3.0 mL. The yellow
color developed was read immediately at 412 nm, and
GSH levels were calculated using molar extinction coeffi-
cient of 1.36 9 104 M-1 cm-1 and expressed as micromole
per milligram protein.
Estimation of superoxide dismutase activity
Cytosolic superoxide dismutase activity was assayed by the
method of Kono (1978). The assay system consisted of
0.1 mM ethylenediamine tetra-acetic acid, 50 mM sodium
carbonate, and 96 mM of nitro blue tetrazolium. In the
123

Med Chem Res
cuvette, 2 mL of above mixture was taken, and to it,
0.05 mL of postmitochondrial supernatant and 0.05 mL of
hydroxylamine hydrochloride (adjusted to pH 6 with
sodium hydroxide) were added. The auto-oxidation of
hydroxylamine was observed by measuring the change in
optical density at 560 nm for 2 min at 30/60 s intervals.
The superoxide dismutase activity was expressed as units
per milligram protein.
crystallized complex, which reorients side chain hydroxyl
groups and alleviates potential steric clashes. The complex
obtained was minimized using OPLS_2005 force field
(Kaminski et al., 2001) with Polak-Ribiere Conjugate
Gradient (PRCG) algorithm. The minimization was ter-
minated either completion of 5,000 steps (or) after the
energy gradient converged below 0.05 kcal/mol. In the
absence of a co-crystallized complex of inhibitor bound to
COX-1, the binding site was defined using the crystal
coordinates of SC-558, which was superimposed to the
same frame of residues of COX-1 using DaliLite (http://
www.ebi.ac.uk/DaliLite).
Estimation of catalase
Catalase activity was assayed by the method of Claiborne
(1985). In brief, the assay mixture consisted of 1.95 mL
phosphate buffer (0.05 M, pH 7), 1 mL hydrogen peroxide
(0.019 M), and 0.05 mL postmitochondrial supernatant
(10 %) in a final volume of 3 mL. Changes in absorbance
were recorded at 240 nm. Catalase activity was quantified
using the millimolar extinction coefficient of hydrogen
peroxide (0.07 mM) and expressed as micromoles of
hydrogen peroxide decomposed per minute per milligram
protein.
Preparation of ligands
Structures of the ligands (4a–4i) were sketched using built
¨
panel of Maestro (Version 9.2, Schrodinger, 2011) and
¨
taken in.mae format. LigPrep (Version 2.5, Schrodinger,
¨
2011) is a utility of Schrodinger software suit that com-
bines tools for generating 3D structures from 1D (Smiles)
and 2D (SDF) representation, searching for tautomers,
steric isomers and perform a geometry minimization of the
ligands. Molecular Mechanics Force Fields (OPLS_2005)
with default settings were employed for the ligand
minimization.
Statistical analysis
Statistical analysis was carried out using one-way analysis
of variance (ANOVA). In all cases, post hoc comparisons
of the means of individual groups were performed using
Dunnett’s test. A significance level of p \ 0.05 denoted the
significance in all cases.
Docking
Docking studies were carried out using the above men-
tioned prepared proteins of COX-1 (3KK6) and COX-2
(1CX2) and ligands (4a–4i), by employing Glide docking
Docking studies
¨
program (Version 5.7, Schrodinger, 2011) following the
The molecular docking studies were carried out using Dell
Precision workstation T3400 (Intel Core2 Duo Processor,
4 GB RAM, 250 GB hard disk, and Nvidia Quodro FX
4500 graphics card). The docking poses were taken by
reported procedure (Friesner et al., 2004).
Calculation of Prime MM-GBSA descriptors
¨
using GLIDE (Glide, version 5.7, Schrodinger, 2011).
The Prime MM-GBSA approach (Prime version 3.0,
¨
Schrodinger, 2011) is used to predict the free energy of
Preparation of proteins
binding for a receptor and a set of ligands. MM-GBSA is
an acronym for a method that combines OPLS molecular
mechanics energies (EMM), an SGB solvation model for
polar solvation (GSGB), and a nonpolar solvation term
(GNP) composed of the nonpolar solvent accessible surface
area and vdW interactions. The total free energy of binding
is then expressed as:
The crystal structure of COX-1 complexed with celecoxib
(PDB Code: 3KK6) and COX-2 complexed with SC-558
(PDB Code: 1CX2) was taken from the Protein Data Bank
(www.rcsb.org) and prepared using protein preparation
¨
wizard tool (Schrodinger Inc., 2011) which performs the
following steps: assigning of bond orders, addition of
hydrogens, optimization of hydrogen bonds by flipping
amino side chains, correction of charges, and minimization
of the protein complex. All the bound water molecules,
ligand, and cofactors were also removed from the protein.
The tool neutralized the side chains that are not close to the
binding cavity and do not participate in salt bridges. This
step is then followed by restrained minimization of co-
DGbind ¼ Gcomplex ꢁ ðGprotein þ Gligand
where
Þ
G ¼ EMM þ GSGB þ GNP
The ligand in the unbound state is minimized in SGB
solvent, but is not otherwise sampled. In the calculation of
123

Med Chem Res
the complex, the ligand is minimized in the context of the
receptor. The protein is currently held fixed in all calcu-
lations. The following descriptors were generated by the
Prime MM-GBSA approach:
References
Bandyopadhyay U, Das D, Banerjee RK (1999) Reactive oxygen
species—oxidative damage and pathogenesis. Curr Sci 77:
658–665
Bernacki K, Kalyanaraman C, Jacobson MP (2005) Virtual ligand
screening against Escherichia coli dihydrofolate reductase:
improving docking enrichment physics-based methods. J Biomol
Screen 10:675–681
MM-GBSA_DGbind Ligand binding energy, DGbind
MM-GBSA_E_complex Energy of the complex,
Gcomplex
Bhosle D, Bharambe S, Garrola N, Dhaneshwar SS (2006) Mutual
prodrug concept: fundamentals and applications. Indian J Pharm
Sci 68:286–294
Chandrasekharan NV, Dai H, Roos KL, Evanson NK, Tomsik J, Elton
TS, Simmons DL (2002) COX-3, a cyclooxygenase-1 variant
inhibited by acetaminophen and other analgesic/antipyretic
drugs: cloning, structure, and expression. Proc Natl Acad Sci
USA 99:13926–13931
Claiborne A (1985) Catalase activity. In: Greenwald RA (ed)
Handbook of methods for oxygen radical research. CRC, Boca
Raton, pp 283–284
Clark DE, Pickett SD (2000) Computational methods for the
prediction of ‘‘drug-likeness’’. Drug Discov Today 5:49–58
Costa G, Francisco V, Lopes CM, Cruz TM, Batista TM (2012)
Intracellular signaling pathways modulated by phenolic com-
pounds: application for New Anti-Inflammatory Drugs Discov-
ery. Curr Med Chem 19(18):2876–2900
MM-GBSA_E_protein Energy of the receptor without
the ligand, Gprotein
MM-GBSA_E_ligand Energy of the unbound ligand,
Gligand
To set up the calculation, pose viewer file (generated
after docking with Glide) was used to consider the receptor
and source of ligands. After choosing the receptor and
ligands, by using pose viewer file (pv.maegz), the program
Prime MM-GBSA that generate the descriptors were run
with default options that were chosen to produce reasonable
descriptors. The MM-GB/SA scoring along with the
experimental binding affinities data of COX-1 and COX-2
is presented in Table S1 (Supplementary Section).
Cotelle N (2001) Role of flavonoids in oxidative stress. Curr Top Med
Chem 1:569–590
Calculation of QikProp descriptors for prediction
of ADME properties
Dogne JM, Supuran CT, Pratico D (2005) Adverse cardiovascular
effects of coxibs. J Med Chem 48:2251–2257
Friesner RA, Banks JL, Murphy RB, Halgren TA, Klicic JJ, Mainz
DT, Repasky MP, Knoll EH, Shelley M, Perry JK, Shaw DE,
Francis P, Shenkin PS (2004) Glide: a new approach for rapid,
accurate docking and scoring. 1. Method and assessment of
docking accuracy. J Med Chem 47(7):1739–1749
Gibson T (1988) Nonsteroidal antiinflammatory drugs—another look.
Br J Rheumatol 27:87–90
QikProp is a quick, accurate, easy-to-use absorption,
distribution, metabolism, and excretion (ADME) predic-
tion program design to produce certain descriptors related
to ADME. QikProp predicts physically significant
descriptors and pharmaceutically relevant properties of
organic molecules, either individually or in batches.
QikProp has two modes: normal mode and fast mode. In
fast mode, certain time-consuming calculations are
omitted, some properties are not predicted, and some have
different values. In this study, QikProp was run in normal
processing mode with default options (QikProp version
¨
Glide, version 5.7, Schrodinger, LLC, New York, NY, 2011
Hairsine P (1989) Direct measurement of partition coefficient by
shake flask technique and gradient elution HPLC. Lab Pract
38:73–75
Hawkey CJ (1999) COX-2 inhibitors. Lancet 353:307–314
Huang N, Kalyanaraman C, Bernacki K, Jacobson MP (2006a)
Molecular mechanics methods for predicting protein-ligand
binding. Phys Chem Chem Phys 8:5166–5177
Huang N, Kalyanaraman C, Irwin JJ, Jacobson MP (2006b) Physics-
based scoring of protein ligand complexes: enrichment of known
inhibitors in large-scale virtual screening. J Chem Inf Model
46:243–253
Jollow DJ, Mitchell JR, Zampaglione N, Gillette JR (1974) Bromo-
benze induced liver necrosis: protective role of glutathione and
evidence for 3,4-bromobenzenoxide as the hepatotoxic metab-
olite. Pharmacology 11:151–169
Kaminski GA, Friesner RA, Tirado-Rives J et al (2001) Evaluation
and reparametrization of the OPLS-AA force field for proteins
via comparison with accurate quantum chemical calculations on
peptides. J Phys Chem B 105:6474–6487
Kono Y (1978) Generation of superoxide radical during autoxidation
of hydroxylamine and an assay for superoxide dismutase. Arch
Biochem Biophys 186:189–195
¨
3.4, Schrodinger, 2011; Ravindranathan et al., 2010). To
set up the calculation, pose viewer file (generated after
docking with Glide) was used to consider the receptor and
source of ligands. After choosing the receptor and ligands,
by using pose viewer file (pv.maegz), the program Qik-
Prop that generate the descriptors were run with default
options that were chosen to produce reasonable descrip-
tors. The selected properties that are known to influence
metabolism, cell permeation, and bioavailability are pre-
sented in Table 6.
Acknowledgments The authors M.S.D. and P.K.D. are thankful to
UGC and CSIR New Delhi, India for financial support in the form of
Research for Meritorious Students (RFMS) Scheme and Senior
Research Fellowship (SRF). Authors profusely thank to Avtar Singh,
SAIF (CIL), PU for carrying out the NMR studies. Special tribute to
Late Professor Pritam Dev Sharma, one of the investigators.
Kroes BH, van den Berg AJ, Quarles van Ufford HC, van Dijk H,
Labadie RP (1992) Anti-inflammatory activity of gallic acid.
Planta Med 58:499–504
Kuhn B, Kollman PA (2000) Binding of a diverse set of ligands to
avidin and streptavidin: an accurate quantitative prediction of
123

Med Chem Res
¨
Protein Preparation Wizard; Epik version 2.2, Schrodinger, LLC,
New York, NY, 2011
QikProp, version 3.4, Schrodinger, LLC, New York, NY, 2011
Rastogi L, Patnaik GK, Dikshit M (1998) Pylorus ligation induced
gastric mucosal injury in rats. Pharmacol Res 38:125–132
Ravindranathan KP, Mandiyan V, Ekkati AR, Bae JH, Schlessinger J,
Jorgensen WL (2010) Discovery of novel fibroblast growth
factor receptor 1 kinase inhibitors by structure-based virtual
screening. J Med Chem 53:1662–1672
their relative affinities by a combination of molecular mechanics
and continuum solvation models. J Med Chem 43:3786–3791
Leppanen J, Huuskonen J, Nevalainen T, Gynther J, Taipale H,
Jarvinen T (2002) Design and synthesis of a novel L-dopa
entacapone codrug. J Med Chem 45:1379–1382
Li X, Li Y, Xu W (2006) Design, synthesis, and evaluation of novel
galloyl pyrrolidine derivatives as potential anti- tumor agents.
Bioorg Med Chem 14:1287–1293
¨
¨
LigPrep, version 2.5, Schrodinger, LLC, New York, NY, 2011
Lipinski CA, Lombardo F, Dominy BW, Feeney PJ (2001) Exper-
imental and computational approaches to estimate solubility and
permeability in drug discovery and development settings. Adv
Drug Deliv Rev 46:3–26
Lu Z, Nie G, Balton PS, Tang H, Zaho B (2006) Structure–activity
relationship analysis of antioxidant ability and neuroprotective
effect of gallic acid derivatives. Neurochem Int 48:263–274
Lyne PD, Lamb ML, Saeh JC (2006) Accurate prediction of the
relative potencies of members of a series of kinase inhibitors
using molecular docking and MM-GBSA scoring. J Med Chem
49:4805–4808
Ruan K, Yang S, Van Sant KA, Likos JJ (2009) Application of
Hadamard spectroscopy to automated structure verification in
high-throughput NMR. Magn Reson Chem 47(8):693–700
Sawraj S, Bhardawaj TR, Sharma PD (2012) Design, synthesis, and
evaluation of novel indomethacin–antioxidant codrugs as gas-
trosparing NSAIDs. Med Chem Res 21:834–843
Schnitzer TJ (2001) COX-2-specific inhibitors: are they safe? J Am
Med 110:46–49
Shay M, Kamarov SA, Fels D, Meraaze D, Gruenistein H, Siplet H
(1945) A simple method for the uniform production of gastric
ulceration in the rat. Gastroenterology 5:43–61
¨
MacroModel, version 9.9, Schrodinger, LLC, New York, NY, 2011
Singh G, Sharma PD (1994) Mutual prodrugs—a recent trend in
prodrug design. Indian J Pharm Sci 56:69–79
Still WC, Tempczyk A, Hawley RC, Hendrickson TA (1990) General
treatment of solvation for molecular mechanics. J Am Chem Soc
112:6127
Madhukar M, Sawraj S, Sharma PD (2010) Design, synthesis and
evaluation of mutual prodrug of 4-biphenylacetic acid and
quercetin tetramethyl ether (BPA-QTME) as gastrosparing
NSAID. Eur J Med Chem 45:2591–2596
¨
Maestro, version 9.2, Schrodinger, LLC, New York, NY, 2011
Streng WH (ed) (2001) Characterization of compounds in solution-
theory and practice. Kluwer Academic Plenum, New York
Vane JR (1971) Inhibition of prostaglandin synthesis as a mechanism
of action for aspirin-like drugs. Nat New Biol 231:232–235
Vane JR, Botting RM (1998) Anti-inflammatory drugs and their
mechanism of action. Inflamm Res 47:S78–S87
Martin MJ, La-casa C, Alarcon-de-la-Lastra C, Cabeza J, Villegas J,
Motilva Y (1998) Anti-oxidant mechanisms involved in gastro-
protective effects of quercetin. Z Naturforsch C 53:82–88
Mathieu A, Blum J (1981) Eugenol derivatives useful in medicine. Fr.
Demande, Patent 2466458
Mc Carthy MD (1989) Nonsteroidal antiinflammatory drug-induced
ulcers: management by traditional therapies. Gastroenterology
96:662–674
Milanino R, Concari E, Conforti A, Marrella M, Franco L, Moretti U,
Velo G, Rainsford KD, Bressan M (1988) Synthesis and anti-
inflammatory effect of some bis (2-benzimidazolyl) thiolthers
and their cooper (II) chelates, orally administered to rat. Eur J
Med Chem 23:217–224
Orrenius S, Gogvadze V, Zhivotovsky B (2007) Mitochondrial
oxidative stress: implications for cell death. Annu Rev Pharma-
col Toxicol 47:143–183
Pal C, Bindu S, Dey S, Alam A, Goyal M, Iqbal MS, Maity P,
Adhikari SS, Bandyopadhyay U (2010) Gallic acid prevents
nonsteroidal anti-inflammatory drug-induced gastropathy in rat
by blocking oxidative stress and apoptosis. Free Radic Biol Med
49(2):258–267
Wadhwa LK, Sharma PD (1995) Glycolamide esters of 6-methoxy-2-
naphthylacetic acid as potential prodrugs—physicochemical
properties, chemical stability and enzymatic hydrolysis. Int J
Pharm 118:31–39
Warner TD, Giuliano F, Vojnovic I, Bukada A, Mitchell JA, Vane JR
(1999) Nonsteroidal drug selectivities for cyclooxygenase-1
rather than cyclooxygenase-2 are associated with human gastro-
intestinal toxicity—a full in vitro analysis. Proc Natl Acad Sci
USA 96:7563–7568
Wills ED (1965) Mechanisms of lipid peroxide formation in tissues
role of metals and haematin proteins in the catalysis of the
oxidation of unsaturated fatty acids. Biochim Biophys Acta
98:238–251
Winter CA, Risley GA, Nuss GW (1962) Carrageenan-induced edema
in hind paw of the rat as an assay for antiinflammatory drugs.
Proc Soc Exp Biol Med 111:544–547
Pepe RO (1951) Preparation of trimethylgallates of methoxyphenols
by methylation of trimethylgallic monoesters of diphenols.
Archivos de Farmacia y Bioquimica del Tucuman 5:297–300
Perron NR, Brumaghim JL (2009) A review of the iron-binding
mechanism for polyphenol antioxidant activity. Cell Biochem
Biophys 53:75–100
Pontiki E, Hadjipavlou-Litina D, Litinas K, Nicolotti O, Carotti A
(2011) Design, synthesis and pharmacobiological evaluation of
novel acrylic acid derivatives acting as lipoxygenase and
cyclooxygenase-1 inhibitors with antioxidant and anti-inflam-
matory activities. Eur J Med Chem 46:191–200
Wolfe MM, Lichtenstein DR, Singh G (1999) Gastrointestinal
toxicity of non steroidal antiinflammatory drugs. N Engl J Med
340:1888–1899
Xie W, Robertson DL, Simmons DL (1992) Mitogen-inducible
prostaglandin G/H synthase: a new target to be a potential source
of inhibitors of NF-jB activation. Drug Dev Res 25:249–265
Yalkowsky SH, Morozowich W (1980) In: Ariens EJ (ed) Drug
design IX. Academic Press, London, pp 121–185
Yalkowsky SH, Morzowich W, Banerjee S (eds) (1992) Aqueous
solubility-methods of estimation for organic compounds. Marcel
Dekker, New York
¨
Prime, version 3.0, Schrodinger, LLC, New York, NY, 2011
123