Synthesis and biological evaluation of phenyl-1H-1,2,3-triazole derivatives as anti-inflammatory agents

Article abstract of DOI:10.1016/j.bioorg.2015.01.003

Rapid and efficient synthesis of a phenyl-1H-1,2,3-triazole library enabled cost-effective biological testing of a range of novel non-steroidal anti-inflammatory drugs with potential for improved drug efficacy and toxicity profiles. Anti-inflammatory acti

Full text of DOI:10.1016/j.bioorg.2015.01.003

Accepted Manuscript
Synthesis and biological evaluation of phenyl-1H-1,2,3-triazole derivatives as
anti-inflammatory agents
Tae Woo Kim, Yeonjoong Yong, Soon Young Shin, Hyeryoung Jung, Kwan
Ha Park, Young Han Lee, Yoongho Lim, Kang-Yeoun Jung
PII:
S0045-2068(15)00004-8
DOI:
http://dx.doi.org/10.1016/j.bioorg.2015.01.003
YBIOO 1783
Reference:
Bioorganic Chemistry
To appear in:
Received Date:
23 September 2014
Please cite this article as: T.W. Kim, Y. Yong, S.Y. Shin, H. Jung, K.H. Park, Y.H. Lee, Y. Lim, K-Y. Jung, Synthesis
and biological evaluation of phenyl-1H-1,2,3-triazole derivatives as anti-inflammatory agents, Bioorganic
Chemistry (2015), doi: http://dx.doi.org/10.1016/j.bioorg.2015.01.003
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis and biological evaluation of phenyl-1H-1,2,3-triazole derivatives as anti-
inflammatory agents
Tae Woo Kim1,†, Yeonjoong Yong2,†, Soon Young Shin3, †, Hyeryoung Jung2, Kwan Ha
Park4, Young Han Lee3, Yoongho Lim2,* and Kang-Yeoun Jung1,*
1Department of Biochemical Engineering, Gangneung-Wonju National University, Gangwon
210-702, Korea
2Division of Bioscience and Biotechnology, BMIC, Konkuk University, Seoul 143-701, Korea
3Department of Biological Sciences, Konkuk University, Seoul 143-701, Korea
4Department of Aquatic Life Medicine, Kunsan National University, Kunsan 573-701, Korea
†These authors contributed equally to this work.
*Corresponding authors.
Prof. Yoongho Lim
Division of Bioscience and Biotechnology, Konkuk University
Hwayang-Dong 1, Kwangjin-Ku, Seoul, Korea 143-701
phone : 82-2-450-3760
fax : 82-2-454-3760
Email: yoongho@konkuk.ac.kr
Prof. Kang-Yeoun Jung
Department of Biochemical Engineering, Gangneung-Wonju National University
1

Gangwon 210-702, Korea
phone : 82-33-640-2403
Email: kyjung@gwnu.ac.kr
2

ABSTRACT
Rapid and efficient synthesis of a phenyl-1H-1,2,3-triazole library enabled cost-effective
biological testing of a range of novel non-steroidal anti-inflammatory drugs with potential for
improved drug efficacy and toxicity profiles. Anti-inflammatory activities of the phenyl-1H-
1,2,3-triazole analogs synthesized in this report were assessed using the xylene-induced ear
edema model in mice. At least four analogs, 2a, 2b, 2c, and 4a, showed more potent effects
than the reference anti-inflammatory drug diclofenac at the same dose of 25 mg/kg. To
explore relationships between the structural properties of phenyl-1H-1,2,3-triazole analogs
and their anti-inflammatory activities in xylene-induced ear edema, comparative molecular
field analysis was performed, and pharmacophores showing good anti-inflammatory activities
were identified based on an analysis of contour maps obtained from comparative molecular
field analysis. The anti-inflammatory effect on the molecular level was tested by the
expression of tumor necrosis factor-alpha induced COX-2 using Western blots. Because the
addition of the analog 2c caused the expression change of TNF-α induced COX-2, the
molecular binding mode between 2c and COX-2 was elucidated using in silico docking.
key words: diclofenac; anti-inflammatory drug; NSAID; QSAR; in silico docking; COX-2
3

1. Introduction
Non-steroidal anti-inflammatory drugs (NSAIDs) are among the most widely used
drugs in the world, accounting for 35% of the global market for prescription pain medications
[1,2]. Pharmacological activity of NSAIDs results from the suppression of prostaglandin H2
(PGH2) synthesis. PGH2 mediates many biological functions and is synthesized from
arachidonic acid by cyclooxygenase (COX) enzymes. NSAIDs operate by inhibiting COX
enzymes, which limits the production of PGH2 and ultimately results in anti-inflammatory,
analgesic, and antipyretic effects [3]. Diclofenac, a nonselective benzene acetic acid
derivative NSAID, has been prescribed for the treatment of chronic and acute pain for more
than 30 years [4]. Unfortunately, PGH2 also mediates other functions such as gastrointestinal
cytoprotection and nonselective COX enzyme inhibition, and has been associated with
bleeding, ulceration, and perforation [5]. The discovery that a second COX enzyme, referred
to as COX-2, is the primary site of pro-inflammatory PG synthesis led to the development of
COX-2 selective NSAIDs with reduced toxicity [5]. COX-2 selective inhibitors entered the
market in 1999 with the introduction of celecoxib (sold as Celebrex), followed by many other
COX-2 selective inhibitors [6] However, Vioxx was voluntarily withdrawn from the market
in 2004 due to concerns regarding adverse cardiovascular effects [7]. Subsequently, Bextra,
another COX-2 selective inhibitor, was withdrawn for similar reasons [8]. Because of the
tremendous success and the potential for severe adverse effects of COX-2 selective
inhibitors, many studies have explored novel compounds with improved toxicity profiles. In
particular, due to the success of diclofenac as a nonselective COX inhibitor, multiple efforts
have been made to develop COX-2 selective analogs [9]. Lumiracoxib, marketed under the
name Prexige until it was also withdrawn from the market due to hepatotoxicity, showed the
4

highest COX-2 selectivity of any NSAID [10]. Relatively minor changes in chemical
composition, such as the differences between diclofenac and lumiracoxcib (Fig. 1),
significantly affect COX selectivity, anti-inflammatory activity, and toxicity.
Although small variations in NSAID analogs can have profound effects, the shape of
the binding pockets of COX enzymes dictates that the basic structure of COX-2 selective
inhibitors are similar [5]. Thus, rapid and efficient methods to synthesize a variety of
similarly shaped analogs will enable cost-effective development and testing of a library of
potentially important compounds. The approach presented in this report uses 2-aminobenzyl
alcohol as a starting material. Rather than connecting a second aromatic or other moiety
through the single nitrogen, the nitrogen is converted to an azide, which is then converted to a
triazole using a click chemistry azide alkyne cycloaddition reaction. Connecting the second
moiety through a triazole moiety rather than through a single nitrogen facilitates efficient
synthesis of a variety of diclofenac analogs. The stability of triazole products also allows for
oxidation and Arbuzov reactions on the alcohol moiety, further increasing the variety of
products that can be synthesized using this approach.
To facilitate the discovery of novel NSAIDs with improved drug efficacy and better
toxicity profiles than diclofenac, we used this approach to synthesize a series of phenyl-1H-
1,2,3-triazole phosphonate and carboxylic acid derivatives (Fig. 1). Commercially available
2-aminobenzyl alcohol was used as a starting material to produce the triazole phosphonate
and carboxylic acid derivatives (3a–h). Oxidation of the alcohol precursors produced
relatively pure carboxylic acid derivatives in seven of eight reactions. Compound 2d was
unique in that it yielded a cyclic amide product (3d). During the synthesis of benzylic
phosphonates and carboxylic acids utilizing conventional heating, higher reaction
temperatures led to the formation of more minor products. When microwave radiation was
5

used, only phosphonate and carboxylic acid derivatives were formed, even at higher
temperatures.
Anti-inflammatory activities of the 24 phenyl-1H-1,2,3-triazole analogs synthesized
in this report were assessed using the xylene-induced ear edema model in mice [11]. To
explore relationships between the structural properties of phenyl-1H-1,2,3-triazole analogs
and their anti-inflammatory activities in xylene-induced ear edema, comparative molecular
field analysis (CoMFA) was performed. The pharmacophores with good anti-inflammatory
activities were determined based on the analysis of contour maps obtained from CoMFA. To
elucidate the anti-inflammatory effect on the molecular level, the expression of tumor
necrosis factor-alpha (TNF-α) induced COX-2 was analyzed by using Western blots for the
analog 2c showing the best anti-inflammatory activity in xylene-induced ear edema. Because
the addition of the analog 2c caused the expression change of TNF-α induced COX-2, the
molecular binding mode between 2c and COX-2 was investigated using in silico docking.
Several analogs synthesized here showed effects more potent than the reference anti-
inflammatory drug diclofenac at the same dose, and our results can be used to develop novel
anti-inflammatory drugs.
2. Results and discussion
2.1. Chemistry
By replacing the nitrogen bridge found in diclofenac and most diclofenac analogs
with a triazole moiety, we rapidly and efficiently developed a library of phenyl-1H-1,2,3-
triazole analog NSAID candidates. All compounds in Table 1 were synthesized in good
6

yields from commercially available 2-aminobenzyl alcohol using click chemistry and
Arbuzov reactions. With the exception of compound 3d, which is discussed below, all
carboxylic acid derivatives were obtained in yields of 84% or higher. Importantly, the
composition of the substituent on the triazole had very little effect on the yield. The yield for
each of the Arbuzov reactions was lower than for the corresponding oxidation with the
exception of 2d and 3d. However, yields of the oxidation and phosphonate additions were not
correlated. For example, compound 2b had the highest yield for oxidation and the lowest
yield for phosphonate addition.
2.2. Biological studies
2.2.1. In vivo anti-inflammatory studies
Of the 24 compounds synthesized and tested for anti-inflammatory activity, 14
showed inhibitory action in both edematous increases of ear thickness and ear weight in mice.
Four compounds, 2a, 2b, 2c, and 4a, showed effects more potent than the reference anti-
inflammatory drug diclofenac at a dose of 25 mg/kg. Activities of 10 compounds were
similar to that of diclofenac. Inhibitory activities of test compounds on xylene-induced ear
edema in mice are listed in Table 2 and their graph containing error bars with standard
deviations of the compounds tested in this research is shown in Fig. 2.
2.2.2. Western blot analysis
7

In this report, the inhibitory effects of 24 phenyl-1H-1,2,3-triazole analogs on the
xylene-induced increase in ear weight of mice were tested to examine anti-inflammatory
activities. It is well established that tumor necrosis factor-alpha (TNF-α), a major
inflammatory cytokine, up-regulates COX-2 expression via nuclear factor kappa-light-chain-
enhancer of activated B cells (NF-κB) activation, which in turn restrict NF-κB activity by a
negative feed-back mechanism. Thus, when COX-2 is inhibited by diclofenac, TNF-α-
induced COX-2 expression is strongly enhanced due to increased NF-κB activity [12]. To
confirm the effect of analog 2c on the inhibition of COX-2, we determined COX-2
expression following TNF-α in the absence or presence of analog 2c. Diclofenac was used as
a positive control. As expected, diclofenac alone did not have any significant effect on the
accumulation of COX-2 protein, however, enhanced TNF-α-induced accumulation of COX-2
protein (Fig. 3). Treatment with TNF-α plus compound 2c more strongly accumulated COX-
2 protein than that treated with TNF-α plus diclofenac.
2.3. quantitative structure–activity relationships
The common characteristic of 23 analogs was a 1-phenyl-1H-1,2,3-triazole
component (Table 1). The X-ray crystallographic structure of N-[(4-phenyl-1H-1,2,3-triazol-
1-yl)acetyl]-beta-D-glucopyranosylamine containing phenyl-1H-1,2,3-triazole was deposited
in the Protein Data Bank (PDB) as an inhibitor of glycogen phosphorylase (2PYI.pdb) [13].
Other groups, excluding phenyl-1H-1,2,3-triazole, were deleted using the Sybyl 7.3 program
(Tripos, St. Louis, MO). The initial structures of phenyl-1H-1,2,3-triazole analogs were
constructed based on the crystallographic structure of phenyl-1H-1,2,3-triazole and were
subjected to energy minimization. Since rotational bonds exist, systematic conformational
8

searches were performed using the grid search method provided by the Sybyl program.
During the grid search, the selected bonds were rotated from 0° to 360° in 15° increments. At
each increment, the conformation was minimized using the Tripos force field and Gästeiger–
Hückel charge. The maximum iteration for the minimization was set to 1000. The
minimization process was terminated at the convergence criteria of the total energy (0.05
kcal/mol∙Å) [14].
The anti-inflammatory activities in a xylene-induced ear edema model were used as
biological data for QSAR calculations. In addition, the inhibition values for xylene-induced
weight increases in the ear were employed. The biological test was performed on 24
compounds synthesized in this report. Since one compound, 4d, was not testable since it
resulted in animal death within 15 min after administration, 23 compounds were used for
QSAR calculations.
In total, 23 compounds were randomly divided into two groups: a training set to
create QSAR models and a test set to validate the models. The test set (derivatives 4c, 3f, 2d,
and 2g in Table 1) was chosen using hierarchical clustering analysis, in which structurally
similar analogs were grouped [15]. The majority of derivatives chosen for the test set
belonged to the separated structural groups (Suppl. Fig. 1). Therefore, our test set could be
used to validate whether the QSAR models were reliable.
In this report, to explore relationships between the structural properties of phenyl-
1H-1,2,3-triazole analogs and their anti-inflammatory activities in xylene-induced ear edema,
CoMFA was performed, which depends on the relative positioning of the compounds in the
fixed lattice. Therefore, alignments of compounds contained in the training set were
important. Phenyl-1H-1,2,3-triazole analogs selected for a training set were put in a 2.0 Å
grid and aligned using the Sybyl/DATABASE alignment module, where the atom-based root-
9

mean-square-fit method was adapted. Compound 2-[1-{2-(hydroxymethyl)phenyl}-1H-1,2,3-
triazol-4-yl]butan-2-ol (2c) showing the most potent inhibitory activity was used as a
template. As shown in Suppl. Fig. 2, all analogs aligned well. Several CoMFA models were
generated using the Sybyl program, and the model showing the best cross-validated
correlation coefficient (q2 = 0.599) with a region-focusing method with a weight of 0.3 was
chosen.
To establish a linear relationship between the biological activity and the resulting
field matrix of the compounds, partial least-squares (PLS) analysis was performed. The cross-
validated analysis was performed using the leave-one-out (LOO) method [16]. The final non-
cross-validated (r2 = 0.956) analysis was performed using the optimal number of components,
6. For PLS analysis, the standard error of estimate and F value were 7.239 and 43.078,
respectively. To examine the CoMFA model, the inhibitory activities of xylene-induced
increases in ear weight of the training set were predicted and compared to the experimental
data (Suppl. Table 1). A plot of the predicted values against the experimental data is shown in
Suppl. Fig. 3. The residual values between the experimental and predicted values for the
training set ranged from 0.4% to 21.8% (Suppl. Table 1).To validate the CoMFA model, four
compounds were selected as a test set based on hierarchical clustering analysis. The residual
values between the experimental data and the values predicted from CoMFA ranged between
26.2% and 47.6%. Since the biological data were obtained from animal experiments, this
error range is acceptable.
To visualize relationships between the structural properties of phenyl-1H-1,2,3-
triazole analogs and their inhibitory effects of the xylene-induced increase in ear weight,
CoMFA contour maps were generated using Sybyl 7.3 and presented as contour plots around
2c (Fig. 4). The steric and electrostatic field descriptors contributed 39.4% and 60.6%,
10

respectively. For the steric field, the sterically bulky favored region contributed 20% and the
disfavored region contributed 80%. In the CoMFA contour plots, the steric field contours are
shown in green (more bulk-favored) and yellow (less bulk-favored), while the electrostatic
field contours are shown in red (electronegative substituent-favored) and blue (electropositive
substituent-favored). The electronegative favored region contributed 30% and the
electropositive favored region contributed 70%. Since the 1-methyl position of 2-butanol
connected to the 2-triazole component of 2c, the presence of yellow contours suggested that
this is a less bulky-favored steric region. Although the bulky-favored regions are shown
around 2-butanol, the characteristics of less bulky-favored regions are predominant because
the sterically bulky favored region contributed 20% and the disfavored region contributed
80%. At the ortho position of 1-o-tolyl-1H-1,2,3-triazole, the green contour suggested that
the bulky favored region is adventitious for the inhibitory activity compared with 3 and 4
series compounds. Comparing analogs 2a–h, electronegative groups contribute more
inhibitory effects; the average activity of derivatives 2a, 2b, 2f, 2c, and 2h was 60.1, and that
of derivatives 2e, 2d, and 2g was 117.2. The substituent of the ortho position of 1-o-tolyl-1H-
1,2,3-triazole did not favor electronegative groups. Comparing derivatives 4a–h, 3a–h, and
2a–h, derivatives 3a–h with carboxyl group showed lower inhibitory effects. Compared with
functional groups at the ortho position of 1-phenyl-1H-1,2,3-triazole and 2-(1H-1,2,3-
triazole-4-yl)butan-2-ol, the effect of the less bulky functional group at 2-(1H-1,2,3-triazole-
4-yl)butan-2-ol was dominant, excluding 2f, 3f, and 4f. Compared with the dependence of
the fluorine position, the inhibition effect was not consistent, although the para position of
fluorine was introduced to prevent drug oxidation. In addition, compared with 2e, 2f, 3e, 3f,
4e, and 4f based on differences in the functional groups, 2f, 3f, and 4f were favored as a
substituent to induce more electrostatic fields, which is shown in red (Fig. 4).
11

Pharmacophores obtained from the CoMFA contour maps showing good inhibitory
effects on xylene-induced increases in ear weight are summarized in Fig. 5. The
electronegative favored region at the ortho position of the phenyl ring contained in 1-phenyl-
1H-1,2,3-triazole, the electropositive and bulky favored regions at the meta position, and the
electronegative and less bulky regions at the C-4 position of triazole are important for
activity.
2.4. Molecular docking
Because Western blot analysis demonstrated phenyl-1H-1,2,3-triazole analog 2c
causes the expression of TNF-α induced COX-2 expression, the molecular binding mode
between 2c and COX-2 was investigated using in silico docking study on an Intel Core 2
Quad Q6600 (2.4 GHz) Linux PC workstation with Sybyl7.3 software (Tripos). The 3D
structure of mouse COX-2 was found in the PDB as 1PXX.pdb, which consisted of a
homotetramer and four ligands including diclofenac [17]. Since four polypeptide chains of a
homotetramer were identical, chain A (1PXX-A) was chosen for in silico docking.
Diclofenac was removed using the Sybyl program and apo-1PXX-A was obtained, whose
solution structure was determined using energy minimization. To confirm whether deletion of
declofenac disturbed the 3D structure of the protein, apo-1PXX-A was compared with the
crystallographic structure of 1PXX-A. The root-mean-square deviation value between 1PXX-
A and apo-1PXX-A was 0.35 Å, and the apo-protein was used for docking experiments. For
the flexible docking procedure, the docking radius was set to 2.1 Å. When diclofenac was
docked into apo-1PXX-A, the binding energy was –14.59 kcal/mol. The binding site of
diclofenac was analyzed using the LigPlot program [15]. Ten residues within the binding site
12

of diclofenac analyzed using LigPlot included Tyr348, Val349, Leu352, Leu384, Trp387,
Met522, Gly526, Ala527 (hydrophobic interactions), Tyr385, and Ser530 (H-bonds). Three
H-bonds were observed between two oxygens of the carboxylic group of diclofenac and
hydroxyl groups of Ser530 and Tyr385, with distances of 2.65 Å, 2.91 Å, and 2.73 Å,
respectively (Fig. 6). Among the 30 complexes obtained from the 30 iteration docking
process, complex 1PXX-A–2c showing the best binding energy (–11.22 kcal/mol) was
chosen. Ten residues surrounding analog 2c included Arg120, Tyr348, Val349, Tyr355,
Trp387, Phe518, Met522, Ala527 (hydrophobic interactions), Tyr385, and Ser530 (H-bonds).
Unlike complex 1PXX-A–diclofenac, two H-bonds were observed between hydroxyl group
of 2c and hydroxyl groups of Ser530 and Tyr385, with distances of 3.28 Å and 2.75 Å,
respectively (Fig. 6). By comparing complex 1PXX-A–diclofenac with that of 1PXX-A–2c,
seven residues including Tyr348, Val349, Tyr385, Trp387, Met522, Ala527, and Ser530
were found in both binding sites. The 3D image of the 1PXX-A–diclofenac complex was
constructed using PyMOL software (The PyMOL Molecular Graphics System, version 1.0r1;
Schrödinger, LLC, Boston, MA). Three H-bonds between two oxygens of the carboxylic
group of diclofenac and hydroxyl groups of Ser530 and Tyr385 are shown in Fig. 7A, and
their distances are 2.65 Å, 2.91 Å, and 2.73 Å, respectively. Likewise, the 3D image of the
1PXX-A–2c complex was constructed. Two H-bonds between hydroxyl group of 2c and
hydroxyl groups of Ser530 and Tyr385 are shown in Fig. 7B, and their distances are 3.28 Å
and 2.75 Å, respectively.
3. Conclusion
13

A series of phosphonate and carboxylic acid analogs of phenyl-1H-1,2,3-triazole
were synthesized using click chemistry and Arbuzov reactions. The substituents on the
triazole had a limited impact on the yield of the oxidation and phosphonation reactions,
indicating that this procedure can be used to synthesize a wide variety of phenyl-1H-1,2,3-
triazole analogs. Reaction temperature affected the ratio of major product to minor product,
with higher temperatures resulting in more minor product. When microwave heating was
used, even at higher temperatures, only the phosphonate and carboxylic acid derivatives were
produced. Synthesis of the carboxylic acid derivative of compound 2d was unique, resulting
in a cyclic amide in good yield. Pharmacophores obtained from CoMFA contour maps
showing good anti-inflammatory activities included an electronegative favored region at the
ortho position of the phenyl ring containing 1-phenyl-1H-1,2,3-triazole, an electropositive
and bulky favored region at the meta position, and an electronegative and less bulky region at
the C-4 position of triazole. The anti-inflammatory effect on the molecular level was
evaluated by the expression of TNF-α induced COX-2 using Western blots for the analog 2c
showing the best activity in xylene-induced ear edema. Because the addition of the analog 2c
enhanced TNFα-induced COX-2 expression, the molecular binding mode between 2c and
COX-2 was investigated using in silico docking. Four compounds, 2a, 2b, 2c, and 4a,
showed more potent effects than the reference anti-inflammatory drug diclofenac at the same
dose.
4. Experimental
4.1. General
14

For preparation of phenyl-1H-1,2,3-triazole analogs dichloromethane and Et3N
(Sigma-Aldrich, St. Louis, MO) were distilled from CaH2 (Sigma-Aldrich, St. Louis, MO)
immediately prior to use. All nonaqueous reactions were conducted in flame-dried glassware
under a nitrogen atmosphere with magnetic stirring. Nuclear magnetic resonance (NMR)
spectra were obtained on a Lamda 300 spectrometer (JEOL, Tokyo, Japan) and recorded at
1
300 MHz for H (75 MHz for 13C) with CDCl3 (Sigma-Aldrich, St. Louis, MO) as solvent
and (CH3)4Si(1H) (Sigma-Aldrich, St. Louis, MO) or CDCl3 (13C, 77.0 ppm) as internal
standards unless otherwise noted. All 31P NMR chemical shifts were reported in ppm relative
to 85% H3PO4 (external standard, Sigma-Aldrich, St. Louis, MO). Fourier transform infrared
(FT-IR) spectra were recorded on an FT-IR 460 series spectrometer (Jasco, Tokyo, Japan).
High-resolution fast atom bombardment (FAB) mass spectra were obtained using a hybrid
liquid chromatography–quadrapole-time-of-flight tandem mass spectrometer at Gangneung-
Wonju National University, Republic of Korea. P < 0.05 with non-paired tests was compared
to the vehicle control.
4.2. Synthesis of triazole derivatives
To prepare triazole derivatives 2a–h a solution of sodium ascorbate (0.4 equiv.) and
CuSO4·5H2O (0.2 equiv.) in H2O was added to the reaction mixture of azides, 1 (1.0 equiv.)
and the corresponding alkynes (4.2 equiv.) in isopropanol. After stirring vigorously at 40 °C
for 16–26 h, the reaction was diluted with distilled H2O (5 mL) and extracted with CH2Cl2.
The combined extracts were dried over anhydrous MgSO4, filtered, and evaporated in vacuo
to yield the crude products 2a–h, with a by-product. This crude product was purified by flash
column chromatography to obtain pure products 2a–h. A solution of sodium nitrite (420 mg,
15

6.09 mmole) in H2O (4 mL) was added dropwise to an ice-cooled solution of 2-aminobenzyl
alcohol (500 g, 4.06 mmol) in distilled H2O (5.0 mL) and concentrated H2SO4 (0.87 mL). The
mixture was then stirred for 30 min at the same temperature. A solution of sodium azide (530
mg, 8.12 mmol) in H2O (3.1 mL) was added slowly and was allowed to stir for 20 h at room
temperature. The reaction mixture was filtered, the solution obtained was extracted with
CH2Cl2, and the combined organic extracts were dried over anhydrous magnesium sulfate
(MgSO4). After removing the solvent under reduced pressure, the crude oil was purified by
flash column chromatography with a gradient solvent system using acetone and chloroform to
yield the desired product, 2-azido benzyl alcohol (1) (440 mg, 2.95 mmol, 72.0%): Rf (4:6 =
acetone:chloroform) = 0.81; 1H NMR δ 4.59 (s, 2H), 7.10–7.15 (m, 2H), and 7.25–7.36 (m,
2H); 13C NMR δ 61.32, 117.91, 124.84, 128.92, 129.06, 131.80, and 137.66.
4.2.1. Synthesis of 2-[4-(2-hydroxypropan-2-yl)-1H-1,2,3-triazol-1-yl]benzyl alcohol (2a)
Pure product 2a (120 mg, 0.52 mmol, 76.9%) was obtained from the reaction mixture
of azides, 1 (100 mg, 0.67 mmol) and 2-methyl-3-butyn-2-ol (0.28 mL, 2.82 mmol) in
isopropanol (3.0 mL): Rf (4:6 = acetone:chloroform) = 0.42; IR (KBr, cm–1) 3375, 3145,
1
2978, 2931, 2365, 1607, 1499, 1235, 1012; H NMR δ 1.72 (s, 6H), 4.47 (s, 2H), 7.37–7.64
(m, 4H), and 7.85 (s, 1H); 13C NMR δ 30.5 (2), 61.9, 68.6, 120.8, 124.3, 129.1, 130.0, 131.7,
135.6 (2), and 141.1; HRFABMS calcd. for C12H15O2N3 (M+Na)+ 256.1062, found 256.1065.
4.2.2. Synthesis of 1-[2-(hydroxymethyl)phenyl]-1H-1,2,3-triazol-4-yl acetate (2b)
The reaction mixture of azides, 1 (200 mg, 1.34 mmol) and propargyl acetate (0.56
mL, 5.63 mmol) in isopropanol (6.0 mL) was used to obtain the pure product 2b (249 mg,
1.01 mmol, 75.2%): Rf (2:8 = acetone:chloroform) = 0.38; IR (KBr, cm–1) 3397, 3148, 2926,
16

2363, 1741, 1653, 1499, 1238, and 1033; 1H NMR δ 2.12 (s, 3H), 4.49 (s, 2H), 5.31 (s, 2H),
7.38–7.66 (m, 4H), and 8.02 (s, 1H); 13C NMR δ 20.8, 57.4, 61.6, 124.4, 125.2, 129.1, 130.1,
131.4, 135.5, 135.7, 143.1, and 170.8; HRFABMS calcd. for C12H13O3N3 (M+Na)+ 270.0855,
found 270.0858.
4.2.3. Synthesis of 2-[1-{2-(hydroxymethyl)phenyl}-1H-1,2,3-triazol-4-yl]butan-2-ol (2c)
The reaction mixture of azides, 1 (200 mg, 1.34 mmol) and 3-methyl-1-pentyn-3-ol
(0.637 mL, 5.63 mmol) in isopropanol (6.0 mL) was used to obtain the pure product 2c (300
mg, 1.21 mmol, 90.3%): Rf (3:7 = acetone:chloroform) = 0.46; IR (KBr, cm–1) 3361, 3159,
2971, 2348, 1460, 1229, and 1043; 1H NMR δ 0.93 (t, J = 7.32 Hz, 3H), 1.67 (s, 3H), 1.92–
13
2.17 (m, 2H), 4.48 (s, 2H), 7.38–7.64 (m, 4H), and 7.84 (s, 1H); C NMR δ 8.3, 28.0, 35.9,
61.9, 71.4, 121.4, 124.3, 129.1, 129.9, 131.6, 135.6, 136.1, and 154.9; HRFABMS calcd. for
C13H17O3N2 (M+Na)+ 270.1218, found 270.1215.
4.2.4. Synthesis of 4.2.4. 2-[4-(1-cyclohexenyl)-1H-1,2,3-triazol-1-yl]benzyl alcohol (2d)
The reaction mixture of azides, 1 (200 mg, 1.34 mmol) and 1-ethynylcyclohexene
(0.67 mL, 5.63 mmol) in isopropanol (6.0 mL) was used to obtain the pure product 2d (288
mg, 1.13 mmol, 84.3%): Rf (1:9 = acetone:chloroform) = 0.50; IR (KBr, cm–1) 3391, 3138,
2934, 1688, 1499, 1231, and 1046; 1H NMR δ 1.76 (m, 4H), 2.25 (m, 2H), 2.45 (m, 2H), 4.48
(s, 1H), 6.63–6.67 (m, 1H), 7.36–7.64 (m, 4H), and 7.77 (s, 1H); 13C NMR δ 22.2, 22.4, 25.4,
26.4, 62.0, 119.5, 124.2, 126.2, 126.8, 129.1, 129.8, 131.7, 135.7, 136.2, and 149.8;
HRFABMS calcd. for C15H17O1N3 (M+Na)+ 278.1269, found 278.1261.
4.2.5. Synthesis of 2-(4-m-tolyl-1H-1,2,3-triazol-1-yl)benzyl alcohol (2e)
17

The pure product 2e (329 mg, 1.08 mmol, 92.6%) was obtained from the reaction
mixture of azides, 1 (200 mg, 1.34 mmol) and 1-ethynyl-3-toluene (0.73 mL, 5.63 mmol) in
isopropanol (6.0 mL): Rf (1:19 = acetone:chloroform) = 0.41; IR (KBr, cm–1) 3375, 3135,
1
2924, 2344, 1654, 1499, 1231 and 1038; H NMR δ 2.47 (s, 3H), 4.54 (d, J = 6.96 Hz, 2H),
7.48 (m, 8H), and 8.14 (s, 1H); 13C NMR δ 21.5, 62.0, 120.8, 122.9, 124.3, 126.5, 128.9,
129.2, 129.4, 129.7, 130.0, 131.8, 135.6, 136.0, 138.8, and 148.2; HRFABMS calcd. for
C16H15O1N3 (M+Na)+ 288.1113, found 288.1110.
4.2.6. Synthesis of 2-[4-(4-fluorophenyl)-1H-1,2,3-triazol-1-yl]benzylalcohol (2f)
The corresponding azides, 1 (200 mg, 1.34 mmol) and 1-ethynyl-4-fluorobenzene
(0.73 mL, 5.63 mmol) in isopropanol (6.0 mL) was treated to obtain the pure product 2f (258
mg, 0.96 mmol, 71.6%): Rf = 0.35 (1:19 = acetone:chloroform); IR (KBr, cm–1) 3390, 3137,
1
2921, 2269, 1611, 1492, 1231, 1040, and 840; H NMR δ 4.54 (s, 2H), 7.13–7.20 (m, 2H),
7.43–7.57 (m, 3H), 7.64–7.07 (m, 1H), 7.85–7.91 (m, 2H), and 8.13 (s, 1H); 13C NMR δ 61.8,
116.3 (d, JC–F = 22.21 Hz, 2C), 120.7, 124.4, 120.1 (d, JC–F = 3.71 Hz), 127.6 (d, JC–F = 8.64
Hz, 2C), 129.2, 130.1, 131.6, 135.6, 135.9, 147.2, and 162.3 (d, JC–F = 246.12 Hz);
HRFABMS calcd. for C15H12ON3F (M+Na)+ 292.0862, found 292.0864.
4.2.7. Synthesis of 2-(4-phenyl-1H-1,2,3-triazol-1-yl)benzyl alcohol (2g)
The corresponding azides, 1 (200 mg, 1.34 mmol) and phenylacetylene (0.62 mL,
5.63 mmol) in isopropanol (6.0 mL) were used to obtain the pure product 2g (299 mg, 1.19
mmol, 88.9%): Rf (1:1 = ethyl acetate:n-hexane) = 0.57; IR (KBr, cm–1) 33978, 3135, 2913,
2271, 1607, 1500, 1231, and 1041; 1H NMR δ 4.53 (s, 2H), 7.35–7.66 (m, 7H), 7.89–7.92 (m,
2H), and 8.17 (s, 1H ); 13C NMR δ 61.8, 120.9, 124.4, 125.8 (2), 128.6, 129.0 (2), 129.1,
18

129.8, 130.0, 131.5, 135.6, 135.9, and 148.0; HRFABMS calcd. for C15H13ON3 (M+Na)+
274.0956, found 274.0959.
4.2.8. Synthesis of [2-{4-(3-fluorophenyl)-1H-1,2,3-triazol-1-yl}phenyl]methanol (2h)
The corresponding azides, 1 (200 mg, 1.34 mmol) and 1-ethynyl-3-fluorobenzene
(0.65 mL, 5.63 mmol) in isopropanol (6.0 mL) were used to obtain the pure product 2h (334
mg, 1.24 mmol, 92.34%): Rf (2:8 = acetone:chloroform) = 0.53; IR (KBr, cm–1) 3398, 3137,
1
2924, 1618, 1501, 1456, 1238, 1182, 1159, and 1040; H NMR δ 4.54 (s, 2H), 7.08 (dd, J =
2.19, 8.24 Hz, 1H), 7.40–7.58 (m, 4H), 7.61–7.69 (m, 3H), and 8.19 (s, 1H); 13C NMR δ
61.8, 112.8 (d, JC–F = 22.82 Hz), 115.5 (d, JC–F = 20.97 Hz), 121.3, 121.4 (d, JC–F = 3.08 Hz),
124.4, 129.2, 130.2, 130.6 (d, JC–F = 8.02 Hz), 131.6, 132.0 (d, JC–F = 8.64 Hz), 135.5, 135.8,
147.0 (d, JC–F = 3.08 Hz), and 161.7 (d, JC–F = 244.28 Hz); HRFABMS calcd. for
C15H12ON3F (M+Na)+ 292.0862, found 292.0864.
4.3. Synthesis of carboxylic acid derivatives
In order to prepare carboxylic acid derivatives 3a–h when triazole derivatives were
oxidized at room temperature, decomposition of the triazole moiety was observed. Therefore,
triazole derivatives (2a–h) prepared from 2-aminobenzyl alcohol were treated with an excess
of Jones reagent at 0ºC to obtain relatively pure carboxylic acid derivatives (3a–h).
Excluding compounds 3a and 3e, recrystallization yielded pure acids (3b, 3c, 3d, 3f, 3g, and
3h) in good yields. Jones reagent was added in a dropwise manner to a triazole 2a–h (1.0
equiv.) solution in acetone (10 mL) cooled to 0ºC until the starting material disappeared, as
monitored by thin-layer chromatography. The reaction was quenched by the addition of
19

isopropanol, followed by saturated NaHCO3. The solution was extracted three times with
Et2O and the combined organics were washed with H2O and brine, dried over MgSO4, and
filtered and evaporated in vacuo to yield the crude products 3a–h (Suppl. Scheme 1). This
crude product was purified by recrystallization to obtain the pure products 3a–h.
4.3.1. Synthesis of 2-[4-(2-hydroxypropan-2-yl)-1H-1,2,3-triazol-1-yl]benzoic acid (3a)
The pure product 3a (180 mg, 0.73 mmol, 84.9%) was obtained from the reaction
mixture of triazole compound 2a (200 mg, 0.86 mmol) in acetone (10.0 mL): Rf (4:6 =
acetone:chloroform) = 0.17; yield = 84.9%; IR (KBr, cm–1) 3379, 3150, 2981, 2920, 2852,
2489, 2348, 1712, 1503, 1239, and 1060; 1H NMR δ 1.68 (s, 6H), 7.51 (dd, J = 1.47, 7.71 Hz,
13
1H), 7.60–7.72 (m, 2H), 7.84 (s, 1H), and 8.06 (dd, J = 1.47, 7.68 Hz, 1H); C NMR δ 30.0
(2), 68.2, 122.4, 127.0, 128.4, 130.2, 131.5, 132.7, 136.3, 155.7, and 167.3; HRFABMS
calcd. for C12H13O3N3 (M+Na)+ 270.0855, found 270.0854.
4.3.2. Synthesis of 1-[2-(hydroxymethyl)phenyl]-1H-1,2,3-triazol-4-yl acetate (3b)
The pure product 3b (200 mg, 0.76 mmol, 93.8%) was obtained from the reaction
mixture of triazole compound 2b (200 mg, 0.81 mmol) in acetone (10.0 mL): Rf (2:8 =
acetone:chloroform) = 0.14; yield = 93.8%; IR (KBr, cm–1) 3402, 3161, 3003, 2883, 2348,
1724, 1506, 1395, 1238, and 1047; 1H NMR δ 2.11 (s, 3H), 5.29 (s, 2H), 7.48–7.52 (m, 1H),
7.62–7.76 (m, 2H), 8.05 (s, 1H), and 8.07–8.81 (m, 1H); 13C NMR δ 20.9, 57.6, 126.7, 127.3,
128.3, 130.6, 131.8, 133.1, 136.2, 142.6, 167.0, and 171.7; HRFABMS calcd. for
C12H11O4N3 (M+Na)+ 284.0647, found 284.0643.
4.3.3. Synthesis of 2-[4-(2-hydroxybutan-2-yl)-1H-1,2,3-triazol-1-yl]benzoic acid (3c)
20

The pure product 3c (194 mg, 0.69 mmol, 92.6%) was obtained from the reaction
mixture of triazole compound 2c (200 mg, 0.75 mmol) in acetone (10.0 mL): Rf (4:6 =
acetone:chloroform) = 0.10; yield = 92.6%; IR (KBr, cm–1) 3428, 3106, 2922, 2360, 1706,
1466, 1267, and 1052; 1H NMR δ 0.87 (t, J = 7.32 Hz, 3H), 1.62 (s, 3H), 1.90–2.00 (m, 2H),
7.57 (dd, J = 1.08, 7.68 Hz, 1H), 7.64–7.78 (m, 2H), and 8.00 (s, 1H), 8.04 (dd, J = 1.57,
7.50 Hz, 1H); 13C NMR δ 8.6, 28.3, 36.6, 72.0, 124.4, 127.8, 129.8, 131.1, 132.1, 133.6,
137.3, 155.8, and 168.2; HRFABMS calcd. for C13H15O3N3 (M+Na)+ 284.1011, found
284.1014.
4.3.4. Synthesis of (1Z,3Z,4Z)-3-(2-oxocyclohexylidene)benzo[f][1,2,5]triazocin-6(3H)-one
(3d)
The pure product 3d (163 mg, 0.57 mmol, 73.1%) was obtained from the reaction
mixture of triazole compound 2d (200 mg, 0.78 mmol) in acetone (10.0 mL): Rf (1:9 =
acetone:chloroform) = 0.03; yield = 73.1%; IR (KBr, cm–1) 3367, 2974, 2893, 2363, 1655,
1
1381, 1089, and 1048; H NMR δ 1.74–1.83 (m, 4H), 2.38 (t, J = 7.14 Hz, 2H), 3.16 (t, J =
6.96 Hz, 2H), 7.58 (dd, J = 1.47, 7.32 Hz, 1H), 7.70–7.81 (m, 2H), 8.13 (dd, J = 1.47, 7.32
Hz, 1H), and 8.71 (s, 1H); 13C NMR δ 24.2, 25.3, 34.6, 40.1, 128.1, 129.0, 129.7, 131.6,
132.4, 133.8, 136.6, 147.8, 167.3, 170.8, and 195.7; HRFABMS calcd. for C15H13O2N3
(M+Na)+ 290.0905, found 290.0901.
4.3.5. Synthesis of 2-(4-m-tolyl-1H-1,2,3-triazol-1-yl)benzoic acid (3e)
The pure product 3e (194 mg, 0.69 mmol, 92.6%) was obtained from the reaction
mixture of triazole compound 2e (200 mg, 0.75 mmol) in acetone (10.0 mL): Rf (1:19 =
acetone:chloroform) = 0.06; yield = 92.6%; IR (KBr, cm–1) 3399, 3133, 2921, 2491, 1715,
21

1602, 1505, 267, 1082, and 1043; 1H NMR δ 2.26 (s, 3H), 7.14 (d, J = 7.68 Hz, 1H), 7.29 (t,
13
J = 7.68 Hz, 1H), 7.57–7.76 (m, 5H), 8.02 (dd, J = 1.47, 7.68 Hz, 1H), and 8.53 (s, 1H); C
NMR δ 21.5, 123.9, 124.1, 127.4, 128.0, 129.8, 129.9, 130.2, 131.4, 131.4, 132.3, 133.9,
137.4, 139.9, 148.9, and 168.3; HRFABMS calcd. for C16H13O2N3 (M+Na)+ 302.0905, found
302.0902.
4.3.6. Synthesis of 2-[4-(4-fluorophenyl)-1H-1,2,3-triazol-1-yl]benzoic acid (3f)
The pure product 3f (187 mg, 0.66 mmol, 88.0%) was obtained from the reaction
mixture of triazole compound 2f (200 mg, 0.75 mmol) in acetone (10.0 mL): Rf (1:19 =
acetone:chloroform) = 0.1; yield = 88.0%; IR (KBr, cm–1) 3402, 3123, 2922, 2360, 1716,
1
1602, 1561, 1493, 1301, 1267, 1236, 1158, 1145, and 1078; H NMR δ 7.11 (t, J = 8.79 Hz,
2H), 7.45–7.54 (m, 3H), 7.79–7.84 (m, 3H), and 8.21 (s, 1H); 13C NMR δ 115.9, 116.2,
122.3, 125.7, 126.5, 127.6, 127.7, 129.9, 130.0, 130.2, 130.3, 134.4, 146.7, 161.3, and 176.0;
HRFABMS calcd. for C15H10O2N3F1 (M+Na)+ 306.0655, found 306.0652.
4.3.7. Synthesis of 2-(4-phenyl-1H-1,2,3-triazol-1-yl)benzoic acid (3g)
The pure product 3g (191 mg, 0.72 mmol, 90.0%) was obtained from the reaction
mixture of triazole compound 2g (200 mg, 0.80 mmol) in acetone (10.0 mL): Rf (1:1 = ethyl
acetate:n-hexane) = 0.17; yield = 90.0%; IR (KBr, cm–1) 3386, 3133, 2922, 2349, 1706, 1453,
and 1076; 1H NMR δ 7.3–7.49 (m, 3H), 7.58–7.89 (m, 5H), 8.11 (dd, J = 1.83, 7.71 Hz, 1H)
13
and 8.33 (d, J = 2.94 Hz, 1H); C NMR δ 123.5, 126.7 (2), 127.9, 129.2, 129.4, 129.9 (2),
131.0, 131.2, 132.4, 133.7, 137.0, 148.5, and 168.0; HRFABMS calcd. for C15H11O2N3
(M+Na)+ 288.0749, found 288.0749.
22

4.3.8. Synthesis of 2-[4-(3-fluorophenyl)-1H-1,2,3-triazol-1-yl]benzoic acid (3h)
The pure product 3h (192 mg, 0.68 mmol, 92.2%) was obtained from the reaction
mixture of triazole compound 2h (200 mg, 0.74 mmol) in acetone (10.0 mL): Rf (2:8 =
acetone:chloroform) = 0.1; yield = 92.2%; IR (KBr, cm–1) 3423, 3144, 2922, 2360, 2283,
1909, 1716, 1590, 1479, 1300, 1274, and 1240; 1H NMR δ 7.30–7.49 (m, 3H), 7.58–7.89 (m,
13
5H), 8.11 (dd, JC–F = 1.83, 7.71 Hz, 1H) and 8.33 (d, JC–F = 2.94 Hz, 1H); C NMR δ 113.4
(d, JC–F = 22.82 Hz), 116.0 (d, JC–F = 20.97 Hz), 122.6 (d, JC–F = 2.47 Hz), 124.7, 128.1,
129.7, 131.5, 131.9 (d, JC–F = 8.64 Hz), 132.4, 133.93, 134.0, 137.4, 147.6 (d, JC–F = 2.47 Hz),
164.7 (d, JC–F = 242.43 Hz), and 168.1; HRFABMS calcd. for C15H10O2N3F (M+Na)+
306.0655, found 306.0659.
4.4. Synthesis of phosphonate derivatives
For preparation of phosphonate derivatives 4c and 4h, triazole 2a–h (1.0 equiv.) was
mixed with triethyl phosphite (20 equiv.) at room temperature. Iodine (1.0 equiv.) was added
to the mixture of 2a–h and P(OEt)3 at 0ºC. After stirring under an inert atmosphere for 5 min
at the same temperature, the reaction mixture was warmed to either room temperature or 80ºC
and stirred for 2–8 h. Excess triethyl phosphite was removed under reduced pressure to obtain
crude product. Crude product was diluted with 1:1 (ether:acetone), washed with saturated
NaHCO3 and brine, dried over anhydrous MgSO4, and concentrated in vacuo (Suppl. Scheme
2). The pure products, 4a–h, were obtained by flash column chromatography using gradient
solvent system. The synthetic methods of compounds 4a–h except 4c and 4h have been
published previously [18].
23

4.4.1.
Synthesis
of
diethyl
2-[4-(2-hydroxybutan-2-yl)-1H-1,2,3-triazol-1-
yl]benzylphosphonate (4c)
Triazole 3c (100 mg, 0.40 mmol) was used to obtain pure product 4c (110 mg, 0.30
mmol, 75.0%): Rf (1:9 = ethyl acetate:methylene chloride) = 0.2; yield = 75.0%; IR (KBr,
1
cm–1) 3425, 2924, 2866, 2333, 1747, 1639, 1454, and 1049; H NMR δ 0.93 (t, J = 7.5 Hz,
3H), 1.25 (t, J = 6.96 Hz, 6H), 1.94–2.00 (m, 2H), 3.08 (d, J = 21.96 Hz, 2H), 3.94–4.04 (m,
13
4H), 7.40–7.56 (m, 4H), and 8.06 (s, 1H); C NMR δ 8.3, 16.3, 16.4, 27.9, 28.8 (d, JC–P
=
137.57 Hz), 36.0, 62.3, 62.4, 71.2, 123.0, 126.9, 127.8, 128.1, 129.9, 131.7, 131.8, and 154.3;
31P NMR (85% H3PO4) δ 24.96 ppm; HRFABMS calcd. for C17H26O4N3P (M+Na)+
390.1575, found 390.1578.
4.4.2. Synthesis of diethyl 2-[4-(3-fluorophenyl)-1H-1,2,3-triazol-1-yl] benzyl phosphonate
(4h)
Triazole 3h (100 mg, 0.37 mmol) was used to obtain the pure product 4h (101 mg,
0.26 mmol, 70.27%): Rf (1:9 = ethyl acetate:methylene chloride) = 0.3; yield = 70.3%; IR
(KBr, cm–1) 3429, 2924, 2839, 2345, 1639, 1585, 1454, 1265, 1034, and 960; 1H NMR δ 1.24
(t, J = 6.96 Hz, 6H), 3.13 (d, J = 21.96 Hz, 2H), 3.96–4.07 (m, 4H), 7.03–7.09 (m, 1H), 7.38–
7.60 (m, 5H), 7.65–7.71 (m, 2H), and 8.55 (s, 1H); 13C NMR δ 16.3, 16.4, 29.0 (d, JC–P
138.18 Hz), 62.4, 62.5, 112.7 (d, JC–F = 22.82 Hz), 115.1 (d, JC–F = 20.98 Hz), 121.4 (d, JC–F
=
=
3.08 Hz), 123.2, 126.9 (d, JC–P = 3.08 Hz), 127.7 (d, JC-P= 8.64 Hz), 128.2 (d, JC-P = 3.70 Hz),
130.0 (d, JC-P = 3.70 Hz), 130.5 (d, JC–F = 8.63 Hz), 131.9 (d, JC–P = 4.93 Hz), 132.6 (d, JC–F
=
8.63 Hz), 136.7 (d, JC–P = 7.40 Hz), 146.6 (d, JC–F = 3.08 Hz), and 163.3 (d, JC–F = 244.28
31
Hz); P NMR (85% H3PO4) δ 24.98 ppm; HRFABMS calcd. for C19H21O4N3PF (M+Na)+
412.1218, found 412.1215.
24

4.5. In vivo anti-inflammatory studies
Biological assay was performed for anti-inflammatory activities of the test
compounds were assessed with the xylene-induced ear edema model in mice [19]. Mice
(specific-pathogen free ICR strain, male, 24.0–28.0 g; Damool Science, Daejeon, Republic of
Korea) were injected with positive control or test substances at 25 mg/kg (suspended in 0.5%
carboxymethyl cellulose-Na solution, 100 μL/100 g body weight rate), ip, 30 min prior to the
induction of ear edema with xylene. Xylene (100%; Junsei Chemical, Tokyo, Japan) (10 μL
each) was applied onto both the interior and posterior surface of the right ear lobe for edema
induction. The left ear was used as a non-edematous control. Thirty minutes later, animals
were killed by CO2 gas anesthesia, and ear lobes were circularly dissected out with a 6-mm-
diameter metal ear punch. The circular ear tissues were measured for their weight and
thickness. The inhibition was calculated using the following equation : Inhibition (%) = [1-
(measured value of thickness / vehicle mean)] x 100.
4.6. Western blot analysis
For detection of COX-2 expression, MDA-MB-231 cells (American Type Culture
Collection, Rockville, MD) cultured in Dulbecco's modified eagle medium (DMEM)
supplemented with 10% (v/v) heat-inactivated fetal bovine serum (HyClone, Logan, UT)
were treated with 10 ng/ml TNF-α in the absence or presence of analog 2c (20 and 40 μM)
for 12 h. The cells were lysed in a buffer consisting of 20 mM 4-(2-hydroxyethyl)-1-
piperazineethanesulfonic acid (HEPES, pH 7.2), 1% Triton X-100, 10% glycerol, 150 mM
25

NaCl, 10 μg/mL leupeptin, and 1 mM phenylmethylsulfonyl fluoride (PMSF). The protein
extracts (20 μg each) were separated by 10% sodium dodecyl sulfate -polyacrylamide gel
electrophoresis and transferred to nitrocellulose membranes. The blots were incubated with
the corresponding primary antibodies and developed using an enhanced chemiluminescence
detection system (Amersham Pharmacia Biotech, Piscataway, NJ).
4.7. Quantitative structure–activity relationships
For quantitative structure–activity relationships (QSAR) the biological test of the
xylene-induced ear edema model in mice was performed for 24 compounds synthesized in
this report. Because one compound, 4d, was could not be tested since it resulted in animal
death within 15 min after its administration, 23 compounds were used for QSAR calculations.
All QSAR calculations were performed on an Intel Core 2 Quad Q6600 (2.4 GHz) Linux PC
workstation with Sybyl7.3 software (Tripos). All QSAR experiments were performed as
described previously [20].
4.8. Molecular docking
Docking studies between phenyl-1H-1,2,3-triazole analog 2c and target proteins in
silico were performed on an Intel Core 2 Quad Q6600 (2.4 GHz) Linux PC workstation with
Sybyl 7.3 software (Tripos). The binding modes were analyzed using the LigPlot program
[21] and the images of the binding complexes were viewed using the PyMol software (The
PyMOL Molecular Graphics System, Version 1.0r1, Schrödinger, LLC).
26

Acknowledgments
This work was supported by the Basic Science Research Program through the National
Research Foundation of Korea (NRF) funded by the Ministry of Education, Science, and
Technology (Grant number: 2011-0010745) and the next generation Biogreen21 program
(RDA, PJ009532). SY Shin was supported by the KU Research Professor Program of
Konkuk University.
References
1. C.A. Velazquez, Q.H. Chen, M.L. Citro, L.K. Keefer, E.E. Knaus, J. Med. Chem. 51
(2008) 1954-1961.
2. I. Melnikova, Nat. Rev. Drug Discov. 9 (2010) 589-590.
3. C. Velazquez, P.N. Praveen Rao, E.E. Knaus, J. Med. Chem. 48 (2005) 4061-4067.
4. J.F. Renard, D. Arslan, N. Garbacki, B. Pirotte, X. de Leval, J. Med. Chem. 52 (2009)
5864-71.
5. A.L. Blobaum, L.J. Marnett, J. Med. Chem. 50 (2007) 1425-1441.
6. J.E. Frampton, G.M. Keating, Drugs 67 (2007) 2433-2472.
7. S.K. Khetan, T.J. Collins, Chem. Rev. 107 (2007) 2319-2364.
8. C.A. Rouzer, Chem. Res. Toxicol. 2 (2008) 274-275.
9. U.K. Bandarage, L. Chen, X. Fang, D.S. Garvey, A. Glavin, D.R. Janero, L.G. Letts, G.J.
Mercer, J.K. Saha, J.D. Schroeder, M.J. Shumway, S.W. Tam, J. Med. Chem. 43 (2000)
4005-4016.
10. S. Tacconelli, M.L. Capone, P. Patrignani, Curr. Pharm. Des. 10 (2004) 589-601.
27

11. C. Romay, N. Ledon, R. Gonzalez, Inflamm. Res. 47 (1998) 334-338.
12. B. Poligone, A.S. Baldwin, J. Biol. Chem. 276 (2001) 38658-38664.
13. K.M. Alexacou, J.M. Hayes, C. Tiraidis, S.E. Zographos, D.D. Leonidas, E.D. Chrysina,
G. Archontis, N.G. Oikonomakos, J.V. Paul, B. Varghese, D. Loganathan, Proteins 71
(2008) 1307-1323.
14. S.Y. Shin, H. Jung, S. Ahn, D. Hwang, H. Yoon, J. Hyun, Y. Yong, H.J. Cho, D. Koh,
Y.H. Lee, Y. Lim, Bioorg. Med. Chem. 22 (2014) 1809-1820.
15. S. Pirhadi, J.B. Ghasemi, Eur. J. Med. Chem. 45 (2010) 4897-4903.
16. J. Hyun, S.Y. Shin, K.M. So, Y.H. Lee, Y. Lim, Bioorg. Med. Chem. Lett. 22 (2012)
2664-2669.
17. S.W. Rowlinson, J.R. Kiefer, J.J. Prusakiewicz, J.L. Pawlitz, K.R. Kozak, A.S. Kalgutkar,
W.C. Stallings, R.G. Kurumbail, L.J. Marnett, J. Biol. Chem. 278 (2003) 45763-45769.
18. T.W. Kim, T. Islam, K.Y. Jung, J. Ind. Eng. Chem. 16 (2010) 461-466.
19. J. Kou, Y. Ni, N. Li, J. Wang, L. Liu, Z.H. Jiang, Biol. Pharm. Bull. 28 (2005) 176-180.
20. S.Y. Shin, Y. Woo, J. Hyun, Y. Yong, D. Koh, Y.H. Lee, Y. Lim, Bioorg. Med. Chem.
Lett. 21 (2011) 6036-6041.
21. A.C. Wallace, R.A. Laskowski, J.M. Thornton, Protein Eng. 8 (1995) 127-134.
28

Figure legends
Fig. 1. Structures of diclofenac and lumiracoxcib.
Fig. 2. Increments in ear weight (%) containing error bars with standard deviations of the
compounds tested in this research.
Fig. 3: Effect of analog 2c on TNF-α-induced COX-2 expression. MDA-MB-231 cells were
treated with 10 ng/ml TNF-α in the absence or presence of analog 2c or diclofenac (Df) for
12 h. The cells were lysed and the soluble fractions were separated by SDS-polyacrylamide
gel and subjected to Western blotting with anti-COX-2 antibody. GAPDH was used as an
internal control to show equal protein loading.
Fig. 4: CoMFA contour maps generated using Sybyl 7.3. The steric and electrostatic field
descriptors contribute 39.4% and 60.6%, respectively. The steric field contours are shown in
green (more bulk-favored) and yellow (less bulk-favored), while the electrostatic field
contours are shown in red (electronegative substituent-favored) and blue (electropositive
substituent-favored).
Fig. 5: Pharmacophores obtained from the CoMFA contour maps showing good inhibitory
effects on xylene-induced increases in ear weight. Half circles denote favored regions.
Fig. 6: (left) Ten residues residing in the binding site of the 1PXX-A–diclofenac complex
analyzed using LigPlot: (right) Ten residues surrounding analog 2c. Seven residues including
Tyr348, Val349, Tyr385, Trp387, Met522, Ala527, and Ser530 found in both binding sites
are marked in red color. The residues with or without half circles denote hydrophobic
interactions or hydrogen bonds, respectively.
Fig. 7: (A) The 3D image of the 1PXX-A–diclofenac complex constructed using PyMOL.
The compound colored in yellow is diclofenac. Hydrogen bonds between two oxygens of
29